In our discussions at Yale about the EAD, a topic which is perhaps more
philosophical than technical (though it has technical implications) has
arisen. I promised to bring it to this forum for comment.
Specifically, is there an important,
possibly critical distinction to be made between the intellectual
structure of series/ subseries/etc. that exists "above" the physical
unit(s) being described and which are assigned unique, unit-level
identity (e.g. a unit number or title), which we're obscuring by simply
tagging that physical unit as another form of <c> (numbered or not)?
In fact, the use of the tags <unitloc>, <unittitle> and <unitdate>
within a <c> would suggest that there is. While it is true that one
can use attribute definitions to specify the character of data in a
given <c>, would we be better off if we provided a distinct tag? This
type of distinction between the headings hierarchy and the physical
units subordinated to it was provided for in the structure of the
Matlists, in the FINDAID EAD. Is this a feature we'd be well advised to
Of course, between collections and institutions listings describe
material at various levels of detail (collection down through folder,
even item), so the physical unit being described will vary. But at some
level (so this argument goes) there is a qualitative difference
between the physical unit and the headings structure which surrounds it,
and is applicable (at least in most cases) to a number of physical units.
If this is true, then it seems that a different tag ought to be used to
designate the physical unit, contained within and subordinate to the
headings the structure (in EAD terms the <c> and tagged in a
distinctive manner, not just defined by the attributes assigned to it.
It should be noted that this fairly precise distinction between unique
physical units and the associated headings structure is at the heart of
the Beinecke's finding aid structure and style, and is essential to
the various macros we use to process the finding aid text file and
generate various outputs (paper printouts, folder and box labels,
diagnostic lists to verify headings sequence and placement within the
finding aid list, continuation statements and series/subseries headers
in the finding aid, etc.) But we believe that this distinction is
one that reaches beyond local practice and applies in some fundamental
way to the logic of archival finding aids in a more general sense and
should, therefore, be accounted for in a more pronounced fashion than
is provided for in the EAD's recursive <c> with attribute definition.
We're interested in hearing your thoughts on this, both as to the
philosophical distinction and the actual, or likely, practical
implications for the EAD.
Beinecke Rare Book and Manuscript Library
[log in to unmask]