IASCL - Child Language Bulletin - Vol 24, No 2: December 2004
IN THIS ISSUE
IASCL HOMEPAGE CONTENTS
July 25-29, 2005
Crosslinguistic and Intercultural Aspects of Unimpaired and Impaired Language
A Window on Universal and Language Particular Learning Mechanisms
The Xth meeting of IASCL is hosted by the Freie Universität in Berlin, one of Germany´s most important research universities, in cooperation with the Humboldt-Universität, the Universität Potsdam and the Zentrum für Allgemeine Sprachwissenschaft, Typologie und Universalienforschung Berlin.
Please refer to the following URL: http://www.ctw-congress.de/iascl/ for updates on the programme schedule early in 2005.
NEW LANGUAGE CORPORA
Katherine Demuth from Brown University has recently donated a new Sesotho corpus. This is the first Bantu language corpus in CHILDES and also the first corpus from an indigenous African language. Audio files are also available, but they are not yet linked to the transcripts.
The Demuth Sesotho Corpus was compiled by Katherine Demuth in the southern African country of Lesotho from 1980-82. Data was collected in a small Lesotho mountain village of 550 people in the district of Mokhotlong, where it was possible to establish close rapport with both the children and their families. The Corpus contains a longitudinal study of four target children’s language development as they interacted with members of the extended family including mothers and/or grandmothers, an uncle and occasionally the father (in one family), and especially older siblings, cousins, and peers. These target children are: Hlobohang (boy) 2;1-3;0, Litlhare (girl) 2;1-3;2, ‘Neuoe (girl) 2;4-2;9, and Tsebo (girl) 3;8-4;1. The two older girls were cousins living in the same household, and where therefore recorded together.
Monthly recordings of spontaneous speech consisted of 3-4 hours each over approximately one year, resulting in a corpus of 98 hours of speech containing approximately 13,250 utterances containing lexical verbs or approximately 1/2 million morphemes.
Broad phonemic transcription was conducted by Katherine Demuth with the assistance of the mothers and grandmothers as soon as recording sessions were complete. These transcripts were then verified independently by a researcher at the National University of Lesotho. The original transcription was by hand. The data were subsequently computerized and one third of the corpus hand-tagged by Sesotho speakers at Brown University in the 1990’s. A computational morphological parser was then developed (with the assistance of Mark Johnson, Brown University) to tag the remaining part of the corpus, files were then converted to CHILDES format, and the audio tapes were digitized. These still remain to be linked to the transcripts.
The corpus should therefore be accessible to and of broad interest for researchers of child language, Bantu linguistic structures, and computational linguists interested in morphological parsing and/or machine translation. Collection of this corpus was supported in part by Fulbright and SSRC (Social Science Research Council) dissertation funding. Computerization and tagging of the corpus has been supported in part by NSF grants BNS-08709938 and SBR-9727897. Katherine Demuth wishes to thank all who have assisted with this research over the years, and the children and families who provided the data.
Those wishing to use this corpus should notify Katherine_Demuth@brown.edu, and cite the following reference:
Chinese narrative development
A new corpus studying narrative development in Chinese-speaking children has been donated by Chienju Chang of National Taiwan Normal University.
The study examines 24 children between the ages of 3 and 6 using methods for toy play and narrative elicitation from Hemphill et al. (1994). The files are currently in Chinese characters, a full %mor line will be eventually be added.
New video corpus
A new video corpus has been donated by Virginia Yip, Stephen Matthews, and Uta Lam at the Chinese University of Hong Kong. This corpus tracks the language development of Alicia from 1;9 to 3;6. One set of 38 transcripts, all linked to the video, traces her development in Cantonese and another set of 38 transcripts from the same time period traces her development in English. Three other files represent the code-mixing or switching of Cantonese and English. These files can be located for browsing at http://xml.talkbank.org:8888/talkbank/file/CHILDES/Biling/YipMatthews/
The relevant directories are AliciaCan, AliciaEng, and AliciaMixed. The zip files of the transcripts are in http://childes.psy.cmu.edu/data/biling/YipMatthews.zip.
The video files themselves are at http://childes.psy.cmu.edu/data/video/YipMatthews/
This is the first child in the ever-growing YipMatthews corpus with a full video record. Charlotte, Kathryn, and Llywellyn have complete audio but only Alicia has complete linkage to video. However, audio files for Alicia and the other children are also available at http://childes.psy.cmu.edu/data/audio/YipMatthews/
Brian MacWhinney and his CHILDES team have now automatically tagged all the American English corpora, three of the Spanish corpora, most of the Japanese corpora, and the Cantonese corpora. They will continue by working on the English corpora from the UK, two of which are already done, and then Italian and perhaps some Hungarian.
Brain MacWhinney has recently completed training of a POST database for disambiguation of the Spanish morphological tags.The current version of the MOR tagger and the span.db POST database are available from the CHILDES server. The files in the Aguirre, Ornat, Marrero, and ColMex directories are now fully tagged and disambiguated.
The disambiguator is not making any mistakes, however, there are a variety of transcription errors remaining in these files. For example, a common problem is omission of the accent on éste and ésta which leads to them being treated as demonstratives.
You can either view these files one by one through the browsable XML facility or else download the whole directories as zip files. If anyone is interested in going through these files or other Spanish files to either clean up problems or note consistent gaps, that would be quite helpful.
Brian MacWhinney and his CHILDES team intend to gradually process the remaining Spanish corpora through MOR over the next year or two.
New training software
Brian MacWhinney corrected the training set for the POSTTRAIN program. A new program called TRNFIX that spots all discrepancies between POST and MOR is being currently used. TRNFIX allows the user to decide which one was right.
The Brown Eve corpus is the current training corpus. Using just the adult utterances in that corpus, the eng.db file used by POST was created. Training on maternal input gives better accuracy than training on maternal input plus child sentences or just child sentences.
Using the new eng.db, Brian MacWhinney went through all of the Eng-USA and Eng-UK corpora (with 6 exceptions) and produced a new disambiguated %mor line that should be now more accurate than the previous one. The major errors in the previous version were in the area of confusions of determiners with pronouns and verbs with auxiliaries.
The corpora which are still “in the shop” vis a vis MOR are: Clark, Hall, Forrester, Howe, Manchester, and Wells. After these six are finished, work will be extended to the English corpora in the narrative and clinical directories.
On the first anniversary of Prof. Elizabeth Bates’s death some of her colleagues
and close friends share their memories of a an exceptional scholar and a formidable
REMEMBRANCES AND LEGACIES
Judith Johnston, Brian MacWhinney,
Donna Thal and Beverly Wulfeck
December 13, 2004 marked the one-year anniversary of the death of our dear friend and collaborator Elizabeth Bates. During this past year colleagues around the world have honored Liz through tributes, symposia, publications, and memorials. Each of these highlighted her extraordinary scientific contributions to many fields including language development, cognitive neuroscience, and communicative disorders. When asked to contribute an article about Liz for the Child Language Bulletin, it seemed fitting to focus on her legacy.
Carnegie Mellon University
I first met Liz in 1974 when she was visiting Dan Slobin at Berkeley. The next year we both took up positions in Colorado - she at the University of Colorado and I at the University of Denver. We then began a life-long collaboration leading to the formulation of a model of language learning, processing, and loss called the Competition Model. The core insight underlying this model was that, during language development, the strengths of the neural encoding of words, cues, and forms come to match their external cue validity in the input. The concept of cue validity was far from original, going back to Brunswik, Bayes, and others. But we were the first to apply the concept consistently to language processing and the first to provide evidence of its broad applicability.
Alongside the core notion of cue validity was the notion of language as a limited processing channel that Liz had picked up from Dan Slobin. Cues compete intensely within this narrow channel, and grammar emerges as a resolution of these competitive forces. We worked together with colleagues from over 20 countries to apply these ideas crosslinguistically. We showed how even languages as related as English and German differ markedly in their configuration of cues to sentence processing. In the 1990s we built both symbolic and neural network models to account for the details of these interactions.
In the 1970s, we also explored related aspects of the pragmatic grounding of language. Building on ideas from Piaget, Bruner, and Searle, Liz was able to show how symbolic functions intercorrelate and interact in the first two years of life, against a background of remarkable individual variation in linguistic and cognitive style. I focused my attention on the ways in which languages mark pragmatic and discourse notions such as perspective, subject, and topic. Although we both realized the importance of these various pragmatic factors for language learning, we never succeeded in fully incorporating these ideas into the mechanics of the Competition Model.
Beginning with the publication of “Rethinking Innateness” in 1997, we began to realize that linking our earlier pragmatic insights to the Competition Model would require formulation of a fuller, mechanistic account of embodied cognition. The approach that we began to sketch out, each in our own terms, supplanted the earlier opposition between nativism and empiricism with the newer emergentist paradigm. This paradigm relied on ideas from dynamic systems, neural networks, data-driven linguistics, neuroscience, and functional linguistics. In Liz’s terms, “The really interesting interactions are when you take black and white and you put them together and you get red. Much of child development has to do with that type of … emergent product.”
Liz was passionate about science and about ideas. From her Jesuit teachers, she learned the importance of stating ideas in their clearest and most falsifiable form. Although none of us can hope to match her energy and eloquence, many of us who were persuaded by her vision now feel deeply committed to exploring its further intellectual consequences in the context of an emergentist theory of embodied language and cognition.
REMEMBERING A CLINICAL EXPLORATION
University of British Columbia
In 1978 or thereabouts, Liz and I collaborated in the development of a continuing education short course on clinical pragmatics. Although initially I had been reluctant to add another area of study to my extended doctoral program, a series of conversations with Liz had convinced me that the pragmatic perspective on language development would make a crucial contribution to the practice of speech-language pathology (SLP). Liz was obviously the ideal person to book as a conference speaker on this topic. As she and I talked over the possibilities, the challenge was in the packaging. How could we make a fast-talking, theoretically loaded, presentation by Liz accessible to the “average” clinician? Our solution was to use a dialogue format rather than a lecture. Liz would talk for 10 minutes, then I would step up and say “that reminds me of this kid I saw in the clinic last week” or “the thing that I really notice about (topic) from a clinical perspective is…“. Back and forth, we would trade the podium in this fashion for 3-4 hours. Liz’s role was to present the basic framework of developmental pragmatics along with pertinent research findings. My role was to highlight concepts, illustrate practical applications, and ask for clarification when I suspected the audience may have missed a point. It wasn’t quite Laurel and Hardy, but judging from the number of invitations we received, it was a lively and useful show.
In this short course, and in her subsequent work, Liz influenced the practice of developmental SLP in both broad and specific ways. Most generally, she helped us see that the goal of intervention is not the mere acquisition of grammatical knowledge but the ability to use this knowledge in effective communication acts. This functional perspective permits us to recognize and describe the communicative strengths of some children whose grammatical repertory is limited, and the communicative deficits of others who speak grammatically but are inappropriate in context. More specifically, Liz’s work on pre-verbal communication gave new purpose to our observation of nonverbal 2 and 3 year olds. Instead of merely noting that they are not yet talking, we now routinely assess the complexity and range of their nonverbal communicative functions – an assessment that is particularly important for young children with autism. Also specifically, Liz’s research on the predictive value of vocabulary size not only led to a clinical assessment tool (MacArthur-Bates CDI), but is motivating a new emphasis on early vocabulary development.
I don’t know whether our work on the collaborative short course was Liz’s introduction to the world of developmental language disorders. It at least provided an extended opportunity for her to learn about clinical issues and to consider the relevance of data from atypical development – explorations that ultimately benefited both practice and theory.
ELIZABETH BATES’ CONTRIBUTION TO CLINICAL PRACTICE WITH LANGUAGE IMPAIRED CHILDREN
San Diego State University
Elizabeth Bates was a leader in the functionalist psycholinguistics movement. She was committed to making her work available to applied fields such as Speech-Language Pathology as well as the theoretical fields in which she was one of the brightest stars. A prolific writer, with research programs in bilingualism, adult aphasia, and child language, she published 10 books and more than 200 articles with content that included normal and impaired populations. Three of her books had a powerful impacted clinical practice with language impaired children. These are Language and Context, The Emergence of Symbols, and From First Words to Grammar.
In the 1960s and early 1970s, the field of psycholinguistics was dominated by formal aspects of syntax. Language was conceptualized as a system with no apparent relation to other domains such as perception, cognition, and social interaction. Language “developed” when sufficient exposure to the adult grammatical system activated the innate grammatical knowledge: it was not learned. In those three books Liz showed us how language could emerge from earlier nonlinguistic communicative acts and she provided a model for how language could be learned. She revolutionized our thinking about language development and language disorders by demonstrating how language structure could develop out of the process of communicating, the interdependence of language and nonlinguistic cognition, the continuity of prelinguistic cognitive development and later language development, and the continuity of linguistic and grammatical development in young children.
Liz introduced the notion of prelinguistic performative acts (protoimperatives and protodeclaratives) that were built on intentionality and the concepts of cause and effect (which develop at around 10 months of age). The conceptualization of language as a tool led to the realization that opportunities to use the tool were necessary for the development of language skills. That, in turn, led to the development of more naturalistic assessment and intervention contexts in which children could experience the power of the language tool.
Her descriptions of the imperative and declarative protoforms helped clinicians to see the connection to comparable linguistic forms, and to set up interventions for the purpose of helping children do the same. Assessment of protoforms is now expected in clinical practice (see Paul, 2001), and research has established the usefulness of teaching these forms to a wide range of linguistically impaired children (see, for example, Yoder & Warren, 1999, 2002; Wetherby & Prizant, 1996). The relations between language and gesture that Liz demonstrated in The Emergence of Symbols led directly to studies showing that use of gestures was related to better language development in typically developing (Acredolo & Goodwyn, 1988) and late-talking (Thal & Bates, 1988; Thal, Tobias, & Morrison, 1991; Thal & Tobias, 1994) toddlers. This effectively changed the practice (in use by speech-language pathologists) of discouraging gesture-use by language impaired children and led to programs for incorporating gestures into early language training (Acredolo, Goodwyn, and Abrams, 2002; Capone, 2003).
The continuity between vocabulary and grammar was demonstrated in From First Words to Grammar. The parental questionnaire developed for that book was the seed for MacArthur-Bates Communicative Development Inventories. These Inventories have made a significant impact on clinical practice with infants and toddlers by providing valid and reliable norms for early language development using a time-efficient method. Their adaptation to 33 different languages has also given us one of the few tools that can be used for bilingual evaluations and cross-linguistic comparisons of language development in infants and toddlers.
A LEGACY OF MENTORSHIP AND COLLABORATION
San Diego State University
It is clear that Liz’s legacy will be in the scientific contributions she made to so many disciplines. In a prolific career over three decades, Liz conducted studies in over 20 languages, on four continents, and authored or co-authored 10 books and over 200 articles.
However, her influence extends beyond theories, models and ideas. Liz was an extraordinary mentor and generous collaborator. I consider myself lucky to have had both relationships with her. Over the past 10 years alone, Liz mentored more than 8 post-docs and 24 pre-doctoral students, across many disciplines and from many countries. During her memorial service in San Diego on February 15, 2004, current and former students reflected on the many ways Liz influenced and nurtured their professional and personal development.
Former student Arshavir Blackwell noted that “beyond Liz’s many intellectual achievements and contributions to the field, one of the things that stands out most in my mind, after working with her for six years, is how fun, how full of life she was, what a witty and incisive way she had of expressing herself and describing the world around her, and what a great sense of humor she had, ---particularly about herself.” Another former doctoral student Arturo Hernandez remarked on Liz’s impact on her students, long after graduation. “Her imprint was so cemented into my intellect that I always went back to her for guidance. My return trips were intellectual vacations. A chance to rekindle the fire and recharge my batteries.” UCSD doctoral student Ayse Saygin noted Liz’s love of collaborating with students. “Liz taught me to not only focus on enjoying the daily process of being a researcher, but also to openly share a love of research with others.” Finally, former student and close collaborator Fred Dick expressed sentiments widely held by Liz’s other former students: “She made us think harder, write more, write better, and argue with more passion. She was a challenge, she was complex and engaging, and she was downright fun - and she changed us forever, and much for the better.”
In addition to mentoring students, Liz established numerous exciting collaborations in the US and with colleagues around the world, and she was generous with her time and resources. Most importantly, Liz inspired us. Colleague Nina Dronkers aptly described Liz as someone who “managed to master several different fields and integrate them into fresh new perspectives that made us step out of the traditional molds of our fields and dare to think outside the box.” Another long-time colleague, Jeff Elman described Liz’s energy, vision, and love for collaborations. “She was a builder …of theories, of projects and programs, and of communities.” Indeed, Liz’s greatest professional joy came from discussions at scientific conferences, seminars, and meetings, as well as through extensive (and often intense) email exchanges carried out with colleagues around the world. While many scientists engage in such activities, Liz was one of those rare individuals who did it in the international arena, and with such flair!
So, as we complete our first year without Liz, we mourn her passing because she was our cherished friend and we will miss her. However, we know she lives on in so many ways. Liz may have lost her struggle with cancer, but because of her insatiable scientific curiosity, incredible intelligence, and indefatigable drive she has left indelible marks on her students and colleagues and a priceless legacy for generations of scientists to come.
NEWS FROM THE PUBLICATION COMMITTEE
The fourth volume in the Trends in Language Acquisition Research has just gone into production! Editors are Paul Fletcher and Jon Miller. The theme of the volume they prepared links in with the theme of our last IASCL conference in Madison, which was run by Jon Miller, and co-organized with SRCLD. The new TiLAR volume has the promising title: Bridging the Gap: Language Disorders and Developmental Theory. Contributors are Leonard Abbeduto, Robin Chapman, Paul Fletcher, Michael Garman, Morton Ann Gernsbacher, Heather Geye, Steven Gillis, Paul Govaerts, Deborah James, Amy Lederberg, Jon Miller, Karen Schauwers, Patricia Spencer, Vesna Stojanovik, Stephanie Stokes, Michael Thomas, Susan Ellis Weismer and Anita Wong.
We thank all the contributors for their good work.
We are also grateful to the TiLAR Advisory Board (in this case: Jean Berko Gleason, Ruth Berman, Philip Dale and Brian MacWhinney) for giving very helpful feedback.
We also thank the publisher, John Benjamins, for their continued support.
Remember: All IASCL members who were in Madison and those who joined the IASCL later will receive a copy of this new book as part of their membership. We are certain that this book will be a much appreciated addition to your personal library.
Annick De Houwer and Steven Gillis
Barron-Hauwaert, Suzanne (2004).
Language strategies for bilingual families.
The one-parent-one-language approach. Clevedon: Multilingual Matters.
Segmental-prosodic interaction in phonological development: A comparative investigation.
Special double issue of the Canadian Journal of Linguistics/Revue Canadienne de Linguistique 48(3-4) (2003).
Guest editors: Heather Goad (McGill University) and Yvan Rose (Memorial University)
Goldstein, B. (ed.) (2004).
Bilingual Language Development and Disorders in Spanish-English Speakers.
Baltimore: Brookes Publishing.
Workshop: Frequency effects in language acquisition (Frequenzeffekte
DGfS-Jahrestagung (Annual meeting of the German linguistic society)
Theoretical and Experimental Neuropyschology (TENNET XVI)
Emergence of Language Abilities: ontogeny and phylogeny
Seventh Annual International Conference of the Japanese Society for Language Sciences
Deadline for abstracts: 15/02/05
The IASCL is a worldwide organisation, which means that is aims to serve child language researchers in all countries of the world. Child language research is important everywhere, both from a theoretical perspective (cf. for instance the significance of cross-linguistic evidence) and from a more applied point of view (cf. for instance the need for good description to allow for the assessment of language learning problems). Unfortunately financial considerations are often a hindrance to the development of scientific disciplines in countries with severe economic problems. The IASCL has always been supportive of would-be IASCL members working in such countries by waiving membership fees for them.
IASCL funds are limited, though. In the past, donations from regular IASCL members have been very helpful in supporting colleagues from economically disadvantaged countries. In order to continue offering that support, your donations are very welcome indeed. Each donation, whatever the amount, will be acknowledged by a receipt signed by the IASCL Treasurer (useful perhaps for tax purposes). You may send donations in either pounds sterling or American dollars.
(1) Cheques in pounds sterling payable to IASCL can be sent to:
Dr Anna Theakston
University of Manchester
Department of Psychology
Manchester M13 9PL
Cash payments in pounds sterling can also be made by prior arrangement with Dr Theakston (firstname.lastname@example.org) at the above address.
(2) For American dollar amounts, please send your donations to:
Prof. Judith Becker Bryant
IASCL Assistant Treasurer
Department of Psychology, PCD 4118G
University of South Florida
Tampa, FL 33620-7200
The IASCL as a whole will be sure to benefit from the more diversified nature of its membership as a result of your donations. Many thanks in advance!
Anna Theakston, IASCL Treasurer
The Child Language Bulletin is the newsletter of the International Association for the Study of Child Language. It is distributed free to all members of IASCL and it is published twice a year.
The Bulletin is available on the IASCL Web page http://iascl-www.uia.ac.be
and all members of the association will receive an e- email message each time
a new issue of the Bulletin is published. A hard copy of the Bulletin will
only be sent to those members who ask for it by sending a message to the editor.
I encourage members to submit news and information that might be relevant to our research community. I would especially like to hear from doctoral students on new theses being completed. They are often a wonderful source of new data and new ideas that are not always easily accessible before publication.
Please do send any items to the address below. I am looking forward to your submissions!
Dr Ludovica Serratrice
University of Manchester
Faculty of Education
Language and Literacy Studies
Manchester M13 9PL