This question raises several issues, none of which has a simple solution.
Dave Wells has already noted that political, ethnographic and
linguistic groupings have no simple relationship. It is therefore not
a good idea to try to combine them, because you are likely to create
inconsistencies. "Spectrum : the UK museum documentation standard"
(Cambridge : The Museum Documentation Association, 1994) specifies
distinct units of information for the following:
Place (several units of information for this, including:)
Place name
Place name type
Place coordinates
Place coordinate type
Place - geographical feature
Place - geographical feature type
. . . etc.
People
People's tribe
People's culture
People's linguistic group
So my first recommendation is not to try to create "ethno-linguistic
groupings which complement the existing geo-political groupings" but
to treat these as separate items of information with their own
structures and authority files. Each authority file can, and should,
refer to the others, using "related term" references and scope notes.
They can in fact just be distinct facets within a single thesaurus.
My second recommendation follows from this, and that is to forget the
concept of a "number of levels". I don't know whether your software
limits what you can do, but if at all possible you should store the
relationships between terms in a separate thesaurus or authority file
and not repeat them in the catalogue record for every object.
In your example there should be no need to store a hierarchy of place
names such as "Asia - South Asia - India - Assam" in the catalogue
record for every teapot. Your thesaurus should hold this hierarchy,
and a teapot need only be recorded as coming from "Assam" - the
narrowest applicable item in the list. The software should allow you
to search for all objects from India or Asia by expanding the search
term to include all its narrower terms.
In some cases you might need to go down several steps to achieve
sufficient specificity. If you have a hundred objects from Assam you
might well want to index them at the level of towns and villages; if
you have only three objects from Ethiopia the country name alone
might be sufficiently specific.
My third recommendation is to distinguish between "description" and
"access points". The controlled vocabulary we have been talking about
above provides for access points or "index terms" which make a search
more reliable and efficient. Once you have retrieved a record, you
should be able to read a full description of the object, which may
include as much detail as is known about where it came from, what its
history is, and so on, in a structured or a textual or visual form.
This description should meet the worries of curators who think that
by imposing a controlled vocabularly you are "losing historical or
original information". You must preserve this, but separately from
the access points.
The above discussion has concentrated on place, but the same principles
apply when dealing with cultural or linguistic groups.
A few other specific points:
1. Look at the Anglo-American Cataloguing Rules, chapter 23, for
guidance in deciding the best form of geographic name to use,
particularly the part about making additions in parentheses to
distinguish identical names of different places or to clarify where
places are by adding the name of a containing place.
2. Different spellings are handled in a thesaurus by deciding which
one is "preferred" and referring to it from others: e.g. Asanti _see_
Ashanti. If your software is good it will allow you to search on
either term, telling you when it makes a substitution.
3. Groupings such as "Arab culture area" should be handled by the
thesaurus hierarchy. But be careful in defining narrower terms to
a term like this - specific countries may not be permanently and
inherently part of this "area", and if they are not they should not
be made narrower terms (unless your thesaurus has a date dimension,
which is not common). Is this a grouping of countries or a grouping
by race?
4. I don't know of a ready-made thesaurus for the subjects you need,
but you could explore the following in looking for one:
a. Museum of Mankind (part of British Museum). They have done a lot
of computerisation of their ethnography collections and have used a
thesaurus. Worth contacting them.
b. Look at the tables in volume 1 of the Dewey Decimal Classification.
Table 5 is "Racial, ethnic, national groups" and Table 6 is "Languages".
This is not a thesaurus, but worth looking at for hierarchical
relationships.
c. The UK MARC manual, 3rd ed., appendix B: language codes. (Wetherby :
British Library, 1990.) Based on USMARC code list for languages
(Washington : Library of Congress, 1987). Gives a long list of
languages with preferred and non-preferred forms, though not a
hierarchy as such.
d. Internet resources, e.g. "The human languages page"
<http://www.dcs.warwick.ac.uk/~bear/Language-Page.html>, which is a
mirror of the WWW Virtual Library page for languages,
<http://www.willamette.edu/~tjones/Language-Page.html>.
This has lots of links to pages about languages, including the
"Universal Survey of Languages: A Collaborative Project"
<http://www.teleport.com:80/~napoleon/index.html>.
This has not got very far yet, but it has the beginning of a tree
of languages "by family".
There are ethnographic resources on the Internet too, but I have not
had time to look for them and this message is already too long.
Have you asked your librarian to do a literature search for you?
I'd be glad to discuss further, and interested to know what other
people think.
Regards
Leonard Will
--
Dr Leonard D Will Tel: +44 181 372 0092
Information Management Consultant Fax: +44 181 372 0094
27 Calshot Way, ENFIELD, Middlesex Email: [log in to unmask]
EN2 7BQ, United Kingdom
|
