On the uses of encyclopedias

I’ve finally finished reading Andrew Brown’s A Brief History of Encyclopaedias. I continued to find it both interesting and annoying all the way through. (It’s less than 120 pages long, but it took me a few days to read it because I kept falling asleep. That’s not the book’s fault; I was inexplicably sleepy all weekend.)

Here’s one particular thing that it left me thinking about:

Brown repeatedly talks about alphabetical order as if it were a sort of fragmentation method, a way to disconnect related topics from each other, a kind of randomization system. As if you start out with a coherently structured set of articles and then you fragment them by putting them in alphabetical order.

Which baffled me. To me, the entire point of alphabetical order is to make it easier for readers to find the material that they’re looking for. It certainly results in some things that are semantically related to each other not being anywhere near each other in the encyclopedia, but I feel like that’s not a major problem if you’re using the encyclopedia as a reference work, looking up specific topics (and following cross-references) rather than trying to read the whole encyclopedia front to back.

But Brown quoted various people who’ve been involved with making encyclopedias who seemed to share his idea about the evils (albeit possibly necessary ones) of alphabetical order.

So I’m left in the probably untenable position of feeling that some creators of encyclopedias didn’t understand what encyclopedias were for. Which probably means that I’m wrong.

But I think I can retreat to a safer position by making a more general statement:

When you’re writing something, it’s a good idea to consider who your audience is and what modes of interaction with your work you want to make easier.

(A bunch of the works that Brown describes in this book are not things that I would consider to be encyclopedias at all. Several of them were (if I understand right) designed to teach rulers how to rule well. Some of them were vast compendia of quotations. One of them was a sort of thesaurus—a collection of phrases that you could use as synonyms for each other so as not to repeat yourself when writing. Those are all interesting kinds of works, but they’re not collections of general-interest articles on a wide range of topics, written for ordinary people to look stuff up in.

But here again, some encyclopedia makers have disagreed with me about what an encyclopedia is or should be. In the first edition of Britannica, for example, there was a 39-page article on “Farriery,” apparently intended to teach farriers how to do their jobs. That seems to me to be at odds with the purposes of an encyclopedia, but its presence in what I would have considered the canonical example of an English-language encyclopedia suggests that I’m wrong.)

Anyway, there are also other aspects of the problem that I feel like Brown ignores. For example:

  • He doesn’t talk at all about taxonomies and such—about the structure of a collection of articles on a variety of topics. He seems to take for granted that organization by subject is better than alphabetical order, but he ignores questions like “How do you decide which subjects are connected to each other?” and “How do you organize a collection of subjects that are connected to each other?”
  • He ignores the fact that in a linear medium like text on paper, you have to put things in some linear sequence. If you’re ordering by topic, how do you decide which topic comes first, which second, and so on? There are lots of good answers to that question, but if you don’t have some answer, you’re going to end up with a random-feeling ordering.
  • He mentions indexes in passing, but doesn’t talk about their value as a way to offer alphabetization for ease of reference even when the main body of the work is not in alphabetical order.
  • He talks a little about Wikipedia, but he neglects to say a relevant thing about electronic encyclopedias: you no longer need to put their entries in alphabetical order, because now readers use search (and follow links) to find the topics they’re looking for. The entries are stored in a database somewhere (not necessarily in any particular order that would make sense to a human), and it’s up to the search interface to obtain them and show them to the reader.

…I suppose a shorter version of my post here would be to say that I wish that Brown had consulted a librarian and/or a tech writer. :)


(P.S.: I know that lots of people use alphabetical order as a tool for serendipity of browsing (especially in paper dictionaries)—if you look up a given word, you might come across another interesting word nearby. I like that too, but for me most of the value of that serendipity can be had by telling the computer to show you a random entry.)

(P.P.S.: I also know that some people do read through encyclopedias front-to-back. But I don’t feel like that’s the main intended use for an encyclopedia.)

Join the Conversation