Sequencing and Its Consequences: Path ... - Semantic Scholar

Sequencing and Its Consequences: Path Dependence and the Relationships between Genetics and Medicalization1 Sara Shostak and Peter Conrad Brandeis University Allan V. Horwitz Rutgers University

Both advocacy for and critiques of the Human Genome Project assume a self-sustaining relationship between genetics and medicalization. However, this assumption ignores the ways in which the meanings of genetic research are conditional on its position in sequences of events. Based on analyses of three conditions for which at least one putative gene or genetic marker has been identified, this article argues that critical junctures in the institutional stabilization of phenotypes and the mechanisms that sustain such classifications over time configure the practices and meanings of genetic research. Path dependence is critical to understanding the lack of consistent fit between genetics and medicalization. Medical concepts and interventions are inextricably a part of how people in contemporary societies give order and meaning to our worlds. Health professionals now define and treat as medical conditions many human behaviors once considered deviant—including alcoholism, mental disor-

1 This research was supported by the National Science Foundation (award no. 035381), the Robert Wood Johnson Health and Society Scholars Program at Columbia University, the Netherlands Institute for Advanced Study, and the Theodore and Jane Norman Fund for Faculty Research and Creative Projects in Arts and Sciences at Brandeis University. We gratefully acknowledge the insightful questions, comments, and suggestions of Peter Bearman, Jason Beckfield, Wendy Cadge, David Cunningham, Steven Epstein, Brian Fair, Jeremy Freese, Karen Hansen, Jocelyn Viterna, participants in the Genetics and Social Structure Conference at Columbia University (March 2006), and the AJS reviewers. We thank Elizabeth Ginsberg for research assistance. Direct correspondence to Sara Shostak, Department of Sociology, Brandeis University, MS 071, Waltham, Massachusetts 02454-9110. E-mail: [email protected]

䉷 2008 by The University of Chicago. All rights reserved. 0002-9602/2008/114suppl-0010$10.00

AJS Volume 114 Suppl. (2008): S287–S316

S287

American Journal of Sociology ders, opiate addictions, eating disorders, sexual dysfunction, and child abuse. A range of common life processes have been medicalized as well, including menstruation, infertility, childbirth, menopause, aging, and death. Medicalization can also be seen at work in the elaboration of new categories of human problems, such as idiopathic short stature, social anxiety disorder, and adult attention deficit hyperactivity disorder (ADHD). Moreover, medical expertise has had a legitimating role in ordering dimensions of social life at both the micro- (e.g., personal hygiene) and macrolevels (e.g., urban planning) (Rose 2007a). Consequently, the growth of medical jurisdiction and its extension into myriad domains of human life has been “one of the most potent transformations of the last half of the twentieth century in the West” (Clarke et al. 2003, p. 161), profoundly shaping our social relationships, notions of the good life, and “the kinds of people that we are” (Rose 2007a, p. 700). A central rationale for the recent massive public and private investments in research in human genetics and genomics (e.g., the Human Genome Project [HGP]) is that advances in genetic and genomic knowledge will increase our capacity to prevent and cure common diseases, especially by making it possible for medical interventions to be tailored to individuals and their genetic risks (Hedgecoe 2004; Guttmacher and Collins 2005). Its advocates predict that the HGP will revolutionize the practice of medicine in the 21st century (Guttmacher and Collins 2005). Accordingly, relationships among genetic information, medical categories and practices, and “human kinds” (Hacking 1999) are emerging as focal concerns of social scientists (Clarke et al. 2003; Duster 2006; Almeling 2007; Rose 2007b). Central to many sociological analyses of genetics is the concept of geneticization, the process by which “differences between individuals are reduced to their DNA codes, with most disorders, behaviours, and physiological variations defined, at least in part, as genetic in origin” (Lippman 1991, p. 19). Geneticization represents not only “a way of thinking” about human differences but also “a way of doing,” as genetic technologies are “applied to diagnose, treat, and categorize conditions previously identified in other ways” (Lippman 1998, pp. 64–65). The concept of geneticization is rooted in that of medicalization and “derives guidance” (Lippman 1991, p. 27) from the literature in medicalization that is critical of the effects of the expansion of medical jurisdictions (Press 2006). This conceptual affinity is reflected especially in critical assessments of genetics, which warn of the potential of geneticization to contribute to the medicalization of social problems and thereby to undermine macrosocial explanations of differences in individual and group outcomes (Nelkin and Lindee 1995; Conrad 2000; Duster 2006). Given the association of genetic information and medical innovation S288

Genetics and Medicalization in biomedical researchers’ advocacy for the HGP, as well as social scientific critiques of genetics and geneticization for contributing to the medicalization of social problems, it may appear that there is an elective affinity (Weber [1905] 1958) among genetic information, geneticization, and medicalization, with genetic information leading inexorably to geneticization and medicalization.2 This assumption has only recently been challenged in empirical studies, which suggest, for example, that extant medical practices set limits on the relationship between genetics and geneticization, even for conditions with simple genetic etiologies, such as polycystic kidney disease, an autosomal dominant trait for which genetic testing is available (Cox and Starzomski 2004). This article contributes to a growing body of social scientific literature that looks at the effects of genetic information in specific social and temporal contexts and, in so doing, illuminates the dimensions of social organization and process in and through which genetic information comes to matter. Specifically, we find that the dynamics of path dependence (Goldstone 1998; Thelen 1999; Mahoney 2000; Pierson 2000) surrounding extant definitions of phenotypes and the institutions in which they are located shape the social location and meanings of subsequent genetic research and its applications. Based on analyses of three conditions for which at least one putative gene or genetic marker has been identified—depression, homosexuality, and susceptibility to chemical exposures—we demonstrate that phenotypes are stabilized within institutional and structural patterns that are reproduced over time and that shape the relationship among genetics, geneticization, and medicalization. Thus, the meanings of genetic research must be understood as conditional on its position in a sequence of events (Abbott 1992; Bearman, Faris, and Moody 1999). Analysis of these three cases also highlights the fact that structural patterns surrounding extant definitions of phenotypes shape the meanings of genetic discoveries in varied ways. In the case of depression, the prior medicalization and standardization of the phenotype in the third Diagnostic and Statistical Manual of Mental Disorders (DSM-III; APA 1980) provides a specific disease entity, extensive structural support, and financial incentives for genetic research, including research to identify persons who can be defined as genetically at risk for depression and who might benefit from prophylactic interventions. Here, self-reinforcing processes support continued medicalization and ensure that when scientists find associations between genes and depression, the genes are understood to 2

Medicalization, like “most words ending in ‘tion’ . . . [is] ambiguous between process and product, between the way one gets there and the result” (Hacking 1999, p. 36). In this article, we use the word medicalization to refer to a social process rather than an outcome.

S289

American Journal of Sociology be causes of a disease condition. In contrast, homosexuality was officially demedicalized in 1973, when it was deleted from the DSM, with significant political, cultural, and institutional consequences. As a result, scientists frame studies of genetic etiologies of homosexuality as research into “a naturally occurring” and “normal variation” in human behavior (Hamer et al. 1993, pp. 321, 326) that has no relation to medical jurisdictions. The phenotypes associated with susceptibility to chemicals are located in two institutional domains, clinical practice and environmental regulation, where they figure in very different kinds of risk assessment. This provides an opportunity to examine processes of increasing returns within institutional domains for the same genotypes, as well as exchange across them. Together, these analyses demonstrate that critical junctures in the institutional stabilization of phenotypes and the mechanisms that sustain such classifications over time configure the consequences of genetic information vis-a`-vis both geneticization and medicalization.

PATH DEPENDENCE

There is wide variation in how social scientists conceptualize path dependence (Pierson 2000, pp. 252–53; Thelen 2003, p. 221). An expansive, and influential, definition asserts that “what happened at an earlier point in time will affect the possible outcomes of a sequence of events occurring at a later point in time” (Sewell 1996, pp. 262–63; for critiques, see Thelen 1999, p. 202; Mahoney 2000, p. 510; Pierson 2000, p. 252). A more restrictive definition states that path-dependent analyses focus exclusively on sequences with the following characteristics: (1) they are causal processes that are highly sensitive to events that take place in the early stages of an overall historical sequence; (2) these early historical events are “contingent occurrences that cannot be explained on the basis of prior events or initial conditions”; (3) once contingent events take place, relatively deterministic causal patterns emerge (Mahoney 2000, pp. 510–11).3 Scholars have responded to this conceptual variation with a number of strategies, including embracing a definition that is situated between the expansive and restrictive versions (Eden 2004), focusing on specific parts of path-dependent arguments and identifying the kinds of empirical phenomena that are subject to such processes (Pierson 2000), and distinguishing more clearly between mechanisms of reproduction and the logic of change at work in particular instances (Thelen 2003). The definitional assumptions in this article are as follows. First, as a

3 As a consequence of this variation, path dependence may appear either as pervasive in society and politics or as an extremely rare occurrence (Thelen 1999, p. 220).

S290

Genetics and Medicalization form of narrative, path dependence is “intrinsically temporal in both construction and explanatory logic” (Griffin 1993, p. 1109; see also Abbott 1992). Specifically, path dependence describes how “key decisions at choice points produce outcomes that set history on a course whose mechanisms of production make the initial sequence unstoppable . . . [and] from which it is difficult, or even impossible, to return” (Katznelson 2003, p. 290).4 Second, decisions made at a choice points or “critical junctures” are important because they lead to “the creation of institutional or structural patterns that endure over time” (Mahoney 2001, p. 6). Third, analyses of path dependence are required to identify “choices or conditions that foreclosed options and steered history in one or another directions” and the mechanisms that reproduce the outcome of a historical turning point (Haydu 1998, p. 352). While the dynamic of increasing returns is often identified as a primary mechanism of reproduction, others may include a wide range of functional, power, and legitimation mechanisms (Mahoney 2000, pp. 517–26), such as incentive structures, conventions, normative appeals, and elite power (Katznelson 2003, p. 292). Finally, following Abbott’s lead, path dependence is sensitive to issues of timing and sequencing, as it recognizes that the timing of an event may be crucial to outcomes (Abbott 1992; 1995; 2001). Particularly, events in the earlier parts of a sequence often matter much more than those in later parts (Abbott 1992), and “an event that happens ‘too late’ may have no effect, although it might have been of great consequence if the timing was different” (Pierson 2000, p. 263). Although this study draws from the conceptual and methodological armamentarium of path dependence analyses, it benefits also from assessments of the limitations of this approach (Haydu 1998). Particularly, in this analysis, critical junctures need not be completely contingent, as outcomes at given switch points may be “products of the past, rather than historical accidents” (Haydu 1998, p. 354; see also Hall 2003; Eden 2004, p. 287). Additionally, the past can be a source of creativity as well as constraint; choices made at critical junctures not only foreclose options but “may also lead to and shape the switch points confronted by later generations, drawing fault-lines along which later crises erupt and creating options for new solutions” (Haydu 1998, p. 357). Relatedly, while the forces that shape a critical juncture are often quite different than the mechanisms that sustain it over time (Mahoney 2001), this does not preclude the possibility that sometimes “history’s switchmen come along for the ride” (Haydu 1998, p. 353). These insights enable us to elucidate variations in the sequences at the center of our cases. 4 This does not mean that particular alternatives are permanently locked in (Pierson 2000, p. 265; see also our discussion on contingent futures below).

S291

American Journal of Sociology DATA AND METHODS

Identifying whether, when, and how genetic information is taken up in processes of geneticization and medicalization offers an important vantage point for elucidating how extant dimensions of social organization shape the meanings and effects of scientific knowledge. Our analysis addresses this question through analysis of three cases: depression, homosexuality, and variation in responses to chemical exposures. These cases were selected to maximize variation across three domains of social structure that we theorized to be relevant to our analysis (see table 1). First, we selected phenotypes for their particular relationships to medical jurisdiction.5 Specifically, depression was defined as a medical condition prior to molecular genetic research on its etiology, homosexuality was once but is no longer constructed as a medical condition, and variation in responses to chemical exposures has been taken up in the medical arena only when those chemicals are pharmaceuticals. Second, we chose phenotypes that have historically been located in different professional jurisdictions, including psychology, psychiatry, pharmacology, and environmental regulation and policy making. Third, we selected conditions that displayed variance in the degree to which they serve as a basis for individual identity or social organization. That is, we selected cases for which a phenotype is the basis for a human kind (depression and homosexuality) and for which the phenotype does not have a prominent role in contemporary identities or social organization (variation in responses to chemical exposures). This range of variation facilitates identification of “similarities and differences [across cases] . . . and the processes that create, maintain, and change patterned behavior” (Vaughan 1992, p. 181). At the same time, these traits share an important commonality. Even while a degree of heritability may have been suspected or attributed to them at different moments in time, they are not discursively constructed as “genetic diseases,” nor are they firmly located in the networks of medical specialists, genetic counselors, genetic testing, disease registries, genetically oriented advocacy groups, and so forth that surround, especially, highly penetrant single-gene disorders. Consequently, genetic information, geneticization, and medicalization are not already intertwined in these cases but can be examined in temporal sequences unfolding in and through dimensions of social structure. Path dependence implies that “we cannot understand the significance 5 We use the terms trait, phenotype, and condition to refer to the foci of the case studies. These are imperfect designations for complex phenomena that are discursively defined and enacted (Mol 2002) in multiple ways and in varied social and historical contexts. Our use of these terms does not represent endorsement of any particular position regarding the etiology or ontology of these phenomena.

S292

Genetics and Medicalization TABLE 1 Cases (Phenotype/Gene or Marker)

Phenotype Is Object of Medical Practice

Phenotype Is a Basis of Individual or Group Identity Yes

Yes . . . . . . . . . . . . . . . . . . . . . . .

Depression/5-HTT

No . . . . . . . . . . . . . . . . . . . . . . .

Homosexuality/Xq28 chromosomal region

No Susceptibility to pharmaceuticals/NAT1, GST1, etc. Susceptibility to environmental chemicals/NAT1, GST1, etc.

of a particular social variable without understanding ‘how it got there’— the path it took” (Pierson 2000, p. 252; see also Abbott 1992). Therefore, in each case, we identify antecedent conditions, critical junctures, mechanisms that reproduce critical junctures, the emergence of molecular genetic research, and the events following identification of a gene or marker for the trait. Our data come both from the peer-reviewed literature for each phenotype and associated genotypes and from historical accounts.

PHENOTYPES, GENOTYPES, AND SOCIAL STRUCTURE: THREE CASES

Depression From the earliest description of depression in ancient Greece and Rome until a major revision of the DSM in 1980, observers recognized two distinct ways in which its symptoms emerged (Jackson 1986). First, depressive symptoms could be with cause, that is, normal, proportionate responses to serious losses such as deaths of intimates, economic reversals, disappointments in attaining valued life goals, and the like. Historically, appropriate responses to depressions that were with cause typically focused on changing the circumstances that led to them. Second, symptoms that were without cause either arose in the absence of situations that would normally produce sadness or were of disproportionate magnitude or duration to their provoking causes. Because both types shared the same phenotypic symptoms, the essential difference between depressive conditions that were with or without cause lay in the fact that the social environment evoked and maintained the first type, while factors internal to the individual produced the second type (Horwitz and Wakefield 2007). During the period roughly between 1920 and 1970, a psychosocial model of depression dominated clinical practice, psychiatric training, and research; the study of presumably brain-based conditions was marginal to the field (Grob 1991). The major funder for research and training on S293

American Journal of Sociology depression and other mental disorders, the National Institute of Mental Health (NIMH), founded in 1949, focused on psychological and social processes related to mental health but provided minimal funding for biologically oriented research (Kolb, Frazier, and Dirovata 2000; Schooler 2007). Most of this research assumed that depression was more likely to be a product of poor environmental circumstances than of biological or genetic defects. During the 1970s, the psychiatric profession underwent a serious crisis of legitimacy. The psychosocial model did not value precise diagnoses, and the definitions in psychiatry’s official diagnostic manual, the DSMII (APA 1968), were very general and cursory. Neither clinicians nor researchers could reliably use such definitions, and so psychiatric classifications of mental illness were idiosyncratic and varied widely across individual diagnosticians (Kirk and Kutchins 1992). The unreliability of such diagnostic criteria subjected the psychiatric profession to much criticism and even ridicule (Szasz 1961; Rosenhan 1973) and hampered efforts to measure the efficacy of interventions (Lakoff 2005). Moreover, the psychosocial model did not provide a solid grounding for why psychiatrists, as opposed to many other professionals, including clinical psychologists, counselors, social workers, and nurses, should have professional dominance over the treatment of mental illness. Psychiatry, which had always had a shaky position within the prestige hierarchy in medicine, was in danger of losing its legitimacy as a scientific discipline. To deal with this dire situation, a group of research-oriented psychiatrists concluded that only clear, precise, and reproducible definitions of the entities it studied, comparable to those studied in other branches of medicine, could serve as the foundation for a truly scientific discipline. They explicitly contrasted their scientific approach, which they claimed was rooted in empirically supported facts, to what they characterized as the unprovable theories of the psychoanalysts that had dominated the earlier DSM-I (APA 1952) and DSM-II. The empiricists controlled the development of the new edition of the official diagnostic manual, the DSM-III. This manual replaced the amorphous conditions of psychodynamic psychiatry with several hundred specific definitions of various types of mental illnesses that relied on the characteristic symptoms of each entity (Horwitz 2002). Because of the empiricists’ desire to purge the psychodynamic assumptions from the new manual, a core principle of the DSM-III was that these definitions could not assume any particular etiology of symptoms, including genetic causes. The definition of major depression, for example, required the presence during a two-week period of five from a list of nine explicitly delineated symptoms, one of which must be either “depressed mood” or “diminished interest/pleasure in life.” S294

Genetics and Medicalization This strategy allowed the researchers to claim theoretical neutrality and gain support from clinicians who held a broad range of orientations. The DSM-III initiated a number of developments that resulted in a thoroughgoing medicalization of depression. Although no criteria existed to validate diagnoses of depression, the new symptom-based definition was conceptually analogous to the medical conditions that were the objects of other medical specialties. In addition, depression became a single phenotype that combined the previously distinct categories of depressive conditions with and without cause and treated both as diseases. The DSMIII ensured that all depressive conditions, including those with social, psychological, or biological causes, would be considered diseases; at that time, genetic causes had no special standing. The internal dynamics of the psychiatric profession, namely, its desire to enhance its position within the prestige hierarchy of medicine and its status relative to competing mental health professions, led to the specific medicalization of depression in the DSM-III. Following this critical juncture, a variety of institutional and structural patterns in health-care delivery, research and development, and advocacy reinforced this medicalized definition of depression. At the institutional level, health-care providers would only treat conditions that had DSM diagnoses; likewise, insurance companies and federal programs would only pay for the treatment of depressive conditions that met DSM diagnostic criteria (Horwitz 2002). The NIMH channeled research funding to studies that used the new diagnostic criteria and shifted its research focus away from psychosocial factors and toward the biological and biochemical abnormalities that might lead to depression (Kolb et al. 2000). The development of a new class of medications, the selective serotonin reuptake inhibitors (SSRIs), which were developed and targeted for the psychological consequences of the broad range of psychosocial conditions that the DSM-III definition of depression encompassed, reinforced the medicalization of depression. These drugs act to raise levels of serotonin in the brain, and they lent credence to the idea that a deficiency of this neurochemical was at the root of many depressive conditions (Kramer 1993). The later approval of direct-to-consumer advertisements in 1997 unleashed a barrage of messages to the public that promoted the idea that depression stemmed from problems with brain chemistry that could be successfully managed through drug treatments. The use of SSRIs soared during the 1990s, while other types of treatment, especially longterm psychotherapy, declined (Olfson et al. 2002). Advocacy groups such as the National Alliance for the Mentally Ill (NAMI) also aggressively promoted the medicalized view that depression stemmed from a chemical imbalance in the brain. Groups like NAMI S295

American Journal of Sociology embraced the DSM-III diagnosis for its potential to legitimate claims to reimbursement parity with physical illnesses, decrease the stigma that accrues to mental disorders, and deflect responsibility for the causes of mental disorders away from families. By the turn of the century, psychiatric training, research, and practice centered entirely on the medical definitions of the DSM, while socially and psychodynamically oriented psychiatrists were marginalized (Blazer 2005). Genetic research was not an important factor associated with the medicalization of depression. Although researchers had conducted twin and adoption studies of depression since the 1930s, such behavioral genetics studies produced highly ambiguous results, and by their nature, they could not identify specific genes (Horwitz 2002, pp. 144–45). Only toward the end of the 20th century were researchers able to use technologies that allowed them to connect specific genes to depression. Consequently, the social structural processes that maintain the categorization of depression as a major mental illness are part of the conditions of possibility for contemporary genetic research. For example, genetic research on depression assumes that it is a disease of the brain, as evident in the extensive research on the 5-HTT gene, which controls the way that serotonin passes messages through brain cells. The results of a large study, which indicated that the short allele of the 5-HTT gene interacts with highly stressful environments to produce depression, have been especially influential (Caspi et al. 2003). This study found that, among people who experienced four or more stressful life events, 43% with two short alleles and 33% with one short allele developed major depression, compared to only 17% of those with two long alleles. These findings, which showed an interaction between a gene and the social environment, reintroduced the study of the environment as a cause of depression and motivated numerous attempts at replication (e.g., Kendler et al. 2005; Surtees et al. 2006). Environmental factors, however, are being investigated only as triggers of genetic influences (e.g., Moffitt, Caspi, and Rutter 2005). The preexisting medical framework for depression serves to amplify genetic effects and minimize environmental ones. By the time that researchers were able to connect specific genes to depressive symptoms, these genes could only be seen as influencing the development of a diseased condition. Homosexuality In the 19th century, medical and scientific researchers began to investigate homosexuals as a human kind whose past, case history, and “indiscreet anatomy and possibly mysterious physiology” might reveal the etiology of their “species” (Foucault 1980, p. 43). Since that time, researchers have S296

Genetics and Medicalization focused on the body as the locus of homosexuality, even as hereditarian theorists, Freudian psychoanalysts, and other researchers have disagreed about the relative importance of specific biological or psychological mechanisms (Terry 1999). The first edition of the DSM (APA 1952) placed homosexuality among the “sociopathic personality disturbances,” establishing it officially as a mental illness. In 1968, the DSM-II characterized homosexuality under “Personality Disorders and Certain Other Non-Psychotic Disorders,” under the specific heading of “sexual deviation” (APA 1968) The classification of homosexuality as a psychopathology was challenged by researchers, such as Albert Kinsey and Evelyn Hooker, whose findings suggested that homosexuality was a more common than expected variant of human sexuality and that there were many varieties of sexual expression (Bayer 1988, pp. 43–66). While these findings were largely ignored by psychoanalysts, homophile organizations embraced them. During the late 1960s, activists mobilized the research of Kinsey, Hooker, and others (e.g., Szasz 1961) to contest the psychiatric designation of homosexuality as a psychopathology and to draw attention to the harmful effects of purportedly therapeutic psychiatric treatments (Bayer 1988). Like the other civil rights movements of that era, the gay rights movement organized both to end institutionalized oppression and to promote pride in a previously stigmatized and marginalized identity. Consequently, the profession of psychiatry, and especially its designation of homosexuality as a mental illness, were focal points of gay rights activism. Beginning in the late 1960s, activists began to organize picket lines outside the public lectures of psychiatrists who described homosexuality as a disease. In 1970, gay rights activists brought their protest to the annual meeting of the American Psychiatric Association (APA), and soon thereafter the APA became a major target of the gay rights movement. Activists demanded the deletion of homosexuality from the DSM. The campaign to remove homosexuality from the DSM involved both gay rights activists and allied psychiatrists and was vigorously opposed by psychoanalytic psychiatrists. Proponents for deleting homosexuality from the DSM challenged the scientific research on homosexuality, pointed to the low “success” rates for treatment, and contended that the legitimacy of psychiatric practice required narrow, scientific definitional criteria. On December 15, 1973, the board of trustees of the APA approved the deletion of homosexuality from the DSM, stating that “homosexuality . . . by itself does not necessarily constitute a psychiatric disorder” (Bayer 1988, p. 137). At the same time, the board passed a civil rights proposal deploring “all public and private discrimination against homosexuals.” In 1974, as part of a referendum forced by the psychoanalytic psychiatrists who opposed the board’s decision, the membership of the APA affirmed the deletion. S297

American Journal of Sociology By decreeing that homosexuality is not a psychopathology, the APA removed a major source of legitimation for discrimination, persecution, and forced therapy and enabled the gay rights movement to argue even more forcefully against discrimination in diverse domains of social life, including employment, immigration, and family formation. Demedicalization also created an institutional space for the emergence of gay, lesbian, bisexual, and transgender activist groups within the professions of psychiatry, psychology, and social work; the APA itself has participated in civil rights efforts targeted at laws that justified the discredited designation of homosexuality as a form of psychopathology (Bayer 1988, p. 161). Moreover, in subsequent decades, the gay rights movement became institutionalized, establishing thousands of gay advocacy organizations, religious institutions, law offices, health clinics, community centers, and cultural venues (D’Emilio 2002). Activism in the 1990s challenged the notion that sexuality could be understood through the presumptive binary categories of homosexuality and heterosexuality and supported a proliferation of multiple proudly “queer” identities (Gamson and Moone 2004). Consequently, notwithstanding the efforts of groups such as the National Association for Research and Therapy of Homosexuality, which advocates “reparative therapy” for homosexuals, and the cultural and political assaults of the religious right, the demedicalization of homosexuality has proved robust (Conrad 2007).6 The APA’s decision that homosexuality is not a psychopathology left untouched questions about the etiologies of sexual behaviors and preferences, which researchers have continued to theorize as embodied traits. In the 1980s, researchers began to apply the techniques of behavioral genetics to investigate the potential heritability of homosexuality (Pillard et al. 1981). These studies, which preserved psychiatry’s binary definition of homosexual and heterosexual phenotypes, found evidence of heritability in both men (Bailey and Pillard 1991) and women (Bailey et al. 1993). In the 1990s, neuroscientific and genetic researchers claimed to identify specific biological bases of homosexual behavior. LeVay (1991), a neuroscientist, published a study in Science reporting that the hypothalamus in the brains of (presumed) gay men who had died from AIDS tended to be smaller than those of a comparison group, suggesting that gay men may have a different neurophysiology than heterosexual men. Articles about the “gay brain” were front-page news, despite the limitations of the

6

A review conducted by an APA ad hoc committee as part of the revision of the DSM in 1986 found that the diagnosis “ego-dystonic homosexuality” (a revision of “sexual orientation disturbance”) was seldom used in research or treatment, leading the committee to recommend its deletion; the current DSM makes no specific reference to homosexuality.

S298

Genetics and Medicalization study design (Conrad and Markens 2001). In 1993, Dean Hamer, a geneticist at the National Cancer Institute who was studying whether homosexual men had a genetic susceptibility to Kaposi’s sarcoma (Terry 1999, p. 391), reported in Science that “it appears that [the chromosomal region] Xq28 contains a gene that contributes to homosexual orientation in males” (Hamer et al. 1993, p. 325). Hamer’s study received widespread publicity, and Xq28 soon became known widely as “the gay gene” (Conrad and Markens 2001).7 Both LeVay and Hamer have become public advocates for the notion that (male) homosexuality is a biologically based “normal variant in human behavior” (Hamer et al. 1993, p. 261; LeVay and Hamer 1994). Moreover, they contend that scientific research will “help dispel the myths about homosexuality that in the past have clouded the image of lesbians and gay men” (LeVay and Hamer 1994, p. 49). Queer activists have criticized LeVay and Hamer’s research for assuming binary oppositions between homosexuality and heterosexuality, seeking to locate cultural constructions at the molecular level, and basing arguments for human rights in genetics or biology (Terry 1999, pp. 397– 98). Nonetheless, since the 1990s, the notion that many homosexuals are born gay has attained increasing credibility in the gay community (D’Emilio 2002) and replaced the more voluntaristic notion of a gay lifestyle in many writings about homosexual identity (Brookey 2002). There is also wide support for biological explanations of homosexuality in the U.S. population (Conrad 2007, p. 110). At this time, there is no consensus on the costs and benefits of geneticization, as some activists argue that establishing a biogenic cause of sexual orientation will increase social acceptance and serve as a basis for civil rights claims (D’Emilio 2002, pp. 10–12), while others express concern that it could enable and justify pharmacological interventions to prevent homosexual behavior (Brookey 2002). Nonetheless, geneticization and medicalization remain uncoupled in this debate. Susceptibility to Chemical Exposures Scientists have suspected a genetic basis for variation in responses to food and drugs since at least the early 1900s. In his research on “the chemistry of the individual,” British physician and scientist Archibald Garrod sought to identify “the differences in type which are so important in . . . the idiosyncrasies with regard to drugs and food” and which underlay “the proverbial saying that what is one man’s meat is another man’s poison.” (Garrod 1932, p. 3). Garrod’s research on alcaptonuria, a relatively rare 7 Only Hamer’s lab has replicated this finding with a different sample (Hu et al. 1995), and one major attempt to replicate the research failed (Rice et al. 1999).

S299

American Journal of Sociology condition caused by an autosomal recessive genetic mutation, constituted the first proof of Mendelian genetics in humans (Mancinelli, Cronin, and Sade´e 2000) and provided evidence for disease states that are caused by “inborn errors of metabolism.” In the decades immediately following Garrod’s research, inborn errors of metabolism were studied primarily by scientists interested in exploring patterns of Mendelian inheritance, and many of the outcomes of the errors under study were of limited clinical relevance (e.g., variations in capacity to taste or smell particular substances) (Weber 2001). During World War II, when military doctors administered antimalarial drugs to thousands of soldiers, “susceptible phenotypes” became a subject of systematic clinical and scientific concern (Jensen 1962). At this time, doctors observed that a proportion of U.S. and British servicemen receiving antimalarials experienced “serious untoward side effects,” especially acute hemolytic anemia (Jensen 1962). Army physicians were particularly intrigued by apparent subpopulation variation in these responses, noting that “pamaquine caused hemolysis in 5–10% of American Negroes but rarely in Caucasians” (Tarlov et al. 1962, p. 214). In response to these clinical observations, the U.S. Army and the Office of the Surgeon General provided funding to the University of Chicago Army Malaria Research Unit to identify the basis of this response and to chart its clinical course (Dern et al. 1954). This research led to the identification of glucose-6phosphate dehydrogenase (G6PD) deficiency (Carson et al. 1956), which is an enzymatic deficiency that affects individuals’ metabolism, making them vulnerable to hemolytic anemia when exposed to oxidative drugs such as pamaquine (Beutler 1994). Scientists hailed the identification of G6PD deficiency “as an example of a disease which, in the past, was looked upon as a reaction of idiosyncratic nature . . . and which is now better defined in terms of genetic abnormality and biochemical mediation of disease” (Jensen 1962, p. 212). At about the same time that American researchers characterized G6PD deficiency as a disease, researchers demonstrated that genetic control of a drug-metabolizing enzyme, N-acetyltransferase (NAT), was responsible for clinical variation in response to procainamide, a local anesthetic that caused prolonged apnea and death in some patients (Kalow 1968). This research established that differences in drug-metabolizing enzymes, as well as metabolic disorders such as G6PD deficiency, could cause differences in responses to pharmaceuticals. Building on this insight, scientists proceeded to identify genetic control of drug metabolism as the source of serious adverse reactions to a variety of drugs (Calabrese 1984). Although it is common practice to differentiate between chemicals we ingest intentionally (e.g., drugs) and those to which we are exposed involuntarily (e.g., environmental pollutants), the basic physiological proS300

Genetics and Medicalization cesses of metabolism and excretion of these chemicals are often shared. As early as 1957, scientists proposed that the individual variations observed in responses to drugs might be related to susceptibility or resistance to conditions other than “drug idiosyncrasies” (Motulsky 1957, p. 836) When Vogel coined the term pharmacogenetics, he asserted that genetic susceptibilities to drug toxicities were likely to be associated with other kinds of susceptibility (Calabrese 1996). In 1971, Brewer proposed the term ecogenetics, noting that, based on the insights of pharmacogenetics, “we can be sure that other pollutants are even now finding genetically susceptible targets” (Brewer 1971, p. 93). Proponents of the term ecogenetics intended it to encompass research on genetic susceptibilities to pharmaceuticals, environmental chemicals, food and food additives, allergens, and infectious agents.8 From the mid-1970s through the mid1980s, ecogenetic research garnered attention in both government and academia (NAS 1975; Omenn and Gelboin 1983); in the late 1980s and early 1990s, the University of Cincinnati and the University of Washington established centers for the study of ecogenetics. Scientists interested in genetic susceptibilities to occupational and environmental chemicals borrowed techniques from pharmacogenetics to study both industrial chemicals and air pollutants (e.g., Mountain 1963). The early recommendations of industrial hygienists that genetic testing be used to protect “the hypersensitive worker” (Stokinger, Mountain, and Scheel 1968) also borrowed from the clinical orientation of pharmacogenetics. This pattern of borrowing is evident in more recent research as well. For example, the first study to relate a genetic polymorphism in NAT1 to a high-activity NAT1 phenotype in humans and to establish that the phenotype increased risk of a disease, given a specific environmental exposure, reframed drug-metabolizing enzymes (NAT1, NAT2 and GSTM1, GSTT1) as “carcinogen metabolizing enzymes” (Bell et al. 1993). However, research on genetic susceptibilities to occupational and environmental exposures never has received the massive private investment nor the broad biomedical interest that pharmacogenetics and, later, pharmacogenomics garnered (Hedgecoe 2004). Instead, even as these two domains of knowledge production have been linked by the exchange of techniques and technology, their paths run in and through two different institutional domains. Specifically, while pharmacogenetics and pharmacogenomics are oriented to the clinic, ecogenetics and, later, environmental genomics are oriented to environmental health risk assessment and regulation. Risk assessment coalesced as a central activity of the federal government in the 1970s, at about the same time as the emergence of ecogenetics. 8

Gilbert Omenn, personal communication, November 14, 2001.

S301

American Journal of Sociology At that time, Congress ordered federal agencies—most importantly, the newly formed Environmental Protection Agency (EPA) and the older Food and Drug Administration (FDA)—to take responsibility for complex predictive analyses of the risks and benefits of hazardous substances and technologies (Jasanoff 1995). Significantly, their mandate included not only the control of technologies and substances that were known to be harmful, but also the identification and regulation of hazards that could, if not regulated, pose serious threats to health, safety, or the environment. This greatly expanded the discretionary power of the agencies, and, not surprisingly, these new programs of regulation quickly became sites of legal contestations about the nature and precision of scientific knowledge, measurement of risks, and techniques of risk management (Jasanoff 1995, p. 71). As a result of these and other political pressures brought by both the environmental movement and activist scientists (Frickel 2004), the federal government invested in research capabilities at the National Institute of Environmental Health Sciences (NIEHS) and, in 1978, established the National Toxicology Program (NTP). Congress charged the NTP with coordinating toxicology testing programs within the federal government, strengthening the science base in toxicology, developing and validating improved testing methods, and providing information about potentially toxic chemicals to health, regulatory, and research agencies, scientific and medical communities, and the public (NTP 2002). The adversarial and litigious nature of the regulatory process in the United States has continued to shape research agendas, practices, and rationales for investment in the environmental health sciences. Many environmental health scientists in the federal government describe their work as “largely driven by issues that relate to safety of consumer products, occupational exposures, human exposure from substances in the environment, as well as the effects of chemicals on environmental species” (Schwetz 2001, p. 3), and its practitioners describe toxicology, in particular, as “most importantly . . . part of the risk assessment process” (Smith 2001, p. 281). Relatedly, scientists offer the improvement of techniques of risk assessment as a warrant for investment in innovative lines of inquiry in environmental health research, including those based in genetics and genomics (Frickel 2004; Shostak 2005). The NTP tracks yearly how its studies have been used in risk assessment and regulation.9 Thus, even as some scientists and public health practitioners express optimism that information about high-risk phenotypes will be translated for applications in clinical settings and public health practice (Christiani 9 This information is gathered at http://ntp.niehs.nih.gov/index.cfm?objectidp 720166B7-BDB7-CEBA-F4E7383AB26F5BEE (accessed March 30, 2008).

S302

Genetics and Medicalization 1996), at the institutional level, the justification for research on genetic susceptibility to environmental chemicals has focused on its potential “to reduce the guesswork in risk assessment” and “increase markedly” the “ability to issue fair and appropriate regulations concerning human hazards” (Olden and Wilson 2000). In a statement to Congress about the importance of the NIEHS Environmental Genome Project (EGP), an effort that seeks to identify “functionally important variations in DNA sequence, common polymorphisms, in known genes that are likely to be influenced by environmental exposures” (NIEHS 1997), the institute’s director stated:10 Presently, environmental health regulatory agencies craft rules as if “onesize-fits-all.” However, we know that individuals can vary by more than two-thousand fold in their capacity to repair or prevent damage following exposure to toxic agents in the environment. Knowledge of the prevalence of susceptibility genes would take much of the guesswork out of environmental health decision-making.

The current regulatory process rarely requires any characterization of genetic susceptibilities to environmental exposures (Wakefield 2002). However, NIEHS and NTP scientists and administrators are now engaged in the process of translating research about genetic susceptibilities to chemical exposures for applications in risk assessment and policy making (Shostak 2005). The traditional orientation of the environmental health sciences to contributing to the regulation of chemicals in the ambient environment (e.g., air, water, soil) also shapes the political meanings of genetic research on individual susceptibility and its potential clinical applications. For example, environmental health and justice activists have already expressed their opposition to any uses of genetic information to shift the perception of who is responsible for environmental health problems (i.e., from polluters to genetically susceptible individuals living in polluted environments) or to individualize the focus of environmental health analyses and interventions (i.e., replacing public policies designed to protect communities with medical monitoring and treatment of at-risk individuals who have been exposed to environmental contaminants) (Shostak 2004). The institutional matrix (Hacking 1999; Pierson 2000) in which genetic susceptibilities to chemicals are located coordinates applications oriented to population-level risk assessment and regulation, rather those that focus on individual-level clinical applications.

10 Dr. Kenneth Olden, testimony to the House Committee on Appropriations, Subcommittee on Labor, Health and Human Services, and Education, April 2001.

S303

American Journal of Sociology DISCUSSION: THE CONSEQUENCES OF SEQUENCING

The biological sciences, including genetics and genomics, are critical components of the system of abstract knowledge that supports medicine’s jurisdictional claims (Abbott 1988). Because of this fact, one might predict that the discovery of genes associated with specific traits would quickly be deployed as a means of advancing the medicalization of those traits. However, the cases of depression, homosexuality, and variation in responses to chemical exposures suggest that the relationships among genetic information, geneticization, and medicalization are conditional and path dependent. Specifically, the sequence of events in which phenotypes are defined, standardized, and institutionalized as social facts shapes how genetic information may enter into geneticization and medicalization. These prior instantiations of the phenotype are consequential because they set in motion specific reinforcing mechanisms that configure the conditions of possibility for subsequent genetic research and its applications.

Critical Junctures This analysis identifies two different types of critical junctures. In the cases of depression and homosexuality, medical categorization of the condition is the key choice point. In contrast, in the case of susceptibility to chemical exposures, the critical juncture is outside the jurisdiction of medicine; the emergence of the risk assessment paradigm for environmental health regulation and policy is consequential precisely because it locates susceptibility phenotypes in institutions with missions and frames that historically have prioritized providing empirical data for use in the assessment and regulation of population-level environmental health risks. For depression, genetic causes or risks are associated with a condition that is defined as a disease entity because of the medicalization of depression in the DSM-III in 1980 and its consolidation during the 1980s and 1990s. This consolidation combined symptoms of depression that had been thought of as with cause and those that were without cause into a single phenotype that is defined and treated as a medical condition. Had these prior developments not occurred, many depressive conditions might have been considered products of distressing social circumstances, personality dispositions, or spiritual states, as they had been before 1980. The genes for major depressive disorder might have been viewed as analogous to genes for other orientations, such as political dispositions or religious beliefs, and not to those for mental disorders (Alford, Funk, and Hibbing 2005; Koenig et al. 2005). That much depression is now seen as a genetically caused disease is the result of cultural definitions, institutional forces, and political and economic interests that arose decades ago. These S304

Genetics and Medicalization earlier events ensure that genes associated with depression are understood to be causes of a disease condition. However, absent the critical definition of the DSM-III, the products of these genes would likely have been understood as something different than the disease with which they now are associated. The deletion of homosexuality from the DSM was similarly a consequence of a confluence of cultural changes, professional concerns, and political and economic interests that supported the emergence of the gay liberation movement, undermined the authority of psychoanalytic frameworks, and motivated psychiatrists to redefine their profession’s system of diagnosis and treatment. When the APA decreed that homosexuality is not a psychopathology, it removed it from medical jurisdictions. The subsequent burgeoning of gay, lesbian, bisexual, transgender, and intersex (GLBTQI) communities in the United States, the queering of identity, and the elaboration of an expansive human rights agenda established further significant bulwarks against medicalization (D’Emilio 2002). Here, geneticization and medicalization are decoupled. Scientists who study the biological bases for homosexuality frame their research as part of efforts to understand the biological underpinnings of normal variants of human sexual behavior and to dispel pernicious myths about gay men and lesbians (Hamer et al. 1993; LeVay and Hamer 1994). The path-dependent sequence shaping the consequences of research on susceptibility to chemicals begins with the rise of risk assessment as a central strategy for environmental regulation, and especially with the congressional mandates vesting the NTP and NIEHS with responsibility for strengthening the science base in toxicology, developing and validating improved testing methods, and providing information about potentially toxic chemicals to health, regulatory, and research agencies. As a consequence of these events, research on genetic susceptibilities to chemicals in the environment is located in organizations whose missions and frames (Eden 2004) have historically been oriented to improving risk assessment and regulation, rather than clinical interventions; in contrast, in the context of pharmacogenetics and pharmacogenomics, researchers are studying the same genotypes and phenotypes in pursuit of clinical applications. This critical juncture has not constrained environmental health scientists from pursuing research also focused at the individual level or with clinical applications; however, it means that such research must be coordinated with extant arrangements in a dense institutional matrix. Reinforcement Mechanisms and Conditions of Possibility A wide variety of subsequent mechanisms reinforce the effects of the critical junctures we describe. These reproductive mechanisms, in turn, S305

American Journal of Sociology configure the possible relationships among genetic information, geneticization, and medicalization. The critical junctures that define and locate phenotypes in specific structural patterns determine, in part, which professions “hold the jurisdiction” of diagnosis, inference, and treatment of a particular condition (Abbott 1988, pp. 39–40). As evident in the case of depression, when medicine holds the jurisdiction of a condition, institutional mechanisms specific to biomedicine reinforce the effects of the critical juncture. For example, the medicalization of depression is reproduced in regulations that require health insurance to pay for the treatment of major depression, in the policies of health insurance companies that reimburse for treatments for depression, and by public (NIMH) and private (e.g., pharmaceutical) investment in biomedical research on depression. Furthermore, because the genes for depression can only be defined as genes underlying an extraordinarily widespread disorder, they constitute an attractive target for scientific research. The mechanisms that sustain the medicalization of depression also structure the range of applications for genetic research on depression. Specifically, because depression is institutionally stabilized as a major mental illness, once a gene is identified as a risk factor for depression, all people who have this gene may be candidates for intervention, even if they don’t all actually have a depressive condition.11 Thus, genetic information about depression advances geneticization and medicalization via the gradient argument (Abbott 1988, p. 101); that is, by creating the possibility of a new type of condition—being at risk for depression—that might be prevented. For example, genetic tests could identify at-risk individuals, who could then be placed on long-term regimes of drug therapies (Rose 2007b). The market for such products is potentially huge, and market incentives support investment in the preventive management of at-risk groups (Lakoff 2005).12 Biotechnology companies are making large investments in techniques for uncovering possible genes for depression (Schooler 2007), and pharmaceutical companies are striving to develop a new class of more refined, genetically tailored antidepressant medications to replace the SSRIs whose patents are now expiring (Rose 2007b). In contrast, when a trait has been demedicalized, as in the case of homosexuality, medical professionals will lack a credible basis for using 11 Lakoff’s (2005) analysis of the myriad challenges faced by genetic researchers who wish to study a psychiatric disorder that is not stabilized as a biologically based mental illness offers an instructive counterexample. 12 For example, in the study by Caspi et al. (2003), 17% of the sample had genotypes with two copies of the short allele of the 5-HTT gene, and an additional 51% had one copy of the short allele. In theory, then, two-thirds of the population has a genotype that is associated with a susceptibility to depression.

S306

Genetics and Medicalization genetic information to claim it as part of their jurisdiction, even in the presence of research suggesting a genetic etiology. Credibility here refers to “not just social authority rooted in the division of labor or in organizational hierarchies . . . but also cultural authority, which rests on an actor’s capacity to offer what is taken as truth” (Epstein 1995, p. 411). The case of homosexuality vividly demonstrates how social movement activism can reinforce a critical juncture, especially by shifting regimes of credibility. Relatedly, this case provides evidence for Haydu’s (1998, p. 353) claim that sometimes “history’s switchmen come along for the ride.” That is, in contrast to the case of depression, in which there is very little redundancy in events preceding and following the critical juncture, in the case of homosexuality, social movement organization, mobilization, and institutionalization are woven throughout the sequence in which genetic research is embedded. Here also, the mechanisms that sustain the critical juncture shape the possibilities and meanings of subsequent genetic research. Scientists who study possible biological bases of homosexuality may receive public funding to investigate homosexuality as a variant of human sexual behavior, but not as an illness, and there are no significant market incentives or private-sector investments for this research. Additionally, the queering of identity in the GLBTQI community raises questions about the definition of phenotypes in genetic research (Terry 1999). At the same time, research on the biological bases of homosexuality is closely followed and debated within GLBTQI communities, and while there is variation in opinions about the gay gene, GLBTQI organizations are vigilant in their strong opposition to medical framings of sexual orientation (D’Emilio 2002). Thus, while genetic research may advance the geneticization of homosexuality, myriad social structural barriers work against the possibility that genetic research will motivate processes of medicalization. In the case of variation in susceptibility to chemical exposures, the mechanisms that shape the relationship among genetics research, geneticization, and medicalization operate in and through the institutional matrix that connects environmental health science, risk assessment, and regulation. Since the 1970s, reasoning about and preventing environmentally associated illnesses has been located in the jurisdiction of environmental health risk assessment and regulation. The production of knowledge for applications in risk assessment, regulation, and policy making is a major legitimating rationale for the NIEHS, the primary public funder of research on variation in responses to chemicals in the environment, and the NTP. The structure of these institutions and their formal and informal relationships to the regulatory agencies mean that both learning and coordination effects favor the continued production of knowledge S307

American Journal of Sociology that can be used in risk assessment and regulation, rather than as a basis for medical treatments. As a consequence of these organization-level commitments, interinstitutional arrangements, and their implications for funding for environmental health research, genetic research in the environmental health sciences must be articulated with processes of risk assessment and regulation. Thus, while some environmental health scientists express enthusiasm for the potential clinical applications of genetic information about susceptibilities to environmental exposures (Shostak and Rehel 2007), these strategies are incommensurate with the current logic of control for environmental health and illness, and activists already contest their legitimacy (Shostak 2004). Nonetheless, this analysis suggests that changes in the scientific credibility, political support, and associated funding for contemporary practices of environmental health risk assessment and regulation could undermine the organizational commitments that have positioned susceptibility to environmental exposures as relevant primarily to public policy, rather than to clinical practice. There is evidence for such trends in the environmental health arena (Shostak and Rehel 2007), due in part to market-driven knowledge production practices in pharmacogenomics and their applications in the regulation of pharmaceuticals;13 here, learning in one research and regulatory matrix may lower the costs of coordination and learning in another. Thus, by highlighting the specific mechanisms that sustain trajectories, this analysis also identifies how future events may reconfigure relationships among genetics, geneticization, and medicalization. Contingent Futures While path dependence shapes the relationships among genetic research, geneticization, and medicalization, paths are not permanently locked in and impervious to change; events that erode or overwhelm the mechanisms of reproduction that generate continuity can create opportunities for dramatic change (Pierson 2000, p. 265). Additionally, future “case breaking” events may transform the meaning and form of the cases in which they are embedded (Bearman et al. 1999, p. 506). Nonetheless, by describing how critical junctures set specific paths in motion and iden13 An environmental health scientist commented in an interview with one of the authors that, in contrast to pharmacogenomics, “in the environment, it’s more for the public good, without the big bucks waiting at the end of the line,” which caused concern at the NIEHS that “the environmental health sciences are lagging behind the pharmacological sciences pretty dramatically” (interview conducted by Shostak, July 2002). The NIEHS genomics programs launched in the late 1990s were, in part, a response to such concerns (Shostak 2005).

S308

Genetics and Medicalization tifying the mechanisms of reproduction that have generated continuity in these three cases thus far, this analysis also highlights how social structural changes can reconfigure relationships between genetic science and biomedical practices.14 For example, as the case of depression most vividly shows, when a condition has been medicalized, genetic knowledge may result in the emergence of new phenotypes through the identification of persons at risk for illness. Persons at risk are an “interactive kind” (Hacking 1999, p. 130), as the at-risk designation may be used by individuals so identified to reshape their identities and forms of life, including lifestyle, diet, leisure activities, alcohol consumption, smoking, utilization of health services, and understandings of kinship (Novas and Rose 2000, p. 490; Clarke et al. 2003). Consequently, people who identify as at risk may generate new forms of biosocial (Rabinow 1996) organization, as they come together “for mutual support, joint advocacy, and . . . activism” (Hacking 2006, pp. 91–92). Moreover, elucidating categories of persons at risk of disease may also legitimate genetic research into associated behavioral risk factors (e.g., study of the genetic underpinnings of cigarette smoking), leading to the proliferation of at-risk groups (Press 2006). Research on genetic risks for conditions institutionally stabilized as medical conditions may also contribute to the proliferation of new forms of at-risk phenotypes. For example, one goal of pharmacogenomic research on depression is to specify how different combinations of genes can produce comparable phenotypic expressions of depression that are distinct conditions at the molecular level. The goal here is to develop both genetic tests that will identify a variety of susceptibilities to depression and targeted drugs for each type of susceptibility (Rose 2007b, p. 18). Together, these new forms of social organization may reshape extant understandings of phenotypes, dimensions of social structure, and the pathways that configure relationships among genetic information, geneticization, and medical practice. This article has not described every possible sequence shaping relationships between genetics and medicalization. Our analysis clearly suggests that alternate sequences of events can be expected to produce relationships among genetics, geneticization, and medicalization different than those considered herein. For example, individuals who are suffering from a contested illness may find that genetic influences can serve as a resource for establishing the legitimacy of claims to medicalization, bolstering otherwise shaky epistemic foundations. This is particularly true 14 A central claim of the biomedicalization thesis is that technoscientific innovations have caused mesolevel institutional changes that are reconfiguring biomedicine “from the inside out” (Clarke et al. 2003, p. 162).

S309

American Journal of Sociology for contested illnesses, which are characterized by conflict among scientists, health-care practitioners, and people who are suffering from medically unexplained physical symptoms; such phenotypes tend to be clinically ambiguous, and biomedical practitioners and the institutions that support biomedical research and health-care provision view them with skepticism (Zavestoski et al. 2004). The social legitimacy and intellectual plausibility of contemporary disease categorizations often hinge on identification of a biological mechanism (Rosenberg 1997). Therefore, in the case of contested illnesses, patients, scientists, and medical professionals can use the identification of a genetic causal factor to set in motion mechanisms that can initiate and sustain medicalization. For example, in popular media coverage of the identification of three genetic variants associated with restless legs syndrome, scientists stated that the genetic discoveries provided “the prerequisite” for studying “the biological basis of restless legs syndrome,” while a doctor who treats the condition asserted that it was now “on a firmer footing” (Wade 2007). Following such cases over time will elucidate how alternative orderings of events configure the consequences of genetic research.

CONCLUSION

Despite the association of genetic research and medical interventions in both biomedical scientists’ advocacy for the Human Genome Project and social scientists’ critiques thereof, genetic information does not always lead to geneticization, nor does geneticization inevitably lead to medicalization. Rather, there is a lack of consistent fit among genetics, geneticization, and medicalization. Examining this lack of consistent fit reveals that genetic information takes its meaning from its embeddedness in different moments in sequences of events and their social structural consequences. The cases in this article highlight three specific path-dependent sequences with consequences for the relationship among genetics, geneticization, and medicalization. First, when medicalization precedes molecular genetic research, as with the medicalization of depression in 1980, geneticization and medicalization are likely to have a self-sustaining relationship. In contrast, when demedicalization precedes genetic research, as in the case of homosexuality, geneticization is decoupled from medicalization. Finally, as demonstrated by the case of susceptibility to chemicals, when phenotypes are located in two different institutional matrices, research on the same genotypes will have different epistemic and political consequences vis-a`-vis medicalization; at the same time, this case suggests the possibility of coordination across institutional matrices over time. The S310

Genetics and Medicalization sequences, critical junctures, and mechanisms of reproduction central to these cases vary. However, together they highlight that, even amid the genetic revolution, the paths history has taken shape the conditions of social action; their placement in chronological sequences of events gives particular institutions, practices, and ideas the potential to influence outcomes at later times (Stinchcombe 2005, pp. 6–7). The insights of path dependence are prominent in comparative historical studies of national regime change (Mahoney 2001) and are congruent with recent efforts in the sociology of science to focus analysis on how “the resources that actors may use are . . . structured by pre-existing arrangements” (Frickel and Moore 2006, p. 10). However, social studies of science and medicine have paid only limited attention to the dynamics of path dependence. This analysis demonstrates the importance of sequence analyses, in general, and analysis of path dependence, in particular, to understanding how events become embedded and maintained in dimensions of social structure that, in turn, shape the possibilities, meanings, and consequences of genetic research. Additionally, analysis of the relationships among genetic information, geneticization, and medicalization reveals variations in the forms of the sequences that configure the consequences of genetic information. The case of depression evinces a relatively thin and strongly driven pathway, with very little redundancy. In contrast, the path taken in the case of homosexuality has a braided quality; it is not simply sequential, as the actions of gay rights activists and advocates have defined the path iteratively over time (Haydu 1998). The case of susceptibility to chemical exposures evinces the branching quality that often characterizes pathdependent sequences. Thus, “thinking with genetics” here provides one means of answering the “call for sociological explanations that recognize historical contingency, multiple and mutable patterns of causality, and the causal importance of temporality itself” (Haydu 1998, p. 349). REFERENCES Abbott, Andrew. 1988. The System of Professions: An Essay on the Division of Expert Labor. Chicago: University of Chicago Press. ———. 1992. “From Causes to Events: Notes on Narrative Positivism.” Sociological Methods and Research 20 (4): 428–55. ———. 1995. “Sequence Analysis: New Methods for Old Ideas.” Annual Review of Sociology 21:93–113. ———. 2001. Time Matters: On Theory and Method. Chicago: University of Chicago Press. Alford, John R., Carolyn L. Funk, and John R. Hibbing. 2005. “Are Political Orientations Genetically Transmitted?” American Political Science Review 99: 153–67. Almeling, Renee. 2007. “Selling Genes, Selling Gender: Egg Agencies, Sperm Banks,

S311

American Journal of Sociology and the Medical Market in Genetic Material.” American Sociological Review 72 (3): 319–40. APA (American Psychiatric Association). 1952. Diagnostic and Statistical Manual of Mental Disorders. Washington, D.C.: APA. ———. 1968. Diagnostic and Statistical Manual of Mental Disorders, 2d ed. (DSMII). Washington, D.C.: APA. ———. 1980. Diagnostic and Statistical Manual of Mental Disorders, 3d ed. (DSMIII). Washington, D.C.: APA. Bailey, J. Michael, and Richard C. Pillard. 1991. “A Genetic Study of Male Sexual Orientation.” Archives of General Psychiatry 48:1089–96. Bailey, J. Michael, Richard C. Pillard, M. C. Neale, and Y. Agyei. 1993. “Heritable Factors Influence Sexual Orientation in Women.” Archives of General Psychiatry 50 (3): 217–23. Bayer, Ronald. 1988. Homosexuality and American Psychiatry: The Politics of Diagnosis. New York: Basic Books. Bearman, Peter, Robert Faris, and James Moody. 1999. “Blocking the Future: New Solutions for Old Problems in Historical Sociology.” Social Science History 23 (4): 501–33. Bell, Douglas A., Jack A. Taylor, David F. Paulson, Cary N. Robertson, James L. Mohler, and George W. Lucier. 1993. “Genetic Risk and Carcinogen Exposure: A Common Inherited Defect of the Carcinogen-Metabolism Gene Glutathione STransferase M1 (GSTM1) That Increases Susceptibility to Bladder Cancer.” Journal of the National Cancer Institute 85:1159–64. Beutler, Ernest. 1994. “G6PD Deficiency.” Blood 84 (11): 3613–36. Blazer, Dan G. 2005. The Age of Melancholy: Major Depression and Its Social Origins. New York: Routledge. Brewer, George J. 1971. “Human Ecology, an Expanding Role for the Human Geneticist.” American Journal of Human Genetics 23:92–94. Brookey, Robert Alan. 2002. Reinventing the Male Homosexual: The Rhetoric and Power of the Gay Gene. Bloomington: University of Indiana Press. Calabrese, Edward J. 1984. Ecogenetics: Genetic Variation in Susceptibility to Environmental Agents. New York: John Wiley & Sons. ———. 1996. “Biochemical Individuality: The Next Generation.” Regulatory Toxicology and Pharmacology 24:S58–S67. Carson, Paul E., C. Larkin Flanagan, C. E. Ickes, and Alf S. Alving. 1956. “Enzymatic Deficiency in Primaquine-Sensitive Erythrocytes.” Science 124:484. Caspi, Avshalom, Karen Sugden, Terrie E. Moffitt, et al. 2003. “Influence of Life Stress on Depression: Moderation by a Polymorphism in the 5-HTT Gene.” Science 301: 386–89. Christiani, David. 1996. “Utilization of Biomarker Data for Clinical and Environmental Intervention.” Environmental Health Perspectives 104:921–25. Clarke, Adele E., Janet K. Shim, Laura Mamo, Jennifer Ruth Fosket, and Jennifer R. Fishman. 2003. “Biomedicalization: Technoscientific Transformations of Health, Illness and U.S. Biomedicine.” American Sociological Review 68:161–94. Conrad, Peter. 2000. “Medicalization, Genetics, and Human Problems.” Pp. 322–33 in Handbook of Medical Sociology, edited by Chloe E. Bird, Peter Conrad, and Allen M. Fremont. Upper Saddle River, N.J.: Prentice Hall. ———. 2007. The Medicalization of Society: On the Transformation of Human Conditions into Treatable Disorders. Baltimore: Johns Hopkins University Press. Conrad, Peter, and Susan Markens. 2001. “Constructing the ‘Gay Gene’ in the News: Optimism and Skepticism in the US and British Press.” Health Care Analysis 5: 373–400. Cox, Susan M., and Rosalie C. Starzomski. 2004. “Genes and Geneticization? The

S312

Genetics and Medicalization Social Construction of Autosomal Dominant Polycystic Kidney Disease.” New Genetics and Society 23:137–66. D’Emilio, John. 2002. The World Turned: Essays on Gay History, Politics, and Culture. Durham, N.C.: Duke University Press. Dern, R. J., I. M. Weinstein, G. V. LeRoy, D. W. Talmage, and A. S. Alving. 1954. “The Hemolytic Effect of Primaquine: The Localization of the Drug-Induced Hemolytic Effect in Primaquine-Sensitive Individuals.” Journal of Laboratory and Clinical Medicine 43:303–9. Duster, Troy. 2006. “Comparative Perspectives and Competing Explanations: Taking on the Newly Configured Reductionist Challenge to Sociology.” American Sociological Review 71:1–15. Eden, Lynn. 2004. Whole World on Fire: Organizations, Knowledge, and Nuclear Weapons Devastation. Ithaca, N.Y.: Cornell University Press. Epstein, Steven. 1995. “The Construction of Lay Expertise: AIDS Activism and the Forging of Credibility in the Reform of Clinical Trials.” Science, Technology and Human Values 20 (4): 408–37. Foucault, Michel. 1980. The History of Sexuality, vol. 1. New York: Vintage. Frickel, Scott. 2004. Chemical Consequences: Environmental Mutagens, Scientist Activism, and the Rise of Genetic Toxicology. New Brunswick, N.J.: Rutgers University Press. Frickel, Scott, and Kelly Moore. 2006. The New Political Sociology of Science: Institutions, Networks, and Power. Madison: University of Wisconsin Press. Gamson, Joshua, and Dawn Moone. 2004. “The Sociology of Sexualities: Queer and Beyond.” Annual Review of Sociology 30:47–64. Garrod, Archibald. 1932. The Inborn Factors of Disease. Oxford: Oxford University Press. Goldstone, Jack A. 1998. “Initial Conditions, General Laws, Path Dependence, and Explanation in Historical Sociology.” American Journal of Sociology 104 (3): 829–45. Griffin, Larry. 1993. “Narrative, Event Structure Analysis, and Causal Interpretation in Historical Sociology.” American Journal of Sociology 98 (5): 1094–1133. Grob, Gerald N. 1991. From Asylum to Community: Mental Health Policy in Modern America. Princeton, N.J.: Princeton University Press. Guttmacher, Alan E., and Francis S. Collins. 2005. “Realizing the Promise of Genomics in Biomedical Research.” Journal of the American Medical Association 294 (11): 1399–1402. Hacking, Ian. 1999. The Social Construction of What? Cambridge, Mass.: Harvard University Press. ———. 2006. “Genetics, Biosocial Groups, and the Future of Identity.” Daedalus Fall 2006: 81–96. Hall, Peter A. 2003. “Aligning Ontology and Methodology in Comparative Research.” Pp. 373–404 in Comparative Historical Analysis in the Social Sciences, edited by James Mahoney and Dietrich Rueschemeyer. Cambridge: Cambridge University Press. Hamer, Dean, Stella Hu, Victoria L. Magnuson, Nan Hu, and Angela M. L. Pattatucci. 1993. “A Linkage between DNA Markers on the X Chromosome and Male Sexual Orientation.” Science 261:321–27. Haydu, Jeffrey. 1998. “Making Use of the Past: Time Periods as Cases to Compare and as Sequences of Problem Solving.” American Journal of Sociology 104 (2): 339–71. Hedgecoe, Adam M. 2004. The Politics of Personalised Medicine: Pharmacogenetics in the Clinic. Cambridge: Cambridge University Press. Horwitz, Allan V. 2002. Creating Mental Illness. Chicago: University of Chicago Press. Horwitz, Allan V., and Jerome C. Wakefield. 2007. The Loss of Sadness: How

S313

American Journal of Sociology Psychiatry Is Transforming Ordinary Misery into Depressive Disorder. New York: Oxford University Press. Hu, Stella, Angela M. L. Pattatucci, Chavis Peterson, et al. 1995. “Linkage between Sexual Orientation and Chromosome Xq28 in Males but Not in Females.” Nature Genetics 11:248–56. Jackson, Stanley W. 1986. Melancholia and Depression: From Hippocratic Times to Modern Times. New Haven, Conn.: Yale University Press. ———. 1995. Science at the Bar: Law, Science, and Technology in America. Cambridge, Mass.: Harvard University Press. Jensen, Wallace N. 1962. “Hereditary and Chemically Induced Anemia.” Archives of Environmental Health 5:212–16. Kalow, Werner. 1968. “Drug Metabolism Enzymes. Pharmacogenetics in Animals and Man.” Annals of the New York Academy of Sciences 151 (2): 694–98. Katznelson, Ira. 2003. “Periodization and Preferences: Reflections on Purposive Action in Comparative Historical Social Science.” Pp. 270–301 in Comparative Historical Analysis in the Social Sciences, edited by James Mahoney and Dietrich Rueschemeyer. Cambridge: Cambridge University Press. Kendler, Kenneth S., Jonathan W. Kuhn, Jen Vittum, Carol A. Prescott, and Brien Riley. 2005. “The Interaction of Stressful Life Events and a Serotonin Transporter Polymorphism in the Prediction of Episodes of Major Depression: A Replication.” Archives of General Psychiatry 65:529–35. Kirk, Stuart A., and Herb Kutchins. 1992. The Selling of DSM: The Rhetoric of Science in Psychiatry. New York: Aldine de Gruyter. Koenig, Laura B., Matt McGue, Robert F. Krueger, and Thomas J. Bouchard, Jr. 2005. “Genetic and Environmental Influences on Religiousness: Findings for Retrospective and Current Religiousness Ratings.” Journal of Personality 73:471–88. Kolb, Lawrence C., Shervert H. Frazier, and Paul Dirovata. 2000. “The National Institute of Mental Health: Its Influence on Psychiatry and the Nation’s Mental Health.” Pp. 207–31 in American Psychiatry after World War II: 1944–1994, edited by R. W. Menninger and J. C. Nemiah. Washington, D.C.: American Psychiatric Press. Kramer, Peter. 1993. Listening to Prozac. New York: Viking. Lakoff, Andrew. 2005. Pharmaceutical Reason: Knowledge and Value in Global Psychiatry. New York: Cambridge University Press. LeVay, Simon. 1991. “A Difference in the Hypothalamic Structure between Heterosexual and Homosexual Men.” Science 253:1034–37. LeVay, Simon, and Dean H. Hamer. 1994. “Evidence for a Biological Influence in Male Homosexuality.” Scientific American, May, 45–50. Lippman, Abby. 1991. “Prenatal Genetic Testing and Screening: Constructing Needs and Reinforcing Inequities.” American Journal of Law and Medicine 17:15–50. ———. 1998. “The Politics of Health: Geneticization versus Health Promotion.” Pp. 64–82 in The Politics of Women’s Health: Exploring Agency and Autonomy, edited by Susan Sherwin. Philadelphia: Temple University Press. Mahoney, James. 2000. “Path Dependence in Historical Sociology.” Theory and Society 29 (4): 507–48. ———. 2001. The Legacies of Liberalism: Path Dependence and Political Regimes in Central America. Baltimore: Johns Hopkins University Press. Mancinelli, Laviero, Maureen Cronin, and Wolfgang Sade´e. 2000. “Pharmacogenomics: The Promise of Personalized Medicine.” AAPS PharmSci 2 (1): art. 4. http:// www.aapsj.org/view.asp?artpps020104. Moffitt, Terrie E., Avshalom Caspi, and Michael Rutter. 2005. “Strategy for Investigating Interactions between Measured Genes and Measured Environments.” Archives of General Psychiatry 62:473–81.

S314

Genetics and Medicalization Mol, Annemarie. 2002. The Body Multiple: Ontology in Medical Practice. Durham, N.C.: Duke University Press. Motulsky, Arno. 1957. “Drug Reactions, Enzymes, and Biochemical Genetics.” Journal of the American Medical Association 165:835–37. Mountain, John T. 1963. “Detecting Hypersusceptibility to Toxic Substances.” Archives of Environmental Health 6:357–65. NAS (National Academy of Sciences). 1975. Principles for Evaluating Chemicals in the Environment. Washington, D.C.: NAS. Nelkin, Dorothy, and M. Susan Lindee. 1995. The DNA Mystique: The Gene as a Cultural Icon. New York: W.H. Freeman. NIEHS (National Institute of Environmental Health Sciences). 1997. Environmental Genome Project Symposium. Bethesda, Md.: National Institutes of Health. Novas, Carlos, and Nikolas Rose. 2000. “Genetic Risk and the Birth of the Somatic Individual.” Economy and Society 29:485–513. NTP (National Toxicology Program). 2002. Current Directions and Evolving Strategies. Washington, D.C.: Department of Health and Human Services. Olden, Kenneth, and Samuel Wilson. 2000. “Environmental Health and Genomics: Visions and Implications.” Nature Reviews Genetics 1:149–53. Olfson, Mark, Steven C. Marcus, Benjamin Druss, Lynn Elinson, Terri Tanielian, and Harold A. Pincus. 2002. “National Trends in the Outpatient Treatment of Depression.” Journal of the American Medical Association 287:203–9. Omenn, Gilbert S., and H. V. Gelboin. 1983. Genetic Variability in Responses to Chemical Exposure. New York: Cold Spring Harbor Laboratory. Pierson, Paul. 2000. “Increasing Returns, Path Dependence, and the Study of Politics.” American Political Science Review 94 (2): 251–67. Pillard, Richard C., Jeanette Poumadere, and Ruth A. Carretta. 1981. “Is Homosexuality Familial? A Review, Some Data, and a Suggestion.” Archives of Sexual Behavior 10 (5): 465–75. Press, Nancy. 2006. “Social Construction and Medicalization: Behavioral Genetics in Context.” Pp. 131–49 in Wrestling with Behavioral Genetics: Science, Ethics, and Public Conversation, edited by Erik Parens, Audrey R. Chapman, and Nancy Press. Baltimore: Johns Hopkins University Press. Rabinow, Paul. 1996. “Artificiality and Enlightenment: From Sociobiology to Biosociality.” Pp. 91–111 in Essays on the Anthropology of Reason. Princeton, N.J.: Princeton University Press. Rice, George, Carol Anderson, Neil Risch, and George Ebers. 1999. “Male Homosexuality: Absence of Linkage to Microsatellite Markers at Xq28.” Science 284:665–67. Rose, Nikolas. 2007a. “Beyond Medicalisation.” Lancet 369:700–702. ———. 2007b. The Politics of Life Itself: Biomedicine, Power, and Subjectivity in the Twenty-First Century. Princeton, N.J.: Princeton University Press. Rosenberg, Charles. 1997. “Framing Disease: Illness, Society, and History.” Pp. xii–xxvi in Framing Disease: Studies in Cultural History, edited by Charles Rosenberg and Janet Golden. New Brunswick, N.J.: Rutgers University Press. Rosenhan, David L. 1973. “On Being Sane in Insane Places.” Science 179:250–58. Schooler, Carmi. 2007. “The Changing Role(s) of Sociology (and Psychology) in the National Institute of Mental Health Intramural Research Program.” Pp. 55–63 in Mental Health, Social Mirror, edited by W. R. Avison, J. D. McLeod, and B. A. Pescosolido. New York: Springer. Schwetz, Bernard A. 2001. “Toxicology at the Food and Drug Administration: New Century, New Challenges.” International Journal of Toxicology 20:3–8. Sewell, William H., Jr. 1996. “Three Temporalities: Toward an Eventful Sociology.” Pp. 245–80 in The Historic Turn in the Human Sciences, edited by Terrence J. McDonald. Ann Arbor: University of Michigan Press.

S315

American Journal of Sociology Shostak, Sara. 2004. “Environmental Justice and Genomics: Acting on the Futures of Environmental Health.” Science as Culture 13:539–62. ———. 2005. “The Emergence of Toxicogenomics: A Case Study of Molecularization.” Social Studies of Science 35:367–403. Shostak, S., and E. Rehel. 2007. “Changing the Subject: Science, Subjectivity, and the Structure of ‘Ethical Problems.’” Pp. 323–46 in Advances in Medical Sociology: Sociological Perspectives on Bioethical Issues, edited by Barbara Katz Rothman, Elizabeth Armstrong, and Rebecca Tiger. Oxford: Elsevier/JAI Press. Smith, Lewis L. 2001. “Key Challenges for Toxicologists in the 21st Century.” Trends in Pharmacological Sciences 22:281–95. Stinchcombe, Arthur. 2005. The Logic of Social Research. Chicago: University of Chicago Press. Stokinger, H. E., J. T. Mountain, and L. D. Scheel. 1968. “Pharmacogenetics in the Detection of the Hypersusceptible Worker.” Annals of the New York Academy of Sciences 151:968–76. Surtees, Paul G., Nicholas W. J. Wainwright, Saffron A. G. Willis-Owen, Robert Luben, Nicholas E. Day, and Jonathan Flint. 2006. “Social Adversity, the Serotonin Transporter (5-HTTLPR) Polymorphism and Depressive Disorder.” Biological Psychiatry 59:224–29. Szasz, T. 1961. The Myth of Mental Illness. New York: Hoebel-Harper. Tarlov, Alvin R., George J. Brewer, Paul E. Carson, and Alf S. Alving. 1962. “Primaquine Sensitivity: Glucose-6-Phosphate Dehydrogenase Deficiency; An Inborn Error of Metabolism of Medical and Biological Significance.” Archives of Internal Medicine 109:209–34. Terry, Jennifer. 1999. An American Obsession: Science, Medicine and Homosexuality in Modern Society. Chicago: University of Chicago Press. Thelen, Kathleen. 1999. “Historical Institutionalism in Comparative Politics.” Annual Review of Political Science 2:369–404. ———. 2003. “How Institutions Evolve: Insights from Comparative Historical Analysis.” Pp. 208–40 in Comparative Historical Analysis in the Social Sciences, edited by James Mahoney and Dietrich Rueschemeyer. Cambridge: Cambridge University Press. Vaughan, Diane. 1992. “Theory Elaboration: The Heuristics of Case Analysis.” Pp. 173–202 in What Is a Case? Exploring the Foundations of Social Inquiry, edited by Charles Ragin and Howard S. Becker. Cambridge: Cambridge University Press. Wade, Nicholas. 2007. “Scientists Find Genetic Link for a Disorder (Next Respect?).” New York Times, July 19. Wakefield, Jerome C. 2002. “Environmental Genome Project: Focusing on Differences to Understand the Whole.” Environmental Health Perspectives 110:756–60. Weber, Max. (1905) 1958. Translated by T. Parsons. The Protestant Ethic and the Spirit of Capitalism. New York: Charles Scribner’s Sons. Weber, Wendell. 2001. “The Legacy of Pharmacogenetics and Potential Applications.” Mutation Research 479:1–18. Zavestoski, S., P. Brown, S. McCormick, B. Mayer, M. D’Ottavi, and J.C. Lucove. 2004. “Patient Activism and the Struggle for Diagnosis: Gulf War Illnesses and Other Medically Unexplained Physical Symptoms in the US.” Social Science and Medicine 58:161–75.

S316