Venue (hybrid mode): Gothenburg, Sweden - University of Gothenburg, Humanisten.
Visiting address: Renströmsgatan 6, Gothenburg (Room J330. Follow the signs)
Zoom-link: https://gu-se.zoom.us/j/63867291284
Dates: 20-21 April, 2023
Deadline for registration & abstract submission: 10 March, 2023 (For online attendees without presentations, registration is open till April, 18.)
Participation is free. For registered onsite participants, lunches and coffee breaks will be offered free of charge.
NEW! The book of abstracts is now published: Book of abstracts
Description
Språkbanken Text and HumInfra are organizing a workshop on tools and resources aimed at research on profiling second language vocabulary and grammar, to take place on April 20-21, 2023 in Gothenburg, Sweden.
Efforts are made to create electronic resources that could help analyze learner language to feed into research and development projects within Teaching and Didactics, Second Language Acquisition (SLA) and Assessment, Lexicography, Intelligent CALL applications, automatic analysis of learner language, automatic exercise generation, course book writing and various applied derivative products, such as learning apps. Such resources, called second language profiles, are research infrastructure components that often focus on second language (L2) learners, covering descriptions/listings of vocabulary, grammar and morphology learners acquire by level of proficiency. We witness the appearance of such resources, e.g. English Profile, Teacher's Tools for Estonian and Swedish L2 profile, however, since such resources are quite few and, most importantly, new to users/researchers, we need a forum where we can discuss them, practically test them, explore their possibilities for research together and exchange experiences.
In this workshop, through presentations, discussions, demos and hands-on sessions, we intend:
- to establish some common practices, among others, how to use L2 profiles practically and how to apply them to different research questions
- to showcase scenarios for using them in research & teaching and outline benefits of their use
- to exchange experiences and lessons learnt between projects that have created such resources, reflecting on the ways to improve user-friendliness and identifying missing functionalities
- to inspire other languages to work on similar resources stimulating the creation of a new family of resources. Projects of that kind can hardly be "re-run" to fix errors that are discovered post-factum, and therefore planning such projects "right from the start" is a prerequisite. With the experiences from the L2 profile projects for Swedish, Estonian and English, we intend to offer our expertise (à la "advisory role") and help researchers in other countries set up their projects.
Call for presentations / participation
We invite researchers and teachers working with profiling second language lexical, grammatical and other types of competencies to join our workshop. If you intend to participate, please, fill in this registration form by March, 10, 2023. It is not obligatory, but possible to present your work during the workshop. In that case, please, add a short abstract (100-500 words), in the same registration form. We will come back to you with details shortly after March, 10.
Some of the topics of interest:
- descriptions of existing resources, work-in-progress or planned resources
- examples of practical use of such resources
- lexical profiles
- grammatical profiles
- morphological profiles
- word families
- linguistic complexity studies based on L2 profiles
- etc.
Some areas of interest:
- lexicography
- teaching
- (Second) Language Acquisition
- language assessment
- Intelligent Computer-Assisted language learning
- Natural Language Processing
- language learning app development
- etc.
Program
(All through the workshop a few group rooms will be available for discussions between the participants)
'
| Day 1, 20 April, 2023 Venue: University of Gothenburg, Humanisten campus, J330 |
|
| Lexical profiling - approaches, resources, applications | |
| 9:00-9:10 | Welcome and information [Elena Volodina] [Slides] |
| 9:10-9:55 | Invited talk [Chair: Elena Volodina] Jelena Kallas, Institute of the Estonian language (Estonia) Aligning leaners' dictionaries with the CEFR: the case of Estonian Vocabulary and Grammar Profiles [Slides] [Video] |
| 9:55-10:00 | Short break |
| Session 1. Chair: Elena Volodina | |
| 10:00-10:20 | Nina Hicks, University of Fribourg (Switzerland) (online) Lexical features in adolescents' writing: Insights from the trilingual parallel corpus SWIKO [Slides] |
| 10:20-10:40 | Coffee break |
| Session 2. Chair: Julia Prentice | |
| 10:40-11:00 | Kris Heylen, Dutch Language Institute (The Netherlands) Ilan Kernerman, Lexicala by K dictionaries (Israel) Carole Tiberius, Dutch Language Institute (The Netherlands) Linking CEFR-based learner profiles to lexicographic data [Slides] |
| 11:00-11:20 | Bernardo Stearns, University of Galway (Ireland) Using Learner language models for lexical profile [Slides] |
| 11:20-11:40 | Mojca Stritar Kučuk, University of Ljubljana (Slovenia) A cross-section of linguistic competence of South Slavic university students learning Slovene as L2 [Slides] |
| 11:40-11:45 | Short break |
| 11:45-12:30 | Organizer presentation and demo [Chair: Jelena Kallas] David Alfter, University of Gothenburg (Sweden) Swedish Lexical profile [Slides] [Video] |
| 12:30-14:00 | Lunch University canteen |
| Grammatical profiling - approaches, resources, applications | |
| 14:00-14:45 | Invited talk [Chair: Therese Lindström Tiedemann] Geraldine Mark, Cardiff University (Wales) Building on insights from the English Grammar Profile: From really good to painfully obvious [Slides] |
| 14:45-14:50 | Short break |
| Session 3. Chair: Aleksandrs Berdicevskis | |
| 14:50-15:10 | Annekatrin Kaivapalu, University of Helsinki (Finland) Profiling learner Finnish and Estonian: interaction of frequency and accuracy as an indicator of language skills [Slides] |
| 15:10-15:30 | David Alfter, University of Gothenburg (Sweden) French Verb profile [Slides] |
| 15:30-16:00 | Coffee break |
| Session 4. Chair: Aleksandrs Berdicevskis | |
| 16:00-16:20 | Nicolas Ballier, Université Paris Cité (France) Grammatical profiling with UD annotation (WiP) [Slides] |
| 16:20-16:40 | Ekaterina Vlasova, University of Helsinki (Finland) Prepositions in L2 Russian [Slides] |
| 16:40-16:45 | Short break |
| 16:45-17:30 | Organizer presentation and demo [Chair: Geraldine Mark] Therese Lindström Tiedemann, University of Helsinki (Finland) Swedish Grammatical Profile [Slides] [Video] |
| 17:30-17:40 | Rounding off Instructions how to get to dinner place [Elena Volodina] |
| 18:00 -... | Dinner at Berzelius Bar & Matsal. Address: Södra Vägen 20, 412 54 Göteborg |
| Day 2, 21 April, 2023 Venue: University of Gothenburg, Humanisten campus, J330 |
|
| Word families, Morphological families and Morphological profiling - approaches, resources, applications | |
| 9:00-9:45 | Invited talk [Chair: Therese Lindström Tiedemann] Gabriele Pallotti, University of Modena and Reggio Emilia (Italy) Profiling complexity: methodological issues and applications to L2 morphology [Slides] [Video] |
| 9:45-9:50 | Short break |
| Session 5. Chair: Kris Heylen | |
| 9:50-10:10 | Maria Belén Diez-Bedmar, University of Jaén (Spain) The FineDesc learner corpus: Making the CEFR/CV more user-friendly: fine-tuning descriptors with Learner Corpus Research results |
| 10:10-10:30 | Francesca La Russa, Sapienza Universitá di Roma (Italy) Maria Roccaforte, Sapienza Universitá di Roma (Italy) Using a learner corpus to design a phraseological syllabus of Italian collocations[Slides] |
| 10:30-10:50 | Coffee break |
| Session 6. Chair: Kris Heylen | |
| 10:50-11:10 | Christina Lindqvist, University of Gothenburg (Sweden) Mårten Ramnäs, University of Gothenburg (Sweden) A Digital Dictionary of Romance Word Families [Slides] |
| 11:10-11:30 | Isidora Glišić, University of Iceland (Iceland) From corpus to profiles: Icelandic L2 corpus [Slides] |
| 11:30-11:40 | Short break |
| 11:40-12:25 | Organizer presentation and demo [Chair: Christina Lindqvist] Elena Volodina, University of Gothenburg (Sweden) Swedish Morphological profile [Slides] [Video] |
| 12:25-13:40 | Lunch University canteen |
| Linguistic complexity – tying it all together | |
| 13:40-14:25 | Invited talk [Chair: Gabriele Pallotti] Aleksandrs Berdicevskis, University of Gothenburg (Sweden) We need to know more about relative complexity and learnability [Video] |
| 14:25-14:45 | Coffee break |
| 14:45-15:30 | Discussion session (University canteen for onsite and Zoom rooms for online participants) Discussion leaders and instructions |
| 15:30-15:45 | Move back to the main room and Zoom |
| 15:45-16:00 | Sum up from small groups [Chair: Therese Lindström Tiedemann] |
| 16:00-16:30 | Rounding off: prospects, next steps [Elena Volodina] |
Invited speakers
Aleksandrs Berdicevskis, University of Gothenburg, Sweden
Title: We need to know more about relative complexity and learnability
Abstract: For the last two decades, language typology and related fields have witnessed a hot debate about language complexity. Several influential theories have emerged that claim that languages are not equally complex, and that the distribution of complexity depends on social factors, such as number of speakers, degree of language contact and number of non-native speakers. I will briefly review some recent evidence in favour and against those theories and argue that whether the theories are correct or not, they make interesting non-trivial hypotheses about mechanisms of language learning and language change. I will then make my main point, which is that these hypotheses cannot be properly addressed without a deep understanding of second language acquisition, most crucially, the concepts of relative complexity ("what is difficult for whom") and learnability. The talk will mostly focus on morphological complexity.
Bio: Aleksandrs Berdicevskis is a researcher in computational linguistics at Språkbanken Text, University of Gothenburg. His research focuses on explanatory approaches to language change and language typology, with a particular attention to language complexity. He currently leads the project Cassandra: Explaining and predicting short-term language change in Modern Swedish.
Jelena Kallas, Eesti Keele Instituut, Estonia
Title: Aligning learners' dictionaries with the CEFR: Estonian Vocabulary and Grammar Profiles
Abstract: In the talk, we will introduce the Estonian Vocabulary Profile and the Estonian Grammar Profile, which are designed to support the CEFR illustrative descriptors scales of linguistic competence with language-specific descriptions. We will focus on the methodology and corpora that were used for the development, trial and validation of the Estonian Grammar Profile. Currently, the profile provides descriptions of grammar competence on the morphology, derivation, phrase and sentence levels, from the pre-A1 level up to the B2 level for young learners, and from A1 to C1 for adult learners. All descriptions are equipped with example sentences compiled either by experts or taken from the coursebook and learner’s corpora.
In addition, we will address the issues related to our attempt to combine this resource with the Estonian learners’ dictionary Sõnaveeb for Learners. The dictionary is compiled in the Dictionary Writing System Ekilex, whose long-term goal is to have a single data source that provides consistent and comprehensive information about Estonian, including CEFR labels. We will report on the work in progress from the point of view of data modelling. Given a construction-based and usage-based understanding of L2 acquisition, we assume that linguistic knowledge at a particular proficiency level is not best described as a set of words and a set of grammatical structures, as is the current practice, but rather as a set of combinations of particular word meanings and forms with particular schematic constructions. This means that the lexicographic resource must include descriptions of grammatical constructions, and that the language proficiency level should be attributed not to lemmas and constructions, but to particular word meanings in particular forms and in particular constructions.
Bio: Jelena Kallas is a Senior Computational Lexicographer – Project Manager at Eesti Keele Instituut (Institute of the Estonian Language, Tallinn, Estonia). She has been involved in various lexicographic projects, including monolingual and bilingual dictionaries, and SLA projects. She is leading the Estonian L2 profile project called "Teacher Toolkit". She is a holder of national team grant "Expanding the scope of a multi-purpose lexicographic resource to grammar and L2 competence (2023-2027)" where she and the other project members will work on the development of a theoretical and methodological framework for the description of grammatical constructions and L2 linguistic competence in an lexicographic resource, relying on a usage-based and construction-based approach to linguistic theory, language acquisition and lexicography.
Gabriele Pallotti, University of Modena and Reggio Emilia, Italy
Title: Profiling complexity: methodological issues and applications to L2 morphology
Abstract: In this talk I will first discuss how linguistic complexity should be theoretically defined and practically operationalized, in a wider context of interlanguage analysis and linguistic profiling. In particular, I will argue that it needs to be kept apart from other constructs such as processing difficulty or developmental timing. Then, I will present an approach to empirically measuring morphological complexity, its conceptual and methodological challenges and how they were addressed in the development of an online morphological complexity analyzer.
Bio: Gabriele Pallotti is a professor of Language teaching methodology at the University of Modena and Reggio Emilia. His research focusses on interlanguage development, linguistic complexity, morphology, L2 interaction, methodology and epistemology in applied linguistics. He coordinates the project Observing interlanguage and is the associate editor of the Eurosla Studies Series (Language Science Press). He has led several national and international projects on language learning and teaching, funded by the National Ministry of Education and the European Union.
Geraldine Mark, Cardiff University, Wales
Title: Building on insights from the English Grammar Profile: From really good to painfully obvious
Abstract: The English Grammar Profile (EGP) Project was a four-year quasi-longitudinal study investigating learner grammar from the Cambridge Learner Corpus (CLC). The main output of the research is the EGP, a free educational online database, which provides a profile of over 1,200 corpus-based grammar competency statements about learner grammar use across the six CEFR levels. In the first part of this talk I’ll describe the methodology that we developed to build the EGP, discuss the key insights from the study and show how the investigation has enhanced our understanding of the developmental nature of grammar acquisition and use. I’ll then look at further ways to explore the data taking a usage-based (UB) approach. UB studies have shown that language users are sensitive to the statistics of repeated patterns in language and that we figure out ‘structural regularities’ in language as we subconsciously tune into mappings of form and meaning (Ellis et al. 2016). Using this large scale proficiency-levelled data I will look at how we can use corpus tools to investigate if and how structural regularities develop in L2 English and how this might offer further insight into learner language development.
Ellis, N. C., Römer, U., & O’Donnell, M. B. (2016). Usage-based approaches to language acquisition and processing: Cognitive and corpus investigations of construction grammar. Oxford: Wiley.
Bio: Geraldine Mark is an applied corpus linguist with experience in research, teaching and learning, publishing and materials design. Her principal interests are in corpus linguistics and its diverse applications, particularly in relation to language development and usage in L1 and L2, data-driven learning, and multi-modal interaction. She is a Visiting Lecturer at the University of Malta, and advises on the FoRCE project, building and analysing a corpus of Maltese English. She is a Research Associate on a multi-modal project, IVO (www.ivohub.com), funded by the UKRI Arts and Humanities Research Council and the Irish Research Council, examining online workplace multi-modal interaction. She is co-author of English Grammar Today (2011, Cambridge University Press, with Ronald Carter, Michael McCarthy and Anne O’Keeffe) and co-principal researcher (with Anne O’Keeffe) of the English Grammar Profile, an online resource profiling L2 grammar development.
Organizers
- Elena Volodina, University of Gothenburg, Gothenburg, Sweden
- Therese Lindström Tiedemann, University of Helsinki, Helsinki, Finland
- David Alfter, Gothenburg Research Infrastructure in Digital Humanities (GRIDH), University of Gothenburg, Gothenburg, Sweden
Contact
- Elena Volodina, < elena dot volodina at svenska dot gu dot se >