Log in

19 April 2009 @ 07:59 pm
Class Warfare 2--The Empire Strikes Back  
Back on August 10, I wrote a post called "Class Warfare," about cataloging (classification) and how it relates to fiction, specifically mine. Today, with the recent debacle on Amazon fading away, this topic is suddenly relevant, and to more than just me and Phyllida.

The Amazon mess, for anybody who missed it, equated books categorized as "gay" or "lesbian" with erotica ("adult content") and temporarily stripped them of their sales ranking, making them impossible to find through subject searches. In my earlier post, I discussed how Amazon seems not to have a "Romance—bisexual" subject heading, which meant that Phyllida was lumped into the "Romance—gay" category, leading to bad feelings from some readers.

What is often misunderstood in discussions of classification and cataloging is the role of the cataloger. As an article in today's "Ideas and Trends" column in The New York Times says:

"It wasn’t the first time that a technological failure or an addled algorithm has spurred accusations of political or social bias. Nor is it likely to be the last. Cataloging by its very nature is an act involving human judgment, and as such has been a source of controversy at least since the Dewey Decimal System of the 19th century. Today, the Internet is built on complex algorithms that categorize things and that are inevitably imperfect."

Now, I'm not saying this lets Amazon off the hook—quite the opposite. But I am saying that outsiders (non librarians) don't realize that the individual library cataloger doesn't get to make up the categories. A comment on my Aug. 10 post compared me to Linnaeus, the famous Swedish taxonomist of the 18th century who developed the system, much modified but still in use today, by which all zoology, from protozoa to primates, is arranged. And in the interest of honesty I had to explain that no, I don't create the categories in my work. For each book, I simply try to find the best fit among existing subject headings.

I work in a science library, which makes things a little easier than working with Everything in the Universe, like in a general public library—or Amazon. But in any library, there is what is called "controlled vocabulary," the subject headings that must be applied consistently to avoid chaos. Subject headings vary between classification schemes. The Library of Congress headings are not the same as the Sears list, for example. But any library that uses the LC subject headings (LCSH) will use the same headings for the same topics or things.

If you're looking for books on Coleoptera, the LCSH subject heading is the common term, Beetles. Type in Coleoptera for a subject search, and the catalog gives you the "See" reference, telling you that in the catalog, "Coleoptera" is not used, and to look under "Beetles" instead. Now that catalogs are online, you don't have to sigh, slam the C drawer shut, pull out the B drawer and start all over again. You just click on the hyperlink.

Does this dampen catalogers' creativity? Undoubtedly. But what's the alternative? Chaos. As "The Public Editor" column put it (NY Times again), writing on style manuals and usage conventions (which is part of the cataloger's work as well):

"Many Times readers do get offended and irate over style issues … and the complaints often involve an accusation that the newspaper is being disrespectful.… Each case illustrates the challenge of maintaining a consistent style in a changing world, where some people read political motives into simple usage conventions, where words once thought acceptable become objectionable, and where other words once objectionable become part of everyday language. A newspaper has to have rules, the linguistic equivalent of driving on the right side of the road and stopping at red lights, to avoid chaos for readers."

Working in a science library does save me from a lot of this kind of complaint. I have yet to get any irate letters from beetles. But in this age of increasing recognition of animal cognition, I expect any day to hear from a chimpanzee annoyed that her proper (Linnaean) scientific name, Pan troglodytes, isn't a subject heading, but instead directs users to the common "Chimpanzees."

It's when we get to human beings that things get rough. The subject heading "African-Americans" has gone through any number of variations over the years, as the acceptable terms changed. One interesting example is "American Indians," still in use and resistant to the more politically correct "Native Americans." For one thing, many of the people in question prefer the "Indians" heading. But it leads to bizarre spin-off headings like "Indian art." No, that's not art from India the country. If you want that you must search under "East India."

And then, as I mentioned before, there's the untamed wilderness of fiction. As I said, most libraries and bookstores don't classify literary fiction, and even the genre fiction is given very broad categories, like Romance and Sci-fi. These aren't subject headings in the way that Chimpanzees or Beetles are. LC doesn't even use these genre terms for the works of fiction themselves, but only for nonfiction works discussing them (in catalogers' terms, books that are about romances or sci-fi). For example, the Library of Congress record for Phyllida used only the one LC heading, "Bisexuals—Fiction." Not "Romance," and not even the specific phrase heading, "Bisexuality in marriage."

Readers on Amazon can assign their own "tags" to books that don't seem to fit well within Amazon's controlled vocabulary. Of the idiosyncratic tags given to Phyllida, my personal favorite is "all gay guys love Phyllida." I realize this was probably meant derisively, but there's something sweet about it. And it does recognize that the book's main character gets along quite well with the gay and bisexual men in the story. Part of her appeal is that she likes them as they are and has no interest in changing them.

But as a subject heading, "all gay guys love Phyllida" doesn't work. A good subject heading has to be potentially applicable to more than one book and to works of other authors. Even if I write a sequel, chances are it won't focus on Phyllida alone or discuss "all gay guys" and their "love" for her. If I am motivated enough to write a whole series about Phyllida, she might conceivably get her own heading: Carrington, Phyllida (fictional character). But that would be the extent of the editorializing.

Cataloging by human beings takes a lot of time and thought, and few libraries allow as much as we'd like. We catalogers have to make decisions based on the cover or jacket copy, the table of contents and an introduction—and even that is pushing it. Most catalogers have quotas. They have to catalog an absurdly high number of books per week, and can't actually read them, any more than the typical librarian's job involves sitting at a circulation desk reading everything in the library while typing catalog cards.

Yes, catalogers make mistakes, but people programming computers make worse ones, especially when applied hastily over enormous databases of diverse material and with ulterior motives. A good cataloging scheme has a broad range of well-defined categories and subject headings, like LC's. By contrast, Amazon's "Internet … algorithms that categorize things … are inevitably imperfect." In a library, staffed by human beings, a book that is a bisexual love story would not have to be crammed into the awkward fit of "gay romance;" and the children's classic Heather Has Two Mommies would not be marked as containing "adult content."

None of this excuses Amazon's latest idiocy. It was a programmer tinkering with an "algorithm," no doubt at the behest of the management, if not Jeff Bezos himself, who made the original "ham-fisted error." What this incident does point up is, in the Internet age, when many people think libraries and the faceless, sexless drudges who work in them are unnecessary, just how much we need them.

Libraries and librarians don't censor. We recognize the existence of nuance, like bisexual, alongside gay and lesbian. We understand the difference between fiction and nonfiction, between pornography and romance, between genuine adult content and children's books. And if you want to find a gay romance novel, we don't ask why, or check your ID, or say, "Sorry, we don't approve of books like that and we're not going to help you find the ones that might have inadvertently slipped through the acquisitions process." We say, "Well, the romance novels are over there, but they're not classified. Here's a reading list compiled by librarians, and you can check the catalog to see if we own them. And have you read…"
Artaxastraartaxastra on April 20th, 2009 07:59 pm (UTC)
I worked in the library for years as an undergrad, and I totally see what you mean. Now, that it's my own books being catalogued, I admit that I find some of the cataloging baffling and occasionally bizarrely inappropriate to content, but that's the nature of the beast!
ann_amalieann_amalie on April 20th, 2009 09:53 pm (UTC)
Thanks for understanding :)

I know this post was long and boring, but there were all these articles in the Times on Sunday that discussed the intersection of cataloging as I do it and the Amazon crap, and I had to write it all out of my system.

I've worked at cataloging for over ten yers now, and a lot of it still baffles me. And even when I get it, a lot of it makes me crazy.

But the alternatives really are worse, as the Amazon thing showed.