Regimes of Perceptibility and the Social Determinants of Health
Introduction
Many early social scientific critiques of the Human Genome Project (HGP) used the concept of geneticization as their point of departure. Indexing myriad potential negative social consequences – including genetic reductionism, determinism, essentialism and fatalism – the concept of geneticization made strong claims about the potential power of genetic information. Social scientists were especially concerned with the power of genetics to shift scientific research, public understanding and scarce resources away from the social and environmental determinants of health. Two decades after the first writings on geneticization and more than a decade into the post-genomic era, the social and environmental determinants of health remain prominent concerns in both scientific and public arenas. For a wide array of scientists, “the environment” – varyingly defined – and its effects on health are primary research foci. We do not see this as a simple repudiation of the geneticization thesis. Rather, we contend that in the current moment, addressing the critical concerns at the heart of the concept of “geneticization” demands new conceptual frameworks.
Simply put, the assumption of the geneticization critique that genes and environments are mutually exclusive explanatory frameworks for human health and illness is an insufficient conceptual basis for contemporary analyses of biomedical research and practice. For while “the environment” is a focal concern across disciplines in the life sciences and social sciences, there is no consensus within or across disciplines as to how it should be conceptualized or operationalized. Indeed, some operationalizations of the environment render it an internal, individual attribute that conveys health risks at the molecular level – very much like a gene. As such, it is now clear that merely focusing research on environmental or social determinants of health does not necessarily lead to social structural understandings of health and illness as implied by the concept of geneticization. Rather, much depends on how genes, environments and their interactions are conceptualized and operationalized in biomedical research.
Consequently, in order to understand contemporary practices in biomedical science, we need analytic frameworks that support inquiry into “historical ontology,” that is, an examination of the history of things – such as “genes,” “the environment” and “the social determinants of health” – as they are formed in relationship to specific epistemological traditions and practices (Mitman, Murphy and Sellers 2004). Drawing on work in the history of science, we deploy the concept of “regimes of perceptibility,” which directs analytic attention to questions about what factors, or levels of analysis, are made visible or obscured by specific scientific techniques and technologies. We make our case for this approach by applying it to three determinants of health – diet, toxic chemicals and stress – that scientists from different disciplines have investigated at varying levels of analysis. Further, we argue that this framework can be extended to ask questions about how a political economy of perception may favor particular conceptualizations and operationalizations of the environment.1 Thus, while moving on from now outdated assumptions embedded in the concept of geneticization, we envision new scholarly engagements that maintain fidelity to some of its central social justice concerns. We begin with a review of the concept of geneticization and the wealth of empirical studies it has generated.
The Geneticization Critique
The concept of geneticization entered social scientific research through the work of Abby Lippman, a professor of epidemiology at McGill University and a dedicated women’s health activist. As introduced by Lippman, geneticization refers to “an ongoing 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” (1991:19). Lippman understood the concept of geneticization as a jumping off point for a broad “social justice critique” of the new genetics. In an interview with the Canadian Women’s Health Network in 1999, she defined geneticization as “the tremendous extent to which ‘genetics’ is taking precedence over how we see health and social problems.”2 Similarly, prominent social scientists and humanists highlighted the potential of genetics to supplant public health approaches to complex diseases and to undermine efforts to identify the social and environmental determinants of health (Conrad 1999; Sherwin and Simpson 1999). In an oft-cited work, Duster (2003) argued that extensive public sector investment in genetic research would disproportionately and negatively impact Blacks by diverting attention and resources away from social and environmental factors that contribute to increasing rates of lung cancer and cardiovascular disease in the African American population (see also Conrad 1999; Duster 2006).3
In these early writings, geneticization further served as connotative short-hand for various related concerns about how new genetic findings might be understood and used, including:
1 genetic reductionism, in which a complex and ecumenical understanding of the causes of human development is supplanted by one in which genes are perceived as the “true cause” of difference4 (Sloan 2000:17),
2 genetic determinism, in which genes are taken as inevitably implying traits and behaviors (Lippman 1992; Alper and Beckwith 1993; Rothman 2001; Nelkin and Lindee 2004),
3 genetic essentialism, in which genetics becomes a dominant way “to explore fundamental questions about human life” and “to talk about guilt and responsibility, power and privilege, intellectual or emotional status” (Nelkin and Lindee 2004:16).
Lippman believed that much of the new genetics was motivated by “two E’s … economics and eugenics.” As a women’s health advocate, she hoped that “there’s enough strength in the women’s community, and enough belief that this is not the way to go, so that we can mobilize and put brakes on geneticization.”5
Unsurprisingly, given the critical stance built into the original formulation of the concept, there has been debate about whether the term “geneticization” can be used to neutrally characterize social phenomena (Freese and Shostak 2009).6 At issue is whether “geneticization” is too “ethically loaded” or “morally circular” to serve as a framework for empirical research (Hedgecoe 1998). Some writers have suggested that it should be understood as “a heuristic tool” in a moral debate (Ten Have 2001). Others have countered that conceptual writings about geneticization are replete with claims that can be evaluated with the tools of empirical research (Hedgecoe 2001) and social scientists have generated a substantial literature towards that end.
This empirically oriented literature reveals that genetic concepts, technologies and practices have multiple and contradictory effects (Freese and Shostak 2009). Analyses of biomedical research articles have found that genetic information can reshape disease categories (Hedgecoe 2002) and that scientists privilege genetic explanations over non-genetic explanations, at least in part because genetic causes are seen as more easily specified and researched (Hedgecoe 2001; see also Shim 2005). Content analyses of newspaper articles find that media coverage of research on gene–environment interaction selectively emphasizes the gene “half” of the interaction and largely ignores environmental causes of disease (Horwitz 2005). There are discrepancies among studies that seek to assess whether geneticization in the print media is increasing. For example, Duster (2003) found an increase in articles invoking a genetic explanation of crime in the late 1970s and early 1980s. Condit, Ofulue and Sheedy (1998) found, over a longer timespan, that, if anything, discourse about heredity may be becoming less deterministic. Research by Phelan and colleagues (2013) demonstrates that news articles discussing racial differences in genetic bases of disease increased significantly between 1985 and 2008 and were significantly less likely than non-health-related articles about race and genetics to discuss social implications. However, Condit (1999) has warned against any presumption that the use of supposedly geneticized metaphors in the media (e.g., “genetic blueprint”) implies that public conceptions of genetic causation are correspondingly deterministic or reductionistic.
There is clear evidence that people attribute individual health and social outcomes to genetics (Shostak et al. 2009). However, research that moves from the relative simplicity of the printed word to the messy realities of clinics, patient advocacy groups and the daily experiences of those living with illness – or the risk of illness – has suggested that neither the lived realities nor the broad implications of genetics are as simple as early writings suggested. Against the notion that geneticization is an inevitable consequence of genetic research, empirical analyses have highlighted the power of daily practices of diagnosis and treatment of disease to enable or impede geneticization, even for conditions with simple (i.e., autosomal dominant) genetic etiologies (Cox and Starzomski 2004; Kerr 2005).7 In contrast to prior conceptualizations of genetics as a deterministic discourse, this research points to the enduring power of local knowledge (Rapp 2000), national health policy regimes (Parthasarathy 2007) and the institutional embeddedness (Shostak, Conrad and Horwitz 2008) and social meanings (Sankar et al. 2004) of medical conditions in shaping lay understandings of and responses to genetic information. In an explicit challenge to the assumption that genetic information will lead to fatalism, a growing cadre of researchers contend that genetic information creates new obligations to act on knowledge to protect health and to maximize quality of life (Novas and Rose 2000; Frosch, Mello and Lerman 2005; Rose 2006; Gibbon 2007).
Moreover, many users of genetic testing appreciate the nuances of probabilistic risk and predictive uncertainty, and are correspondingly circumspect in their interpretations of genetic information. This appears to be the case even in regard to prenatal genetic testing, arguably the clinical setting wherein genetic testing is the most routinized (Markens, Browner and Press 1999; Rapp 2000; Franklin and Roberts 2006; Thompson 2007). Parents of children with genetic conditions (e.g., Klinefelter, Turner and fragile X syndromes) may simultaneously accept the authority of molecular genetic test results and create a space for uncertainty about the condition by emphasizing variation between diagnosed children, the individuality of their diagnosed child and his or her accomplishments, and other ambiguities in prognosis that complicate the significance of genetic information (Whitmarsh et al. 2007; see also Timmermans and Buchbinder 2013). Even when individuals embrace the idea that a disease “runs in the family,” they interpret information about genetic susceptibility in the context of their own beliefs about the multiple causes of illness, patterns of inheritance and observable risk factors in their families (Meiser et al. 2005; Lock et al. 2006; Shostak, Zarhin and Ottman 2011).
A Place for the Environment
In contrast to the early predictions of the geneticization thesis, the genomic revolution has not resulted in a complete erasure or “undoing” (Frickel et al. 2010) of the environment in research on human health and illness. In fact, as noted by researchers across the life sciences and social sciences, genomics has had the paradoxical effect of calling attention to the importance of social and environmental factors vis-à-vis health (Olden and White 2005; Pescosolido 2006; Schwartz and Collins 2007; Bearman 2013). As scholarly observers of the contemporary life sciences have noted, “It is almost ironic that the deeper biologists delve into the human body and the more fine-grained and molecularised their analyses of the body become, the less they are able to ignore the many ties that link the individual body and its molecules to the spatio-temporal contexts within which it dwells” (Niewöhner 2011:290). Within the life sciences, researchers assert that current research makes it absolutely clear that “the genome cannot operate independently of its environmental contexts – both external and internal to the body” (Meaney 2010, in Lock 2013). Indeed, a hallmark of the post-genomic era is the imperative that scientists elucidate the role of the environment in shaping the processes and outcomes of gene action (Shostak and Moinester 2015).
In fact, soon after the completion of the Human Genome Project (HGP) in 2003, the leadership of the National Human Genome Research Institute (NHGRI) began making statements about the importance of understanding gene–environment interactions. In a paper entitled “Welcome to the Genomics Era,” Guttmacher and Collins noted that it “bears repeating” that “even in the genomic era, it is not genes alone but the interplay of genetic and environmental factors that determines phenotype (i.e., health or disease)” (2003:997; emphasis added). In pursuit of better understanding this “interplay,” the NHGRI convened a “Gene Environment Interplay Workshop” in 2010. This meeting brought together 150 scientists, representing a wide array of fields, to evaluate the state-of-the-science in the study of gene–environment interactions in complex diseases, and make recommendations regarding research priorities, challenges and next steps (Bookman et al. 2011). The participants at the workshop called for an “integrative” model of complex diseases, emphasizing particularly the need to bring measures of environmental exposures into biomedical research. Today, the leadership of the NHGRI also advocates for the “integration” of genomic information” with “environmental exposure” in order to generate “a much fuller understanding of disease aetiology” (Green, Guyer and National Human Genome Research Institute 2011:208).
It has not escaped the notice of researchers that this emergent intellectual agenda will require new and perhaps novel conceptualizations – and measurements of – the environment in gene–environment interaction (Boardman, Daw and Freese 2013; Fletcher and Conley 2013). This has led to calls for better measurements of environmental exposures and for the inclusion of exposure data in large databases, including, but not limited to, those being used in genomics research (Wild 2005; Schwartz and Collins 2007; Rappaport and Smith 2010). These calls often emphasize that understanding common, complex diseases requires “that both environmental exposures and genetic variation be reliably measured” (Wild 2005:1847). At the same time, scientists note that the disproportionate investment, to date, in genetics research, means that they often use “cutting edge” genomics technologies to assess genetic variation and gene expression, while using self-report questionnaires to characterize environmental exposures (Rappaport and Smith 2010:460).
However, cross-disciplinary scientific consensus about the importance of understanding “the environment” in gene–environment interaction is not the same as a cross-disciplinary scientific consensus about how the environment should be conceptualized or measured. For example, within the life sciences, the environment may refer to the cell (the environment of the gene), endogenous hormonal profiles (the environment of the cells), indoor or outdoor physical environments (the environment of the human body) or stressful life situations (the social environment).
Simply put, there is a remarkable degree of heterogeneity in the levels of analysis at which a single concept – “the environment” – can be operationalized. Insofar as the geneticization framework implies a mutually exclusive relationship between genes and environments in explanations of human health and illness, this heterogeneity, along with the increasing focus across disciplines on the interplay of genetic and environmental factors, is potentially obscured. By directing analytic attention instead to questions regarding the factors and levels of analysis made more or less visible by specific scientific techniques and technologies, the concept of “regimes of perceptibility” (Murphy 2006) provides greater analytical leverage for investigating the diverse processes and outcomes of contemporary biomedical research about genes, environments and human health and illness.
Regimes of Perceptibility
The concept of “regimes of perceptibility” comes from historian Michelle Murphy’s research on the emergence of Sick Building Syndrome in the 1980s (2006). Murphy’s analysis contrasts scientific, corporate and laypersons’ approaches to apprehending the presence and effects of chemical exposures inside the office buildings where (mostly female) workers reported a constellation of health problems, such as headache, rashes and immune system disorders. She demonstrates how specific techniques of measurement make particular aspects of the environment more and less visible, with consequences for what dimensions of the environment are perceived as more or less real and actionable. As she notes, domains of imperceptibility are the “inevitable result” of the tangible ways that domains of perceptibility are established: “the history of how things come to exist is intrinsically linked to the history of how things come not to exist, or come to exist with only uncertainty or partiality” (Murphy 2006:9). By regularizing, standardizing and sedimenting the contours of perception and imperception within academic disciplines, regimes of perceptibility determine which objects have the opportunity to populate the worlds of the lab, the clinic and the community – and which do not (Murphy 2006).
The concept of “regimes of perceptibility” points to a new set of questions about how genes and environments are studied in research on human health and illness. It suggests that we must ask not only “is it genes or the environment?” but also consider how the environment is conceptualized, how it is operationalized and, consequently, which dimensions of the environment can be seen and which become invisible. Likewise, we may ask questions about how regimes of perceptibility make some actors – whether individual or institutional – more or less credible witnesses and with what consequences for different kinds of action. As such, the concept of regimes of perceptibility also offers analytic leverage on how some objects or situations, such as the environment, become imbued with “complexity” or “uncertainty,” while others, such as genes, are perceived to be more easily ascertained and actionable.8
Perceiving Exposures
We deploy the concept of “regimes of perceptibility” to examine three kinds of environmental exposure – diet, toxics and stress – relevant to health disparities. In each case, we describe different materializations of the environment in contemporary research on human health and illness. We find that some materializations of these environmental exposures include social institutions, contexts and processes, while others render “the environment” an internal, individual attribute. Taken together, these brief case studies demonstrate how environmental exposures may be made more or less social in nature, depending on the techniques and technologies that scientists use to study them. These renderings, in turn, have consequences for how scientists, policymakers and the public may understand human health and seek to ameliorate health disparities. Therefore, we contend that the underlying social justice concerns of the geneticization critique now demand a different kind of inquiry, one focused on regimes of perceptibility and how they shape knowledge production and policymaking regarding the social and environmental determinants of health.
Diet
Nutrition was among the earliest foci of public health efforts in the USA (Rosen 1993). More recently, concern about the rising incidence of obesity has highlighted the role of diet in a wide variety of diseases, including diabetes, cardiovascular disease and certain types of cancer. There are marked health disparities for these diseases, which often include earlier onset of illness and more severe disease among minorities compared to Whites (Williams et al. 2010).
At the same time, a very different line of research – the emergent field of nutritional epigenetics – has rendered diet an “environmental exposure” that is assessed at the molecular level (Landecker 2011:167). Indeed, in nutritional epigenetics, “food stands in for the environment in the dyad of ‘gene–environment interactions’” (Landecker 2011:168). It is against this background that we can make sense of a recent paper in the Annual Review of Nutrition that calls diet “the greatest single source of chemical exposures, including nutrients, nonnutritive chemicals, pesticides, and others” (Jones, Park and Ziegler 2012:186; emphasis added).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
