TERMINOLOGY USERS AND PRODUCERS


ACTORS AND WORKING CONDITIONS


Present situation

The Terms of Reference of the POINTER Project identified a number of different types of actor in the terminology field. Among the most important distinguishing characteristics are:

This matrix view helps provide an accurate representation of the complicated (or fragmented) terminology environment. However, for the purposes of examining and highlighting workflows and specific needs, it can be helpful to classify actors in the terminology field into three main categories: terminology users, terminology producers and others (e.g. teachers, researchers and tool providers). It should be remembered, though, that in practice the lines are often blurred (thus, for example, translators both consult dictionaries and glossaries and "create" or at least document their own terms).

Terminology Users


Among the major applications for which terminology

However, the most important actors who render terminological activities economically viable are those who cannot not prosper without (multilingual) terminology:

Terminology Producers


The bulk of terminology work is actually performed by people active in related professions as a part-time (or even spare-time) activity. In fact, many such "terminologists", particularly those outside large institutions, are not even aware that they are involved in terminological activities and have received no formal training in the subject.

As a result, knowledge of the methodology and procedures behind systematic terminology work is generally limited, even in large enterprises and public authorities.

In addition, since terminology work is performed right across the academic, public and private sectors, it covers an extremely large number of different applications and industries, both within and outside the language sector. This large number and diversity of actors distinguishes the profession from neighbouring fields, such as the comparatively homogeneous speech and written corpora sectors.

This situation is per se a major barrier to the creation and dissemination of reusable resources, as it complicates some of the basic actions needed to achieve co-ordinated action and market development, e.g. identifying, accessing and convincing actors.

Another characteristic of terminological activities is that terminology producers are actually the main users of their own terminology, a situation which may have consequences for quality and confidentiality (man resources are produced for "in-house use only").

Problems

Training is undoubted the major problem affecting terminology producers and users. A number of different, though interrelated, aspects need to be examined.

While translators may be sensitised to general language issues, they often have no formal training in terminology, unless they have graduated recently enough to have taken advantage of the recent integration of terminology work into translation curricula. The same applies, to an even greater extent, to technical and other writers.

In addition, language services providers - and translator-terminologists in particular - often lack the more general communications, business and computing skills necessary to perform their work successfully, and to make their case known to decision-makers, customers and the general public.

Conversely, domain specialists are almost always totally untrained in terminology, and in many cases in general linguistic issues as well. They are also often unwilling to learn since, in addition to the time this would entail, their hierarchical position is almost always considerably higher than that of terminology and/or language specialists.

Nowadays terminology work cannot be carried out effectively and economically without the use of modern language technology. In some cases, however, there is a defensive or even openly hostile attitude towards automation on the part of terminology workers, who regard computers as a threat rather than an enabling technology (this might be related to the fact that they consider themselves as professionals rather than service providers).

This problem may be aggravated further by the implementation of a terminology management system. If employees are expected, in addition to their normal duties, to populate the empty shell with terminological data, they may well experience this as an added burden. What is more, during the long period required to build up the database, they may find themselves working with a tool which does not perform the specified task satisfactorily.

Other related problems are that:

Solutions

The specific solutions proposed to counteract the lack of training and skills are dealt with in detail in Chap. 5.2: "Training".

It should, however, be borne in mind that the variety of users and creators of terminology leads to the necessity to combine a range of (initial and continued) training, as well as information and awareness actions targeted at the various groups and addressing different issues: terminology work, technical and scientific skills, modern information technology techniques and an integrated, team-oriented approach to terminology involving high-level expertise.

Beyond specific actions, the maturity of the whole terminology activity field will improve the situation in the various problem areas identified above. For example, the acceptance of computer tools will be greater and easier once these have reached a sufficient level of maturity and flexibility; the design of these tools may even favour the generalisation of the team-work approach to terminology creation.

Information on the extent of the terminology community itself is also likely to contribute to the appropriate development of the profession towards a better status recognition.

Recommendations

NATIONAL AND REGIONAL ASPECTS OF TERMINOLOGY WORK


Present situation

The national surveys conducted during the course of the POINTER Project show a great variety in the organisation of terminological activities, not only between countries, but also between economic sectors and regions of the same countries, with notable disparities particularly in the commercial availability of terminology. In spite of these variations, the following common factors should be highlighted.

Firstly, some degree of infrastructure for terminology activities exists in almost all European countries, ranging from associations for translators, technical writers or even specifically for terminologists, to public bodies active in terminology work or promotion, and to courses in terminology offered in university, technical and whole-of-life education contexts. Equally, the private sector gives many examples of some form of terminology work being performed.

Secondly, it appears that, although a fair number of such institutions or enterprises exist, very few produce exchangeable and/or saleable terminology. Whilst a terminology services industry is certainly emerging, it is still underdeveloped to almost invisible, even in the most terminologically aware countries. Specialised dictionaries are indeed produced, but generally by publishing houses for commercial reasons and with minimal liaison with the existing embryonic infrastructure, and with virtually no feedback, other than sales, from the potential users.

Thirdly, it would seem that in most countries, very little co-ordination takes place between those engaged in terminological activities, even within a given field of activity or branch of terminology, resulting in duplication and wasted efforts. Not only are the (full-time or part-time) terminologists in industry likely to have little contact with others working in the same or similar fields, but contacts with subject specialists are also often problematic. Those "terminologists" who are in fact reassigned translators (and who sometimes still perform translating activities) are usually in direct contact only with other translators, usually in the same organisation. Across subject fields, and in different sectors, the problem is naturally compounded, so that potential users of terminology such as documentalists, archivists, administrators, trainers and workers in artificial intelligence, not to mention lexicographers and computational linguists, are often not even aware that resources are available.

The reasons for this lack of communication also vary according to the country, but once again, certain common features run through the surveys. There is a general fragmentation of creation and distribution mechanisms at institutional, sector/industry and national levels. Such paper dictionaries as are published have to be sufficiently general in scope to attract a broad public, leaving the more problematical and more numerous specialist areas without adequate printed terminological documentation or with none at all. Similarly the languages used tend to be those of international communication. Term banks tend to be small in size, possibly highly specialised, but difficult to access. These difficulties are compounded by considerations of confidentiality and uncertainty about the legal position as regards the copyright of terminology. This lack of co-ordination and concertedness in terminology work is also the direct result of poor funding in both the private and public sectors, though there are notable industrial exceptions in this respect. The current recession has in fact called a halt to recent expansion of terminological activities in some industries, and some terminological and even translation units, set up in the late 1980's and early 1990's, have been closed down. The economic downturn has, in these cases, sent terminology back to its previous position as a part-time, low-priority activity, ever vulnerable to cutbacks in times of crisis.

Terminology work, just like translation work, is now being outsourced more and more frequently, posing new problems of access to validated sources and restating old problems of reusing terminological work already done. In addition, due to time and cost pressures, much of this terminology work now consists of the compilation of simple word lists, with all the inherent quality problems that this entails (this also applies to much in-house work). Moreover, co-operation is hindered on the one hand by the fact that many of these translators and terminologists are either freelance or working in agencies, and, on the other hand, that their terminology is often (mistakenly) seen as a jealously guarded professional asset.

The current downgrading of certain of these activities can be partly ascribed to the lack of awareness on the part of decision and policy makers of the benefits which terminology can potentially bring. The role which terminology can play commercially and strategically is still not sufficiently grasped at upper management level. The added value of terminology work is not seen, as it is perceived as outside the scope of their specific range of corporate activities, and so there is little motivation at this level for a terminology network or even more local solutions.

The need for terminology is, however, increasingly being felt at other levels, by the producers and users themselves, and in academic and political circles, especially in the context of increased European contact. Where terminology is controlled and consistently used as an interface in businesses or organisations, the result is high-quality communication. The POINTER national surveys show different organisational models for terminology both within the company, between various departments, and with regard to external partners, especially in the export field. Whatever the precise form of organisation, however, the important factor is that terminology is actually used in the multilingual communication process.

Terminology creators are also increasingly aware of the existence of tools which can help them in their task, and realise what could be done if sufficient means were forthcoming. Moreover, the role of terminology in information extraction and knowledge transfer is becoming increasingly evident, although all the consequences are far from having been drawn. Allied with this is an increasing demand for expert consultancy in the creation of terminological resources, and in the use of tools for this.

The differences observed in the different European countries can be partly ascribed to the role which multilingualism plays in each society. Thus, countries which have traditionally and officially been monolingual show a low level of terminology awareness, and thus of terminology production, even in the national language. This is the case of course in the United Kingdom, but also to a lesser extent in Spain, where Castilian Spanish with its status as a world language was long neglected with respect to terminology work in modern fields (however, in the latter case close collaboration between all Spanish speaking countries e.g. via Union Latine considerably enhanced - and enhances - overall terminological activities in the language). Officially multilingual countries have well-developed terminology networks spanning the language divide. This is the case of the two European countries with a long tradition of multilingualism, Switzerland and Belgium, and it should be underlined that in these two countries networks are to be found in both the public sector and the private sector.

Terminology takes on a critical importance where it serves as a tool in language planning, usually in smaller language communities such as Catalan or Basque. In very small language communities, however, and in those without access to resources, these activities tend to be carried out by enthusiasts and have very little echo in the rest of the community. Where a political will has the structures to express itself, as in autonomous regions of Spain, terminology activities are systematic, efficiently disseminated and carried out with the co-operation of many sections of the community, including the business sector. Indeed, much of the current terminological activity which can be observed in the rest of Spain today can be traced back to the model provided in Catalonia, especially in the key field of training. The Nordic countries, each of which has its own small language community or communities, constitute another interesting example, where cross-border co-operation through Nordterm has considerably enhanced terminology activity throughout the region.

France and Germany are somewhat anomalous in this respect. Though both are large, generally monolingual countries, their terminology activities are better developed than in, say, the United Kingdom or Italy. In the case of France, the role of official policy for terminology has been preponderant. The Délégation Générale de la Langue Française, as well as its predecessors, has striven to co-ordinate terminological activities in France, with resulting high awareness of terminology and a relatively high level of training. Support from international French-language terminological networks has also significantly furthered the cause, as is also the case for the Dutch-language association. In Germany, it has not been any explicit national language policy but rather perhaps the tradition of exporting quality goods which has given the main impetus to terminological work, though the influence of the Deutscher Terminologie-Tag e.V., the Bundesverband der Dolmetscher und Übersetzer (BDÜ) e.V. and the Deutsches Institut für Terminologie (DIT) e.V., and of Vienna-based Infoterm, should also be mentioned.

Among the most encouraging indications to come out of the POINTER survey is the growing willingness on the part of translators and terminologists to take part in distribution networks. A much repeated prerequisite, however, is that the network be comprised of and dedicated to the users themselves. Such a network would have to come to grips with the question of confidentiality mentioned above, and work in a clear-cut legal context (for a more detailed discussion of this point, cf. Chap. 4.3: "Terminology Distribution and Exchange").

Problems

In spite of the voluminous inventories of terminology resources which POINTER has catalogued in many countries (and which are not only confined to the major languages), users consider existing resources as largely inadequate, since they are not specialised enough, are insufficiently elaborated, are not up-to-date, are insufficiently validated, are not available in the languages required and are not publicised if they are. Outside the "big four" languages (English, French, German and Spanish - in practice often reduced to the "big two" - i.e. English and the national language), provision of reliable terminology is haphazard at best, and the imbalance between languages is tending to grow. Thus in practice, English has become the lead language for terminology work since the Second World War (although at an international level a multilingual approach is often adopted from the outset). Moreover, in the lesser-used languages translation usually takes place with completely inadequate dictionaries, often using English as a "transfer language". The resulting errors can be widespread and far-reaching. The subject areas in which terminology is supplied also tend to be narrow, concentrating on UDC 6 (applied science and technology, medicine), followed by UDC 3 (social sciences, economics, trade and commerce), with special emphasis put on such fields where international co-operation has forced terminology work to be done, such as in aeronautics and in the telecommunications industry. Even in English, there is room for improvement in coverage, specificity and contemporary relevance.

Terminology resources in printed form date rapidly; it not only takes a considerable time for the books to be printed, but once they are printed, editors show some reluctance in regular updating. Editors of printed dictionaries are also somewhat coy about clearly indicating their validation methods, with the result that the users may be in doubt about the reliability of their data. Compared with general language dictionaries, it is not even easy to locate existing specialised dictionaries, as the inventories which exist tend to be poorly distributed, and editors' catalogues are difficult to consult, as a very high proportion of specialised dictionaries are published by a multitude of firms active in a very wide variety of domains. This means that even if a user knows of a dictionary on a desired subject, it is not always easy to obtain it. The paradox is that the more detailed, specialised and up-to-date terminological data is, the more difficult it is to locate, as it tends to be grey literature (e.g. glossaries at the end of theses, technical reports, or internal documents).

More and more printed dictionaries are coming out in electronic form, such as CD-ROMs, which can be more easily worked into a word processor. However, these dictionaries are far from having all the information, both linguistic and factual, to fulfil the functions that are required of them, in particular in view of computer-assisted translation.

The result is, that, throughout Europe, translators, professional and casual, are making do with inadequate terminological resources, especially when they are outside their own specialism, either because they do not know that better resources exist, or because they are too difficult to find. While a comparison with twenty or thirty years ago would show considerable progress, the importance of technical terminology to a wide number of disciplines and the speed with which knowledge is being invented or revised mean the problem remains extremely serious.

Apart from published dictionaries, little terminological work is reused. Technical standards, such as those published by ISO and national organisations (DIN, AFNOR, etc.), which should be of direct help to terminologists and translators, are often unknown or ignored as irrelevant or prohibitively expensive. Most terminology produced in an organisation is prepared for internal rather than external use, and outside users would have difficulty in assessing its validity.

Those questioned about the desirability of a distribution network, though generally expressing a very favourable opinion, voiced the following fears. Firstly, they were uncertain as to their legal position and that of their company; concern was expressed about copyright, intellectual property rights and liability. Fears exist too about confidentiality and competitive advantage which may be given unwittingly in divulging company terminology, though these appear to be declining. From a practical point of view, it is far from obvious that the terminology produced in one company is fit for external consumption, and a selection would have to be made before dissemination. Over and above initial selection, a method of maintenance would have to be found, and agreement reached on quality and validation procedures. Finally, from an economic point of view, costing and pricing mechanisms would have to be worked out. In their absence, much of the goodwill which is evident in many sectors is condemned to inertia.

Solutions and Recommendations

The solutions which can be envisaged for the problems mentioned above are directly incorporated into the recommendations below. A special case should be made for training, however, as facilities vary greatly from one country to another. All improvements in terminology in general imply the participation of trained personnel, not only translators and terminologists themselves, but also subject experts, standardisers, documentalists, etc. The recommendations for a modular training system and other necessary training measures are outlined in Chap. 5.2: "Training", and Appendix 5: "The POINTER Training Model".

In several countries, national centres for terminology exist already, and these should be strengthened when they work well. Where there is no focal point, one should be created if at all possible, or if not, this new role should be assumed by existing professional associations and networks. In some cases, this will entail addressing or readdressing the thorny problem of language policy in general; a concerted European initiative here, involving national governments, the European Commission, and organisations from the field would be one way of proceeding here. The specific functions that national terminology centres should assume are listed in Chap. 6.2: "The European Terminology Infrastructure - Main Recommendations"; they include the provision of information and brokerage .services, co-ordination and liaison, validation, training, standards promotion, lobbying, research on demand and activity monitoring. As with the measures proposed for the European infrastructure, it should be clear that such national centres can only work successfully in co-operation with others, both within the individual countries concerned and abroad.

In addition, general language (LGP) networks and networks for related languages and language groups already exist in most cases; far more use could be made of them, both in a pan-European context, and on a local level. In all cases, these organisations should be induced to work in close co-operation with ELRA, which, in turn, should be in a position to assist especially in the area of distribution. Broad interest groups associated with terminology, and professional bodies and chambers of commerce, should also be linked closely with these focal points. One of the prime tasks of these focal points should be advice on legal issues and help with commercialising existing terminology. In addition, it is necessary to elaborate guidelines for costing and pricing terminology work as explained in Chap. 2.2: "Economic Aspects of Terminology and Terminology Work".

Networking should be encouraged both between national and regional organisations, and between the private and public sectors. Full use of WWW servers should be made, as there are already very promising terminology services being set up. Sector servers can play a most useful role here in the dissemination of much-needed information. The term banks identified in the POINTER inventory should be encouraged to participate in such an initiative, and mechanisms should be implemented to discuss how this goal can be achieved in collaboration with all interested parties.

The internationalisation which is emerging with the ever-increasing use of the Internet has made companies aware of the huge needs in technical translation and writing in order to exploit to the full the potentialities of this network. This is particularly the case with the emergence of new markets in Eastern Europe and in Asia and Latin America. The servers set up should therefore be particularly attentive to these needs.

Promotion of awareness about terminology should be undertaken among decision and policy makers in both private and public sectors, through the focal points, but without neglecting traditional vectors, such as professional associations, translators associations, etc. It is particularly important that the economic interest of terminology be made clear, its potential in gaining productivity and market share, though it is just as necessary to explain how this potential can best be exploited. It should not be forgotten that the choice and coherence of corporate terminology improve the efficiency of communication within the company and foster a positive image to the outside world. Pilot projects in strategic and highly visible areas such as environmental protection, if possible in collaboration with Eastern Europe, should be supported as a test of the efficiency of terminology.

Another sector which should be sensitised to the needs of terminology is that of publishers of specialised works. If these could be persuaded to improve the quality of their indexes, it would enable the automatic generation of context information, which could then be incorporated into term banks and electronic dictionaries. Similarly, indexes from theses should be systematically exploited in the same way. The whole question of the incorporation of term records into electronic dictionaries merits further study, as this important interface has so far received relatively little attention and merits further funding.

When no national resources exist, such as machine-readable dictionaries in general and special languages, there is a good case for these to be created with some degree of public funding, either seed money, partial subsidy or full funding, as appropriate. More generally, some mechanism has to be found to correct the imbalance which exists between countries with large internal markets and export possibilities, and those other countries whose language is not spoken beyond their own borders and whose home market is far too small to be adequately covered by purely commercial ventures. Care should be taken, however, not to distort the market and to focus on providing support for creation, rather than the direct provision of resources.

Given the need to recycle the mass of underused terminology that should be made more generally available, it will be necessary to elaborate precise validation techniques and choose the tools to go with them. This means that attention should also be paid to validation of terminology tools in general, with special emphasis laid on updating existing work. In addition, the other recommendations made in Chap 4.3: "Distribution and Exchange" should be implemented.

EUROPEAN ASPECTS OF TERMINOLOGY WORK


Introduction

Given the brief of POINTER to produce recommendations for a future European terminology infrastructure, the question of what structures and activities currently exist at this level assumes a special significance. In fact, there are many hundreds of European-level institutions and organisations which are producers, holders and users of terminology, both within the official framework of the European Union and outside it. This chapter provides a brief overview of a few of the major players, and their activities and resources, concentrating on those which are particularly relevant to the proposed terminology infrastructure.

EU Institutions

General Comments


The European Union is one of the major users of translation and terminology in the world. This is partly due to the fact that its legislation is directly applicable in member states, and therefore has to be made available in all official working languages. As a result, staff translators at the European Commission currently produce more than 1 million pages per year, and deal with at least 6 to 7 million terms in the process (on average there are 8 or 9 problematic terms per page). The European Union therefore has a vested interest in promoting the provision and use within its own institutions of efficient, high-quality resources and solutions. As the sections below show, the institutions concerned have invested a great deal of time and effort in building up such applications for internal use. In some cases, these have the potential to serve both as role models for projects outside the confines of the EU institutions and, subject to the resolution of a number of open issues, as a basis for redistribution to a wider audience.

European Commission Terminology Unit


The Terminology Unit is a horizontal service covering both Brussels and Luxembourg and is designed to provide linguistic support in all official languages of the European Union. The main services it offers are the EURODICAUTOM database, help desk services for translators and other officials, research geared towards the production of specialist glossaries, and support measures such as the preparation of translation-oriented terminology resources and the development of computerised methods of terminology work. In addition, staff provide advice on terminological issues for the Commission as a whole, and collaborate with the terminology services in other EU institutions.

As regards the specialist glossaries, over 50 works, many of them in nine languages, have been produced, and a number have been made available outside the Commission. Subject fields range from major treaties such as those of Maastricht and Rome, economic and administrative issues such as VAT and budget terminology, and innovative or central areas of science and technology, such as plasma physics, biotechnology and mining). Quite apart from their role as multilingual resources and as sources of terminology in innovative areas, these glossaries are important in that they frequently document what has been called the "Eurolects"(2), i.e. words and phrases which have their origin within the European Union and for which national equivalents do not exist.

EURODICAUTOM

At the heart of the Commission's terminology work is its multilingual termbank, EURODICAUTOM, which was designed both for use within the Commission and as an inter-institutional service. Started in 1973, it is multilingual and multidirectional (i.e. any language can be combined with any other). According to Commission sources, the termbank at present contains over 630,000 concepts (both terms and phrases) and 200,000 abbreviations and acronyms in the nine official languages valid before January 1995 (although not all terms exist in all languages). In addition, Finnish and Swedish are now being added following their countries' accession to the European Union. The database contains phrases and terms which are EU-specific and have a legal or normative character, together with a large amount of domain-specific terminology. A large number of glossaries, reference works, and standards have been incorporated into the database.

EURODICAUTOM is widely used within the Commission, and the contents are also made available to external users via the Commission's host service, ECHO (the total average number of queries in 1995 was slightly over 4,500 per month, while the monthly ECHO connection hours for the period January to November 1995 ranged from 1147 to 1994)(3). In addition, the service is available via a (currently unofficial and therefore unsupported) World Wide Web site (http://www.uni-frankfurt.de/~kurlanda/eurodicautom.html); an official ECHO Web site is planned for 1997. Selected parts of EURODICAUTOM (some 268 kB of data) have also been published on CD-ROM as part of the TERM-DOK project conducted with TNC; other CD-ROM projects are currently under discussion.

EURAMIS

The goal of the EURAMIS (European Advanced Multilingual Information Services) project is to create a uniform client-server environment making a suite of batch-oriented language and information management applications available to users throughout the Commission (i.e. not just language staff). The project is being run jointly by the European Commission's Translation Service and DG XIII, and is financed out of the Commission's Multilingual Action Plan (MLAP) programme. At the heart of EURAMIS is the Linguistic Resource Database (LRD), an object-oriented database containing a translation memory, termbases and machine translation dictionaries. A number of different databases and views will be provided to cater to the different types of users and their needs (e.g. general users, working groups, individual users). The basic structure of the LRD is built around multilingual aligned texts containing parallel descriptions for the different languages (so-called linguistic objects, or LOs). Depending on the use to which it is to be put and hence the database it comes from, an LO may contain either individual terms or phrases, or complete sentences or paragraphs.

As regards terminology in particular, EURAMIS plans to offer a new, more user-friendly batch access environment for EURODICAUTOM, and possibly for other major databases later on. In addition, a customised terminology tool will support the various different user groups, offering menu-driven input (thus reducing mistakes), maintenance tools, output filters, retrieval profiles and functionality for restricting access. Over time, it is planned to provide automatic creation of text-related glossaries in all official languages, as opposed to merely those source languages currently served by the Commission's machine translation system.

Council of Ministers


The Terminology Service of the Council of Ministers provides assistance to the Council's 500+ translators, who mainly translate working documents (often at short notice) for the many working parties which prepare Council decisions. Terminology work at the Council is strongly oriented towards achieving consistency between and the legal equivalence of versions of texts in different languages. Subjects range over a wide area, from agriculture to nuclear technology.

Much of the terminology produced is available in TIS, the Council's "Terminological Information Service". This internal database contains c. 170,000 entries in up to 13 languages (including Irish, Latin, and the new EU languages Finnish and Swedish), plus c. 10,000 abbreviations. All necessary alphabets, accents and diacritics are used. In the default setting, users consult the computer on a terminology problem in a specific language and are provided with the translation in their own language. Where multiple responses are found, these are weighted according to relevance to the users in question. Translators can search the entire database using Boolean operators and right-hand truncation.

A large team of terminologists (i.e. translators assigned to terminology work) add to and modify the database contents on-line as they research problems for their translator colleagues. Batch updates /indexing, archiving, etc.) are performed at night. Terminologists are also provided with enhanced search capabilities and facilities for making lists. It is planned to migrate the database to a more modern platform in 1997.

Terminology from the Council of Ministers is also made available to EURODICAUTOM.

European Parliament


The European Parliament's terminology database, Euterpe ("Exploitation unifiée de la terminologie au PE"), is a Multiterm-based application designed to provide the Parliament's c. 450 translators with terminology they need for their daily work which cannot be found in EURODICAUTOM. As such, it relies heavily on the analysis of the Official Journal of the European Communities (OJ), plus thematic areas such as AIDS, educational systems, and government institutions at all levels. In addition to providing an integrated environment for staff translators working with text processors, the terminology staff also use the database to produce multilingual and bilingual paper glossaries for use by other groups within the European Parliament, by other institutions, and by freelances. In-house staff also have on-line access to EURODICAUTOM data. Updates of selected sections of the database, which now contains some 40,000 entries in total, are transferred to EURODICAUTOM once a month(4).

European Court of Justice


The European Court of Justice also provides terminological support for translators and other staff members. The JUDIT database contains the Court's legal terminology (both individual words and phrases), including many terms reflecting internal structures and procedures. It has been created by the Council's terminologists working in collaboration with linguistically trained/aware lawyers in the relevant divisions. Where appropriate national equivalents for particular terms may also be given. The multilingual THECLA thesaurus offers translations in up to nine languages of terms from a wide variety of areas which frequently occur in Court material.

Problems


EURODICAUTOM

Although EURODICAUTOM has a relatively large base of regular and satisfied users within the European institutions, reception outside it has been mixed. In particular, the following criticisms have been levied:

General

At a more general level, it would seem that co-operation between the institutions of the European Union, and between them and other European bodies, could be improved. This is, in fact, the objective of recent inter-institutional discussions, which have resulted in a decision in favour of a single database for all institutions (a project that will probably take several years to implement).

Equally, mechanisms for co-operation between the European Union and external instances such as international, regional and national networks, governmental and commercial organisations and projects could be expanded or improved in many cases (cf. Chap 3.4: "International Aspects of Terminology Work" for some concrete examples). Were this to be achieved, much greater use could be made of the opportunities for exchanging and improving terminology, especially in those areas which are outside the Commission's main areas of activity.

Above and beyond this, knowledge of and access to CEC and other institutional resources, although growing, is poor. Given the size of the resources offered by EURODICAUTOM and other databases and glossaries, and the general lack of terminological material outside, this is unfortunate.

Solutions


The status of Europe as a leading multilingual world power, and the pressing need for a multilingual information society have been established beyond doubt. As the prime motor for and architect of the new Europe (a role which is hardly likely to decrease in future), the Commission can and should act as a catalyst in this area. Any proactive language resources policy for Europe must also be extended to cover the EU's internal resources and applications, which should act both as "pump primers" and as paradigms for the new European information society. This clearly includes the facilitation of widespread access within the short to medium term of at least a part of its current resources. Of course, care should be taken to ensure that this does not lead to market distortions, and in particular to the stifling of the new language engineering and language resources industries that the Commission wishes to see established. However, given the size of the problem facing us and the extremely short timeframe in which solutions have to be developed and implemented, such help should be accepted as necessary.

Since EURODICAUTOM is a multilingual tool with considerable (although by no means uniform) merit, it has a certain role to play in the creation of a European terminology infrastructure if modified appropriately. A decision by the Commission to promote unrestricted, efficient and well-publicised access to EURODICAUTOM data would provide a solid operation base for the rapid dissemination of considerable terminological resources. In this context, it is important to distinguish between the various types of data currently contained in the database, with the clear emphasis being on official European Union terminology (e.g. legal and official terms). In other areas, market forces can be expected to operate more readily - indeed, the fact that many subsets of EURODICAUTOM come from external published sources is proof of this.

As we have seen, to achieve such a goal considerable consolidation and validation of EURODICAUTOM data is necessary. In the long term, this could be achieved domain by domain, by making reference to other standard or high-quality data and with the help of domain specialists and terminologists. In addition, the reuse/integration into EURODICAUTOM of those high-quality resources held by other European bodies which have not already been integrated should be regarded as a priority.

The result could be a new CD-ROM resource collection (or domain-specific collections), a comparatively easy and inexpensive task (the structure of EURODICAUTOM is such that information can be classified according to domain or sector). Institutions such as ELRA could then undertake its dissemination, in whole or in part, to a wide audience. Care should be taken to market any such CD-ROM at an affordable price, especially as the size of the potential market means that a commercially attractive number of copies could be pressed, and the data concerned are in essence in the public domain. If this strategy were to be adopted, internal Commission users could be provided with more reliable data, while external users would have access to reliable EU terminology. In the mean time, EURODICAUTOM could still be disseminated as at present and in its present state and undergo gradual improvements.

In addition, once consolidation and validation work has been completed, the generic terminology and multidisciplinary concepts it contains could be isolated and collated with a view to compiling a multilingual sub-base of "general" terminology for use by a wider public. Such a sub-base could also serve as the starting point for the preparation of encyclopaedic bases. Here, too, the cost/benefit analysis for such a project is likely to be positive, while the end product would be likely to have considerable impact on consumers. Not only do specialists such as translators, editors, information and documentation staff and domain specialists require high-quality terminological services, but the general public can also gain considerable benefit from accessing more general information produced using the same terminological approach. Against the backdrop of the free circulation of workers, goods and technologies in Europe, EURODICAUTOM could be a useful source of information e.g. for tourism, to increase understanding of technical brochures, and for use in university and school education. The role of the Commission as the provider of such generic resources should, however, be closely defined to avoid competition with the emerging terminology and language resources market.

In addition, the terminological information contained in EURODICAUTOM but derived from standardisation bodies and other authoritative sources should undergo methodological consolidation to facilitate the process of adding equivalents in other languages (often a difficult task in the originals). At present, these bodies produce high-quality data which cannot always be expanded to create multilingual versions, due to a lack of the minimum information required. Such projects can, of course, only be performed with the consent of and in conjunction with the bodies concerned.

Furthermore, it should not be forgotten that Europe is also oriented towards other parts of the world such as Eastern Europe, Japan, the Middle East, the CIS, China, and South-America. With the implementation of forthcoming European or global telematics networks, EURODICAUTOM can potentially play a leading role in communications in general. For this reason, the Commission needs to actively encourage its consolidation and promotion. This should not be regarded from the point of view of a commercial operation (although the costs could be easily recouped if EURODICAUTOM were to be accessed by a huge number of users) but rather as a public service.

Recommendations


To sum up, the institutions of the European Union should continue a deliberate and proactive policy on internal resources and activities, covering the following major points:

Other European Institutions

Present Situation


There are a large number of other European institutions and projects, both within the ambit of the European Union and independent of it, which are involved in terminological activities of different kinds.

Many of these either produce terminology for their own internal use or for their target audiences, and/or are involved in other aspects of terminological activity such as training, harmonisation, and quality assurance and validation. Increasingly, informal and to some extent formal links are now being created (especially within the individual categories) which could serve as the basis for terminological co-operation and exchange.

The following "mixed bag" of institutions is an attempt to illustrate the diversity of potential activities, sources and contacts.

CEN/CENELEC

CEN (Comité Européen de Normalisation), the general sector European standardisation body, is a superstructure uniting the standards bodies of the 15 members of the European Union, Iceland, Norway and Switzerland, and 11 affiliates from Central, Eastern and Southern Europe. Its sister organisation CENELEC (Comité Européen de Normalisation Electrotechnique) is responsible for the electronics area, while ETSI (European Telecommunications Standards Association) covers the telecommunications sector.

The aim of these standardisation bodies is the harmonisation of standards at a European level as a means of guaranteeing the free exchange of goods and services by removing any technical barriers to trade. European Commission funding is used to translate national standards: nearly 1,500 trilingual (i.e. English, French and German) technology standards have been published, many of which contain terminology. In addition, roughly 40 terminology standards have been published(5). The work is done by technical committees (TCs) comprising representatives of the relevant national standardisation committees. Member states are obliged to take over CEN standards once adopted, either in one of the three original languages or in translation into their national language. This obligation is leading to a sharp reversal in language workflows within the national standards bodies, as drafts are now increasingly created in the CEN languages, and in English in particular.

European Investment Bank (EIB)

Founded in 1958 under the Treaty of Rome as an autonomous public institution within the then European Community, the European Investment Bank (EIB) is the long-term financing arm of the Commission, deploying banking resources to help finance capital projects throughout the Union, with particular emphasis on reducing inequalities between the regions. It also provides funding for projects in non-member countries with which the Union has concluded economic and financial co-operation agreements (e.g. Mediterranean countries and the African, Caribbean and Pacific signatories to the Lomé Conventions). As such, it is an important relay point in the transfer of knowledge and technology to these areas.

The official working languages of the EIB are English and French, with German also being used in particular for reports to the Board of Directors. Terminology work is concentrated on these three languages. The EIB's glossaries are used by the in-house translation department, by other Bank staff (e.g. rapporteurs) and by external service providers. Examples of such areas are capital markets, development banking and technical assistance projects covering a wide variety of industrial and infrastructure sectors. The integration of domain specialists in the terminology creation and documentation process means that the results produced are generally of a very high quality. These glossaries are published in paper form and are also available on-line in-house. There are no plans to release the terminology for general use. Some elements of the glossaries are regarded as confidential and would have to be removed before wider publication.

European Environmental Agency (EEA)

The European Environmental Agency was founded in 1993 with the aim of providing the Commission and the member states of the European Union with objective, reliable and comparable information on problems in the environmental area, thus allowing them to implement the necessary measures.

Terminology work at the EEA is performed by two terminologists with a strong international approach. A multilingual thesaurus covering English, Spanish, German, Austrian, Dutch and Italian terms is currently under development, and this could serve as the basis for the development of a standardised multilingual terminology. However, work is currently impeded by the absence of standards for the exchange of linguistic and terminological information within the working group, and by an absence of consensus on common storage media.

CEDEFOP

CEDEFOP, the European Centre for the Promotion of Vocational Training (in French: Centre européen pour le développement de la formation professionnelle) has prototyped the use of terminology as a cost-saving and quality assurance device in European harmonisation projects, and in conference terminology, project management and database links in particular. In a project running from 1985 - 1993 on "Equivalent Professional Qualifications", 19 sectors comprising roughly 200 job classifications for skilled manual and clerical workers were examined by a multinational committee of employers' and trade union representatives. The use of conference terminologists for this project reduced total costs by one half, since the creation and documentation from the beginning of readily accessible, agreed multilingual terminology greatly increased the transparency of the written and verbal communication processes.

Terminology Training under ERASMUS

A recent initiative to co-ordinate terminology training in higher education across a number of European universities has been undertaken under the auspices of the ERASMUS programme, led by the École d'Interprètes Internationaux, Université de Mons-Hainaut, Belgium (IPC-95-B-4022/09). The network currently involves 15 universities across 9 countries (Universität Hildesheim; Leopold-Franzens-Universität Innsbruck; Friedrich Schiller-Universität Jena; Fachhochschule Köln; Facultad de Traducción e Interpretación, Las Palmas de Gran Canaria; Université de Maine; Universitade de Lisboa; Université de Mons-Hainaut; Katholieke Universiteit Nijmegen; Université de Paris X; Université de Haute Bretagne Rennes 2; University of Surrey; University of Manchester Institute of Science and Technology; Universiteit Utrecht; University of Vaasa).

The network aims to exchange students for periods of four to eight months, enabling students to gather 150 hours of accredited taught courses (Unités de valeur) comprising theoretical and practical work, including the compilation of a 100-term entry computerised terminology. The courses may be followed at the student's home institution or the exchange institution. In addition, it is envisaged that each student who wishes to gain the European Certificate of Terminology will complete a three-month period of practical training at a recognised organisation carrying out terminology work (one such organisation being the European Parliament).

Terminology Organisations and Networks

Several supranational terminology- and language-based networks (e.g. Nordterm, Rat für Deutschsprachige Terminologie, REALITER, RITerm, Rint and LISA - the Localisation Industry Standards Organisation) are currently operating at least partly within Europe. These are described in more detail in Chap 3.4: "International Aspects of Terminology Work".

Professional and Industry Associations

Last but not least, there are a number of European umbrella organisations and professional and industry associations which are potential sources of terminology (e.g. the European Trade Union Confederation, or the European Confederation of the Iron and Steel Industry).

Problems


The list of organisations and activities given serves to illustrate the vast potential at this level. However, as is the case at other levels of activity, there is a considerable lack of information and awareness, both within the institutions concerned and on the part of the outside world, as to the existence and value of their terminological resources.

As regards standardisation in Europe in particular, one problem is the number of different and sometimes even conflicting committees, which lead to duplication of effort and inconsistencies. In addition, the preponderance of domain experts without experience in the methods and practices of terminology work can lead to suboptimal work.

Solutions and Recommendations


Awareness campaigns should be started with the aim of encouraging:

INTERNATIONAL ASPECTS OF TERMINOLOGY WORK


Present situation

Introduction


Any examination of the issues needed to ensure improved co-operation in the field of terminology within Europe would be incomplete without an analysis of the various types of organisation and network already working at a supranational level. Not only do they represent a strong nucleus of existing activities which must be integrated if the proposed terminology infrastructure is to be a success, but they also offer valuable lessons from which to learn.

In practice, the following different categories of organisation and network can be established:

These categories are discussed in more detail in the sections below.

Terminology in International Non-Governmental Organisations


There are hundreds of international institutions engaged in terminological activities, and in resource creation, standardisation and harmonisation in particular, in nearly all domains (cf. [Krom 85], [Hoff et al 96]). The work performed covers the entire spectrum of descriptive and prescriptive approaches, from the publication of terminology which is obligatory throughout the world (e.g. the international aviation terminology produced by the International Civil Aviation Organisation - ICAO) to purely academic description and comparison (e.g. INTERCOCTA, cf. "Descriptive Activities" below). While any selection from such a vast number must necessarily be subjective, the following pages attempt to give a flavour of some of these activities.

Standardisation Bodies
International Electrotechnical Commission

The International Electrotechnical Commission (IEC) in Geneva has been active for many decades in the standardisation of electrotechnical terminology (in fact, its first technical committee, founded in 1908, dealt with terminology). TC1 has the job of standardising and co-ordinating the terminology used by other IEC technical committees and to produce the multilingual International Electrotechnical Vocabulary (IEV). The second edition, which appeared in 1992, runs to 4,500 pages, including alphabetical indexes.

International Organisation for Standardisation

The Geneva-based International Organisation for Standardisation (ISO) performs the following three activities:

For further discussion of the role of standardisation in terminology, cf. Chap. 4.4: "Standardisation".

Quasi-standardisation Work

In addition to ISO and the IEC, many hundreds of organisations world-wide are engaged in the production of terminology which is accepted (willingly or unwillingly) as authoritative by the relevant expert communities. One example is the International Union of Pure and Applied Chemistry (IUPAC), whose numerous commissions prepare nomenclatures, definitions and terms in accordance with the IUPAC guidelines. The resulting chemical terminology is internationally binding for English. A specific system of naming principles and rules also governs the creation of any new chemical term. The systematic terminology systems thus created are also known as chemical nomenclatures.

Other examples of international organisations producing terminology are the International Commission on Glass, with its multilingual "Dictionary of Glassmaking", the International Institute of Refrigeration (IIR), and the International Telecommunication Union (ITU), which produces a database (TERMITE) containing some 59,000 entries in English, French, Spanish and, in a few cases, other languages. Equally, the World Meteorological Organisation published a multilingual vocabulary of over 3,500 terms in English, French, Spanish and Russian in 1992.

Harmonisation

Here again, a large number of organisations and groups are involved in the harmonisation of terminology produced by themselves or others. One example of such work is the internationally harmonised customs terminology, which fulfils an important function in international goods traffic.

A second is the RailLex terminology project involving 29 national railway organisations under the umbrella of the International Union of Railways. The group has produced a CD-ROM, RAILLEXIC, which contains c. 17,000 entries in 11 languages, an information database on railways standards and a number of railway thesauri. Another 9 languages are currently being added.

A third example of harmonisation is the "Incoterms" (i.e. the international rules for the interpretation of trade terms) published by the International Chamber of Commerce (ICC) since 1953. This multilingual terminology, which includes highly specialised set phrases, is the result of a laborious harmonisation process among many countries. However, as business is now increasingly being conducted on-line, the importance of Incoterms is declining at the expense of such standardised systems and EDIFACT and SWIFT (cf. Chap. 4.4: "Standardisation").

Descriptive Activities

In contrast to the natural and exact sciences, the humanities and the social sciences exhibit a low level of standardisation, including that of terminology (jurisprudence and legal practice are a partial - but only a partial - exception to this rule). The reason for this is that these areas are largely characterised by monocultural traditions and specific features, and by pluralistic methods and theories. The primary objective in this area should be to increase the transparency of specialised texts by means of an explanation of the terms and expressions used and by creating descriptive-comparative terminology collections.

One example of work already being performed in this areas is the Committee on Conceptual and Terminological Analysis of the International Social Science Council (INTERCOCTA). This body promotes the creation of glossaries of the specific conceptualisations achieved during theoretical and empirical social science work. It has also developed special methods of descriptive terminology work in the social sciences, and successfully tested them in concrete work on ethnicity and political science terminologies. These methods have also been made available in a form adapted to the needs of computerised co-operative terminology work.

International Terminology Organisations


In addition to ISO/TC 37, there are four organisations whose brief covers the promotion of and provision of information on terminological activities at an unrestrictedly international level (as opposed to what may be described as regional, supranational or language-based concepts):

The following sections examine these organisations, whose headquarters and leadership are in practice quite heavily oriented towards the German-speaking world, in more detail:

Infoterm

Infoterm was founded in 1971 on the basis of a contract between UNESCO and the Austrian Standards Institute (in association with what was then the Federal Ministry of Construction and Technology and the Federal Chamber of Commerce). As the international clearing house and referral centre for terminology, it is a neutral, non-commercial institution that aims to provide information, consultancy and training on all aspects of general importance to terminology at large. Its clientele comprises governmental and other public institutions, international and regional bodies specialising in terminology, and intergovernmental organisations.

At present, Infoterm performs the following main functions:

In addition, Infoterm acts as the organiser, co-organiser and/or co-ordinator for the international activities of all the organisations mentioned in section 3.4.1.3, including ISO/TC 37.

TermNet

In 1978, Infoterm was charged by UNESCO with establishing an international terminology network known as "TermNet". The new association was originally founded in 1980 as an informal network of collaborating institutions and experts, and became a separate legal entity at the end of 1988. It is an international non-profit association with corporate membership. Membership, which is world-wide, has now risen to over 50, while "subnetworks" such as China TermNet have also developed. In addition to providing a forum for discussion, TermNet now acts as a service centre for its members, performing applied R&D projects aimed at promoting the development of the terminology market and the knowledge industries. It also publishes monographs and a regular bulletin, TermNet News.

Gesellschaft für Terminologie und Wissenstransfer (GTW)

The Gesellschaft für Terminologie und Wissenstransfer (Society for Terminology and Knowledge Transfer) was founded in 1986 in the run-up to the first international congress on "Terminology and Knowledge Engineering" (TKE), which it now hosts every three years. The goal of the GTW is to co-operate with individual experts and institutions from research and industry in organising activities aimed at integrating methods and tools for terminology, documentation and computer science. Its longer-term aim is to develop a sound methodological and technical foundation for knowledge processing as a whole. As such, it is increasingly focusing on prenormative research and development activities.

Internationales Institut für Terminologieforschung (IITF)

The Internationales Institut für Terminologieforschung (International Institute for Terminology Research) was founded in 1989 in co-operation with the Institute of Scientific Theory and Research at Vienna University, Austria. It is a co-operation network for academic experts for terminology research and training, and organises joint research and training activities with individuals and institutions in the academic world. The focus of its activities is on basic research in terminology science and terminology teaching, and on training.

Regional and Supranational Terminology Organisations


Regional terminology organisations - and particularly those engaged in terminology planning and harmonisation - have an essential role to play in national science and technology and, in particular, of the transfer of science and technology between countries. This is especially important in newly industrialised countries such as Malaysia and Brazil, in the countries of the former Eastern bloc, and in the case of major language networks. The following sections give a number of examples of all types of organisation.

Language-Based Organisations

In order to share resources, avoid duplicating work and mutually benefit from experience, a number of language-based terminology networks have arisen. In most cases, all aspects of terminological activities are represented by these networks, which bring together theoretical research, training, terminology creation, neology, exchanges, etc.

These networks, which in many cases operate partially within European boundaries and partially outside, cannot be ignored in any future European infrastructure without running the risk of creating serious terminological harmonisation problems with European-based languages, particularly transcontinental ones.

Realiter (Pan-Latin Terminology Network)

Realiter assembles 55 institutions from 16 American and European countries which are inclined towards very specific work concerning six Latin languages (Catalan, Romanian and the four official Latin languages of the European Union) and English.

Created in 1993, Realiter's main objective is to regroup terminological activities and to encourage exchanges between terminology specialists from countries having a common use of Romanian languages. Member institutions share the development and harmonisation of Latin languages, considering their common origin to the extent that they share similar lexical origins and forms. Among these projects, we will keep the following: computer vocabulary in 7 languages, vocabulary on the environment, a list of productive forms in Latin languages, a methodological guide for terminological work, as well as collaboration between students preparing theses and dissertations on terminology, which allows them to contact and have easy access to different terminological resources.

Through Realiter's permanent activities, it carries out terminological inventories and plans a structure of communication in order to create an information service through a hypertext network format (web), which will be transmitted by Internet and will favour, at low cost, the exchange between specialists from different countries. The Union Latine, with the aid of the DGLF of France, is responsible for the secretariat of this network.

Rint (International Network of Neology and of Francophone Terminology)

Rint was created in 1986 and its activities since then have consisted of terminology development and international co-operation on Francophone terminology. Its language policy is designed to support multilingual discourse with an aim to promoting the French language at an international level. At an organisational level, it is composed of a number of national and regional groups, and associated members.

Rint publishes glossaries, bibliographies and directories in the fields of terminology on a regular basis (e.g. a French/English dictionary of intelligence programmes, and a French/English dictionary of protection against the ozone layer). In addition, it is active in the fields of training, theoretical and methodological research in terminology and the development of terminology tools. Another activity concerns the dissemination of information. Rint publishes a new terminology review and two types of terminology inventories: an inventory of work in process and terminology projects and an inventory of recent works. Last but not least, the network monitors terminology developments in Francophone languages, particularly in Africa.

RITerm (Ibero-American Terminology Network)

RITerm, created in 1988, unites institutions involved in terminology work in Ibero-American countries (i.e. Portuguese and Spanish speaking countries in Latin America, and the countries of the Iberian peninsula). With approximately 40 registered members, RITerm is presently undergoing changes. Its objectives, as defined by the RITerm-BD project, include: the creation of a network of terminology data bases, and the establishment of a training programme and a specialised library. RITerm has organised symposia in Caracas (1988), Brazil (1990), San Milan of Cogolla (1992), and Buenos Aires (1994), with the next symposium to be held in Mexico in 1996. This network is financed and controlled by its members although, for certain activities, national and international institutions are requested to help financially. The Universidad Pompeu Fabra de Barcelona is presently responsible for the RITerm secretariat.

Union Latine

Union Latine is an intergovernmental organisation comprising 32 states (8 European, 19 American, 4 African and 1 Asian). In recent years, its programme department II, dedicated to terminological activities, has carried out a number of important terminology surveys in member countries. It has also contributed to the training of 7 national terminology institutions, and 2 regional networks. Its main information service is the widely read multilingual publication "Terminometro". In addition, it publishes repertories and inventories of translation resources and terminology in the Latin world. Union Latine has also actively supported RITerm's activities, notably the establishment of a network of terminology databases in the RITerm and the Union Latine areas, and supported numerous terminology training courses in these countries.

Rat für Deutschsprachige Terminologie (RaDT)

Founded in 1994 under the auspices of the German, Swiss and Austrian UNESCO Commissions, the Rat für Deutschsprachige Terminologie (RaDT - Council for Terminology in German) is concerned with questions of terminology and terminology policy related to the German language and its use in the German-speaking world in general. The 25 or so members of the Council are all leading figures in terminological activities and include representatives of organisations in Switzerland, Austria, Germany, Denmark, South Tyrol, Belgium and Liechtenstein.

Arabterm

Arabterm, the terminology association of the Arab countries was founded with the support of Infoterm in 1989. Efforts are directed at the harmonisation of Arabic terminologies in the different countries, which in some cases are completely different or even contradictory as a result of colonial and post-colonial influence. Another priority is the systematic recording, development and storage of terminology in terminology databases.

EAsTerm

Plans for an East Asia Terminology Research Centre (EAsTerm) were unveiled by the Chinese in 1993, and the institution is due to be founded in late 1997. The background for this idea is the fact that the Chinese language (and above all the Chinese script) has had a greater or lesser influence on a number of languages (such as Japanese, Korean and Vietnamese). In addition, the many Chinese communities outside the People's Republic - especially Taiwan but also Singapore and the Chinese communities in the United States, Europe and elsewhere - do not speak a uniform language. In may cases, the same Chinese characters are used, but with sometimes significantly different meanings. This not only proves a handicap for specialised communication between experts, but is also an inconvenience to other, more general communication between Chinese people living in different countries.

MABBIM

MABBIM is a body set up by Malaysia, Indonesia and Brunei for the purpose of co-ordinating the very intensive language and terminology planning activities in these countries. The common language of these countries (which is known by different names in the individual countries but is Malay in origin) also serves as an instrument for national unification, while ethnic and religious distinctions (above all in Malaysia) have been fully maintained. Infoterm has been providing support for Malaysian terminology planning at the request of the Malaysian government for a number of years.

Geopolitical Regions
Nordterm

The Nordic countries (Denmark, Finland, Iceland, Norway and Sweden) co-operate in a wide range of fields, including that of general language (Nordic Language Council) and terminology (Nordterm). The activities of Nordterm cover all areas of terminology and a number of languages (including minority languages such as Sami, which is spoken by the Lapps). Systematic language and terminology planning has always been a major interest, and international terminology research has received invaluable input from numerous experts in this geographic area.

The CIS and the former CMEA

The Commonwealth of Independent States or CIS (i.e. the former USSR) and the countries previously belonging to the Council for Mutual Economic Assistance (CMEA) have a time-honoured tradition of systematic terminology work and terminology science. Following the political changes in the area, efficient terminology management in multilingual object-oriented terminology databases has become more important than ever before, in particular for scientific and economic purposes, but in many instances for language and terminology planning as well. Requests for assistance from Ukraine and from other countries demonstrate this.

The organisation of an international conference on terminology science and terminology planning in Riga (Latvia) in 1992 served to promote the formulation, improvement or modification as appropriate of terminology policies in numerous countries (e.g. Latvia, Lithuania, Estonia, the Ukraine, and even Russia itself).

The NAFTA and MERCOSUR Countries

The newly created North American Free Trade Area (NAFTA), which comprises Mexico, the USA and Canada and which has English, French and Spanish as its three main languages has become a further important area for regional terminology activities. Particular emphasis is being placed on the co-ordination of terminology standardisation and the preparation of a common terminology database, on multilingual quality assurance terminology and on the application of quality management in terminological activities. MERCOSUR, the common market for South America, is also active in the field of terminology, with Portuguese and Spanish being the major languages.

Terminology in the Specialised International Organisations of the UN System


The Headquarters of the United Nations in New York (UNNY), Geneva (UNOG), and Vienna (UNOV), and the other major special organisations within the United Nations system are almost all active in the preparation and distribution of terminology for their own (internal) purposes. Such activities are generally designed to support their translation services (e.g. by producing multilingual - i.e. English, French, Russian, Spanish, Arabic, Chinese - glossaries and terminology surveys) and documentation departments (thesauri, classification systems, etc.). In addition, they are also involved in a number of cases in preparing or organising the preparation of terminologies which are regarded as authoritative or quasi-authoritative terminologies by their respective users. The latter activities are often carried out in co-operation with the appropriate regional or national organisations, or influence the latter's terminological activities significantly. In a few cases, such terminology is legally binding on the member states of the United Nations. The following sections list some of the more important actors in more detail.

The World Health Organisation (WHO)

The World Health Organisation (WHO) has been very active for many years in the compilation of multilingual auxiliary terminological products such as the nomenclature of diseases and glossaries on acupuncture, as well as in the preparation and utilisation of a thesaurus for rapid, efficient information retrieval. In addition, it has developed a programme to integrate terminology administration, thesaurus utilisation, and the management of bibliographical information (WHOTERM), largely based on recommendations by Infoterm.

United Nations Educational, Scientific and Cultural Organisation (UNESCO)

The "Terminology, Documentation, and References" department of UNESCO (United Nations Educational, Scientific and Cultural Organisation) has become the centre for terminological activities within the organisation. A sophisticated application ("TRD-REF") based on the UNESCO documentation software package MicroISIS has been developed for the management of terminological, bibliographical and factual data in a complex, multilingual database (obtainable via Infoterm).

United Nations Industrial Development Organisation(UNIDO) and International Atomic Energy Agency (IAEA)

Numerous terminological activities are being performed at the United Nations Industrial Development Organisation (UNIDO) and the International Atomic Energy Agency (IAEA) in Vienna. One example is a design for multilingual user interfaces to heterogeneous databases developed for UNIDO. The IAEA also houses a branch office of the Food and Agricultural Organisation (FAO) which manages the AGROVOC thesaurus, based on the MicroISIS application programme "MTM" (Multilingual Thesaurus Manager).

Organisation for Economic Co-operation and Development (OECD)

The Organisation for Economic Co-operation and Development (OECD) is also highly active in the area of terminology, and in particular has created a macrothesaurus in English, French and Spanish. This thesaurus is published and sold also in electronic form and is widely used as a multilingual retrieval tool. There is a co-operation agreement between OECD and Infoterm concerning the co-ordinated development of the thesaurus management programme, which is based on MicroISIS.

Other Individual Activities

The UN specialist organisations located in the United States, such as the World Bank (WB) and the International Monetary Fund (IMF) (the fourth edition of whose Glossary appeared in 1992), also run terminology databases.

In addition, the Economic Council for Europe (UN/ECE) of the United Nations in Geneva is responsible for laying down the methods and formats to be used in electronic data interchange. This is performed by Working Party 4 (WP 4): "EDIFACT - Electronic Data Interchange For Administration, Commerce and Trade", which has set up an expert group for multilingual EDIFACT terminology. This group has developed plans for the design and implementation of multilingual terminology databases for EDIFACT users, a high priority since up to now almost the only terminology available has been in English, a factor which has considerably impeded wider acceptance of the standard.

JIAMCATT

In order to improve co-operation in and co-ordination among UN organisations active in the field of terminology, the Joint Inter-Agency Meeting of Computer-Assisted Terminology and Translation (JIAMCATT) was established a number of years ago within the framework of the Inter-Agency Meeting of Language Arrangements, Documentation and Publication (a high-level co-ordination body). Recent topics included the use and implications of translation memories and voice recognition systems [AITC 95]. In addition, questions concerning the copyright of terminological data and other intellectual property rights and problems in connection with the exchange of terminological data are of constant concern.

Problems

Terminological activities occur in nearly every subject field at international and regional levels, but there is insufficient co-ordination between the players, both across domains and at different levels within a single area. In particular, although the language-based networks have done extremely good work, co-operation across language groups, and mutual knowledge and recognition of activities, has sometimes been suboptimal. As a result, users are faced with inconsistent or poor terminology, while creators regularly duplicate their efforts unnecessarily.

As has been mentioned elsewhere, domain experts performing terminology work (e.g. in specialist and standardisation organisations) are often not aware of the latest terminological tools and methods. This means that, despite their competence and experience, their results are often only useful in the context of the application for which they were developed (i.e. they are monofunctional) and cannot be interchanged easily, if at all.

In addition, there is a general lack of awareness and/or concern among individual expert communities of the terminological needs of other users, thus making it difficult for outsiders (even where these are experts in the same subject field) to identify and access existing resources, and to evaluate their quality and degree of prescription if they do. Thus much international terminology is "grey literature" (it is estimated that less than 20% of existing terminological collections can be accessed using catalogues and information systems).

As a result, users of all kinds, and especially those from outside the domain in question often create their own terms, thereby adding to the general problems of homonymy and synonymy.

Solutions

In order to avoid a duplication of efforts at international and European levels, to create or maximise synergy effects and to ensure a rapid solution to the pressing need within Europe for high-quality terminology, the future European terminology infrastructure should work closely with, rather than attempt to recreate, all existing international and supranational structures, and should take advantage of the information and experience available through them. To this end, all types of international organisation should be approached with regard to active participation in the new European infrastructure, and co-operation on the distribution of their existing resources. Success in this area would provide a solid nucleus of participatory institutions and a corpus of high-quality and accepted terminology in the short to medium term. The Terminology Summit planned for 1996 and the preparatory work leading up to it should be used for this.

Recommendations


1. "Who's Who in Terminology", currently being prepared by a consortium led by Praetorius Ltd. (UK)

2. [Goffin]

3. Internal CEC statistics

4. [Ball] in [Schmitz 93], plus CEC information

5. See [Manu 95]