Leaving my own thoughts on what a ‘sense of place’ means to another time, I’m going to talk a little bit about the practical issues we have encountered when putting geographic context about objects onto our database.
Having plenty of geographical context in the paper accession files, we had to decide how to put that information onto the database, and how to make it searchable. We decided to create a hierarchy of places so that any specific place would, ideally, be linked to a county, region and country. Adlib, the database we’re working with, has a hierarchical capacity which enables us to do this. So, the geographical keyword (i.e. place) ‘Reading’ could be linked to ‘Berkshire’, for example. This brought up the issue, however, of deciding what form our hierarchy should take. Should we use a current list of contemporary administrative units, which include ‘unitary authorities’? Or should we use the still commonly used ceremonial counties? Long discussions threw up more and more ‘but what if…’ problems. How would we put ‘the Cotswolds’ into a hierarchy, for example? What about an object which arrived in the 1950s from Middlesex, a county which completely ceased to exist in the 1960s?
Throughout the course of these discussions, I discovered that my own understanding of the county system in the United Kingdom was woefully inaccurate. In my defence, it really is quite confusing. The ‘County of Herefordshire’, for example, is also a unitary authority, and the ‘City of London’ is apparently also a ceremonial county. Aside from showing up my own lack of geographical understanding though, our discussions did raise an important point – how do our visitors, who will hopefully be using ‘place’ as a way to search and access the collections, understand ‘place’?
Eventually, we decided that the only way to achieve any sort of consistency in our cataloguing was to use a hierarchy based on contemporary administrative boundaries, and we have based ours on the hierarchy used in the online ‘Getty Thesaurus of Geographic Names’. Bearing in mind some of the ‘problem places’ mentioned above, though, we have added in a number of other hierarchy levels. These will hopefully enable us to both more accurately represent the level of contextual detail contained in the accession records, and make the hierarchy fit with as many understandings of ‘place’ as possible.
Places are therefore linked first to their ‘administrative unit’, but then also to their ceremonial county (if relevant) and region. So ‘Reading’ is linked up to ‘Reading [unitary authority]’, which is linked to ‘Berkshire’, which is linked to ‘South East England’, and so on. A lower level may also be added, that of ‘specific locale’, such as farm names, estate names and, where relevant, street names. The problem places such as now non-existent counties and vague areas such as the ‘Cotswolds’ are still entered, but link straight to the country/countries they are in. They are listed as ‘non-preferred terms’, showing that they should only be used when no other information is available, and also have a note which defines the area that they refer to.
It’s certainly not a perfect system, and we still regularly encounter new ‘but what if…’ problems, but we hope that eventually it will enable visitors to access information about ‘place’ in our collections in a way that is both consistent and logical, but also meaningful to them.