19 December 1997

Kristi Kiesling, Chair
Committee on Archival Information Exchange
Society of American Archivists

Comments on the revision of MARBI 97-7: Coding Leader/06 and Leader/08 for archival material
 
 

The Harvard manuscript and archive community disagrees with the basic premise of the original proposal: that there is no need to distinguish the "type"s of material in the traditional sense, and that what is most important is the content, i.e. "text".  We are concerned that the SAA proposal as it was originally drafted would, if approved, undermine our ability as archivists to retrieve all the material in our care from our online catalogs, and would seriously impair the ability of our patrons to locate our material in large, heterogeneous databases.

1.  Leader 08 Type of Control.  Not all manuscripts are described "archivally," therefore "Type of Control" by itself is not enough to identify all manuscript material in a database.

The change in definition that was approved at Midwinter reads in part "the material is described according to archival descriptive rules which focus on the contextual relationship between items [emphasis added]." Harvard has tens of thousands of manuscripts and letters that were acquired singly, and are therefore not described "contextually."

Many libraries have cataloged single manuscripts under AACR1 and AACR2 rules, and these materials cannot properly be described as being under "archival control"; they are not cataloged according to APPM, nor, as single items, can one say that one is describing the relationship between items.

There are many people--staff and researchers--who will want to retrieve both single items and collections of manuscripts in the same search, regardless of the "type of control" we as archivists and curators have used in describing them.

It should also be pointed out that the 040 already accommodates encoding of the descriptive rules followed.

2.  Leader/06 type of record.  We strongly believe that there is utility in being able to distinguish between printed/existing-in-multiple-copies materials, and manuscript/unique materials in the catalog.  It is certainly true that the phrase "manuscript language material" is ambiguous and may cause confusion (although fortunately our patrons are not likely ever to encounter this phrase).  Most people are not exactly sure what a "manuscript" is, let alone "manuscript language material," although they usually know that it differs from a printed book.  However, we have all been living with the ambiguity of what is a "manuscript" for a long time, and we do not think it has caused too many problems.  Researchers generally accept that it includes typescript and computer printouts because they realize that these are created by an individual, by hand, rather than by a method of mass production.

The proposal makes the argument that on the Internet, a single instance may be published widely.  True; and this is an argument for yet another value that could be assigned.  However, this does not change the fact that for the past 500 years of Western history the distinction between "printed" and "manuscript" material has been a useful one, and that retaining this distinction conveys useful information to the scholars who use our collections and on-line catalogs.  The differentiation was also felt to be useful for music and maps, and it would be interesting to hear whether these communities want to lose this distinction as well; in large catalogs such as Harvard's it is useful to distinguish easily between a manuscript score and a published one.

We also think that the term "text" is likely to be as confusing to users of the catalog, and believe that most systems, particularly at institutions with very small collections of manuscript or archival material, will choose to continue to use the familiar display constant "books," or a book-like icon, for "a".

3. 008 for Books.  It is interesting that while this proposal suggests that codes "e" (for manuscript map), or a new code for manuscript music in position 23, would be useful, they make no similar proposal for manuscript "text."  We would be satisfied with the proposed changes if there were an additional code in the 008 to allow Harvard to distinguish manuscript from printed text. Again, for single manuscripts, using the 008 for Mixed Materials would not make sense.

It is wrong to assume, as the original MARBI 97-7 seemed to assume, that all manuscript material can 1) be described archivally, or 2) is part of a collection, or 3) can be described as part of Mixed Material.  In many repositories, there are single manuscripts that 1) have not been described "archivally" using APPM, and are not going to be redescribed; 2) are not part of a collection, and 3) are not part of Mixed Material collections.

While it has been irritating to us to see our records for single manuscripts come up as "books" in the on-line catalog following format integration, this can and will be changed, because format integration does maintain the usefulness of the distinction between printed and manuscript material, and  distinctions in the fixed fields that allow our systems to display such distinctions.  However, if we lose the means of distinguishing manuscripts from other forms of "text," we and researchers who are trying to locate all pertinent manuscript material in the Harvard catalog--or any online catalog--will not be able to do so.  Our fear is that researchers will soon come to associate "mixed materials" with archives and manuscripts (as that will be how many of our collections will be coded) and will overlook all the single manuscript "text"s that pertain to their research.

We urge that SAA, in redrafting its proposal, preserve our ability to group similar materials--single manuscripts, and manuscripts in collections--together in our online catalogs.  We believe that "manuscriptness" should continue to be expressed in the Leader/06, as it is a required element, and contains data already used by online systems in retrieval.  Moving information about manuscripts from a required field in the Leader would necessitate major programming changes before we could get the systems functionality that we now have, and the archival community is small enough that it would be difficult to get priority for that kind of programming.

Sincerely yours,

Harvard University Library Council, Manuscripts and Archives Committee:

Larry Dowler, Harvard College Library
Tim Driscoll, Andover-Harvard Library
Laura Linard, Baker Library
Robin McElheny, Harvard University Archives
Leslie A. Morris, Houghton Library
Eva Moseley, Schlesinger Library
David Warrington, Law School Library