The loss of function words and grammatical morphemes in agrammatism has been investigated by a number of researchers. Different theories have been put forth by Goodglass (1968), Kean (1977), and Grodzinsky (1984), as well as several others. One of the problems they faced was that these function words and morphemes did not form a natural class in syntactic theory. The three theories mentioned will be reviewed. The implications of a new theory of syntax (Fukui 1986) for the study of agrammatic speech will then be examined in the hopes that they will encourage further research in this particular area.