This paper investigates some of the underlying assumptions prevalent in much of the research concerning the language patterns of black children and compares two competing research approaches: the deficit model, which assumes that black children from the ghetto hear very little language, much of it ill-formed, and that they are impoverished in their verbal expression; and the "difference" model, which holds that socially subordinate societies and language varieties are self-contained systems, neither inherently superior nor deficient. Also contained in this paper is an exploratory study of the Bernstein hypothesis, which pertains to restricted and elaborated language codes. This exploratory study observes college students and operationalizes slang as an instance of the restricted code. A statistically significant difference was found between the number of greetings given in the highest context condition and those given in the lowest context condition and between the number of slang phrases used in the highest context condition and those used in the lowest context condition. The findings suggest that context may be an important condition to include in a study that attempts to investigate language systems. (Author/RB)