Just as the writer’s storyboard maps and links book chapters ‘concept maps’, for thesaurus construction, aim to solve problems caused by ambiguous labels or out-dated concepts and aid the revision of old hierarchies or the construction of new ones. By focusing on a narrow problem area of a hierarchy, and converting the relationship between thesaurus terms into a graphical representation, it becomes easier to discuss a problem and consider new structural relationships.
In the case of multilingual thesauri ‘concept maps’ can be useful in the first instance to identify some quality control issues such as ambiguity in the use of a concept in the parent language, or a lack of currency of the concept in the parent language, prior to addressing issues of meaning and translation across multiple languages.
Concept map (Broughton, 2006: 18) *
Recently we applied this method to address a data indexing problem where terms for the use of drugs for recreational purposes and for medicinal purposes had become conflated. Immediately we found a bias in the broader term (BT) for the concept of DRUGS which assumed that both medicinal drugs and drugs for recreational use were manufactured products, when one of the most widely used drugs now, cannabis, is a natural plant. Furthermore, what is currently considered a drug not only includes alcohol and tobacco but also substances in products designed for everyday use such as glue.
When the concepts were first added to the thesaurus it was the case in most western countries, especially those where drug control was a policy issue, that all drugs were manufactured products, non-medicinal drug takers were addicts and all drugs with addictive properties were illegal drugs.
Clearly, within Europe today, what constitutes a crime in the use of drugs will always vary so adding particular drugs to the ‘criminal’ and ‘policy’ topic areas we had identified would result in inconsistency in the thesaurus structure. However, the importance of collecting data on attitudes to, or experience of, drug-taking is of great cross-border research interest. When reference was made to CESSDA data catalogue measurements of ‘drug use’, both criminal and recreational, rather than the chemical properties of drugs themselves, the social science concept of ‘drug-use’ was revealed to be key in the process of conceptual analysis for indexing purposes across all CESSDA archives.
Such a complex problem of thesaurus construction, caused in the main by changes in social practice and social policy, suggests that a path through the resolution of such a problem might be found in the application of facets. Ranganathan’s five primary categories, or facets, denoted by the acronym PMEST (Personality, Matter or Substance, Energy, Space, Time) that became integral in the construction of colon classification schemas, and has classically been used as a thesaurus construction tool for art and architecture in the Art and Architecture Thesaurus (AAT), could possibly provide the indexer of social science data with greater flexibility in the construction of complex concepts.
*Broughton, V. Essential Thesaurus Construction. London, Facet, 2006.