Aside from the maintainability of this technique, it sounds as though it's taking one of the least efficient routes towards doing its job. (I'm not a LISPer, and although I should stop and think about the code there until I grok it before commenting, I'm looking for a lazier answer.)

So, let's say we have two documents, and want to calculate their distance. Each document has 2,000 words. We can either scan each word in each document once, counting them in memory and processing a total of 4,000 words, or we can start with a word in one document, scan both documents for all occurrences of that word, move on to the next word, scan both documents for all occurrences of that word, and so on, for a total of ~N! scans. Given a choice between a really efficient function running in factorial time, or a longer, uglier function running in constant time ... I'd kind of prefer the latter.

Am I way off the mark here? Is there something I misunderstood?