4.4. What next?

Controlled vocabularies have been developed in several countries for national and regional heritage records, museum collections inventories and research programmes (see Sanjuan 2004). For example, the National Monuments Record Thesauri provides links to the Thesaurus of Monument Types, the Archaeological Objects Thesaurus (developed by the Museum Documentation Association), the Thesaurus of Building Materials and the Maritime Thesauri. A list of controlled vocabularies in the United Kingdom is also available from INSCRIPTION, part of the Forum on Information Standards in Heritage (FISH). Other examples exist in Canada, France, and Italy.

Smaller projects developed their own terminologies for ad hoc purposes. Thesauri have been developed for arts and architecture (the most famous being AAT Arts and Architecture Thesaurus of the J.P. Getty Trust, maintained by the Getty Vocabulary Program, available freely online). Few of them are correlated and even fewer are multilingual. They are split among countries and organisations but also among disciplines. Work is very fragmented and sometimes kept within organisations, hidden from end-users and rather used only by staff for cataloguing or indexing. An important amount of scholarship is not accessible as a retrieval and learning tool. In addition, many vocabularies are only available in printed form. Until now, this situation meant wasted efforts, and in effect prevented the use of modern communication technologies in collaborative efforts to achieve multilingual vocabularies.

We need an inventory and assessment of existing controlled vocabularies for cultural heritage, made by an international team of experts. Controlled vocabularies should be published online, as ready made bricks for future construction that others can reuse, repackage, and build services upon - provided that legal and financing problems can be solved. Although not easy, mapping among already complete terminologies is a reasonable approach instead of starting from scratch. Controlled vocabularies are based on home-made classification systems, respond to regional and thematic needs and depend on field scientific theory and fashion at the time of their creation. They get old and need refreshing. Controlled vocabularies should be viewed not only within the context of the projects that created them but as key resources that can become multilingual.

Terminological tools for cultural heritage communication need large-scale co-operation projects. They cannot be just by-products of other projects. Until this area receives proper attention - in concrete facts, not in pious words - we cannot expect significant progress. The Semantic Web needs multilingual mapping tools.

A recent initiative is MICHAEL - Multilingual Inventory of Cultural Heritage in Europe (2004-2007). To clarify, in this case, 'Europe' means France, Italy and the United Kingdom, and 'multilingual' means three languages (guess which). This project developed out of the MINERVA Project for digitisation of cultural and scientific content. Multilingual issues were only surveyed in MINERVA Plus Project by a working group coordinated by the French Ministry of Culture and Communication (see the working plan). An online questionnaire concerning multilingualism of cultural websites in participant countries was meant to gather information on the use of controlled vocabularies too. The results of this survey and its conclusions are not known.

Funded with 3.4 million euros, project MICHAEL's aim is to establish an international online service (Figure 12), in the framework of the eTen European programme, which will allow its users to search, browse and examine multiple national cultural heritage resources from a single, multilingual point of access. The effort of translation is going to be solved by the project instead of the end user. How this service will contribute to the already well-developed 'cultural tourism' in those countries and 'push up the economic value of archaeological and cultural sites as cultural products' (as stipulated amongst its goals) remains to be seen.

Figure 12
Figure 12: Michael Project Homepage

Other expected benefits of the project are: 'interoperability and the use of common standards', 'improved access', 'e-learning', 'social cohesion', 're-use by creative SMEs', 'integration of national initiatives' and so on. All these are mandatory statements if you want to get European funding for a project nowadays. They might become true if money went to developing both tools and content: improving, reshaping, rewriting, interpreting and adding content to the existing resources, and developing new ones. You cannot expect to attract tourists, young people, less educated and researchers with the same narrative. Most of the records in the national cultural heritage databases are boring, meant for specialists and heritage administration. Some of the images serve conservation and protection purposes, not touristic and aesthetic ones. Here multilingual also implies the type of discourse - developing different discourses for different groups of the public should be part of the project's development.

Multilingual access is effective if users get much more than a thin veneer. Money should go to translating content, mapping controlled vocabularies, explaining concepts and regional terminology. Economic and social inclusion is better addressed if cultural heritage is presented online in minorities' languages and for people with disabilities too, in parallel with training courses and presentations to teach less-favoured groups, students and communities how to use these marvellous new tools. Funding should also go to intelligent advertising to let many people know that MICHAEL exists.

But it seems that EC money supports mainly the development of technology (using SDX open source platform), reports and meetings. If content improvement is left entirely to 'the national governments and agencies responsible for cultural heritage', integration of cultural heritage information will remain wishful thinking. With budgetary cuts and plenty of tasks in their own countries, national bodies will hardly find resources to make changes for the sake of integrating with the others. As to new countries joining the consortium and implementing the model at their own cost, we are at risk of remaining with a very narrow definition of 'Europe'. I would be glad to be proved wrong...

For archaeology, projects should be addressed to key areas of interest across Europe, starting with the 'what' (typologies), 'when' (periods) and 'where' (places) - fundamental questions - to arrive at 'why', 'how' and 'from where' conclusions. The following kind of issues must be dealt with:

Martin Doerrs research on 'Thesauri of Historical Periods' standardisation, together with Athina Krisotaki and Steve Stead could become a very useful tool (Doerr 2004). The progress of this project was presented at the CIDOC Conference in Zagreb, May 2005 (Doerr 2005).

Standards on developing controlled vocabularies must be updated. The ISO standards for monolingual and multilingual thesauri are 15 years old or more. Although in the main they remain valid, some revisions are needed to bring them to the digital age. Leonard Will considers that the many new names that have been used recently, such as ontologies, taxonomies, semantic webs and topic maps are not fundamentally different, in that they all consist of a set of concepts and relationships between those concepts. The main requirement for thesauri development standards is to move from words to concepts.

'The problems that arise in creating multilingual thesauri are the same as those that are found when merging or mapping two thesauri in the same language. So long as relationships are between concepts then we can use as many words as necessary to define the scope and meaning of a concept in each language. The difficulty arises in trying to find concise labels to use when showing relationships between concepts or linking concepts to documents (the process of indexing documents)' (Will 2002, 8).

The present fragmented puzzle of terminologies in cultural heritage should become more coherent and merge together. Existing thesauri and glossaries should become available for a larger audience. European-scale research is needed and European funding expected. I know this agenda seems in opposition to what the European Commission regards as a priority: networking, mobility, technological tools. It sounds well and good but the reality is different. We should campaign for real problems and better solutions for real needs.


© Internet Archaeology/Author(s)
University of York legal statements | Terms and Conditions | File last updated: Tue Sep 27 2005