ACAT RDA and Library of Congress Subject Heading

Posting to Autocat

On 12/4/2015 7:24 PM, Loretta Hartsell wrote:
Do you think the Library of Congress Subject Headings will ever connect and parallel to keyword searching? And is the Library of Congress Subject Headings people working with RDA? In my viewpoint RDA is the template for coding and organizing information so the patron can better access and use it. LCSH is just a component within a cataloged RDA bibliographic record. But I don’t see where all of this is coming together to work for us and the patron.

This is a really great question, and in many ways the most important question of all. But as with many really great questions, it packs within it lots of questions at once.

When broken apart, the questions are:

  1. do all academic and professional fields need the same type of subject analysis?
  2. What about subject strings vs. subject descriptors?
  3. LCSH subject strings are pre-coordinated vs. post-coordinated strings, or in other words, the subjects and their subdivisions are in a strict order. Does this still hold today?
  4. What is the utility of cross-references and authorized forms today?
  5. What are the consequences of the “single search box” that libraries are implementing more and more, which include massive numbers of materials that do not–and never will–have LC Subject headings or often–any subject forms at all?
  6. What is happening with subjects in the rest of the information world such as “relevance ranking”? How has that changed user expectation?

I am sure there are other questions as well. I won’t discuss all of those questions here, but I’ll just make a couple of observations.

In other postings, articles and podcasts, I have gone to some trouble to demonstrate that the subject heading strings that we use were designed by our predecessors for a completely different information environment–an environment many of us can hardly imagine today. I suspect that in just a couple of decades, that original information environment will be practically impossible for those people of the future to imagine.

Within that original, 19th-century environment the subject heading strings worked more, or less, successfully. (Our predecessors understood the shortcomings and discussed them at some length) Because the informational environment has changed in such fundamental ways, many of those assumptions no longer hold true and it is up to us to adapt what we make–and what we have always made–to the environment we now inhabit.

In some fields, such as business or the hard sciences, resources made long ago are often considered less important than those made today. Whether that is true or not, it definitely does not hold in other areas where materials created long ago are just as vital as ever. The reason I point this out is that in some fields such as astronomy, what was written in the 18th or 19th century have been made essentially obsolete by more recent discoveries. As a result, when people are searching in those areas, if the earlier results are more difficult to find and/or access, it is not considered a problem. This has a major consequence for people searching certain topics today, who do not want to be bothered by seeing things from the 1800s or even the 1940s or 1960s, but in other realms, materials from those times may be just as important as ever.

As a concrete example, someone wanting to know about Venus probably wants to know the latest information today and is much less interested in Galileo’s “Starry Messenger” (written in 1610) which is of less informational value today, and can actually get in the way of those who want “relevant” information about Venus. But people who want to understand the Renaissance must read some of the writings from that same time, and one of those “must-read” works could be Galileo’s book. While some materials may still be current in some fields, in others they clearly have been superseded.

This may be clear enough, but the problem is: how do you make a single tool, i.e. a library catalog that all people can use today, especially as more and more people come to expect “relevance” ranking? It’s always been a problem and is probably even more difficult today.

Fortunately, analysis shows that people still want topical (i.e. subject) access to information. (See the talk by the head of Google Scholar and my analysis at This fellow found that people search for topics much more than just names and titles. Therefore, lumping both needs together within a single catalog leads to many problems.

If libraries are to make their subjects really useful again–and I think they can and should!–it will take a lot of work and deep consideration. Turning them into URIs won’t solve much.