1. Scientific objectives of the workshop

This workshop will deal with the phenomenon of variation in terminology and, more generally, in specialized language. It will bring together experts from specialized communication, terminology, translation science and computational linguistics, ensuring thereby a pluridisciplinary and cross-disciplinary approach to the phenomenon. We intend to invite experts from terminology science, translation science, computational linguistics, lexicography and computer science (a list of the prospective participants is attached; we there also mark whose participation needs yet to be confirmed; participants without a mark of this kind have already confirmed their willingness to participate in the workshop).

Linguistic variation is normal in general language: examples of variant expressions are synonyms, paraphrases or syntactically different alternative ways to express a given concept. Though expressing typically "the same ideas", variants often belong to different registers, regional varieties, sociolects etc., which is why sociolinguistics has traditionally dealt with this concept, focusing on general language. Regional varieties have been the topic of study in many pluricentric languages: for German, the 'Variantenwörterbuch des Deutschen: die Standardsprache in Österreich, der Schweiz und Deutschland sowie in Liechtenstein, Luxemburg, Ostbelgien und Südtirol' by Ammon et al. (2004) is a typical example of a dictionary of regional varieties.

As far as specialized languages are concerned (Fachsprachen), as well as terminology, variation has for long either been ignored or even banned: concept-based terminology in the sense of Wüster (1991) advocates the principle of monoreferentiality: one concept, one term. If there were several variant terms for one concept, this would introduce ambiguity and thereby be a hindrance for effective and efficient communication. As a consequence, terminological data collections are often set up from a prescriptive point of view: for a given concept, one term is recommended and possible variants are proscribed.

It was probably the quite recent emergence of corpus-based terminology work, as well as the systematic and comparative analysis of specialized texts which triggered a new wave of interest in terminological variation in the research community: together with the recognition of a need for a more descriptive (and possibly thus less prescriptive) approach to specicalized communication and terminology, many researchers got aware of the existence of a broad range of variation phenomena in specialized languages; hereafter, we note just some examples:

- variation induced by differences in domain knowledge and specialized language knowledge in communication partners: expert communication vs. communication with or among lay persons;

- variation induced by the layering of specialized languages: workshop language vs. scientific language vs. marketing speak, etc.

- variation induced by secondary term formation;

- regional variation;

- corporate language and the interest in companies to coin terminological variants in order to have "own" denominations of certain concepts.

Contrary to earlier views on specialized languages, especially in technical domains, it is by now mostly taken for granted that term variation must be accounted for in the description of specialized terminology, and that it has to be represented in terminological data collections, in specialized dictionaries, or in the language resources of natural language processing systems.

Yet the exact extent of the phenomenon and the way in which term variation should be described and represented requires more research: the data category standard ISO 12620:2009, to give but one example of an authoritative document, is relatively vague on the notion of variant:

"A.2.1.9 variant

DESCRIPTION: One of the alternate forms of a term.

EXAMPLE: spelling variants: catalogue (GB), catalog (US)"

The ISO standard does not describe the full range of variation phenomena, nor does it explain whether 'alternate' forms of terms are characterized by any specific properties.

There are a number of theoretical and descriptive, as well as practical and application-oriented research questions which we intend to tackle in detail in the proposed workshop:

- Theoretical and descriptive issues concern the extent of the phenomenon of term variation, its relationship with synonymy and with sociolinguistic factors, as well as with the properties of certain (types of) domains.

- Practically and application-oriented issues concern techniques for identifying term variation in texts, as well as ways of representing variants in terminological data collections and dictionaries: here, we intend to explore in particular the possibilities offered by electronic dictionaries and term banks and by lexical semantic modelling in computational linguistics.