Utilizing an ethnically diverse snowball sample of 31 current or former marijuana smokers, this paper uses recent data to update and expand Howard S. Becker's (1953) theory of using marijuana for pleasure. Although the main tenets of Becker's social-processbased theory of marijuana use have survived the test of time amazingly well, the data suggest that some significant aspects of the theory need to be updated. Most users become intoxicated the first time they smoke marijuana, many no longer have to learn that they are intoxicated through social interaction with other users, and most immediately find the experience pleasurable. Historical changes in marijuana "scenes," rather than fundamental weaknesses, necessitate revisions in Becker's theory.
KEY WORDS: Marijuana, Howard Becker, illicit drug use, drug use.
Howard S. Becker's (1953) theory of "using marijuana for pleasure" is highly regarded in the drug research field. It was one of the first to seek to understand drug use from a social-- process perspective. Becker challenged the dominant theories of the time, which hypothesized that the motivation for drug consumption was the result of some form of psychological predisposition or personal trait of the user. Building upon Lindesmith's (1947) research on heroin (Becker, 1953; see note 5), Becker stood the dominant theory of the day on its head, hypothesizing that the motivations for drug use do not precede the deviant act but follow it; they are learned, formed, and influenced via the social process of engaging in drug use. Lindesmith and Becker's pioneering theoretical contributions in this area led to a considerable body of subsequent social-process-based or "career" drug research (Hunt, 1997).
Becker developed a three-stage social interactive process through which all marijuana users must pass in order to be left "willing and able to use the drug for pleasure when the opportunity presents itself ' (Becker, 1953:236). He proposed that marijuana smokers must learn to: (1) smoke the drug properly, (2) detect that they are intoxicated from the drug, and (3) define this state of intoxication as a pleasurable event. Becker surmised that people would not be motivated to continue to smoke marijuana unless they were able to learn to do so pleasurably.
Drug researchers in the social sciences have relied on the scientific method to build their theories and literature. Replication is an important part of the social scientific method, but it is rarely attempted in the drug research field: "Attempts at replication are particularly important (and rarely attempted) when dealing with contextual models (given the fluidity of contextual reality)" (Hirsch, Conforti & Graney, 1990:498). As such, certain theories may suffer from inadequate replication. I believe that Howard S. Becker's (1953) landmark research on marijuana fits this description.
There is limited research replicating Becker's three-step model of marijuana use, suggesting that the process has changed or is more complex than originally suggested, and/or that the social context surrounding marijuana use has evolved. The only scholars to attempt a direct replication of Becker's work (Hirsch et al., 1990) suggest that a much greater proportion of first-time users become intoxicated during their initial encounter with the drug, a considerable discrepancy from the original study. Adler and Adler (1996) do not evaluate all stages of the theory, but they report that developmental issues in very young marijuana smokers do not allow them to pass through all three stages. Specifically, "tinydopers"-their term for marijuana smokers up to eight years of age-lack the self-awareness skills necessary to recognize that they are intoxicated, even though they outwardly "manifest clear physical symptoms of the effects" of the drug (Adler & Adler, 1996:232). Too few studies exist, and those that do exist contain enough variation and discrepancy to call for another replication to advance the theory. …