Learn how to build a data-driven, knowledge-based enterprise. Register for KMWorld today!

Alphabetical order

This article appears in the issue October 2005 [Volume 14, Issue 9]


   Bookmark and Share

By David Weinberger

Mortimer Adler devotes much of A Guidebook to Learning: For the Lifelong Pursuit of Wisdom, to railing against alphabetical order. He would hate what's going on now.

Adler, who died at 98 in 2001, is chiefly remembered for two huge accomplishments: the Great Books of the Western World series, published in 1952, and the Propaedia, published in 1984, both published by the Encyclopaedia Britannica. Both attempt to give the reader a sense of how ideas connect, which is exactly what alphabetical order does not do.

The Great Books brought together--chronologically--25 million words of 443 works by 76 authors in 52 volumes. In addition, the set included two volumes called Syntopicon, which Adler personally oversaw at great expense. He and his team of 30 indexers and 60 clerical helpers spent 10 years identifying 102 Great Ideas in the Great Books, broken down into 2,987 topics. The team then carefully connected all the dots so that if you wanted to learn about "Justice," for instance, you would be sent to all the places in the 52 volumes where the idea was mentioned.

The series must have sounded like a Great Idea itself but the initial reception was icy. Dwight MacDonald excoriated it in a review in The New Yorker that faulted Adler for his exceptionally boring selection and his refusal to provide introductory guidance to the works. As MacDonald said, Galen was important in the history of medicine, but no one is going to be turned onto the classics by reading Galen's endless and hopelessly wrong nostrums. More important, MacDonald points out how arbitrary and unhelpful is the Syntopicon. For example, under the section on Justice you'll find pointers to Plato and to a single passing quote in Tristram Shandy without any indication that one is more substantial than the other.

Adler's Propaedia, 32 years later, attempts to achieve the same goal. It explicitly attempted to help readers get past the randomness of alphabetical order of The Encyclopaedia Britannica. The Propaedia breaks knowledge into 10 categories and goes down seven levels. It took eight years to complete. Adler explicitly maintained that the list of topics should be thought of as circular, rather than as hierarchical.

Adler was hardly the first person to object to alphabetical order. The earliest: Medieval attempts to pull together knowledge avoided it--with some exceptions--because it failed to represent the grandeur of God's organization of the world. The great French Encyclopedia of the second half of the 18th century ran into its share of doubters for the same reason. In fact, its creators--rationalists and religious skeptics--liked alphabetical order because it kept them from having to build a huge unit on theology just to stay out of trouble.

Samuel Coleridge, another famous alphabetization foe, notably critiqued the Encyclopaedia Britannica as a "huge unconnected miscellany ... in an arrangement determined by the accident of initial letters." Coleridge's own unalphabetized encyclopedia failed.

Obviously, alphabetization isn't the be-all and end-all of organization. Its strength is precisely that it is arbitrary and meaningless ... and that we train our children on it even before they can read. But Adler's argument against alphabetical encyclopedias seems to me to be wrong-headed in two directions. First, encyclopedias already contain entries that synthesize knowledge and show its interrelationships: The entry on Religion gives a broader context than the ones on Judaism, Episcopalianism and Tom Cruise.

Second, great teachers do what Adler attempted, and we love them for it. But the task of institutionalizing and centralizing a single synthesis, such as the Syntopicon or the Propaedia, is bound to fail. When the French encyclopediasts drew the tree of knowledge, they put Theology under Science, and attached to it branches that included superstition, divination and black magic. French theologians of the day would have drawn the tree radically differently. Adler's articulation of the structure knowledge is better read and wiser than the one that you or I would draw, and it has real value. But the fact is that knowledge doesn't have a single natural structure.

The complaint against alphabetizing knowledge isn't totally without justification. If all anyone knew came from reading an alphabetized list of topics, we'd be worried about that person's ability to pull ideas together. But that's generally not how encyclopedias are used. We look things up in them because we already have an interest that is itself part of a larger context. And good encyclopedia articles explain the context of the ideas they're discussing.

Further, take a look at Wikipedia. It is neither arranged by topic nor alphabetized. You find articles by searching for them. But it is hardly without guidance to offer about how knowledge connects. Wikipedia articles contain many, many links to information inside and outside Wikipedia. And category tags cluster articles in ways the readers find useful. Together, these provide a type of context that even Mortimer Adler couldn't.

Of course, it's up to us to click on the links.


David Weinberger edits "The Journal of the Hyperlinked Organization" , e-mail self@evident.com.


Search KMWorld

Connect