Analyzing word frequency in six complete texts, a study investigated how vocabulary can be used to define texts. The texts included three stories from 5th and 6th grade readers, selections from literature anthologies for 8th grade and 12th grade students, and a magazine essay for adults. Results indicated that if particular words occur frequently in a text, they do so because the language requires it. Few words were found to occur in all of the texts; those that did were almost all function words, "be" forms, or pronouns. Analyses of word frequency lists revealed that some content words occurred more than others simply because they referred to common concepts. Overall, findings suggested that the wording of any text is not random and that, in fact, texts "self-control" their vocabulary. These results suggested that (1) readers can build dependable strategies for dealing with words common in a text but less common in the language as a whole; (2) authors and editors should concentrate on relating the content of texts to the audience rather than focus on controlling vocabulary through the use of word lists; and (3) teachers concerned with vocabulary development should focus on functional use of words in the context of real texts rather than resort to decontextualization lists or dictionary exercises. (JD)