Sociology Compass 5/4 (2011): 298–310, 10.1111/j.1751-9020.2011.00363.

Sociological Perspectives on Addiction

Darin Weinberg*
University of Cambridge

Abstract
This article provides a critical survey of sociological research on addiction. It begins with the sem-
inal research of Alfred Lindesmith on heroin addiction then proceeds through discussions of func-
tionalist contributions, research that exemplifies what David Matza called the ‘appreciative’ turn in
the sociology of deviance, rational choice theories, and social constructionist approaches. It is con-
fined to research on addiction in its original meaning as putative enslavement to a substance or
activity rather than merely deviant or disapproved activity more broadly. As will be seen, though,
there is a ubiquitous and theoretically interesting tendency even among those who contend to be
writing about addiction as such to slip into modes of analysis that effectively substitute questions
regarding the social approval of an activity for questions concerning whether it is voluntary or
involuntary. Hence, one purpose of this article is to explore whether, and how, this slippage
might be avoided.

Introduction
In sociology, addiction has been approached from several distinct theoretical vantage
points. Regrettably, the term has often been used interchangeably with other terms
including deviant drug use,1 drug misuse, and drug abuse. Such imprecision results in a
confusion of questions concerning the social approval of various sorts of drug use with
questions concerning whether this use is voluntary. Much of the history of social policy
concerning psychoactive drugs has been predicated, at least ostensibly, on the claim that
these substances possess unusual powers over people and must be regulated to protect cit-
izens from their own personal proclivities to succumb to addictive use. If we are not able
to distinguish claims regarding the putative morality of drug use from claims regarding
people’s ability to control their use, we are poorly equipped to effectively evaluate the
history of policies predicated on the notion that people need protection from putatively
addictive substances. We are also poorly equipped to evaluate social research that either
endorses or rejects this idea. If it is to have any meaning at all, the term addiction cannot
be considered synonymous with terms denoting voluntary drug use.2
This article will be confined to research that speaks to addiction as such rather than
disapproved drug use more broadly. As will be seen, though, there is a ubiquitous and
theoretically interesting tendency even among those who contend to be writing about
addiction as such to slip into modes of analysis that effectively substitute questions regard-
ing the social approval of drug use for questions concerning whether it is voluntary or
involuntary. Hence, one purpose of this article is to explore whether, and how, this slip-
page might be avoided. The article is divided into five sections. The following section
outlines and evaluates critically the seminal contributions of Alfred Lindesmith to the
sociology of addiction. Next I address the work of major functionalists who have sought
to theorize addiction. I then consider the various perspectives on addiction developed by

ª 2011 The Author

Sociology Compass ª 2011 Blackwell Publishing Ltd
 Sociological Perspectives on Addiction 299

those who exemplify what David Matza (1969) dubbed the ‘appreciative’ tradition in the
sociology of deviance. In the further section, I consider the efforts of theorists who have
sought to subsume addiction into the rational choice model of social action more gener-
ally. Finally, I discuss the important contributions to the sociology of addiction made by
social constructionists inspired by the work of people like Michel Foucault and Bruno
Latour. I conclude with a brief synopsis of my own position.

Alfred Lindesmith: the father of the sociology of addiction

The earliest and, by far, most influential sociological research concerned specifically
with addiction was conducted by Alfred Lindesmith (1938, 1940, 1947, 1968). More
than 70 years after its original publication, his theory remains the classic sociological
theory of addiction (Akers 1992; McAuliffe and Gordon 1974; Stephens 1991;
Weinberg 1997a, 1998). Lindesmith’s groundbreaking achievement was to show that in
addition to affecting the body biochemically, drugs, and the experiences their ingestion
produce, are meaningful to people in ways that cannot be reduced to the interactions
of a chemical agent with a human physiological system. Hence, above and beyond
biochemistry, Lindesmith persuasively argued that to understand addiction it is indis-
pensable to consider addicts’ subjective perceptions of drugs, drug effects, and their
wider social lives. Lindesmith noted that users who acquired heroin on the street were
often vulnerable to addictive patterns of use, but those who had been administered
opiates in hospital settings were not so vulnerable. He explained this by suggesting that
whereas both hospital and street users experience physiological withdrawal symptoms
upon cessation of use, only street users become consciously aware of the fact that the
source of their distress lies in their heroin deprivation. Lindesmith argued that by using
drugs specifically to alleviate withdrawal, mere drug users were transformed into
genuine drug addicts.
While Lindesmith’s theory retains its canonical importance, it has been subject to sev-
eral serious critiques. Most fundamentally, his theory relies on an outdated division of
human perception into: (1) brute biological sensations the body passively experiences in
immediate response to its physical environment, and (2) the mind’s active and deliberate
interpretation of those sensations. This voluntaristic understanding of meaning and inter-
pretation profoundly undermines Lindesmith’s capacity to theorize addiction as a loss of
self-control, or as something suffered rather than chosen (Weinberg 1997a). Moreover,
Lindesmith also presumes that physiological withdrawal distress is a necessary pre-requisite
for the onset of addictive behavior. His work reflects the wisdom of an era wherein it
seemed sensible to speak of distinctions between hard drugs (defined by the fact that they
produce physiological withdrawal symptoms) and soft drugs (defined by the fact that they
do not do so). But the era has now passed when people could speak confidently of a dis-
tinction between drugs that produce genuine, or physical, addictions and drugs that pro-
duce only a more nebulous psychological addiction. The single most important catalyst to
this era’s passing was the advent of crack. Crack cocaine is widely recognized as extre-
mely addictive by clinical professionals and non-professionals alike but, oddly enough, it
produces no gross physiological withdrawal symptoms (Gawin 1991).3 The same can also
be said of nicotine, and all of the so-called behavioral addictions like sex, gambling, eat-
ing, etc.
Furthermore, Lindesmith’s reliance on the distinction between physical and psychologi-
cal addiction always suffered another serious analytic problem: relapse. Many consider
relapse, or the resumption of a dis-preferred pattern of behavior despite one’s desire not

ª 2011 The Author Sociology Compass 5/4 (2011): 298–310, 10.1111/j.1751-9020.2011.00363.x

Sociology Compass ª 2011 Blackwell Publishing Ltd
300 Sociological Perspectives on Addiction

to, as the defining mark of addiction. Theories that trade on the distinction between
genuine physical addiction and a less severe psychological addiction cannot remain consis-
tent in their explanations of relapse. Lindesmith explained the resumption of drug use
after withdrawal symptoms have ceased primarily in terms of the former addict’s subcon-
scious generalization of the response to withdrawal distress to other forms of stress (cf.
Lindesmith 1968, 154). This theory is plainly residual in the sense that it pastes a new
subconscious mechanism onto the original physiological withdrawal-plus-knowledge-
of-withdrawal theorem. Moreover, it is not consistently supported by empirical data on
opiate addiction (cf. Robins 1993), and affords no explanation of relapse into the use of
substances like nicotine or crack, which do not produce gross physiological withdrawal
symptoms in the first place. Given the analogous tendency of former crack and nicotine
addicts (to say nothing of behavioral addictions like eating, gambling, and sex) to relapse,
we are well advised to look beyond the generalization of withdrawal distress to ade-
quately understand this process.

The functionalist input

During the mid-twentieth century, functionalist sociologists offered a variety of theories
of addictive behavior that departed in important ways from Lindesmith’s seminal work.
Seeking wholly social structural explanations, these theories shared in common a depar-
ture from Lindesmith’s presumption of a necessary physiological component to addiction.
But in so departing, these theories also slipped from analyzing addiction as a loss of self-
control into analyses of merely deviant drug use.
In his famous essay ’Social Structure and Anomie’, Robert Merton (1938) suggested
that chronic drunkards and drug addicts might exemplify the retreatist adaptation, one
of his five modes of adjustment whereby people adopt ostensibly deviant patterns of
action. According to Merton, addicts could be understood as individuals who believe
in the propriety of both the cultural goals and the institutionalized procedures society
affords for achieving those goals but who cannot produce the desired results by socially
sanctioned means. The result is a retreat from social life into ‘defeatism, quietism, and
resignation’ (Merton 1938, 678). This proposition was developed in the 1960s by Clo-
ward and Ohlin (1960) in what became their influential ‘Double Failure’ hypothesis
regarding addictive behavior. Contra Merton, Cloward and Ohlin suggested addicts
were not opposed to adopting illegitimate means of achieving legitimate cultural goals,
but rather were incapable of using even these means for securing social rewards. Hence,
addicts were double failures in the sense that they failed to achieve by either legitimate
or illegitimate means. Heavy drug use was held to alienate putative addicts from both
mainstream and delinquent sub-cultures thus further reducing opportunities for social
success.
Some functionalists moved beyond explanations of the distribution of addicts across
social structural positions to consider the social psychological processes that motivated
addictive patterns of alcohol or drug use. The best known of these was ‘normative
ambivalence theory’. Normative ambivalence theory argues that drinking or drug prob-
lems will result when, in a culturally heterogeneous society, agents are bombarded with
competing normative orientations to the use of alcohol and ⁄ or drugs. Under such condi-
tions the agent, be it a person or social group, will be unable to make unequivocal
recourse to a consistent normative code concerning the use of alcohol or drugs and will
thus oscillate between approval and disapproval of a given pattern of drug or alcohol
use.4 In his critique of this approach, Robin Room (1976) brilliantly shows how, in the

ª 2011 The Author Sociology Compass 5/4 (2011): 298–310, 10.1111/j.1751-9020.2011.00363.x

Sociology Compass ª 2011 Blackwell Publishing Ltd
 Sociological Perspectives on Addiction 301

ideological and theoretical climate of mid-20th century American sociology, ambivalence

became a chosen theoretical commitment largely without need of supporting empirical
evidence. He wrote,
In essence, then, our argument is that, given a societal model with only two levels, the individ-
ual and the whole system, and given an assumption that norms operate only to constrain indi-
vidual deviant behavior, an explanation of the occurrence and continuation of deviance which
is not completely individualistic – i.e., in terms of an inherently defective deviant individual –
must be made in terms of a repeated defective interaction involving the norm – an explanation
for which ambivalence provides a convenient and apparently supraindividual and explanatory
rubric. (Room 1976, 1053)
There are, however, several significant difficulties with this explanatory rubric. First, it
begs many questions regarding exactly how, when, and why norms come into conflict
and presumes that normative conflict is necessarily a source of stress for those who do
experience it. Secondly, it uncritically assumes that human behavior is somehow directly
determined by social norms, a position that has been cast in considerable doubt (cf. Bour-
dieu 1990; Garfinkel 1967; Wittgenstein 1958). Furthermore, even if it is allowed that
under certain clearly specified conditions normative conflict does produce stress, we are
still left to explain why this stress, in its turn, produces chronically deviant, let alone
involuntary, behavior. Because existing normative ambivalence theories afford us no
guidance with respect to these questions, I agree with Room when he suggests a healthy
skepticism regarding their value as explanations of addiction.
According to functionalists, apparently addictive behavior patterns are best regarded as
eminently rational, if painful and socially notorious, adaptations to social structural condi-
tions. Functionalist approaches tended to stereotype addicts as necessarily socially disad-
vantaged and to sometimes confuse the trappings of poverty with the trappings of
addiction. Moreover, they remained conspicuously silent on the question of whether
addictive behavior ought to be understood as involuntary or merely socially disapproved,
largely forsaking attention to the former in favor of the later. But they did have the virtue
of freeing sociological research from the presumption of a brute biological basis for addic-
tion and of allowing sociologists to entertain the possibility that people might experience
drug problems simply as a result of how they had learned to use these substances to cope
with the social structural circumstances of their lives.

The appreciative turn

As part of a broadly based turn against functionalism in the second half of the 20th
century many sociologists began to distance themselves from what David Matza (1969),
in his book Becoming Deviant, dubbed the ‘correctional’ perspective found in functionalist
theories of deviant drug use and adopt what he called an ‘appreciative’ analytic stance lar-
gely informed by ethnographic and other qualitative types of data. Noting that modern
societies were much more fragmented and conflicted than functionalists had usually
allowed, these researchers advocated an agnostic moral regard for putatively dysfunctional
or deviant behavior and an effort to empathize with those labeled deviant. No longer was
it assumed that behavior reviled by the mainstream was necessarily problematic for those
who themselves engaged in the behavior. Nor was it any longer assumed that the social
mechanisms according to which these behaviors were produced and sustained need reflect
a functional breakdown of either the individual or his or her society (cf. Becker 1953;
Preble and Casey 1969; Sutter 1966).

ª 2011 The Author Sociology Compass 5/4 (2011): 298–310, 10.1111/j.1751-9020.2011.00363.x

Sociology Compass ª 2011 Blackwell Publishing Ltd
302 Sociological Perspectives on Addiction

Instead of investigating the mechanisms of functional breakdown, qualitative sociologists

of deviance turned their attentions to providing richly detailed descriptions of drug cultures
(cf. Bourgois and Schonberg 2009; Finestone 1957; Johnson 1980; Rosenbaum 1981;
Rubington 1968), the social settings of drug activity (cf. Sutter 1969; Wiseman 1970), the
ritual practices attendant on drug use in natural settings (cf. Waldorf et al. 1991; Williams
1992), and the self-identities of drug users (cf. Denzin 1993; Lindesmith 1968; Ray 1961).
These studies form the tip of an enormous iceberg of naturalistic investigations of drug use
and drug users that have vastly enriched our understanding of the meanings attendant on
drugs and drug experiences for those involved in these worlds. They have thereby
extended and elaborated upon Lindesmith’s seminal insight that drugs, and the experiences
their ingestion produce, are meaningful to people in ways that cannot be reduced to
the interactions of a chemical agent with a human physiological system. However, so
concerned were these researchers to avoid the appearance of moralism, they often over-
looked the fact that the loss of self-control over drug use is often taken seriously by drug
users themselves and is not necessarily imposed from without by official agencies of social
control. Indeed, only a small fraction of these studies actually speak explicitly to the nature
of addiction itself. Furthermore, those studies that actually do broach the nature of addic-
tion tend to render it somewhat counterintuitively and ⁄ or in ways that do not jibe well
with much of the available empirical evidence (Room 1984; Weinberg 1997a, 2000,
2002).
A closer look at an exemplary study will demonstrate the point. Ray (1961) is the most
widely cited theoretical statement of how relapse can be understood from a symbolic
interactionist vantage point. Stephens (1991, 57–8) cites Ray’s theory as the primary
resource in his own symbolic interactionist account of relapse. Waldorf (1970, 229) cites
Ray’s theory as the ‘only … attempt to learn anything of the processes of relapse after a
period of abstention’, though he expresses some reservations as to the theory’s explana-
tory power. According to Ray’s theory, the cycle of abstinence and relapse should be
understood as a process during which the former user consciously oscillates between a
commitment to his using and non-using self-concepts. Ray concludes his classic essay,
… socially disjunctive experiences bring about a questioning of the value of an abstainer iden-
tity and promote reflections in which addict and non-addict identities are compared. The
abstainer’s realignment of his values with those of the world of addiction results in the redefini-
tion of self as an addict and has as a consequence the actions necessary to relapse. (Ray 1961,
140)
This theory suggests relapse is a process necessarily involving conscious deliberation and
comparisons between one’s using and non-using identities. As one can see, Ray’s is an
extremely cognitivist, rationalist, and indeed, voluntarist construal of the relapse process.
The relapser is an individual who rationally evaluates the pros and cons of being an addict
versus being an abstainer before deciding to relapse. But does this sound like addiction? If
Ray is correct, what sense is there in thinking that addiction is involuntary, that addicts
are in any way disabled by their addictions, or require any kind of therapeutic assistance
whatsoever? The main difficulty is that Ray provides no place for the visceral compo-
nents of relapse – the embodied and emotional urges a great many people report experi-
encing, to which they themselves very often assign primary responsibility for their
relapses, and which constitute the primary justification for therapeutic assistance. Though
we must by no means adopt our informants’ accounts uncritically (Weinberg 2000), we
must contend better than does Ray with the activities and accounts of those whose lives
and experiences constitute our only sources of data.

ª 2011 The Author Sociology Compass 5/4 (2011): 298–310, 10.1111/j.1751-9020.2011.00363.x

Sociology Compass ª 2011 Blackwell Publishing Ltd
 Sociological Perspectives on Addiction 303

Denzin (1993) is the only symbolic interactionist who has sought to explicitly intro-
duce the lived body and emotions of the substance user into his account of addiction. He
is thus the only symbolic interactionist who has tried to move beyond the disembodied
cognitivism, and its attendant implications of voluntarism, that uniformly mark other
symbolic interactionist contributions to our theoretical understanding of addiction. In his
theory of the ‘alcoholic self’, Denzin suggests that alcoholics suffer from an ‘emotionally
divided self’ wherein ‘the self is divided against itself’ (Denzin 1993, 362). While I would
contend this appreciation of the role of emotion in the addiction process is a significant
advance over previous efforts, it does not succeed in fully overcoming the voluntarism
characteristic of earlier symbolic interactionist approaches. Voluntarism slips back into
Denzin’s theory in his insistence that situations which induce the alcoholic to drink are
consciously evaluated as such on the basis of a ‘fully grounded interpretive system’ (Den-
zin 1993, 67) or ‘lay theory of alcoholism’ (Denzin 1993, 64–8). According to Denzin’s
model, craving, to the extent it is an instance of emotional experience, must involve,
(1) a sense of the feeling in terms of awareness and definition, (2) a sense of the self feeling the
feeling; (3) a revealing of the moral or feeling self through this experience. (Denzin 1983,
404)
Denzin’s model proposes that addictive craving necessarily involves, in addition to one’s
imperious desire to use, a reflexive interpretation of this desire as such, a reflexive
acknowledgment that one is the particular self who is experiencing the craving, and some
type of moral evaluation of that particular self. This model reintroduces the conscious
voluntarism that prevented Ray from grasping the spontaneous visceral character of crav-
ing and forces us to conceive of relapse as always preceded by a reflexive interpretation
of oneself and one’s emotions. This interpretive work is construed not just as a possible
component, but as an analytically necessary component of the relapse process. This theory
disregards the profound power of what George Herbert Mead called ‘the affective side of
all consciousness’ (quoted from Denzin 1984, 423) which, though it occurs beneath the
level of deliberate interpretation, nonetheless steers us in much of our practical life. And
like Ray’s theory, it ultimately requires that relapse be viewed as a deliberate, voluntary
and quasi-rational rejection of the recovering alcoholic identity in favor of the resumption
of drug use (Denzin 1993, 286–7). It thereby fails to explain how people could ever
experience relapse not as a deliberate decision to resume drug use but as a steadfast com-
mitment to a non-addict identity and lifestyle that is painfully and persistently thwarted
by powerful visceral compulsions to use.

Rational choice theories of addiction

I have argued that neither functionalist nor symbolic interactionist theorists have managed
to avoid the implication that addictive behavior reflects a conscious, deliberate, self-gov-
erned, quasi-rational, and, hence, voluntary choice to use drugs. Some economists and
social psychologists actively embrace this position. While rational choice theory has con-
tributed very few genuinely new insights into the mechanisms governing addicted behav-
ior and has been criticized both for its rigid formalism and inconsistency with the
empirical facts (Rojeberg 2004; West 2006), it has nonetheless served to clarify the argu-
ment that addiction manifests nothing other than the cost–benefit calculations of addicts.
Micro-economic theorizing on addiction was inaugurated in earnest by Becker and
Murphy (1988) who argued that addictive behavior reflects a choice made in full

ª 2011 The Author Sociology Compass 5/4 (2011): 298–310, 10.1111/j.1751-9020.2011.00363.x

Sociology Compass ª 2011 Blackwell Publishing Ltd
304 Sociological Perspectives on Addiction

knowledge of its future costs and benefits and simply reflect actors’ preferences and their
wider assessments of their circumstances. Most controversially, Becker and Murphy
(1988) argued that actors’ preferences remain stable over time and that while it may entail
severe costs, this behavior is assessed by the addictive actor as the best among his or her
possible alternatives. As Orphanides and Zervos (1995) point out, this model yields a pic-
ture of the addict without regret and, because s ⁄ he already enjoys perfect foresight,
beyond the reach of any instruction as to the dangers of the addicted path. Orphanides
and Zervos (1995) seek to remedy this consequence by denying the addicted actor is fully
aware of the consequences of embarking upon this path. While introducing the possibility
of learning, regret, and inadvertently falling into addiction, this is all predicated on the
initial ignorance of the actor as to the outcome of their initiating a potentially addictive
course of behavior rather than inconsistent preferences or a failure of rationality.
Others have suggested that the assumption of consistent preferences cannot be sustained
in the case of addiction. Instead, they have introduced a model of the addict as prone not
only to prefer present to future rewards in a consistent pattern but also to ‘hyperbolically
discount’ the prospects of future costs and benefits (cf. Ainslie 1992). Essentially, this
means that we should think of human actors not as fully future oriented cost–benefit ana-
lysts but as often vulnerable to temptations against their better judgments. The greater the
prospect of immediate reward and the greater the expected intensity of that reward, the
harder we find it to remain attentive to our longer term plans. This approach forsakes the
orthodox rational choice model, recasting cost–benefit calculation so as to better accom-
modate the fact that costs and benefits are often valued very differently depending on
their perceived imminence. Others associated with the choice theoretic tradition have
sought to better attune considerations of cost–benefit calculation to both its external envi-
ronmental cue and internal visceral impulse sensitivity (cf. Elster 1999).
Most now agree the argument that addictive behavior is consistent with the axioms of
orthodox rational choice theory is not very credible. While modifications to the ortho-
dox theory have marginally increased the credibility of choice theoretic arguments, even
modified arguments remain tenuously grounded in empirical data and have been criti-
cized for merely providing formal descriptions of idealized narratives of addiction rather
than explaining actual empirical instances of it (Rojeberg 2004). While it has offered little
in the way of positive insight, by explicitly formalizing the thesis that addictive behavior
is the product of rational and voluntary cost–benefit calculations, this tradition has served
to draw into relief the stark difficulties that remain more tacit in other social scientific
traditions that have construed addictive behavior as more or less deliberate and self-gov-
erned. In my view, it is in this light that the broader analytic benefits of this literature are
most properly construed.

Social constructionist contributions

In one sense, all social scientific contributions to our understanding of addiction are social
constructionist insofar as they seek to identify social forces that influence the emergence
and assessment of behavior deemed addictive. In this section, however, I confine my
attention to studies that seek, more radically, to argue that addiction is not only influ-
enced by social factors but is also fundamentally a culture-bound phenomenon – that it is
unintelligible outside the nexus of cultural practices and beliefs within which it is found
(cf. Keane 2002; Room 1985; Rudy 1986; Schneider 1978; Weinberg 2002, 2005; Wie-
ner 1981). Levine (1978), for example, argued the rise of the Temperance movement
reflected a larger cultural revolution that demanded heightened levels of self-control,

ª 2011 The Author Sociology Compass 5/4 (2011): 298–310, 10.1111/j.1751-9020.2011.00363.x

Sociology Compass ª 2011 Blackwell Publishing Ltd
 Sociological Perspectives on Addiction 305

individualism, and accountability to the demands of a capitalist economy all of which

were found incompatible with certain patterns of heavy alcohol use. Hence, the political,
economic, and cultural changes that comprised the transition to Modernity provided the
context in which Temperance ideology, and in particular the notion of ‘addiction’, could
take root and appear reasonable to large numbers of people (see also Alexander 2008;
Ferentzy 2002; O’Malley and Valverde 2004; Reinarman 2005; Valverde 1998).5 Immen-
sely valuable social constructionist studies have also been published on the activities and
influence of science in the modern history of addiction (cf. Acker 2002; Campbell 2007;
Gomart 2002). Other social constructionists have looked at the role played by the crimi-
nal justice system in giving shape to not only the contemporary status of addiction as a
concept but also the lives of those identified as addicts (cf. Duster 1970; Lindesmith
1965; Reinarman & Levine 1997). These studies highlight the importance of acknowl-
edging that the concept of addiction has very often been used to legitimate the stigmati-
zation, marginalization, and persecution of drug users whether or not they have lost
control of their drug use. Such studies leave little doubt that criminalization has itself
caused the lion’s share of the suffering now associated with drug use. However, they
rarely if ever speak to the question of whether some drug users sometimes do in fact lose
control of their drug use, let alone how this might be best understood sociologically.
Some social constructionist studies of addiction approach the phenomenon from a less
historical and more ethnographic and ⁄ or biographical perspective. Based on one of the
most incisive and richly nuanced studies ever conducted with cocaine users and quitters,
Waldorf et al. (1991) show that problematic patterns of drug use as well as difficulties
quitting are overwhelmingly associated with those who lack what they call a ‘stake in
conventional life’. People who have something to lose tend to mitigate the damage drug
use introduces into their lives better than those who do not. As crucial as this insight
remains, it begs the question of whether addiction ought to be construed as simple bad
judgment fostered by difficult circumstances or as a kind of syndrome, something seem-
ingly more alien to us than our own faculties of judgment and by which it is possible to
feel genuinely afflicted. Though I think both scenarios do occur, I don’t think the former
warrants the label ‘loss of self-control’. Therefore, to the extent we wish to theorize how
a learned pattern of behavior can emerge into a syndrome over which we feel little if any
direct control (and by which we can experience affliction), we need a theoretical frame-
work that does not necessarily reduce all learned behavior to a singular faculty of judg-
ment, subjectivity or self.
I would argue that just such a framework is provided by the post-humanist tradition of
social research (Weinberg 2008). The humanist tradition pervades Western thought as it
pertains to politics, morality, law, art, and is very clearly evident throughout the sociolog-
ical research on addiction that I have considered here. Adherents to the humanist tradi-
tion embrace an axiomatic commitment to the principle that the irreducible atoms of
social life must inevitably consist only of human subjects – integrated, intentional, delib-
erative agents possessing interests and investments in cultural frameworks that endow their
worlds with meaning, value, and distinctive rationalities. Post-humanists worry that,
among other things, this imagery reifies human nature and neglects the extent to which
scientific studies of primates, cyborg technologies, and artificial intelligence modify what
it means to be human and, indeed, encourage the breakdown of decisive conceptual
boundaries between the human and the non-human (cf. Haraway 1991; Hayles 1999).
Others suggest that whatever humans might be, our conceptual and practical regard for
them is not fixed but intimately tied to the specific networks of interacting entities, both
human and non-human, that provide for their substantive stabilities and accountabilities

ª 2011 The Author Sociology Compass 5/4 (2011): 298–310, 10.1111/j.1751-9020.2011.00363.x

Sociology Compass ª 2011 Blackwell Publishing Ltd
306 Sociological Perspectives on Addiction

(Callon 1986; Knorr-Cetina 1997; Latour 1996; Pickering 1995). Finally, because they
eschew axiomatic conceptual commitments regarding the essential identity of any things,
including human subjects or selves, post-humanists advise studying situated practical
action directly for clues as to how the identities of things are forged, sustained, modified,
fragmented, influential, or forgotten in any actual case.
I have found these ideas profoundly instructive in my own research on addiction (cf.
Weinberg 1997a,b, 2000, 2005). Quite unlike the largely rationalized depictions of
addicted behavior we find in most social scientific research on addiction, the people I
studied did not uniformly depict addiction as a self-governed activity but often character-
ized it as a deeply troubling and mysterious loss of self control.6 This can be seen, for
example, in the following fieldnote excerpt from my research with homeless people pre-
sumed to suffer from addictions,
I’ve promised myself I wouldn’t use a thousand times and really meant it. And then I use. I
mean it’s like there are two sides of me. The rational reasonable person who knows he’s gonna
die if he keeps on living the way he is and the insane one who just doesn’t care. My reasonable
side of me can be as sure as it wants to be but when those drugs appear in front of me the
insane one takes over and all those reasons I had not to use are just gone. They just disappear.
And I use. It’s like my mind just goes dead and my addiction takes over. I hate myself right
afterwards and I’m completely confused by the fact that I just used. I didn’t want to but I did.
It’s all well and good to say you need to make a commitment but for some of us that’s not
enough. We need something more than that and it doesn’t help for people to be all smug about
how we need to make a commitment and it’s all that simple.
Well known writers on addiction like Stanton Peele (1989) and John Davies (1992) have
suggested that such accounts are not valid descriptions of addiction but merely socially
functional for those who provide and ⁄ or believe them. They note that because they must
have been learned from watching others give them, these kinds of accounts must inevita-
bly reflect the speaker’s ‘vocabularies of motive’ (Mills 1940) rather than a genuine loss of
self-control. I dispute neither the claim that these accounts are functional nor that they
reflect conceptual commitments prevalent in the cultures to which putative addicts
belong. What I do dispute is the presumption that either of these things necessarily fore-
closes on an account being also descriptively valid (Gubrium and Holstein 2009; Haraway
1991). All accounts implicate their authors’ extant conceptual commitments and the uses
to which their authors’ are putting those accounts. However, that does not mean that we
cannot study how our research subjects collectively assess the extent to which they are
supported by empirical evidence. Moreover, the radical reduction of such accounts to
mere expressions of people’s preconceptions and ⁄ or their instrumental interests in provid-
ing them fails to explain either the phenomenology of addiction as a source of suffering
or the ways in which addictions manifest as consequential non-human agents in ongoing
practical action. A post-humanist understanding of addiction is eminently equipped to
overcome both of these serious failings.
Building upon Bruno Latour’s (2004) work on bodily articulation, Emilie Gomart
(2002, 2004) has done much to bring the insights of post-humanism to bear on the study
of addiction. She takes issue with the liberal philosophical antinomy between the human
subject as absolutely free and the human body as mechanically determined, conceptualiz-
ing drug use and addiction as vehicles for the articulation of human subjectivities rather
than inevitably destructive of them. Gomart and Hennion (1999) fruitfully compare drug
users with music aficionados to draw out the parallel processes through which people
learn and prepare themselves to be taken over by drugs or by music. Becoming able to

ª 2011 The Author Sociology Compass 5/4 (2011): 298–310, 10.1111/j.1751-9020.2011.00363.x

Sociology Compass ª 2011 Blackwell Publishing Ltd
 Sociological Perspectives on Addiction 307

succumb to the pleasure of drugs entails learning certain skills and subjective transforma-
tions and decidedly not a merely passive biochemical determinism (see also Becker 1953,
1967). This is quite right. Similarly, Gomart (2002, 2004) describes how in the Blue
Clinic, patients were active and tactical users of drugs rather than mere slaves to them.
Hence, their physiological dependence on substitution drugs is characterized as a ‘gener-
ous constraint’, a physiological constraint on their actions that allows clinic workers to
simultaneously coerce ⁄ seduce drug users into adopting more stable and less risky forms of
life. What is not addressed in these studies, though, and what I am seeking to explore in
my own work, is why and how addictions might become things that cause people to suf-
fer and from which we might, on occasion, seek to free them.

Concluding remarks
Hence, unlike Gomart, I do not conceptualize addiction as a relation between the drug
user’s body and a psychoactive drug. I have deliberately avoided this because the vast
majority of drug users do not become addicted and, as Gomart suggests, to characterize
the behavior of addicts as mechanically determined by drugs themselves is empirically
hopeless. Instead, following the lead of my research subjects, I conceptualize addictions as
non-human agents residing in the bodies of those who are addicted. In opposition to bio-
logically reductionist accounts, Latour (2004, 205) writes, ‘to have a body is to learn to be
affected, meaning ‘‘effectuated’’, moved, put into motion by other entities, humans or
non-humans. If you are not engaged in this learning you become insensitive, dumb, you
drop dead’. It is as a distinctive type of just this kind of ‘learn[ing] to be affected’ that I
want to conceptualize addictions as embodied agents. But whereas Latour (2004) appears
to suggest a normative judgment that any way in which bodies learn to be affected by
the world should be embraced, I want to be a bit more cautious. It seems to me that to
have a body is also to be vulnerable to disease (cf. Mol 2002; Mol and Law 2004). How-
ever, as a post-humanist, I regard diseases not merely as pathological biological mecha-
nisms, but, more generally, as patterns of harmful bodily articulation. That is, patterns of
bodily articulation with which we cannot, or do not want to, identify our selves precisely
because they afflict or endanger those articulations with which we do self-identify. If we
are clear that this is what is meant by disease, then I can see no reason to deny, and every
reason to affirm, that an addiction might be just such a thing.

Short Biography
Darin Weinberg is University Senior Lecturer in the Department of Sociology, Cam-
bridge University, and a Fellow of King’s College, Cambridge. Before coming to Cam-
bridge in 2000, he was an Assistant Professor of Sociology at the University of Florida.
His research focuses primarily on the practical purposes to which concepts of addiction,
mental disorder, and learning disability are applied in various historical and contemporary
contexts. He is particularly interested in how these concepts figure in state-sponsored
campaigns of social welfare and social control, and in what their uses reveal about how
and why people distinguish the social and natural forces held to govern human behavior.
His books include Of Others Inside: Insanity, Addiction, and Belonging in America (Temple,
2005), Talk and Interaction in Social Research Methods (edited with Paul Drew and Geoffrey
Raymond, Sage, 2006), and Qualitative Research Methods (edited, Blackwell 2002). He
received his PhD in Sociology from UCLA, an MSc in Social Philosophy from the Lon-
don School of Economics, and BA in Sociology and Communications from UCSD.

ª 2011 The Author Sociology Compass 5/4 (2011): 298–310, 10.1111/j.1751-9020.2011.00363.x

Sociology Compass ª 2011 Blackwell Publishing Ltd
308 Sociological Perspectives on Addiction

Notes
* Correspondence address: Darin Weinberg, Department of Sociology, University of Cambridge, Free School
Lane, Cambridge CB2 3RQ, UK. E-mail: [email protected]

1
Reference to ‘drugs’ is used in this article as a proxy for all putatively addictive substances and ⁄ or behaviors
whether or not they actually involve drugs.
2
To be clear, my effort here is not to theoretically define the objective nature of addiction once and for all but:
(1) to promote a greater terminological precision in empirical sociological research on addiction, and (2) to exam-
ine the idea that addictions, however defined, might sometimes be experienced as sources of human suffering.
Those who have sought to understand addiction as a type of affliction, or source of suffering, have overwhelm-
ingly done so by defining addiction as something that causes a ‘loss of control’, and so it is the sociology of
addiction as the loss of control that I am here most concerned to explore. To be sure, a great deal of important
sociological research has looked not at the loss of self-control but at the ways in which claims regarding the dan-
gers of addiction have figured in campaigns of social control. As vitally important as it is, this largely social con-
structionist research tends either to completely reject the claim that some people really do sometimes lose control
of their use of drugs and ⁄ or begs the question of how such a loss of control is best understood sociologically. By
extant social constructionist lights, we may sometimes suffer from being defined in one way or another. But the
notion that something other than human subjects, as defining agents, might be the source of suffering has been
more difficult to formulate within the social constructionist frame. My hope is to show that this need not be the
case.
3
By ‘gross physiological withdrawal symptoms’, I mean symptoms like vomiting, cramping, delirium tremens,
runny nose, itchy eyes, and so on, which implicate specific physiological effects of withdrawal. This is in contrast to
more psychological effects like anxiety or stress headaches which are less clearly linked to specific physiological
effects of withdrawal.
4
Though informed by symbolic interaction and using a qualitative approach to studying the meaning of drug
involvement, Ray’s influential theory of relapse to heroin addiction (Ray 1961) is in essence a normative ambiva-
lence theory. Likewise, some learning theories of drug and alcohol problems also exhibit strong affinities with nor-
mative ambivalence theories (cf. Akers 1992).
5
Rouse and Prabha Unnithan (1993) have contrasted responses to alcohol problems in the United States with
those in the Soviet Union and found that while, in the United States, responses are largely informed by Protestant
ideology, responses in the Soviet Union are informed by what they call a Proletarian Ethic. Both approaches, how-
ever, seem to hinge on efforts to discourage unproductiveness in the political economic system.
6
Of course, different commentators, including addicts themselves, often debated the characteristics of their own
and one another’s addictions as well as the degree of influence their addictions had over their personal conduct in
any given instance. But the fact that the characteristics and relative influence of people’s addictions on their
activities are intrinsically contestable hardly disqualifies them from being taken seriously as causal agents. Indeed,
post-humanist social theory teaches us that it is precisely in this resistance to unequivocal description and decisively
unilateral human control that non-human agencies are most robustly in evidence.

References
Acker, Caroline Jean. 2002. Creating the American Junkie. Baltimore, MD: Johns Hopkins University Press
Ainslie, George. 1992. Picoeconomics. Cambridge: Cambridge University Press.
Akers, Ronald L. 1992. Drugs, Alcohol and Society. Belmont, CA: Wadsworth Publishing Company
Alexander, Bruce. 2008. The Globalization of Addiction. Oxford: Oxford University Press
Becker, Gary S. and Kevin M. Murphy. 1988. ‘A Theory of Rational Addiction.’ Journal of Political Economy 96(4):
675–700.
Becker, Howard S. 1953. ‘Becoming a Marijuana User.’ American Journal of Sociology 59: 235–42.
Becker, Howard S. 1967. ‘History, Culture and Subjective Experience: An Explanation of Social Bases of Drug-
Induced Experiences.’ Journal of Health and Social Behavior 8: 163–76.
Bourdieu, Pierre. 1990. The Logic of Practice. Stanford: Stanford University Press.
Bourgois, Philippe and Jeff Schonberg. 2009. Righteous Dopefiend. Berkeley: University of California Press.
Callon, Michel. 1986. ‘Some Elements of a Sociology of Translation.’ Pp. 196–233 in Power, Action and Belief, edi-
ted by J. Law. London: Routledge
Campbell, Nancy D. 2007. Discovering Addiction. Ann Arbor, MI: University of Michigan Press.
Cloward, Richard A. and Lloyd E. Ohlin. 1960. Delinquency and Opportunity. Glencoe, IL: The Free Press.
Davies, John Booth. 1992. The Myth of Addiction. Amsterdam, Netherlands: Harwood.
Denzin, Norman K. 1983. ‘A Note on Emotionality, Self, and Interaction.’ American Journal of Sociology 89(2): 402–9.
Denzin, Norman K. 1984. ‘Reply to Baldwin.’ American Journal of Sociology 90(2): 422–7.
Denzin, Norman K. 1993. The Alcoholic Society. New Brunswick, NJ: Transaction.

ª 2011 The Author Sociology Compass 5/4 (2011): 298–310, 10.1111/j.1751-9020.2011.00363.x

Sociology Compass ª 2011 Blackwell Publishing Ltd
 Sociological Perspectives on Addiction 309

Duster, Troy. 1970. The Legislation of Morality: Law, Drugs and Moral Judgment. New York: The Free Press.
Elster, Jon. 1999. ‘Emotion and Addiction: Neurobiology, Culture, and Choice.’ Pp. 239–76 in Addiction, edited
by J. Elster. New York: Russell Sage Foundation.
Ferentzy, Peter. 2002. ‘Foucault and Addiction.’ Telos 125: 167–91.
Finestone, Harold. 1957. ‘Cats, Kicks, and Color.’ Social Problems 5(1):3–13.
Garfinkel, Harold. 1967. Studies in Ethnomethodology. Englewood Cliffs, NJ: Prentice-Hall.
Gawin, Frank. 1991. ‘Cocaine Addiction: Psychology and Neurology.’ Science 251: 1580–6.
Gomart, Emilie. 2002. ‘Towards a Generous Constraint: Freedom and Coercion in French Addiction Treatment.’
Sociology of Health & Illness 24(5): 517–49.
Gomart, Emilie. 2004. ‘Surprised by Methadone: In Praise of Drug Substitution Treatment in a French Clinic.’
Body & Society 10(2-3): 85–110.
Gomart, Emilie and Antoine Hennion. 1999. ‘A Sociology of Attachment: Music Amateurs, Drug Users.’ Pp. 220–
47 in Actor Network Theory and After, edited by J. Law and J. Hassard. Oxford: Blackwell.
Gubrium, Jaber F. and James A. Holstein. 2009. ‘The Everyday Work and Auspices of Authenticity.’ Pp. 121–38
in Authenticity in Culture, Self and Society, edited by Phillip Vannini and J. Patrick Williams. Burlington, VT: Ash-
gate.
Haraway, Donna J. 1991. Symians, Cyborgs and Women. New York: Routledge.
Hayles, N. Katherine. 1999. How We Became Posthuman. Chicago: University of Chicago Press.
Johnson, Bruce D. 1980. ‘Toward a Theory of Drug Subcultures.’ Pp. 110–9 in Theories on Drug Abuse: Selected
Contemporary Perspectives, edited by D.J. Lettieri, M. Sayers and H.W. Pearson. Rockville, MD: NIDA Research
Monograph 30.
Keane, Helen. 2002. What’s Wrong With Addiction? New York: New York University Press.
Knorr-Cetina, Karin. 1997. ‘Sociality With Objects: Social Relations in Postsocial Knowledge.’ Theory, Culture &
Society 14(4): 1–30.
Latour, Bruno. 1996. ‘On Interobjectivity.’ Mind, Culture and Activity 3(4): 228–45.
Latour, Bruno. 2004. ‘How to Talk About the Body? The Normative Dimensions of Science Studies.’ Body & Soci-
ety 10(2–3): 205–29.
Levine, Harold G. 1978. ‘The Discovery of Addiction: Changing Conceptions of Habitual Drunkenness in Amer-
ica.’ Journal of Studies on Alcohol 39(1): 143–74.
Lindesmith, Alfred R. 1938. ‘A Sociological Theory of Drug Addiction.’ American Journal of Sociology 43: 593–609.
Lindesmith, Alfred R. 1940. ‘The Drug Addict as Psychopath.’ American Sociological Review 5: 914–20.
Lindesmith, Alfred R. 1947. Opiate Addiction. Bloomington: Principia Press.
Lindesmith, Alfred R. 1965. The Addict and the Law. Bloomington: Indiana University Press.
Lindesmith, Alfred R. 1968. Addiction and Opiates. Chicago: Aldine.
Matza, David. 1969. Becoming Deviant. Englewood Cliffs, NJ: Prentice-Hall, Inc.
McAuliffe, William E. and Robert A. Gordon. 1974. ‘A Test of Lindesmith’s Theory of Addiction: The Frequency
of Euphoria Among Long-term Addicts.’ American Journal of Sociology 79: 795–840.
Merton, Robert. 1938. ‘Social Structure and Anomie.’ American Sociological Review 3: 672–82.
Mills, C. Wright. 1940. ‘Situated Actions and Vocabularies of Motive.’ American Sociological Review 5: 904–13.
Mol, Annemarie. 2002. The Body Multiple. Durham, NC: Duke University Press.
Mol, Annemarie and John Law. 2004. ‘Embodied Action, Enacted Bodies: The Example of Hypoglycaemia.’ Body
& Society 10(2–3): 43–62.
O’Malley, Pat and Mariana Valverde. 2004. ‘Pleasure, Freedom and Drugs: The Uses of ‘Pleasure’ in Liberal Gov-
ernance of Drug and Alcohol Consumption.’ Sociology 38(1): 25–42.
Orphanides, Athanasios and David Zervos. 1995. ‘Rational Addiction With Learning and Regret.’ Journal of Political
Economy 103(4): 739–58.
Peele, Stanton. 1989. Diseasing of America. Lexington, MA: Lexington Books.
Pickering, Andrew. 1995. The Mangle of Practice. Chicago: University of Chicago Press.
Preble, Edward and John J. Casey. 1969. ‘Taking Care of Business: The Heroin User’s Life on the Street.’ Interna-
tional Journal of the Addictions 4: 1–24.
Ray, Marsh. 1961. ‘The Cycle of Abstinence and Relapse Among Heroin Addicts.’ Social Problems 9: 132–40.
Reinarman, Craig. 2005. ‘Addiction as Accomplishment: The Discursive Construction of Disease.’ Addiction
Research and Theory 13(4): 307–20.
Reinarman, Craig and Harry G. Levine (eds) 1997. Crack in America Berkeley, CA: University of California Press.
Robins, Lee N. 1993. ‘Vietnam Veterans’ Rapid Recovery from Heroin Addiction: A Fluke or Normal Expecta-
tion?’ Addiction 88: 1041–54.
Rojeberg, Ole. 2004. ‘Taking Absurd Theories Seriously: Economics and the Case of Rational Addiction Theories.’
Philosophy of Science 71: 263–85.
Room, Robin. 1976. ‘Ambivalence as a Sociological Explanation: The Case of Cultural Explanations of Alcohol
Problems.’ American Sociological Review 41: 1047–65.

ª 2011 The Author Sociology Compass 5/4 (2011): 298–310, 10.1111/j.1751-9020.2011.00363.x

Sociology Compass ª 2011 Blackwell Publishing Ltd
310 Sociological Perspectives on Addiction

Room, Robin. 1984. ‘Alcohol and Ethnography: A Case of Problem Deflation?’ Current Anthropology 25(2): 169–
78.
Room, Robin. 1985. ‘Dependence and Society.’ British Journal of Addiction 80(2): 133–9.
Rosenbaum, Marsha. 1981. Women on Heroin. New Brunswick, NJ: Rutgers University Press.
Rouse, Timothy P. and N. Prabha Unnithan. 1993. ‘Comparative Ideologies and Alcoholism: The Protestant and
Proletarian Ethics.’ Social Problems 40(2): 213–27.
Rubington, Earl. 1968. ‘The Bottle Gang.’ Quarterly Journal of Studies on Alcohol 29: 943–55.
Rudy, David R. 1986. Becoming Alcoholic. Carbondale, IL: Southern Illinois University Press.
Schneider, Joseph W. 1978. ‘Deviant Drinking as Disease: Alcoholism as Social Accomplishment.’ Social Problems
25(4): 361–72.
Stephens, Richard C. 1991. The Street Addict Role. Albany, NY: SUNY Press.
Sutter, Alan G. 1966. ‘The World of the Righteous Dope Fiend.’ Issues in Criminology 2(2): 177–223.
Sutter, Alan G. 1969. ‘Worlds of Drug Use on the Street Scene.’ Pp. 802–29 in Delinquency, Crime, and Social Pro-
cess, edited by D. R. Cressey and D. A. Ward. New York: Harper & Row Publishers.
Valverde, Mariana. 1998. Diseases of the Will. Cambridge: Cambridge University Press.
Waldorf, Dan. 1970. ‘Life Without Heroin: Some Social Adjustments During Long-Term Periods of Voluntary
Abstention.’ Social Problems 18(2): 228–43.
Waldorf, Dan, Craig Reinarman and Sheigla Murphy. 1991. Cocaine Changes. Philadelphia, PA: Temple University
Press.
Weinberg, Darin. 1997a. ‘Lindesmith on Addiction: A Critical History of a Classic Theory.’ Sociological Theory
15(2): 150–61.
Weinberg, Darin. 1997b. ‘The Social Construction of Non-Human Agency: The Case of Mental Disorder.’ Social
Problems 44(2): 217–34.
Weinberg, Darin. 1998. ‘Praxis and Addiction: A Reply to Galliher.’ Sociological Theory 16(2): 207–8.
Weinberg, Darin. 2000. ‘‘Out There’: The Ecology of Addiction in Drug Abuse Treatment Discourse.’ Social Prob-
lems 47(4): 606–21.
Weinberg, Darin. 2002. ‘On the Embodiment of Addiction.’ Body & Society 8(4): 1–19.
Weinberg, Darin. 2005. Of Others Inside: Insanity, Addiction and Belonging in America. Philadelphia, PA: Temple Uni-
versity Press.
Weinberg, Darin. 2008. ‘Toward a Post-humanist Understanding of Addiction.’ [‘Hacia un entendimiento post-
humanista de la adicción.’] Politica y Sociedad 45(3): 159–75.
West, Robert. 2006. Theory of Addiction. Oxford: Blackwell.
Wiener, Carolyn. 1981. The Politics of Alcoholism. New Brunswick, NJ: Transaction Books.
Williams, Terry. 1992. Crackhouse. Reading: Addison-Wesley.
Wiseman, Jacqueline P. 1970. Stations of the Lost. Chicago: University of Chicago Press.
Wittgenstein, Ludwig. 1958. Philosophical Investigations. Oxford: Basil Blackwell.

ª 2011 The Author Sociology Compass 5/4 (2011): 298–310, 10.1111/j.1751-9020.2011.00363.x

Sociology Compass ª 2011 Blackwell Publishing Ltd