BEGIN:VCALENDAR VERSION:2.0 PRODID:-//events.la.psu.edu//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/New_York BEGIN:STANDARD DTSTART:20201101T020000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU END:STANDARD BEGIN:DAYLIGHT DTSTART:20200308T020000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU END:DAYLIGHT END:VTIMEZONE BEGIN:VEVENT UID:17862-3f9fa130085f954a352259442b8f4f90@events.la.psu.edu DTSTAMP:20260611T012442Z DTSTART;TZID=America/New_York:20260403T090000 DTEND;TZID=America/New_York:20260403T103000 SUMMARY:Center for Language Science Speaker Series: John Lipski DESCRIPTION:\nIn the Afro-Colombian village of San Basilio de Palenque\, there are ongoing efforts to revitalize the traditional Spanish-lexifie d creole language Palenquero. Currently\, some instruction in Palenquero is part of the school curriculum\, but with few exceptions\, there is n o grammatical description\, no contrasting with Spanish\, no emphasis on actual communication and no corrective feedback loop to guide the desir ed transition to fluency. Nor do traditional adult speakers routinely sp eak Palenquero to young learners. Cultural factors (e.g. reluctance to c riticize community members’ speech) as well as pedagogical limitations ( e.g. lack of second-language teaching experience) are obstacles to the s crutiny of learners’ production. This linguistic environment has produce d outcomes that would be unexpected in other language revitalization sce narios\, constituting a “perfect storm” for unchecked miscues and unanti cipated divergences from the target language. The result is two hived-of f versions of the Palenquero language within the same community. The rea sons for the divergence are clear\, but the precise nature of L2 Palenqu ero remains elusive. The present study describes the results of a series of deep-learning models using recurrent neural networks trained on Pale nquero corpora\, to determine the extent to which L1 vs. L2 Palenquero c an be distinguished without human intervention\, and to model the points of divergence between the two\, including a classification model that t akes a novel utterance as input and predicts the likelihood that it was produced by an L2 speaker. The discussion offers ideas on how the L2 inn ovations can arise and propagate unnoticed. Even with a corpus that is m iniscule in comparison with the input to large-language models\, the res ults are quite encouraging for the use of deep learning techniques to mo del seemingly intractable sociolinguistic environments.\n\nFor more deta ils: https://events.la.psu.edu/event/cls-speaker-series-john-lipski/ X-ALT-DESC;FMTTYPE=text/html:

In the Afro-Col ombian village of San Basilio de Palenque, there are ongoing efforts to revitalize the traditional Spanish-lexified creole language Palenquero. Currently, some instruction in Palenquero is part of the school curricul um, but with few exceptions, there is no grammatical description, no con trasting with Spanish, no emphasis on actual communication and no correc tive feedback loop to guide the desired transition to fluency. Nor do tr aditional adult speakers routinely speak Palenquero to young learners. C ultural factors (e.g. reluctance to criticize community members’ speech) as well as pedagogical limitations (e.g. lack of second-language teachi ng experience) are obstacles to the scrutiny of learners’ production. Th is linguistic environment has produced outcomes that would be unexpected in other language revitalization scenarios, constituting a “perfect sto rm” for unchecked miscues and unanticipated divergences from the target language. The result is two hived-off versions of the Palenquero languag e within the same community. The reasons for the divergence are clear, b ut the precise nature of L2 Palenquero remains elusive. The present stud y describes the results of a series of deep-learning models using recurr ent neural networks trained on Palenquero corpora, to determine the exte nt to which L1 vs. L2 Palenquero can be distinguished without human inte rvention, and to model the points of divergence between the two, includi ng a classification model that takes a novel utterance as input and pred icts the likelihood that it was produced by an L2 speaker. The discussio n offers ideas on how the L2 innovations can arise and propagate unnotic ed. Even with a corpus that is miniscule in comparison with the input to large-language models, the results are quite encouraging for the use of deep learning techniques to model seemingly intractable sociolinguistic environments.

For more details: https://events.la.psu.edu/event /cls-speaker-series-john-lipski/

URL:https://cls.la.psu.edu/news-events/cls-speaker-series/ LOCATION:Foster Auditorium, 102 Paterno Library END:VEVENT END:VCALENDAR