Summary of results of terminology surveys in France

Choice of surveys

The Union Latine has come up with more than 20 surveys at least partially concerning terminology in France or in French (Les enquˆtes de terminologie). Only four are analysed here, those with most accessible results and which which bring out gaps and needs.

Scope of surveys

Some surveys are broader in scope than others (especially those covering translation, translation tools) and are not treated here, whereas others deals with only one aspect.

Aims of surveys

Some surveys aim at a better understanding of the situation of terminology (what exists and what needs there are), others give a concrete result. The French-language term network Rint uses its surveys to update both the Inventories it publishes, and the Banque de terminologie du Qu‚bec bibliography data base. Two surveys (J. Amyot and GOTA) are in fact preliminary feasability studies in view to setting up a terminology network.

Surveys covered

Four general surveys have been analysed :

Three more specific surveys are mentioned :

Sectors covered

Large firms are at the centre of the Jacques Amyot and GOTA survey, whereas TLS and CTN also include smaller businesses and individuals. The translation market is well covered in all surveys. The CTN's survey includes a part devoted to terminology in universities and public research and other official organisations.

Some terminology applicatons are practically not covered at all: interpreting; publishing technical dictionaries; terminology needs of the cognitive sciences, in expert systems in particular; technology watch; quality control. The need for thesaurus for information-documentation are mentioned in GOTA, Rint, TLS, and should be further documented.

Results in terms of concrete outcome

Some surveys give a concrete outcome, such as the permanent inventories of the Rint, like those of the Union Latine or TermNet. Those of the Rint are published by the network in book form and put on the Quebec term bank data base. It would be desirable for this information (Bibliographies, inventories) to be made available in Europe in date-base form. By the same token, it would be useful to have the results of such works as Elisabeth Blanchon's analysis of terminology software systems, already published in Terminometro and TermNet News, in network accessible data base form. For the Pointer project, it should be mentioned that the CTN's surveys on terminology activities also form a data base, which can be made available to the project. In short, all that comes out of the surveys that is worth saving should be put into a European network.

Daniel Gouadec's word of warning on the limitations of surveys made between eight and three years ago should also be mentioned, with the explosion in electronic networks. Particular attention should be given to just how telematic networks and the information society are affecting the day-to-day life of present and future users and producers of terminology. To cite just one example for France, terminology service have already been launched by Rennes 2 university and the Union Latine, and perhaps others.

Results in terms of needs expressed


The qualitative report by TLS brings out urgent needs in translation oriented terminology for les wide-spread languages, and for isolated translators, of whom there are more than had been expected. Training is thus a priority, a conclusion confirmed by the J. Amyot and GOTA surveys.

According to the Rennes survey, terminology comes second to access to technical documentation in translator's priorities. Long periods of time devoted to looking for terminology, in the absence of dictionaries are reported, and most terminology problems are related to understanding of concepts. There would thus appear to be an opening for terminology tools of dictionary form giving access to technical documentation, such as hypertext, which can be constantly updated, and these should be promoted. Translators complain of not having enough time to make up their own term base or to manage it correctly. As far as the need for information on the terms themselves are concerned, translators seem to be content with those normally provided on standard term records.

Terminology Networks

Proposals suggesting a terminology network should be set up iare prominent in J. Amyot and GOTA, but also appear in the other surveys, especially TLS. Most producers/users are for an association, or a grouping according to sector, a list of addresses, or some other form of working together.

It is widely reported that terminologies created in industry are not made more widely available. Two main reasons go towards explaining this state of affairs.

  1. The unfinished state of term records, especially with respect to checking and validating,
  2. Right to privacy (copyright, intellectual property...).


Subject fields at the cutting edge of progress are cited everywhere are needing terminology work (though just what work is not always clear: which languages, what sort of dictionaries, etc.), but specific fields are targeted for special attention by GOTA : energy, economy and finance (banking and insurance), armaments, the motor industry, computer and imaging technology, legal sectors...

The importance of involvement of those concerned with terminology in the industrial setting should not be underestimated. "If we want businesses to participate in a network [of this sort], they must be convinced that the network is designed for them, to respond to their needs, and not to serve some political purpose, even one of language planning." (Algardy p. 6.) This remark is a challenge for those in charge of the project, which includes very few industrial deciders.

All these surveys bring out the imarginal situation that terminology often occupies in a business. The hope of correcting this situation spurs terminologists and translators on to getting organised (associations, networks, forums), but it is enough to convince the firms' deciders? Might it be opportune to stress the role that terminology are called on to play in the cognitive sciences, or in language engineering?

One last practical word. If a survey is to be undertaken, it would be wise to bear in mind that many of the surveyed are fed up with answering too many (terminology) questionnaires. Furthermore, a survey is a long-term undertaking, so it should be carefully targeted, as so little time is available.

31st March 1995