Language-Learning Round Table meeting
cspringer | 29 mai, 2008 10:25
Language-Learning sponsored Round Table meeting, to be held prior to
the 18th European Second Language Association Conference, Aix en Provence, on September 10, 2008
Acquisition orders and Levels of L2 proficiency in the Common European Framework of Reference for Languages
CORE LANGUAGE PROFICIENCY
Jan H. Hulstijn
Five claims for discussion at the 2008 Language Learning Round Table, to be held just prior to the 2008 EUROSLA conference, Aix en Provence. This text is part of a longer text, published in the December 2007 issue of The Modern Language Journal.
As of today, my theory of core language proficiency, still in its infancy, can be formulated in the following five points.
1. First, I propose a global distinction between lower-order and higher-order cognition in language processing. Lower-order linguistic cognition pertains to the largely implicit, unconscious (a) knowledge of speech sounds, prosody, and phonology, (b) syntactic knowledge of a number of lexical items, (c) morphology and syntax, and (d) the automaticity with which this knowledge can be processed. Higher-order cognition pertains to largely, but not exclusively, explicit, conscious knowledge of all sorts of topics (of which vocabulary size is but one rough index), as well as attention allocation, decision making, inferencing ability, and the like. Working-memory capacity is taken as a potentially important moderator variable in the online use of lower- and higher-order cognition. Empirical support for the distinction between lower-order and language-specific cognition, on the one hand, and higher-order and non-language-specific cognition, on the other hand, in L2 learners was produced in several studies conducted at the University of Amsterdam (Schoonen, Hulstijn, & Bossers, 1998; Schoonen, Van Gelderen, De Glopper, Hulstijn, Simis, Snellings, & Stevenson, 2003; Van Gelderen, Schoonen, De Glopper, Hulstijn, Snellings, Simis, & Stevenson, 2003; Van Gelderen, Schoonen, De Glopper, Hulstijn, Simis, Snellings, & Stevenson, 2004).
2. Second, I hypothesize that all humans, even adults of older age or lower educational backgrounds, are capable of implicit learning, in other words, capable of acquiring lower-order cognition in all walks of life (Ellis, 2005; Hasher & Zacks, 1979; Hulstijn, 2002, 2003, 2005; Paradis, 2004; Reber, Allen, & Reber, 1999; Van Gerven, Paas, Van Merriënboer, & Schmidt, 2002). Implicit learning applies also to the acquisition of oral language skills (listening and speaking) in an L2 just as in a first language (L1).
3. Third, I propose the notion of core language proficiency. It differentiates between knowledge and skill. In the case of speech perception (listening) and speech production (speaking), knowledge refers to the mental representation of (a) speech sounds, phonemes, stress, and intonation patterns; (b) lexical items; and (c) morphology and syntax.
Skill refers to the ability to process phonetic, lexical, and grammatical information receptively and productively, accurately and online. The core of language proficiency restricts this knowledge and skill to frequent lexical items and frequent grammatical constructions, that is, to lexical items and syntactic constructions that may occur in any communicative situation, common to all adult NSs regardless of age, educational level, or literacy.
4. From this definition of core language proficiency, the claim can be derived that all (mentally healthy) adult NSs, regardless of differences in age and intellectual functioning, are able to segment and comprehend, both correctly and quickly, isolated utterances consisting of high-frequency lexical phrases and morphosyntactic structures, when these utterances are perceived under normal acoustical conditions. The extent to which this provocative claim is correct with respect to receptive oral language processing (listening), will be investigated in a new project (2007–2011) conducted at the ACLC with 200 adult NSs differing in age (20–65) and educational background (lower level vocational training versus college level), along with 100 NNSs of Dutch.
5. This definition of core language proficiency does not show exactly where the borderline between core and periphery is to be found. I suggest that the demarcation line must be seen in probabilistic terms. On the basis of statistical analyses conducted on corpora of a wide variety of oral and written discourse, it should be possible to define a set of words and multiple-word constructions that have a high probability of occurring in communicative situations that all adult NSs, regardless of age and educational level, are likely to be able to use. In our new project, two of these parameters (utterance length and lexical difficulty) will be manipulated in order to test to what extent the notion of core language proficiency can be empirically upheld.
Posted in Language Learning Roudtable . Commenter: (0). Rétroliens:(0). Permalien
Language-Learning Round Table meeting
cspringer | 29 mai, 2008 10:19
Language-Learning sponsored Round Table meeting, to be held prior to
the 18th European Second Language Association Conference, Aix en Provence, on September 10, 2008
Acquisition orders and Levels of L2 proficiency in the Common European Framework of Reference for Languages
List of posters (names, titles, abstracts)
No1
Presenters: Inge Bartning & Fanny Forsberg, University of Stockholm (Sweden)
Title: Linguistic features characteristic of the CEFR-levels
Abstract
This project investigates the link between the criteria proposed in the CEFR scale and linguistic acquisitional orders. The main purpose is to propose linguistic criteria characteristic of each CEFR level (A2-C1), in terms of morphosyntax, discourse organisation and formulaic sequences, areas which have already been studied in the light of acquisitional orders (Bartning&Schlyter 2004, Hancock 2007, Forsberg 2006).
To this end, written data has been collected from Swedish university students of L2 French. First, the students take the DIALANG test in order for us to determine their supposed CEFR level. They are then asked to perform two written tasks, one common for all the informants, one conceived specifically to test the skills required at their given CEFR-level. The tasks are then evaluated by raters to make sure that the performance match the CEFR critieria stated for the given level. Lastly, the tasks are analyzed in terms of common features for each level.
No2
Presenters: Florencia Franceschina & Maisa Martin
Department of Languages
University of Jyväskylä
Finland
Poster title: A crosslinguistic investigation of negation as a marker of L2 proficiency level in L2 English and L2 Finnish
Abstract
This poster addresses the question of which linguistic features characterise learners’ performance at the different proficiency levels defined in the CEFR (see Round Table RQ1) by looking at the development of L2 English and L2 Finnish negation in written production at levels A1 to C2. We aim to uncover any systematicity that may exist in the development of negation and thus reveal any non-trivial differences between the two typologically different L2s investigated. We will document emergence and mastery of morphosyntactic negative devices and examine their frequency, accuracy and distribution in the written production of learners at each CEFR level, in order to test the DEMfad model (Franceschina 2006) as a tool for tracking diagnostically meaningful changes across communicatively defined proficiency levels. The data come from the CEFLING project (http://www.jyu.fi/hum/laitokset/kielet/cefling) and it includes writing samples from adults and young writers elicited using a range of tasks.
No3
Presenters: Jonas Granfeldt (1) & Pierre Nugues (2)
(1) Centre for Languages and Literature, University of Lund
(2) Department of Computer Science, Lund Institute of Technology, University of Lund
Sweden
Title: Working with the CEFR proficiency levels in written L2 French: Some insights from the Direkt Profil project
Abstract
We bring to this discussion results and experiences from a project on automatic linguistic profiling of L2 French, carried out at Lund university (Sweden) and funded by the national Research Council (2005-2007). In this project we have developed and tested a system called Direkt Profil for automatic assessment of developmental stage using raw learner texts.
In our contribution we would like to focus on practical problems - and possible solutions - involved when working with the CEFR levels on authentic learner data. We will discuss how a combination of the two major techniques in the Direkt Profil system – Natural Language Processing and Machine Learning – could be extended to address research questions related to the CEFR, in particular i) the empirical validation of the CEFR proficiency levels, ii) the identification of target structures for particular CEFR levels and iii) the relation between CEFR levels and claimed developmental stages in L2 French.
No4
Presenters: Folkert Kuiken & Ineke Vedder, University of Amsterdam
Title: Communicative adequacy and linguistic complexity in the written output of L2 learners of Dutch and Italian
Abstract
In the study presented in this poster, we try to establish which linguistic features characterize written L2 production at the B1 level of the CEFR. More in particular, the study investigates the relationship between communicative adequacy and linguistic complexity. Linguistic complexity is operationalized as syntactic complexity, lexical variation, and accuracy of the output, as established both by some global measures and by the judgments of individual raters. In addition, we study the use of some specific target elements, which may characterize the B1 level. Participants in the study include 40 learners of Dutch L2, 40 learners of Italian L2, 15 native speakers of Dutch, and 15 native speakers of Italian, who all wrote two essays.
No5
Presenter : Sabine Lopez
Université Pierre et Marie Curie – Paris VI-
Title: Are there any links between the CEF scale and the developmental stages defined by SLA researches? A study case in FFL.
Abstract
In the field of the French as a Foreign Language, most institutions have developed a whole range of tests, certificates or diplomas referring to the CEF. Is there a relation between the scale proposed by the CEF and the developmental stages defined by SLA research ? Is there convergence or divergence? On the basis of what criteria? The study of a corpus including written and oral tests from the TCF ( Test de Connaissance du Français) will provide the empirical basis of this poster proposal. These productions have been assessed through scales listed by the CEF and classified according to the levels defined in the common framework. Then, we have subjected them to an automatic analyzer "Direkt Profil". Focus will be laid on the study of morphological and syntactical structures or features such as negation, noun phrases or temporality.
No6
Presenter: Guenter Nold
University of Dortmund, Germany
Title: Morpho-syntactic structures at different levels of the CEFR - Results from the Speaking Test of the DESI Research Project
Abstract
A sample of oral speech of L2 learners (N=400) whose speaking proficiency levels range from A1 to B2 on the general speaking scale of the CEFR forms the data base for a morpho-syntactic analysis (types of verbs, concord, tense and aspect markers, complementation of the verb). Both error-free target language structures and structures with errors will be identified and their first appearance and /or their disappearance at the different levels of the CEFR will be reported. It will be discussed which morpho-syntactic structures and which errors can be typically identified with certain levels of the CEFR if it is taken into account that the criteria for rating the respective test items are not primarily determined by the morpho-syntax of speech.
The following will be looked at in greater detail:
(1) Is it possible to determine cut-off points for the field of morpho-syntax at each level of the CEFR?
(2) If so, are these cut-off points in line with SLA developmental features in the morpho-syntax of oral speech in English?
(3) Which facets of language (e.g. lexis) typically interact with the the morpho-syntax of speech at the different levels of the CEFR?
No7
Presenter: Gabriele Pallotti
University of Modena and Reggio Emilia
Title: Scales for profiling interlanguage development and their relationship to CEFR scales
Abstract
In order to relate CEFR scales with interlanguage developmental sequences it is necessary to provide clear, operational and practicable definitions of the latter. The project presented in the poster pursues the following objectives:
a) formulating explicit operational definitions of ‘acquisition’;
b) constructing theoretically valid and practically reliable developmental scales of acquisition for selected linguistic features;
c) devising procedures that will allow teachers and testers to assess learners’ level on these scales in a reasonable amount of time and without extensive technical knowledge;
d) linking this interlanguage-based assessment to that based on CEFR scales.
Data come from a project on L2 Italian involving SLA researchers, testers and teachers. Subjects are children aged 5-13, interviewed through a set of communicative tasks aimed at eliciting a high number of linguistic structures with good diagnostic value. Elicitation and analysis procedures will be presented along with a discussion of how ‘acquisition’ can be operationalized in terms developmental scales.
No8
Rob Schoonen , Nivja de Jong, Margarita Steinel, Arjen Florijn, & Jan Hulstijn
Amsterdam Center for Language and Communication (ACLC), University of Amsterdam (Netherlands)
Title: The compositional nature of speaking ability
Abstract
Being able to speak adequately in a variety of communicative situations is a highly complex skill, composed of various linguistic and psycholinguistic subskills. This raises questions regarding the acquisition of adequate speaking skills, such as: What is the compositional nature of speaking ability in terms of (psycho)linguistic subskills? That is, to what extent can performance on tests of vocabulary and grammatical knowledge, speed of lexical retrieval and sentence production, and speech fluency predict success in speech performance? Furthermore, to what extent is this compositional nature dependent on the kind of communicative speaking tasks involved, and of the speaker’s level of second language proficiency?
In this presentation, we present data from approximately 200 learners of L2 Dutch, performing 8 speaking tasks and several (psycho)linguistic tasks. The speaking tasks covered formal and personal situations requiring descriptive or argumentative speech acts.
No9
Presenter: Claude Springer, Université de Provence (France)
Title: Realisation of an evaluation grid for written production related to the CEF (B1 / C2) for the italian state examination (upper secondary school), Aosta. Action-research project, 2004-2008
Abstract
This poster presents an action-research project (still in progress) in Italy concerning the process of relating the written exam of the state examination to the CEF in a bilingual region (Aosta). This institutional project was designed for the bilingual teachers of Aosta upper secondary schools. It’s main objective is the possible evolution of the marking procedure teachers use and the certification of bilingual competences of upper secondary pupils in a bilingual system. The marking procedure used by the teachers is obviously traditional with an implicit hypothetical model of an advanced bilingual speaker of French who will be able to follow courses in a French university at a high level of proficiency.
This project followed the recommandations of the Manual for relating exams to the CEF and in particular the familiarisation phase. It concerned first voluntary teachers. With this pilot group of teachers we proceeded to the realisation of a grid of descriptors coherent with the CEF but also with the institutional norms fixed for the A level (baccalaureat) examination.
We have been able to test the grid, with a double marking system, for the last two sessions (traditional marking system and the new CEF system). We have been able to collect a significant written corpus. We are therefore able to present our first results regarding the process of relating a traditional exam to the CEF grid but also the possible definition of written academic acquisition profiles. This project has already proved that there is no such thing as a pure advanced bilingual pupil, even in a bilingual system, but different profiles of competences. Our analysis takes into account the notion of discourse « genres » on which the academic formats are based, but also the performance levels of our pupils in terms of pragmatic and linguistic aspects.
Posted in Language Learning Roudtable . Commenter: (0). Rétroliens:(0). Permalien
Language-Learning Round Table meeting
cspringer | 29 mai, 2008 10:12
Language-Learning sponsored Round Table meeting, to be held prior to
the 18th European Second Language Association Conference, Aix en Provence, on September 10, 2008
Acquisition orders and Levels of L2 proficiency in the Common European Framework of Reference for Languages
Organizers
Jan H. Hulstijn, University of Amsterdam
Daniel Véronique, Université de Provence (Aix-Marseille 1)
Round Table rationale
This Round Table aims to link SLA research on acquisition orders with the levels of language proficiency as defined in the Common European Framework of Reference for Languages (CEFR) (Council of Europe, 2001). The CEFR distinguishes six proficiency levels, defined in terms of (1) What a L2 user can do with L2 (communicative tasks), and (2) How well the L2 user can perform tasks, using which type of general and linguistic competencies, assessed along (a) scales combining the What and How well dimensions (pp. 58-96), and (b) a number of “linguistic competencies” (How well only), such as vocabulary range, vocabulary control, grammatical accuracy, phonological control, and sociolinguistic appropriateness (pp. 108–118).
Various questions emerge for SLA researchers:
1. Can we identify target elements (in the domains of morphosyntax, vocabulary, phonology, etc.) that are typical for certain CEFR levels? For instance, which grammatical structures and what vocabulary must a learner of a particular L2 have acquired at the A2 and B1 proficiency levels? To what extent do we find non-trivial differences between languages in this respect? (For example, if Verb Final in subclauses is typical for B1 in German, will we also find that it is typical for B1 in Dutch?)
2. For grammatical acquisition orders already claimed in the SLA literature, do we find links with levels of language proficiency in terms of CEFR scales?
3. Do we find different linguistic profiles for L2 users at the same level? For example, for L2 learners at the B1 level, to what extent can poor morphosyntactic command be compensated with a large vocabulary? And, vice versa, to what extent can a small vocabulary be compensated by a high command of morphosyntax? Which variability is permitted? Where do we observe constraints in the variability of the componential structure of language proficiency? Do we find clusters of subskills that are typical at particular proficiency levels?
4. In linking levels of L2 proficiency and acquisition orders, it might be helpful to distinguish between representation and processing and/or between declarative and procedural knowledge? Is there any relevant empirical evidence for this idea?
Programme Structure
In most Language Learning Round Table meetings held at previous EUROSLA conferences, most time was consumed by presentations of individual speakers, while too little time remained for discussion. In the spirit of what a “round table” meeting aims to be, we program two long discussion sessions around a poster presentation session. Thus, there are no “presentations” in the traditional sense.
1. The first discussion session is devoted to the following question: What are the pros and cons of linking SLA research on acquisition orders to proficiency levels, as defined by the CEFR (for curriculum design and language assessment)? The panel members give their views first (60 minutes); the floor is then open to the entire audience (30 minutes). Panel members are not supposed to give a formal presentation.
2. Then follows a poster session (60 minutes), preferably in the same room (with refreshments available). Panel members and other researchers who have accepted our invitation to present their research, related to the theme of this Round Table meeting, stand at their posters, presenting the highlights of their research. Thus, posters replace traditional presentations.
3. Discussion session 2 (60 minutes) about a list of claims, put forward by Jan Hulstijn, on the notion of language proficiency (see Appendix 2).
Programme Timeline (September 10, 2008, 1:45 – 5:30 p.m.)
1:45 – 2:00 Opening Jan Hulstijn & Daniel Véronique
2:00– 3:30 Discussion session 1, about the pros and cons of linking SLA research on acquisition orders to proficiency levels, as defined by the CEFR Chair: Jan Hulstijn
2:00 – 3:00 Panel members give their views
3:00 – 3:30 Discussion with the audience
3:30 – 4:30 Poster session and refreshments
4:30 – 5:30 Discussion session 2, about the list of claims put forward by Jan Hulstijn. Panel members and audience give their views. Chair: Daniel Véronique
List of panel members (in alphabetical order)
Jonas Granfeldt, Lund University (Sweden)
Maisa Martin, University of Jyväskylä (Finland)
Günter Nold, University of Dortmundt (Germany)
Gabriele Pallotti, University of Modena and Reggio Emilia (Italy)
Rob Schoonen, University of Amsterdam (Netherlands)
Claude Springer, Université de Provence (France)
List of posters (presenters, affiliation, and titles) in alphabetical order of presenters’ last names
1. Inge Bartning & Fanny Forsberg, University of Stockholm (Sweden)
Linguistic features characteristic of the CEFR-levels
2. Florencia Franceschina & Maisa Martin, University of Jyväskylä (Finland)
A crosslinguistic investigation of negation as a marker of L2 proficiency level in L2 English and L2 Finnish
3. Jonas Granfeldt & Pierre Nugues, Lund University (Sweden)
Working with the CEFR proficiency levels in written L2 French: Some insights from the Direkt Profil project
4. Folkert Kuiken & Ineke Vedder, University of Amsterdam (Netherlands)
Communicative adequacy and linguistic complexity in the written output of L2 learners of Dutch and Italian
5. Sabine Lopez, Université Pierre et Marie Curie – Paris VI (France)
Are there any links between the CEF scale and the developmental stages defined by SLA researches? A study case in FFL.
6. Günter Nold, University of Dortmund (Germany)
Morpho-syntactic structures at different levels of the CEFR - Results from the Speaking Test of the DESI Research Project
7. Gabriele Pallotti, University of Modena and Reggio Emilia (Italy)
Scales for profiling interlanguage development and their relationship to CEFR scales
8. Rob Schoonen , Nivja de Jong, Margarita Steinel, Arjen Florijn, & Jan Hulstijn
Amsterdam Center for Language and Communication (ACLC), University of Amsterdam (Netherlands)
The compositional nature of speaking ability
9. Claude Springer, Université de Provence (France)
Realisation of an evaluation grid for written production related to the CEF (B1 / C2) for the italian state examination (upper secondary school), Aosta. Action-research project, 2004-2008
The organizers
1. Jan H. Hulstijn
Professor of Second Language Acquisition
Faculty of Humanities
University of Amsterdam
134 Spuistraat
1012 VB Amsterdam, the Netherlands
Tel.: +31 20 525-4616/4615
Fax: +31 20 525-4429
Email: hulstijn@uva.nl
Homepage: http://home.medewerker.uva.nl/j.h.hulstijn/
2. G. Daniel Véronique
Professor of French Linguistics and Creole Studies
UFR Lettres, Arts, Communication et Sciences du langage
Université de Provence, Centre d’Aix
29 Avenue Robert Schuman
13621 Aix-en-Provence CEDEX 1
Tel. +33 (0) 4 42 95 35 58/ 59
Fax. +33 (0) 4 42 95 35 58/ 59
Email : georges.veronique@orange.fr
Posted in Language Learning Roudtable . Commenter: (0). Rétroliens:(0). Permalien