EU Lingua D Project (1999 – 2002)

•  http://www.hum.uit.no/a/svenonius/lingua/

IGLO – Intercomprehension of Germanic Languages Online
Project team Salzburg: o. Univ. Prof. Dr. Hubert Haider
Research stipends: Mag. Caroline Moldaschl | Mag. Petra Zauner

A. Background information on the Coordinator and project partners

A.1
The Coordinating institution is the University of Tromsø (Universitetet i Tromsø), which has about 6,500 students and 1,800 staff. The Coordinator of the project, Dr. Peter Svenonius, is an associate professor in the English Department. He will be primarily responsible for the English language material. The person responsible for Norwegian, Professor Tarald Taraldsen, is a full professor in the Linguistics Department. Professor Taraldsen will also be responsible for developing a cognate network, linking cognate words across the seven languages.

The State University of North-Rhine Westphalia (Fernuniversität – Gesamthochschule in Hagen) has about 55,000 students and 1,600 staff (700 academic). Dr. Gerhard Kischel is the head of the Department of Intercultural Multilingualism in the Institute for European Studies. He has recently finished running a three-year Lingua D project (Projekt zur Interkulturellen Mehrsprachigkeit). Dr. Kischel will be primarily responsible for selection of texts, and for overseeing the actual programming work.

The University of Lund (Lunds universitet) has about 31,000 students and 6,200 staff (4,100 academic). Professor Christer Platzack is a professor in the Nordic department and is the Socrates coordinator for his institution. He will be responsible for work relating to the Swedish language, as well as for the determination of the points of parametric variation around which the explanatory grammars will be organized.

The University of Salzburg (Universität Salzburg) has 12,500 students and 1,100 staff (500 academic). Professor Hubert Haider is professor and chair of the Department of General and Applied Linguistics and is the Socrates coordinator for his institution. He will be primarily responsible for the work on the German language and will in addition have primary responsibility for overseeing the development of exercise formats for use in the program.

The Copenhagen Business School (Handelshøjskole København) has 13,500 students and 800 staff. Dr. Sabine Kirchmeier-Andersen is associate professor and chair of the Department of Computational Linguistics. She worked on the EU project Eurotra (Machine translation) from 1987 -1991, both in the Danish branch and in one of the central task forces. Dr. Kirchmeier-Andersen will be responsible for work having to do with Danish but also for aspects of the computational interface and the development of guidelines for vocabulary lists and online dictionaries to be used in the project. She will also organize the Summary meeting in November.

The University of Antwerp (Universiteit Antwerpen) has about 9,000 students and 1,500 staff. Dr. Sven van Elst, project manager at the Centre for Language and Speech, has been the project manager for the Lingua D projects LinC (An interactive approach to language and culture) and LOKI (Integration through Language and Culture). Dr. van Elst will be concerned with work on the Dutch language. He will also have primary responsibility for developing the structure of the overall course syllabus. In addition, he will organize the Start-up meeting in February.

The University of Iceland (Háskóli Íslands) has about 6,000 students and 500 staff. Dr. Sigrí›ur Sigurjónsdóttir is associate professor in the Department of Icelandic Language and Literature. Dr. Sigurjónsdóttir will have primary responsibility for coordinating Dr. Kirchmeier-Andersen’s work on the vocabulary lists and Professor Taraldsen’s cognate network. In addition, she and her colleague Dr. Jóhannes Gísli Jónsson of the Department of Linguistics will be responsible for the work on the Icelandic language and will also coordinate the Review meeting in Reykjavík in May.

The legal status of all of the above-named institutions is that of institution of higher learning.

A.2 Demonstrate that the partnership has the expertise required to undertake the roles identified above, and specify in particular expertise within the partnership relating to:

• foreign language teaching (include certification, if relevant)
• new technologies, where relevant
• dissemination
• analysis of the needs of the target groups
• the coordination of the project

The group is made up of people with diverse backgrounds and talents, reflecting the many aspects of the project.

First, there is (at least) one linguistically trained native speaker for each of the seven languages to be covered. For each language, this native speaker linguist will oversee the production of all material relating to that language (Dr. Svenonius for English, the others in obvious country-language pairings). Each has extensive contacts among the research community in his or her respective field.

Several of these native speaker linguists also have several years’ experience teaching their native language as a foreign language. This applies to Drs. Svenonius, van Elst, and Kirchmeier-Andersen and Professor Haider. All of the members work in institutions of higher learning and have many years of teaching experience of one kind or another (Professors Taraldsen and Platzack primarily in theoretical linguistics, Dr. Sigurjónsdóttir in applied linguistics).

All of the members of the project are computer-savvy and several maintain their own websites. In addition, Professor Haider was a member and chief investigator of the SFB (special research initiative, funded by the DFG) ‘Linguistic Foundations for Computational Linguistics: Language processing technology,’ from 1987 until 1996.

Drs. Kischel, van Elst, and Kirchmeier-Andersen are especially experienced in developing computer-based aids for language teaching. Dr. Kischel was the project manager for a successful three-year Lingua D project (Projekt zur Interkulturellen Mehrsprachigkeit) and Dr. van Elst has managed two (LinC and LOKI). Dr. Kirchmeier-Andersen has worked with a team which developed an innovative language teaching aid called VIA (Visual Interactive Analysis) under the VISUM project, financed (at 282,500 Euros) from the Danish Ministry of Education. She has also worked on several other computational linguistics projects, notably with EUROTRA, in the development of grammars and online dictionaries for multilingual machine translation.

Professors Taraldsen, Platzack, and Haider and Drs. Svenonius and Jónsson have wide-ranging experience in theoretical linguistics, especially in the comparative study of Germanic syntax. All are active in the field, having published articles in important journals and delivered papers at significant conferences in the recent past. Their participation in the project will ensure that the theoretical underpinnings of the project are solid and up-to-date.

Professor Platzack´s book (with Professor Anders Holmberg of the University of Tromsø), The Role of Inflection in Scandinavian Syntax (Oxford University Press, 1995), is a signal effort in the field at identifying systematically the differences among the Scandinavian languages in the parameter-based framework proposed by Noam Chomsky.

Dr. Sigurjónsdóttir is an active researcher in the field of applied linguistics and will be able to contribute her expertise in language acquisition to the project.

A.3 Append a copy of the draft cooperation agreement between the partners (see Guidelines for Applicants 1998 and Addendum 1999).

Appended: Copies of agreement from Tromsø, Salzburg, Copenhagen, Antwerp, and Reykjavík as pages 16-20.

B. Description of the project

B.1 Identify the main component(s) of the project, by choosing one or more from the following:

• urricula
• new materials
• adaptation of materials
• certification/recognition of skills
• other (please specify)

The primary goal of the project is to develop a program which functions as a foreign language distance course. As a matter of fact, the program will constitute forty-two distance courses in one, as it will enable a user to select any of the seven languages included (see point B.2 below) as the start language (?L1?) and any of the remaining six as the target language (?L2?). This means that a speaker of German can use the program to learn Swedish, a speaker of Swedish can use the program to learn Icelandic, and so on. In addition, a speaker who already knows Swedish as a second language can set L1 to Swedish and L2 to Norwegian in order to use Swedish as a stepping stone to Norwegian. The program will then highlight contrasts and provide explanatory commentary in Swedish.

A secondary goal of the project is to create a reference source for comparative Germanic syntax, an area of linguistics which has achieved a great deal in recent years. The level of grammatical explanation available in the program will ensure that the program is of interest to philologists and other linguists. The fact that the product is computer-based makes this feasible; a conventional grammar book which contains substantial detail risks overwhelming the student; but in the computer, layers of successively more detailed explanation can be hidden until requested.

Thus, the program will contain graded texts in each of the seven languages; to maximize the possibility of using the program as a tool for comparison, the same texts will be used (thus, a hypothetical German speaker who knows some Icelandic might set L1 to Icelandic, L2 to Norwegian, to see the similarities and differences between the two languages, but could reset L1 to German upon encountering difficulties owing to lack of fluency in Icelandic, and get German explanations for the same text).

In addition, the program will contain forty-two reference grammars, including forty-two sets of grammatical explanations for various grammatical points encountered in the texts used; there will be seven vocabulary lists and several sets of exercises assisting the learner in acquiring receptive competence (the exact number of sets of exercises will have to be determined; for example, six L1?s will require exercises designed to teach the Icelandic case system, but only the three non-Scandinavian languages will require drills teaching the recognition of the definite suffix, since all of the Scandinavian languages employ it).

B.2 Target languages

The project includes nearly all of the Germanic languages which are national languages of some country:
Norwegian, Swedish, Danish, Icelandic, English, Dutch, and German (but not Luxembourgish or Afrikaans) . This includes several less widely taught languages (the Scandinavian ones) but excludes the smallest languages, for example, Faroese and Frisian, and those which are not national languages, for example Yiddish and Low German.

The modular nature of the program makes it possible that additional languages could be added at a later date, but there are currently no plans to do so.

B.3 Countries in which language learners will benefit from the product

The product will be of benefit to speakers of any of the languages included in the project; owing to the inclusion of English, this includes a large portion of the world’s population, though we expect the product to be of practical use especially to people located in Europe, in particular northern Europe.

B.4 The types of users for whom the product is designed

The product is designed for adults, or at least reasonably mature individuals who have the motivation and inclination to study a foreign language on their own. It is designed for self-study and presupposes some familiarity with foreign language learning, e.g. names of parts of speech and other basic grammatical terminology. In its primary role (see post B.1), it is not intended for specialists but for non-specialists, in fact it should be usable by complete beginners in the target language. It is designed to rapidly confer partial ability, rather than fluency, in the target language, especially comprehension (reading comprehension, in the initial stages of the project, spoken comprehension at a later stage). The product should eventually take learners to advanced levels of comprehension, though obviously this will take longer going from, say, Dutch to Icelandic than from Dutch to German.

However, as noted in post B.1 above, a secondary goal of the project is to include the kind of grammatical information that would be of interest to a linguist studying the variation across the Germanic languages. In this respect the program has appeal for specialists.

B.5 Identify the main characteristics of the products of your project. Please use keywords, for example cultural dimension, preparation for mobility, early learning of languages, self-learning, etc.

The product of the project will be web-based software (although it will be a simple matter to make it available on CD-ROM for those who prefer to own the product). It is designed for self-learning, providing individuals with partial comprehension skills in Germanic languages. It is based on principles of intercomprehension, using the individual’s knowledge of one language as a gateway to knowledge of similar languages.

B.6 Indicate which of the priorities described in the Guidelines for Applicants 1998 and Addendum 1999 the project pursues.

The project develops a tool facilitating the learning of languages, specifically the Germanic languages. It is specifically designed to enable users to develop „partial competence, in particular receptive competence, by means of exploiting the similarities between languages belonging to the same linguistic family“ (from the Guidelines, p. 78).

In addition, it makes „pertinent use of new information and communication technologies and in particular open and distance learning approaches“ (ibid, p. 79); furthermore, the project „promote[s] self-learning“ (ibid, p. 79) by its very nature.

B.7 Specify the pedagogical and didactic approaches and methods which will be developed through the materials to be produced. Describe the methodologies and education approaches which will be used to develop these materials.

The course will be based on reading courses developed earlier in receptive multilingualism, e.g. the EuRom-4 approach (cf. Klein & Stegmann 1999, Klein 1999) or related efforts for the Slavic languages (cf. Meißner & Reinfried 1998, Zybatow 1999).

Dr. Kischel in particular has experience designing courses aimed at open distance learners; a successful course of this kind is equally appropriate for autonomous learners at conventional universities. Such learners require highly motivating, interactive multimedia material. This will be addressed by including pre-, during-, and post-reading tasks related to the comprehension of the texts, with grammatical explanations when appropriate (this means that the grammatical sketches and vocabulary lists used must be closely tailored to the texts chosen).

The focus on comprehension (cf. B.1 above), rather than production, suits our project ideally to the realm of computer-based self-learning, since it massively simplifies the feedback issue: students can be tested on their understanding of foreign-language material, rather than on the correctness of their foreign-language production (e.g. multiple-choice quizzes can be fully automated, in contrast to essay-writing exercises).

In accordance with earlier work in receptive multilingualism, we will limit ourselves (in the initial phase) to a single text category. In our current thinking on this matter, Social Sciences texts may be appropriate: scientific texts normally have a relatively poor grammatical and morphological structure and a limited vocabulary (perhaps 10-12,000 words) with a high incidence of international vocabulary.

The reading approach to be used is the ‘interactive’ one, stemming from the anglo-american research work on reading from the late seventies and early eighties. This approach links two different lines of thinking, the older bottom-up and a younger top-down approach. The bottom-up approach teaches the syntax, morphology, and vocabulary of the foreign language, and is well-suited to training skills. The top-down approach taps in to the knowledge a learner already has, knowledge stemming both from language learning and previous studies (and in our case, familiarity with another Germanic language). The interactive approach is a mixture of these two, and takes the reading process to be interactive: the learner combines previous knowledge with guessing and reading strategies.

The didacticians’ task in designing a course of this kind lies in facilitating the training of reading and guessing strategies through specific types of task and exercises. The texts, the vocabulary, the exercises, and the reference grammar have to be well designed in order to allow the learner to move through the material at his or her own pace, subject to intellectual capacities, motivation, and time constraints.

B.8 Where applicable, describe the role of technology in the project. Please provide the technical specifications of the equipment/media involved in the production and the use of the end products (see also C.6 below).

The product will be web-based software, written in Java according to current plans, with a user-friendly interface accessible through any of the standard browsers. The initial model for the design will be the computer-based reading courses designed at the Open University of Hagen, tailored to the specific needs of our project.

C. / D. Background for the project and justification

C.1 Describe the aims and objectives of the project.

The project has two objectives, one primary and one secondary (cf. also post B.1 above).

The primary objective is to develop a program which functions as a foreign language distance course, teaching comprehension skills for Germanic languages. More specifically, the program will constitute forty-two distance courses in one, since a user may choose any of the seven languages included (cf. post B.2 above) as the language of instruction and any of the remaining six as language to be learned. This will make the program massively useful for a wide spectrum of people, including workers in international offices who need simple reading competence in a variety languages in order to sort incoming mail, students and researchers who need advanced reading competence in a single language in order to read original texts which are not available in translation, and many many others. We envision later stages of the project as including training in oral comprehension as well, but focus initially only on reading.

The secondary objective of the project is to create a reference source for comparative Germanic syntax, utilizing the results of the study of parametric and microparametric variation. For example, Dutch and German are so-called OV languages, since at a certain level of abstraction, the object precedes the verb (as in embedded clauses). The remaining languages in the project are VO languages. The VO-OV distinction is one parameter of variation, and other properties of a language are expected to correlate with it; for example, in an OV language, an auxiliary verb may follow a main verb, while in a VO language this is never the case (Greenberg 1963, Dryer 1992). Another parameter has to do with subject agreement: in a language with ‘rich’ subject agreement, subjects may sometimes be omitted, while in a language with ‘poor’ subject agreement, they may generally not (cf. Holmberg & Platzack 1995, Vikner 1995 for discussion and definition of terms). Here the division among the Germanic languages cuts across the VO-OV parameter: German and Icelandic have strong agreement, and the remaining languages have weak agreement.

The organization of the grammars around these principles will make the program of interest to all linguists interested in parametric variation who know one of the languages of the program, even if they have no interest in actually learning the other languages for practical purposes.

C.2 Explain the rationale of and background to the project (the current situation, the need for improvement, etc.). Explain the need for the project, with reference, where appropriate, to existing foreign language curricula, certificates, and teaching materials.

Software simply does not exist for many, if not most, of the forty-two language pairs we are including. Though books exist for teaching, for example, Danish to Swedes, this has never been put into software format before.

Furthermore, currently available language-teaching software does not focus on comprehension, as we do, and in general it does not focus on broad intercomprehension, using the similarities among several closely related languages to promote rapid learning. In this we are following to some extent the lead of the EuroCom project in Frankfurt, whose proponents we are cooperating with.

C.3 Demonstrate that at least one of the partners has a direct interest in implementing the results of the project.

Each of the institutions involved engages in the teaching of at least some of the languages involved in the project.

Thus the end product is of direct interest to all of the institutions involved. Even more pertinently, many students will benefit from better comprehending languages that they are not actively studying, an ability that gives them greater access to original documents and source materials which have not been translated.

In addition, the comparative aspect of the project makes it of great interest to the theoreticians involved (Professors Haider, Taraldsen, Platzack, and Drs. Jónsson and Svenonius).

C.4 Give a brief account of any preparatory or previous work on the project.

The idea for a pan-Germanic Lingua D project was originally conceived by Friedrich Wittib of Socrates and Ingebjørg Birkeland of the Norwegian Center for International University Cooperation. The idea was aired at a meeting in Brussels in 1998 including Dr. Svenonius, Dr. Kischel, Dr. van Elst, and Professor Haider, under the auspices of Friedrich Wittib and a colleague of Ingebjørg Birkeland’s, Arne Aarseth. At that meeting it was decided to go ahead with a project including seven languages (cf. post B.2 above). It was further decided to gather one researcher for each language for a preparatory visit to plan a project. The preparatory visit took place in Tromsø in April 1999; all of the persons mentioned in post A.1 above were there except for Dr. Sigurjónsdóttir. In addition, Arne Aarseth was able to attend. Most of the details in the current application were worked out at that meeting.

C.5 Give the outcomes of any preliminary needs analysis undertaken (attach appropriate but succinct supporting documents), and list a relevant recent bibliography. If further needs analysis is planned, specify the methodology to be adopted.

No needs analysis studies have been undertaken, but we feel that the need for individuals knowing one Germanic language to learn one or several others is evident; in fact, the development of receptive competence in related languages is a Lingua D priority (cf. post B.6).

C.6 Describe the role of new technologies and/or Open and Distance Learning techniques (if they are components of the projects). Justify how they are appropriate to the pedagogical framework of the project and pedagogical goals of the taget group(s) concerned.

There are three ways in which our project is ideally suited to a software implementation, one of which also makes the project especially ideal for web-based software specifically.

First, as noted in post B.7 above, the fact that we are focusing on comprehension and not production means that we do not have to include tests or exercises directed at determining the accuracy of a student’s foreign-language output. Since output is inherently creative, the exact nature of output varies greatly and this makes it very difficult to test in a program. We are spared this problem entirely as we can allow a student to test him- or herself only in comprehension, for example in a multiple-choice type format.

A second point that makes it natural for us to use software is the scope and complexity of the project. Given the forty-two facets of the program, a paper version would either consist of forty-two massively redundant volumes, or seven large and unwieldy volumes with extensive internal cross-referencing. Software is the ideal medium in which to locate such multi-dimensional complexes as this one.

Finally, the fact that we are using web-based software, and making our product available on the net, means that it can be expanded and updated in the future. This could be in the form of adding more detailed grammatical information, adding texts from different genres, adding additional languages, or taking the reading courses to new levels of advanced learning.

C.7 Explain what you consider to be innovative about the project (in relation to the foreign language teaching/training of the target group(s) identifed).

There are three things which are innovative about this project. One is its scope: we are including seven Germanic languages, and allowing any of the seven to be the window into any of the other six, yielding forty-two facets in all.

Another innovative point is that this course, or rather these forty-two courses, will be focused on conferring receptive competence to the learner, rather than generative competence; that is, the focus will be on teaching learners to understand, rather than to produce, a foreign language. The model is the intercomprehension which exists among the Scandinavian countries, whereby speakers produce their own language but comprehend the others (cf. e.g. Börestam Uhlmann 1999). Initial experiments have been conducted at extending this kind of intercomprehension from Scandinavian to Dutch (Elert 1997). We want to pursue this line much further, including (virtually) all of the Germanic languages in one Sprachbund.

The third innovative factor is closely connected with the first two: we are applying the findings of the parametric analysis of comparative Germanic syntax to identify and illuminate the points of variation among the languages.

Thus, it brings the fruits of the last few decades of research in theoretical linguistics into the purview of language teaching; it represents a bridge from the theoretical to the applied. It is this third aspect that we hope will make the project interesting not only to learners but also to language researchers.

D.4 Outline your plans for piloting the development of activities/products, showing clearly who will be involved and how the work will be undertaken.

The primary division of labor is among language lines: there is (at least) one member for each language involved.

However, other specializations are also represented in the group. One sub-group consisting of Drs. Kischel, Kirchmeier-Andersen, Sigurjónsdóttir, and van Elst (the coordinator for the sub-group) will develop the master syllabus, to be used for each of the individual language courses. Another working group consisting of Professors Taraldsen, Platzack, and Haider and Drs. Svenonius and Jónsson will determine the form that the various grammars should take, based on the parametric principles of recent work in comparative Germanic syntax. Dr. Kischel takes primarily responsibility for selecting the texts to be used. Once these three decisions have been made, the identity of the texts, the form of the course syllabus, and the structure of the grammars, work can begin on developing the individual materials for each language, and on determining the structure of the program. In February, the various participants will come together at a meeting in Antwerp to coordinate efforts.

After the Antwerp meeting, Dr. Kischel will oversee the design of the computational model (along with Drs. Kirchmeier-Andersen and van Elst and the software engineer). Dr. Kirchmeier-Andersen’s primary responsibility will be in developing desiderata for vocabulary lists and online dictionaries for the project. Professor Taraldsen will plan a cognate network, a map of the different cognates present in the texts and a strategy for relating them. Dr. Sigurjónsdóttir will coordinate the vocabulary lists and the cognate network, as they emerge. Professor Haider, with input from Dr. Kischel, will develop the format for exercises to be used by the learners in the program, in accordance with the blueprints set out in the master syllabus. Professor Platzack, assisted by Drs. Jónsson and Svenonius, will further develop the structure of the grammars, in accordance with the results of the Antwerp meeting.

In May, a summary meeting will take place in Reykjavík, where the results of the different sub-projects will be presented. Decisions will be made about the final stages of the project.

After the May meeting, there are two major activities. One is the development of the computational prototype and software manual, for which Dr. Kischel and the software engineer have primary responsibility. The software manual will function partly as an instructional for the prototype but partly as a blueprint for the final software to be completed at the end of the three-year project. The other major activity is the production of materials in each of the seven languages according to the desiderata established in the first half of the year for the grammars, vocabulary lists, exercises, and overall syllabus. Here the group splits up into the seven different languages (Professor Haider taking primary responsibility for German, Dr. Kischel being occupied with the software).

D.5 Describe how the project and, in particular the quality of its products and other outcomes will be monitored and evaluated.

Testing will be undertaken systematically underway. The end product after the first year will be a prototype which could immediately be made available to the public (as a beta-version) on the web. A feedback questionnaire would be one way to help us stay abreast of bugs in the program and unwieldy aspects of the interface. Testing will intensify in the second year of the project, presumably using students at our respective universities as guinea pigs.

E. Dissemination

E Describe the arrangements planned for the publication/marketing and dissemination of results. In particular, describe how the Coordinator and project partners will contribute to the dissemination process, and mention dissemination networks which could make the materials easier to access and obtain for the target group(s) across Europe.

If funding is sufficient, we would like to make the product available for free, either for download or for direct access if a powerful enough server can be obtained. CD-ROMs could be purchased for cost by those with poor internet connections or who for whatever reason prefer to have a permanent copy of the program.

If funding requires, we are prepared to charge a modest fee for the product, including registration at the web site and for the CD-ROM and documentation. Dr. van Elst has experience financing projects through the marketing of the final product.