The purpose of this paper is to outline a new theory of drug abuse, that is, the cultural-identity theory. It seeks to inform substance abuse etiology by understanding how individual (i.e., micro) and environmental (i.e., meso and macro) phenomena influence the construction of drug-related identities and drug abuse. The theory proposes that drug abuse is an outcome of a drug-related identity change process featuring three micro-level (personal marginalization, ego identity discomfort, and lost control in defining an identity), two meso-level (social marginalization and identification with a drug subcultural group), and three macro-level (economic opportunity, educational opportunity, and popular culture) concepts. Together, they describe a motivation for drug-related identity change and an opportunity structure for that change. The theoretical model contains 12 hypothetical relationships that describe the links between the micro, meso, and macro-level concepts. The paper begins with a review of leading theories of drug use and abuse that identifies a niche the culturalidentity theory proposes to fill. Suggestions for empirical investigation of the theory conclude the paper.
Sociology of Crime, Law, and Deviance, Volume 1, pages 233-262. Copyright O 1998 by JAI Press Inc. All rights of reproduction in any form reserved. ISBN: 0-7623-0282-8
234 TAMMY L. ANDERSON
INTRODUCTION Current data on drug abuse paints an increasingly dismal picture as we approach the twenty-first century. Both the adult (middle-aged people mostly in their 30s and 40s) and the school-aged drug abuser population has grown in size following years of relative stability. Currently, at least 4.6% of high school seniors are daily marijuana users, as are about 1% of eighth graders, suggesting that the average abuser is getting younger (Johnston, O'Malley, and Bachman 1995). Still other prevalence data estimate that 12.4 million people used illicit drugs in the past month during 1994 (SAMHSA 1995) and that in 1993, 2.1 million were heavy cocaine users and between 444,000 and 600,000 were heavy heroin users (Rhodes et al. 1995).
Constrained by the rhetoric and ideology of the latest "War on Drugs," discourse about the social problems associated with drug abuse in the last quarter of this century has suffered from political campaigns targeting less-problematic drug use, especially that of "softer" drugs (e.g., marijuana). Such thinking and policymaking overlooks the considerable scholarly evidence that distinguishes drug use and drug abuse as separate phenomena, suggesting different explanations for each. Males (1997, p. 5) recently explained:
Current drug policy ignores the lessons of the 1960s: Moderate drug, particularly marijuana, experimentation is normal and widespread among the young-and will abate without frantic suppression measures. However, the smaller number of habitual users of harder drugs and multiple drugs require urgent and focused attention regardless of their ages.
The current leading etiological theories concerning drug use do not effectively resolve this matter. They have tended to focus more on drug users rather than abusers or not to distinguish between the two. Also, they have depended on micro-level factors (individuals and small groups) which can result in pathologizing the problem or in further stigmatizing the population in question. The purpose of this paper is to offer a new theory-that is, the cultural-identity theory of drug abuse-that attempts to broaden this literature.
Unlike the leading etiological theories (see below), the cultural-identity theory is specifically concerned with drug abuse. It also attempts to link three levels of analysis (i.e., micro, meso, and macro) for a more comprehensive explanation of drug abuse. This kind of approach departs from most existing work, which usually has elaborated explanatory factors at one level of analysis, especially the micro level, instead of exploring and/or specifying links between phenomena measured at the micro, meso (mid-range social group), and macro (broad sociocultural) levels that affect lived experience on a daily basis. As such, it promises to increase our understanding of how the relationship between individual and environmental factors explains drug abuse and drug-related identity change so that improved prevention and treatment strategies can be forthcoming.
A Cultural-Identity Theory of Drug Abuse 235
A REVIEW OF LEADING ETIOLOGICAL THEORIES Etiological reviews (Lettieri, Sayers, and Pearson 1980; Hawkins, Catalano, and Miller 1992; Petraitis, Flay, and Miller 1995) or special journals devoted to this topic (e.g., Journal of Drug Issues, Spring 1996) have identified the following eight theories as dominant in the field: (1) problem behavior theory (Jessor and Jessor 1977; Jessor, Donovan, and Costa 1991; Donovan 1996); (2) the theory of reasoned action (Ajzen 1985; Ajzen and Fishbein 1980; Fishbein and Ajzen 1975); (3) social learning theory (Akers 1977; Akers and Lee 1996; Bandura 1977); (4) social control theory (Hirschi 1969; Kandel 1980, 1985, 1996); (5) selfderogation theory (Kaplan 1975, 1996; Kaplan, Martin, and Robbins 1984, 1986; Kaplan and Johnson 1992); (6) the integrated delinquency model (Elliott, Huizinga. and Ageton 1985); (7) social development theory (Catalano and Hawkins forthcoming; Hawkins and Weis 1985; Catalano et al. 1996); and (8) the theory of triadic influence (Flay and Petraitis 1994). The following paragraphs briefly review four of these theories (i.e., self-derogation theory, the integrated delinquency model, social development theory, and the theory of triadic influence) in an attempt to forge links between various individual and more environmental factors, integrate ideas from the first four listed theories, and offer a useful comparison to the cultural-identity theory. I
Four matters frame this review. They are limitations of extant etiological theory that the cultural-identity theory attempts to address, including confusion surrounding drug use and drug abuse, failure to link the micro, meso, and macro levels of analysis, and an overreliance on narrowly defined peer group concepts and behavioral outcomes.
Drug Use versus Drug Abuse
The persistent focus in etiological theory on drug use instead of drug abuse may be due, in part, to complications that arise from attempts to identify a large enough pool of drug abusers through survey research methods that employ a prospective design with national population samples. Scholars investigating the four theories mentioned above have typically employed this kind of methodology. Such population studies tend to focus on drug use while clinical studies focus more on drug abuse (Institute of Medicine 1996). The result is that most studies of psychosocial etiological factors do not distinguish between the two.
Moreover, there seems to be a tendency in etiological theory to use the words "use" and "abuse" together or interchangeably. Many also attribute credibility to the "gateway" theory of drug use (i.e., that the use of "softer" drugs eventually leads to the abuse of "harder" ones) without discussing the processes involved in that transition. This assumption or oversight is repeatedly called into question by other research. For instance, the Institute of Medicine (1996, p. 117) has noted that it may be incorrect to assume that the factors involved in the initiation of drug use
236 TAMMY L. ANDERSON
are important in the escalation of it. Also, Johnston and colleagues show large discrepancies between monthly or annual use and daily use rates (which can serve as a proxy for abuse) and suggest that only a small portion of eighth graders, for instance, who use drugs (about 21 % in 1995) would likely become daily abusers of them by twelfth grade (4.6%), all else being equal (Johnston, O'Malley, and Bachman 1995). Furthermore, Waldorf, Reinarman, and Murphy (1991) reported that the NIDA surveys, for example, show that the majority of cocaine u 'rs in the U.S. are "ceremonial" users (i.e., they use it occasionally) and do not become abusers. Drug prevention strategies would be greatly improved if "risk" or "protective" factors that distinguished between drug users and drug abusers in the larger population could be identified.
The cultural-identity treats drug use and abuse as separate phenomena. The theoretical tenets described below are meant to explain how certain environmental and individual factors interact to influence drug-related identity change for people who abuse or reach crisis points with drugs. The theory does not propose to do the same for those who use drugs in a non-abusive or unproblematic fashion, which several studies (Waldorf et al. 1991; Granfield and Cloud 1996) have shown is possible over an extended period of time. It would, therefore, be inaccurate to call the cultural-identity theory an explanation of drug use. On the contrary, the theory seeks to address the scholarly inattention to diverse etiologies for drug use and abuse by simply focusing on drug abuse.
Linking the Micro, Meso, and Macro Levels
Etiological theory's overreliance on micro-level influences has hindered its ability to understand and, therefore, to predict how macro-level or sociocultural factors cause drug abuse. The scarcity of work that offers a theoretical or empirical account on the link between micro and macro phenomena further speaks to this matter. Elsewhere, U.S. ethnographers (e.g., Bourgois 1989, 1996; Hamid 1991 a, 1991b, 1991c; Waterston 1993; Dunlap and Johnson 1992) and scholars from the Birmingham School of Cultural Studies (Brake 1985; Clarke et al. 1976; Hebdige 1979; Willis 1976, 1977; McRobbie 1991) have described how various macrolevel phenomena, like economic and social inequality, and social marginalization result in many social ills, including drug abuse. However, this more macrofocused work has not yet explained how it is that most people who experience such conditions do not become drug abusers. Recent theories like Flay's theory of triadic influence stand out as an exception and are discussed in more detail below.
The cultural-identity theory attempts to bridge these two literatures. It elaborates on micro-, meso-, and macro-level phenomena and the links between them. It seeks to avoid overly reductionist and deterministic claims by insisting on an integrative environmental and individual explanation that guards against a micro or macro-level bias. Furthermore, an approach like this could result in more a more integrated and comprehensive prevention strategy consisting of several
A Cultural-Identity Theory of Drug Abuse 237
programs at different levels (i.e., schools, families, and communities) instead of single programs located at one level of analysis.
Peer Groups versus Drug Subcultures in Abuse Etiology
Third. these four theories, especially the work of Kandel, Elliott, and Dishion, pinpoint peer groups as a major explanatory factor in drug use. However, the question of how peer culture operates to influence drug use versus drug abuse is left unanswered. Kaplan et al. (1986) have noted that the peer-based research has not effectively addressed the role of peer groups in the transition from drug use to abuse. This kind of inquiry is important for drug prevention policies, especially recently, since Gorman (1996) has noted the influence of the peer group variable in prevention policies.
The cultural-identity theory focuses on drug subcultural groups (e.g., potheads, dopers, gangs, etc.) instead of peer groups, which highlights identification with specific social groups and the patterns of activity among them (see also Fishkin et al. 1993; Mosbach and Leventhal 1988). It also addresses the subcultural meaning attached to specific drugs (see McRobbie 1991; Hebdige 1979; Willis 1976), which transcends any particular set of individuals and persists over time. These drug-related meaning systems and the identities that youth create in group settings may be the more important factors for drug abuse etiology and prevention.
Identity versus Behavioral-Oriented Variables
Fourth, most etiological theories focus solely on predicting behavioral outcomes-for example, frequency or amount of drug use. The cultural-identity theory studies, in addition, the acquisition of drug-related identities and other identity changes during involvement with drugs-for example, drug-related identity change (see Anderson 1993; Anderson and Mott 1998). Behaviors like actual drug abuse are conceptualized as part of the identity change process from non-drug user to drug abuser.
Interactionists exploring deviant behaviors and careers have noted the importance of identity change in the initiation, persistence, and termination of drug addiction (Ray 1968; Biernacki 1986; Becker 1963; Pearson 1987; Waldorf 1983; Hawkins and Wacker 1983; Jorquez 1983; Waldorf et al. 1991), alcoholism (Denzin 1987; Brown 1991), crime (Shover 1983; Miesenhelder 1982; Schmid and Jones 1991), mental illness (Goffman 1961), and obesity (Dehger and Hughes 1992). This work consistently underscores the notion that "undesirable" behaviors escalate with increased "deviant" identification and lesson with identity change toward non-deviance.
The cultural-identity theory acknowledges this relationship between identity (e.g., drug-related identities) and behavior (drug abuse). It speculates that drug related identification may ultimately distinguish drug use from drug abuse. Future
238 TAMMY L. ANDERSON
empirical work seeks to determine if drug-related identity change mediates the relationship between such behaviors. This approach allows for a deeper understanding of the drives toward drug abuse and my offer new approaches to drug abuse prevention strategies.
Kaplan's Self-Derogation Theory Of the eight etiological theories mentioned above, the cultural-identity theory
has most in common with Kaplan's theory of self-derogation. Kaplan (1975,
1996) and colleagues (Kaplan and Johnson 1992; Kaplan, Robbins, and Martin
1984, 1986) maintain that self-derogation plays a central role in determining drug
use and abuse. For them, negative feelings and statements about oneself and the
socially devaluing experiences that set it up motivate individuals to behave in ways
that minimize self-derogation and maximize positive self attitudes. They propose
that this explains why individuals reject the normative structure and embrace that which is "deviant" (e.g., drug use, drug peers, and drug subcultures). Culturalidentity theory both compliments and enhances these premises. It proposes that negative self-evaluations are part of the etiological process, but articulates a specific mechanism through which such negative self-evaluations lead to drug abuse. It is two social and largely external factors (i.e., personal and social marginalization) that help produce this identity discomfort and can lead to drug-related identity change. Kaplan and colleagues have not delineated the same. These theoretical differences about identity or self-definition and the sources of it could account for an important risk factor that distinguishes drug users from abusers.
The most significant difference between self-derogation theory (and the other four theories discussed below) and the cultural-identity theory, however, pertains to the presence of meso- and macro-level concepts in the explanatory model. Kaplan (Kaplan and Johnson 1992; Kaplan 1996) has recently discussed more macro-level influences (e.g., social controls) on drug abuse, but his theoretical model does not yet contain specific concepts and, therefore, does not directly discuss a link between them. Finally, another major difference between the two is Kaplan's focus on "deviant" acts (see Kaplan 1996) rather than identities and identity change.
Elliott's Integrated Delinquency Model Elliott et al. (1985) have proposed an integrated sociological theory of drug use that draws from social control theory (Hirschi 1969), strain theory (Merton 1938, 1957), and social learning theory (Akers 1977). They posit that strong bonding with "deviant" peers is the primary cause of drug use. "Deviant" peer bonding, they maintain, is a result of weak conventional bonds with parents and school, prior delinquent behavior, and social disorganization.
A Cultural-identity Theory of Drug Abuse 239 Given the recent significance of the peer group concept in drug prevention efforts (Gorman 1996) and in etiological research (see Ennett and Bauman
1991; Kandel 1996), it makes sense to explore more about the ways in which youth peer groups influence or cause undesirable behavior in individuals. Elliott's focus on bonding with "deviant" peers differs substantially from the cultural-identity theory's emphasis on drug subcultural groups. It is important to understand whom individuals interact with (e.g., who become their friends), because research and theory has repeatedly shown that such associations and attachments are predictors of behavior. This was part of what Bandura called "modeling"; individuals often model or do what people around them do-for example, use drugs. However, both drug users and abusers usually report bonding with "deviant" peers (see Kandel 1996 for a discussion of the overstatement of the impact of "deviant" peers in drug research). Therefore, the overall meaning-system that the peers embody and the new definition of the self that individuals get from interacting in "alternative" youth subcultures might comprise the more important explanatory factor and may ultimately distinguish between drug use and drug abuse. For instance, Ennett and Bauman (1991) and Harton and Latane (1997) have recently noted the importance of the "social approval" function that peers play in the etiological process instead of an individual's association with them.
Hawkins and Catalano's Social Development Theory Social development theory is also an integrated approach that combines social learning, control, and differential association theories. It has much in common with Elliott's approach, with the exception that it elaborates on the developmental processes in both pro-social and anti-social (e.g., drug use) behavior. Bonds develop between the individual and socializing agents (family, schools, religious and community institutions, and their peers) during development. An individual's behavior will, therefore, be antisocial or pro-social depending on the behavior, norms, and values of those socializing agents to which he or she is bonded.
Like Kaplan's and Elliott's theories, the general model of social development theory does not specify micro-, meso-, and macro-level variables. Instead, Hawkins, Catalano, and colleagues focus on the individual and his or her perceptions about opportunities and interactions in the environment around them. Jessor and Jessor (1973) made an earlier observation on the possible ways in which factors outside the individual (environmental or "contextual") may impact behavior. They concluded that the perceived environment was a more important predictor of individual behavior than the "actual" or "objective" one. Social development theory is in agreement on this matter.
240 TAMMY L. ANDERSON
Flay's Theory of Triadic Influence Flay and Petraitis' (1994) theory of triadic influence (TTI) is a complex web of factors that prior research has tied to drug use. To date, TTI is the only drug use theory that acknowledges the various streams (i.e., cultural, social-situational, and interpersonal/biological) that affect all human behavior. Furthermore, within each causal stream, Flay and colleagues identify three levels of influence-proximal, distal, and ultimate. Doing so has provided them with a theoretical model that can investigate sophisticated kinds of causal relationships-for example, mediating and moderating ones-on drug use. This is a second quality of TTI that distinguishes it from the theories described above.
The cultural-identity theory differs from TTI in two major ways. First, TTI does not specify which variables or relationships predict drug use versus drug abuse. Second, TTI is a theory about behavioral outcomes. It is not a statement about the drug-related identity change process that begins with no drug use for many and ends in drug abuse for some.
A CULTURAL-IDENTITY THEORY OF DRUG ABUSE The cultural-identity theory proposes that drug abuse is an outcome of a drugrelated identity change process that is set in motion by three micro-level (personal marginalization, ego identity discomfort, and lost control in defining an identity), two meso-level (social marginalization and identification with a drug subcultural group), and three macro-level (economic opportunity, educational opportunity, and popular culture) concepts'-. These concepts describe a motivation for drugrelated identity change and an opportunity structure for that change. The theoretical model contains 12 hypothetical relationships, which are diagramed in Figure 1.
Hypothesis 1. Personal marginalization has a direct and positive effect on ego identity discomfort (before drug use or during early childhood and adolescence).
Hypothesis 2. Personal marginalization has a direct and positive effect on lost control in defining an identity.
Hypothesis 3. E,go identity discomfort (before drug use or during early childhood and adolescence) has a direct and positive effect on lost control in defining an identity.
Hypothesis 4. Lost control in defining an identity has a direct and positive effect on identification with a drug subculture.
A Cultural-Identity Theory of Drug Abuse 241
Hypothesis 5. Ego identity discomfort (before drug use or during early childhood and adolescence) has a direct and positive effect on identification with a drug subculture.
Hypothesis 6.Identification with a drug subculture has a direct and negative effect on later ego identity discomfort (i.e., during drug abuse).
Hypothesis 7. Economic opportunity has a direct and negative effect on social marginalization.
Hypothesis 8.Educational opportunity has a direct and negative effect on social marginalization.
Hypothesis 9. Popular culture has a direct and positive effect on identification with a drug subculture.
Hypothesis 10. Social marginalization has a direct and positive effect on identification with a drug subculture.
Hypothesis 11. Social marginalization has a positive correlational relationship with ego identity discomfort (before drug use or during early childhood and adolescence).
Hypothesis 12. Earlier Ego identity discomfort (before drug use or during early childhood and adolescence) has a positive and direct relationship with later ego identity discomfort (i.e., during drug abuse).
To date, research on this theory has taken place in clinical-type settings, that is, with people involved in treatment programs and who had reached crisis points with drug abuse (Anderson 1991, 1994, 1998a, 1998b, 1998c; Anderson and Mott 1998). Therefore, the theory started with observations from more clinical samples, and current work on it attempts to develop it in order to explain the larger population of drug abusers. Still other components of the theory come from literature reviews on drug abuse and addiction (Anderson 1995).
Cultural-identity theory uses a multi-dimensional definition of drug abuse that distinguishes it from drug use. This definition features the following: (1) a pattern of regular and heavy use over a significant period of time, (2) a set of drug-related problems (at work or with interpersonal relationships, one's own health, and formal social control agencies), (3) previous and failed attempts to terminate drug consumption. and (4) self-identification as having a drug and/or alcohol problem.
The theory maintains that the concepts and hypothesized relationships are significant for drug abusers. It may be that drug users experience some part of this process but may exit it at different points. They may also experience high "risk" on some concepts after other "risk" factors have disappeared. This is a conclusion that Waldorf et al. (1991) reached with heavy cocaine users. The cultural-identity theory does not propose to explain progressions in drug use behaviors per se, such
A Cultural-Identity Theory of Drug Abuse 243 as simple frequency or pattern changes in the consumption of drugs, or movement from no drug use to drug abuse. However, the change in behavioral indices of no drug use and drug abuse are implied by the theory.