When meaning is added to words the problems are multiplied but the potential for their use is far greater. When related to each other, words can be formed into complex hierarchical structures or thesauri that can be the heart of powerful information search systems, if the data comply to the standards set out in the thesaurus. In the UK a Thesaurus of Monument Types (TMT) has been created (NMR Monument Type Thesaurus) that provides a deep structure of terms for vocabulary control use. Even within the same language the creation of such a thesaurus required considerable debate over the definitions used, a debate that it is recognised will continue for certain controversial terms. The TMT is valid only for England and many terms would require adding or redefining to meet the needs of Scotland, for example. These problems would be multiplied again if an attempt was made to create a thesaurus for the whole of Europe.
Some work has been carried out by projects that took on or are taking on specific areas within archaeology or heritage management. The first of these was constructed by the Bronze Age Campaign Council of Europe project. The Bronze Age Campaign project created a multilingual thesaurus of terms relating to the Bronze Age, as at 1994. This work was to gain almost legendary status in the networks that were developing an interest in the application of digital technologies to archaeology and heritage management, but little more happened for some years. The ArchTerra project (1999–2001) brought Internet technologies to partners in Poland, Hungary and Romania and also created a thesaurus as part of their work published online in 2000 by Martijn van Leusen. Following these projects, a much larger and more ambitious project was begun to create a multi-lingual thesaurus of heritage management terms in Europe. The HEREIN project was funded by the main EU funding stream – the Framework programme – and was supported over two stages, the last of which was completed in 2002.
All of these thesaurus projects have moved European archaeology and heritage management a little closer to Hansen's dream of shared data. The ARENA project worked on the practical implementation of shared data and had to find common ground between databases that had been constructed independently of any such overarching thesauri and controlled vocabulary. To make the data searchable on a WHAT theme, a top level of terms had to be agreed that could be mapped onto the terms used in the individual databases. This meant that once again the question had to be asked: 'how far can we go?' Further depth in searching would have required the creation of a much broader thesaurus than the single set of eighteen terms that were used. The HEREIN projects had already demonstrated the degree of time and expenditure required to attain this depth.
It is clear that much work will be required in Europe to facilitate real depth to searching huge data-sets of sites and monuments data. The same is also true for searching data on a WHEN theme. The same disparities in meaning encountered when considering what a site or monument is, are also found when we try to search by period. The ARENA project simplified the issue once again to facilitate data sharing and interoperable searching. A set of twelve periods, based upon the English MIDAS standard, were adopted by the partners, and once again mapped onto any local variations and sub-periods in each data-set.
The problems of meaning for period are multiplied because each period has a different temporal range in almost all nations in Europe, and indeed in many regions. A visual representation (see Fig. 2 below) was used by the ARENA portal search to demonstrate the range or absence of some periods in some countries. This also raised the issue of periods as absolute date ranges or periods as cultural terms. For example, the Iron Age in Denmark has a much longer absolute time span than the Iron Age in the UK, and an ARENA portal search for Iron Age records will retrieve hits from the equivalent cultural period in each country which may be of very different chronological age. This solution had to be adopted because the underlying data did not contain sufficient information to allow a search based upon absolute date. While the ability to recover, for example, 'all European sites occupied in AD 500' might have been desirable from a research perspective, it would effectively have required individual reclassification and recoding of each source record.
Figure 2: Working example of the ARENA period select function, using a graphic display to show variations in period in different geographic locations
© Internet Archaeology/Author(s)
University of York legal statements | Terms and Conditions
| File last updated: Tue Sep 27 2005