First, thanks to John Palmer who wrote at 09:46 on 2/01/2000 with his support.

Some further comments are interspersed below.

Earlier at 14:26 11/01/2000 -0000, Brown, Duncan wrote:
>To take Jeremy's first three points in turn:
>
>(1) One of the difficulties with parameters is that while "nothing
>earlier/later..." is a good rule of thumb, there is plenty within these date
>parameters which can culturally be assigned to different periods. As an
>example, the late Neolithic/early Bronze Age is a minefield of varying
>interpretation, additionally complicated by the cultural phenomenon of the
>"Beaker period". Absolute dating, technology and cultural evidence do not
>fit together to form neat period divisions, and when you introduce
>geographical variables it gets even worse.
>
>(2) Quite right. But problematical.

A sliding scale would be able to cope both with floating sub-periods and controversial boundaries. Searches for either `Late Neolithic' or `Early Bronze Age' could both be set to return references for 'Beaker Period'. Recording of evidence relating to the cusp should have double entries, perhaps (or perhaps not?) at the recorder's discretion - i.e. classic Late Neolithic material need only be described as such.

>
>(3) This might be impossible to do if only high-level period terms are used.
>See the example given for (1). 
>
>While I think we are all agreeing in general terms that a hierarchical
>thesaurus is required to enable chrono-cultural ("period") classification,
>we still need to consider the types of terms we use to describe these
>periods, and in what context we use them. I refer back to my first mailing. 

I think we should include local chronological terms and even relatively short periods, like wars (or even battles for which the duration might be recorded down to the hour of the day!). 

The BIAB list is a simplified version of a much longer draft list that includes such stuff as English Monarchs and Roman Emperors.

__________________________________________________
At 14:24 11/01/2000 +0000, Trevor Reynolds wrote:
>
>For example if you are searching for Roman period sites do you want to retrieve a site from Australia dating to 400AD?

Yes - if I were, for example, looking for corroborative evidence of cosmogenic catastrophies affecting human populations world wide.

>A possible approach is to:
>
>     Agree a standard list of terms with scope notes 
>
>              e.g. Roman  SN: The period when an area is part of the Roman Empire.
>
>     Agree a standard that says that each database entry (object, monument etc.) should have any appropriate periods and an "earliest date" and "latest date" attached.

The MIDAS scheme includes maximum and minimum dates as well as period. In any case, the `earliest date' and `latest date' for an item will be often be inferred by default from the period in which it is thought to belong.
______________________________________________

Martyn Barber makes a lot of sense (at 17:21 on 11/01/2000). But as I see it, periods are simply constructs we use to help us make sense of the evidence - like stratigraphic 'contexts' (oops - I don't want to start that debate!). Periods are only subjective approximations that allow us to bundle together a variety of phenomena that we believe may be related. The tendency to invest periods with great synthetic significance should be resisted, and words to that effect should appear as a preface to any thesaurus of period terms.

The definitions of the periods must open to challenge and public debate - but this cannot usefully take place until the current definitions (however multifarious) are be clearly stated.


Jeremy


___________________________________________________

Jeremy Oetgen

British and Irish Archaeological Bibliography

New address: c/o The British Academy, 10 Carlton House Terrace, London SW1Y 5AH, England.
Tel: +(44)(0)171 969 5444 [or (0)20 7969 5444]
E-mail: [log in to unmask]
E-mail (personal only): [log in to unmask]
Internet: http://www.britarch.ac.uk/biab/index.html
___________________________________________________


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%