An analysis of spoken language discourse cannot be limited to merely the words used in the discourse and their grammatical arrangement. It is well known, for example, that speakers commonly gesture as they speak. These gestures often provide either independent information or important clues needed for a clear understanding of the message of the speaker. Pointing at a book when saying ‘this book’ identifies the book being discussed. Without the deictic gesture, the particular book being described might not be identifiable. Other gestures may provide information which is complementary to the spoken words. In either case, the addressee must combine the grammatically coded spoken information with the gestural information to produce a message which incorporates both. American Sign Language (ASL) is a language produced by gestures of the hands, face, and body. In general, these gestures have been analyzed as parallel to the gestures of articulators in the vocal tract. That is, they have been considered to be parts of the words and morphemes which make up grammatical constructions. Our analysis of a brief ASL narrative shows that, even in the case of a language produced by gestures of the hands, body, and face, the distinction between ‘grammatically coded meaning’ and non-grammatically coded ‘meaningful gesture’ is equally appropriate. That is, sign language discourse also consists not only of grammatically structured arrangements of signs, but also includes meaningful gestures. The interweaving of these two sources of meaning is so extensive in ASL that certain categories of signs are not normally produced without a deictic gesture.