What is a thesaurus ?

A thesaurus is the vocabulary of a controlled indexing language formally organized so that the a priori relationships between concepts are made explicit.
Like a wordlist, a thesaurus is a tool based on a standardized terminology which helps the user to select occurrences in a database following a certain logic.
Yet, it distinguishes itself from a wordlist by the following points:

• it allows terms belonging to the same class to be grouped into hierarchies and makes it possible to establish relationships between these terms and terms from another class.
• if there is a choice between several terms to express the same concept, a single preferred term will be chosen and the user will be directed to it when selecting non-preferred terms.
• the hierarchical relationship allows the user to have access to wider or narrower concepts within the same class.
• the thesaurus is a dynamic tool which can be updated by the addition, amendment and deletion of terms or relationships.

Despite the complexity of its development, a thesaurus offers real advantages in comparison with a wordlist. For instance, the efficiency of the selection of terms is improved and duplication is eliminated by the hierarchical and associative structure. A thesaurus also allows a group of users to make use of a similar system for indexing whatever the level of precision their research may require.

The establishment of a multilingual thesaurus does not correspond to the juxtaposition of N monolingual thesaurus/i. It relies on the construction of a common instrument where all languages have equal status; because of this, it is necessary to establish as many hierarchical, associative or equivalence relationships as possible between the terms chosen in each language, but at the same time, it is important to try and respect specificities proper to each culture.