"Last, there has been some spectacularly misguided and misinformed discussion of the need to create 'master records' for works that are manifested in different physical forms. It is hard for me to believe that this notion has been put about by people who are cataloguers. Let me spell it out. Descriptions are of physical objects (and, nowadays, of defined assemblages of electronic data). It is literally impossible to have a single description of two or more different physical objects…"When I first read this aside in Michael Gorman's highly charged article in opposition to the cataloging rules that would succeed the AACR2 rules (that he edited), I was shocked that anyone would say that cataloging is primarily a "description of physical objects." I thought of library catalogs as being about content, about knowledge. But as Gorman surely has a finely honed grasp of the purposes behind library cataloging, it seemed best not to dismiss such a statement, and I marked it in my copy of the book and tucked it away in my memory.
Michael Gorman, AACR3? Not! in: Schottlaender, Brian. The Future of the Descriptive Cataloging Rules: Papers from the Alcts Preconference, Aacr2000, American Library Association Annual Conference, Chicago, June 22, 1995. Chicago: American Library Association, 1998. p. 27
It has come back to me as I've pondered not only RDA and the state of library catalogs, but in my attempts to explain library cataloging to non-librarians. At the meeting regarding the question of bibliographic metadata and copyright, I described library cataloging as analogous to a medical diagnosis: a great deal of testing, expert knowledge, and judgment result in a few scribbled lines in a medical file and a prescription. If you consider these latter two the "metadata" of the situation, you see that what is visible is merely the tip of the iceberg, with a great deal of intellectual activity hidden below the surface. What I didn't mention at that time, because it wasn't yet clear to me, was the role of observation in the two activities being compared. A good physician knows how to observe and analyze a patient, and a good cataloger knows how to observe and analyze a cultural artifact.
Cataloging rules actually instruct their users on how to observe. In fact, the very first rule in AACR2 (1.0.A) defines the sources of information for the catalog entry: the preferred source of information is always the thing being described. In essence the thing itself is the primary informant for the catalog record.
There are a couple of important things we can conclude from this. The first is that the act of cataloging is an act of describing what is being observed. This makes cataloging something like the act of a biologist who is describing a specimen before her. In theory, if both librarians and biologists follow the rules of their disciplines, the same specimen or artifact would be described similarly by two different professionals. (In fact, there are always edge cases that defy simple application of the rules, but these are also the cases that make the professional activity interesting.)
The next important aspect about library cataloging is that the content of the catalog record is in large part the expression of those who created the artifact itself, not that of the cataloger. As RDA (chapter 2) says:
"The elements reflect the information typically used by the producers of resources to identify their products—title, statement of responsibility, edition statement, etc. "Significant parts of the cataloging description are either quotes from the thing or paraphrases of observable content. I am unaware that anyone has ever challenged the right of catalogers to copy this information from the artifact to the catalog record.
What I have addressed to this point follows a fairly strict definition of "descriptive cataloging" and presumably not terribly far from the division between description and access that is made by RDA, or from the division of AACR2 into description and headings. The access or heading portion of the catalog record adds information beyond the observations of the physical piece. Headings and access points are standardized forms of proper names (including persons, corporations, government bodies, and some titles). The standardized form of the name serves as an identifier for the named entity, and also normalizes display. Here are a few examples:
On the artifact: J. R. R. Tolkien
Heading: Tolkien, J. R. R. (John Ronald Reuel), 1892-1973
On the artifact: Beethoven's Ninth Symphony
Heading: Symphonies, no. 9, op. 125, D minor
On the artifact: T. C. BoyleNote that some of these add more information than is on the actual piece, and that additional information requires research. That doesn't mean necessarily that the additional information requires the level of creativity that qualifies it for copyright protection. This is one of the areas of bibliographic metadata that needs to be analyzed further. However, I think we can conclude that some portion of "descriptive cataloging" consists primarily of observations about real world objects; and some portion normalizes those observations to create standard identifiers for bibliographic entities that exist entirely independently of the cataloging act.
Heading: Boyle, T. Coraghessan.
You have undoubtedly noticed that I have not mentioned subject headings or classification in this post. Subject analysis, although recorded on the same catalog entries as the bibliographic data, is a separate activity in the cataloging workflow, and is not covered by the above-mentioned cataloging rules. As a topic in the "metadata and copyright" discussion it should be covered separately.
* My thanks to Tom Baker who, as I struggled to find another way to say "bibliographic description" suggested that catalogers make observations about things.