Getting Bigger: Children's Bodies,Genes and Environments

Childhood obesity and precocious puberty: ¹ what's the problem?

Puberty and obesity are two sites of human growth and development exhibiting changes at the population level that are perceived to be problematic in their own right and are considered harbingers of catastrophic futures. In the case of puberty, many argue that there has been a significant drop in the normal age of onset across the last two centuries (from 17 in 1830 to under 14 in 1960 and today to closer to 10) (Parent et al., 2003: 673; Ellis, 2004: 926): a fact that is increasingly conceptualized not as an outcome of improving child health (better nutrition, more exercise), but as an indicator of wide-spread pathology, exemplified by an accompanying increase in the frequency of what is called ‘precocious’ puberty. Such pubertal development, occurring before 6–9 years of age, ² is widely regarded as requiring medical treatment for several reasons; negative effects on height, increased risk of early sexual activities and drug-taking, and social difficulties associated with being physically ‘out of sync’ with emotional age and the development of age-peers (Parent et al., 2003; Herman-Giddens, et al., 2004; Palmert and Boepple, 2001). Such broad shifts in the age of onset of puberty (both ‘normal’ and ‘pathological’) are understood as indicating significant change in the hormonal conditions of human (and non-human) life (often linked to increases in exposure to toxic chemicals), raising concern about the future of human reproduction itself. Reviewing the clinical literature on puberty and menarche, for example, DiVall and Radovick (2008: 25) mention that a recent anniversary issue of Science magazine listed ‘What triggers puberty?’ as ‘one of the 100 most compelling questions facing science in the next century’ and note that puberty is ‘a process that is necessary for the very propagation of our species’. Changes in puberty then, and particularly an increase in precocious puberty, are thought to constitute a threat to human survival.

Rising obesity rates among children (and in adults) are also seen as signalling a health crisis of epidemic proportions, with children who have reached extremes in body size held up as warnings of what lies ahead, both for them as individuals and at the population level (James et al., 2001; Rossner, 2002; Wardle, 2005). The government publication Healthy Weight, Healthy Lives: A Cross Government Strategy for England (Cross Government Obesity Unit, 2008: x) argues that one third of children and two thirds of adults in England are currently either overweight or obese, predicting that this will rise to two thirds of children and almost nine tenths of adults by 2050. Children are identified as particular cause for concern ‘because of evidence suggesting a ‘conveyor-belt’ effect in which excess weight in childhood continues into adulthood’ (2008: 1).

Precocious puberty and obesity are linked by a number of common features. Firstly, both are understood to be the products of the interaction of genes and the environment (Weinsier et al., 1998; Palmert and Boepple, 2001; Loos and Bouchard, 2003; Astrup et al., 2004). In both cases, it is presumed that genetics cannot provide an adequate or total explanation for recent trends, since changes at the genetic level do not occur within such a short timescale; but neither has it proved possible, in either case, for scientists or journalists to relinquish attempts at genetic explanations. Genes cannot be blamed, yet are always invoked. As we will show, the explanatory space that this paradox creates is filled in both cases with discourses of genetic predisposition in interaction with environments. However, the specific nature of those interactions remains elusive, with genes figured as static, underlying factors that are acted upon by external forces in indeterminate ways.

Secondly, precocious puberty and childhood obesity not only share their roles as harbinger of future crisis, but are also conceptualized as biologically connected processes, with weight gain seen as a potential consequence of puberty, especially for girls, and excess body fat figured as a potential contributor to precocious puberty (see, for example, Ebling, 2005; Graber et al., 1999; DiVall and Radovick, 2008: 24; Lee et al., 2007). Furthermore, both the precociously pubertal body and the obese body are understood as bodies out of control. In both cases, this figuration is linked to particular ideas about gender-and age- ‘appropriate’ physical appearance and the need for normative performances of gender for children to achieve happiness and health. In particular perceived failure to achieve, or deviation from, normative standards of femininity exposes girls to significant medical interventions to ‘treat’ their pathologised bodies. This connection is discussed in the later part of the chapter.

Thirdly, both obesity and precocious puberty are poorly understood. Although biomedical discourse speaks confidently about what happens when a child goes through puberty, little is known about what initiates those changes, in both ‘normal’ and ‘pathological’ cases (Ebling, 2005; Hermann-Giddens, Kaplowitz and Wasserman, 2004; Gianetti and Seminara, 2008; DiVall and Radovick, 2008). Similarly, the bodily mechanisms of weight loss and gain are poorly understood; an effective understanding of mechanisms of appetite and satiety, for example, remains elusive (Shell, 2002; Loos and Bouchard, 2003; Astrup et al., 2004; Gard and Wright, 2005). These shared uncertainties provide a further basis for interrogating the complexity of the relationship between the conceptual domains (biology, society, nature, genetics) within which ‘getting bigger’ is negotiated scientifically.

However, there are also important differences between obesity and precocious puberty that create openings for rethinking the relationships between bodies, genes and environments in, and across, both contexts. Firstly, while both puberty and obesity are seen as processes of becoming, and while the environments that are seen as creating both of these ‘problems’ are social, in puberty discourses, biological and metabolic processes remain central. Obesity, on the other hand, is conceptualized not as a bodily process, but as a problematic bodily state (that can be ‘diagnosed’ via a simple calculation of kg/m²⁾). Puberty, then, is something that we go through, while obesity is something that we are. Popular discourse on obesity, as a result, tends to jump from behaviours (consumption/inactivity) to obesity (the problematized bodily state), without focusing on the processes (such as metabolic responses) which the body undergoes whilst getting bigger. However, it is also the case that within the medical context, obesity is increasingly conceptualized in metabolic terms – for example, in the recent re-naming of the British Obesity Surgery Society as the British Obesity and Metabolic Surgery Society ³ – and research focused on the bodily process of weight gain, with the goal of producing mechanism-specific drug interventions (see, for example, Bray and Tartaglia, 2000; Lichtenbelt et al., 2009).

Secondly, precocious puberty and childhood obesity are framed as very different kinds of ‘problems’. In puberty, ‘normal’ development is understood as largely genetically determined (some are even prepared to put a figure on this: Wehkalampi et al. (2008) argue that 86 per cent of variance of pubertal timing in girls is genetic, compared to 82 per cent for boys), whilst ‘abnormal’ puberty is seen as caused by the action of the environment – which could mean anything from nutrition to endocrine disrupting chemicals to the absence of a father figure – on those ‘normal’ genes and gene pathways (Parent et al., 2003; Banerjee and Clayton, 2007; Gianetti and Seminara, 2008; Roseweir and Millar, 2009). In this sense, in precocious puberty, ‘normal’ gene pathways or cascades (the most frequently cited being the kisspeptin/ KISS1R system) are disrupted by external influences. In contrast, obesity is commonly conceptualized as resulting from an undesirable genetic propensity to fatness that is enabled by environmental and social factors. Weinsier et al. (1998: 148) describe this in terms of ‘previously silent’ genetic variants becoming manifest: ‘Simply put, our genes permit us to become obese; the environment determines if we become obese’. This is linked to the third key difference between precocious puberty and obesity – the moral freight that each condition brings with it, both for the individuals involved, and in the case of children, for those responsible for their care.

When precocious puberty occurs, it is conceptualized as a tragedy for individuals, but one for which those caring for them are not responsible (see, for example, O'Sullivan et al., 2002). If the environment is to blame, this is not an environment that is understood as controllable by adults in any ‘simple’ way. ⁴ Conversely, obesity in children is seen as the result of the moral failure of adults, although blame can also be seen to be falling increasingly on children, who are exhorted to take individual responsibility for their size (see, for example, Evans et al., 2003; Evans et al., 2004). The opposing discourses of victimhood (the tragic disruption of ‘normal’ gene pathways in precocious puberty) and moral failure (the enablement of genetic propensities towards obesity) underpin the differential moral framings of the two ‘problems’. This distinction also highlights the fact that it is the timing of puberty that is seen as problematic, rather than puberty per se. Obesity, on the other hand, is always conceptualized as problematic and undesirable, with the focus on children positioned as an attempt to head adult obesity off at the pass, rather than to see children through a particular stage in their development. These differences in framing signal the different freight of individualized responsibility that each condition carries and raise questions about how such freight might be better acknowledged and accounted for in medical and scientific research, as well as scientific and popular discourses. In the final part of this chapter, we suggest a theorization of gene/environment systems that allows such questions to be centrally positioned rather than sidelined.

Following the genes and gene pathways in our two case studies

Although, as argued in the introduction to this volume, simple ‘gene for’ claims are rarely considered sustainable in contemporary genetics, in our two case studies the search for a genetic basis has not been abandoned, with considerable effort and funding being invested in establishing the mechanisms and proportions of genetic influence and the relationships between different genes and/or gene pathways and particular environments. Twin and adoption studies are frequently cited as evidence of heritable genetic predisposition to obesity; Friedman (2004: 563), for example, argues that ‘the heritability of obesity is equivalent to that of height and greater than that of almost every other condition’. However, the reported heritability of obesity from these studies varies enormously. Loos and Bouchard (2003: 403), for example, found estimates ranging from 5 per cent to 90 per cent, with study design and the kinds of families studied generating dramatically varying results. Nevertheless, they are unwilling to relinquish genetically determined obesity, listing multiple ‘candidate genes’, and setting out four categories of genetic determination for obesity that reflect different degrees of gene-environment interaction: genetic obesity (rare monogenetic disorders), strong genetic predisposition, slight genetic predisposition and genetic resistance to obesity/obesogenic environments. A very small number of monogenetic obesity disorders have been identified, but these are very rare (<0.1% of the population according to Frayling et al. (2007: 889)) and are therefore unable, according to Loos and Bouchard (2003: 409) to ‘explain the magnitude of the obesity problem that industrialized societies are facing today’. Instead, as is evident in the term they use – ‘polygenic/common forms of obesity’ (p. 409) – most obesity is seen to be aetiologically complex. This recognition of complexity drives a search to identify multiple candidate genes and their interactions with other genes and environments. According to the 2005 update to the Obesity Gene Map (Rankinen et al., 2006), 253 human obesity quantitative trait loci (QTL) ⁵ have been identified (up from 208 in 2003 (Snyder et al., 2004), and just 44 in 1999 (Chagnon et al., 2000)), marking the increasing scientific focus on identifying the genetic pathways through which obesity develops. In 2007, Frayling et al. published a paper in Science announcing the identification of the FTO (fat mass and obesity) gene, stating that the 16 per cent of adults who are homozygous for the risk allele in their study weighed approximately 3 kg more than those without it. This was taken up excitably in the press, signalling a strong public appetite for these genetic explanations: ‘Obesity is not just gluttony – it may be in your genes’ announced the Guardian (Randerson, 2007); and, less moderately, in the Daily Mail: ‘Proof: There IS a gene that makes you fat’ (McRae, 2007). The research team, themselves, however, reflected far more uncertainty in their own assessment of their discovery, pointing out that ‘FTO is a gene of unknown function in an unknown pathway […]’ (Frayling et al., 2007: 893).

In early puberty, one particular gene pathway has received much attention since being identified in 2003 as critical to the initiation of normal puberty. This pathway, involving the hormone kisspeptin and its receptor gene KISS1R, has been extensively studied in mice, monkeys, sheep and in particular small groups of humans (case studies of individuals or families experiencing ‘pathological’ puberty), with some arguing that KISS1R can be understood as ‘crucial’ in relation to puberty (Gianetti and Seminara, 2008: 299; Roseweir and Millar, 2009), whilst others place it only amidst a longer list of candidate genes (DiVall and Radovick, 2008: 24). Although the kisspeptin/ KISS1R system might be necessary for normal puberty to occur, this gene pathway cannot explain either normal or abnormal puberty: as Messager (2005) argues, it remains unclear what causes the initial secretion of kisspeptin in the brain. Despite this, media accounts of this work stated that ‘A team of researchers … recently pinpointed a gene that they believe controls puberty through the regulation of a protein called GPR54 [the earlier name of kisspeptin]’ (Roberts, 2005, emphasis added) or, more lightheartedly, described kisspeptin as the ‘chemical kiss that turns kids into adolescents’ (McKie, 2005). ‘It is sealed with a kiss,’ one report stated, ‘Researchers have found that a protein called kisspeptin triggers the cascade of biochemical changes that leads to puberty and turns children into hormonally challenged adolescents’ (McKie, 2005).

In both conditions, ‘genes’ or ‘genetic predispositions’ are thought to operate in interactions with each other and/or the environment (including metabolisms). In obesity research, this fact is repeatedly cited as an obstacle to effective testing for a propensity towards the condition. Marti et al. (2004: S29) argue, for example, that ‘The occurrence of gene x gene and gene x environmental factors interactions makes it more difficult to interpret the specific role of genetics and lifestyle factors’. Complexity and interaction here are figured as muddying the waters, obscuring underlying truths and certainties which are deemed to be knowable in the future. Indeed, one of the common features of the articles we reviewed is a caveat about the limitations of what can be known from a particular research finding and the need for further research (see, for example, Frayling et al., 2007; Marti et al., 2004).

These models of gene-environment render individuals genetically susceptible to pathology in some environments. Sometimes, this is interpreted as having the ‘“right” genes in the “wrong environment”’ (Rosmond, 2004: 180), as in the case of the so-called ‘thrifty gene’ – a presumed genetic propensity for the body to store fat as insurance against famine (Neel, 1999; Spiegelman and Flier, 2001). ⁶ Indeed, Hill and Peters (1998: 1371) suggest that increases in body mass in order to restore energy balance could be seen ‘not as a result of defective physiology, but as the natural response to the environment’ (emphasis added). This model positions genes and environments as fundamentally mismatched, mirroring the nature-nurture dichotomy discussed above, with the ‘natural’ response figured here as appropriate, and even positive. From a different perspective, a ‘faulty’ gene is given expression through exposure to environmental factors; as Kopelman (2000: 635) suggests, ‘implicit to the susceptible gene hypothesis is the role of environmental factors that unmask latent tendencies to develop obesity’. ⁷

In each case, ‘genes load the gun, but environmental factors pull the trigger’ (Perrin and Lee, 2007: 308), either by enabling an undesired propensity or by being unable to constrain it. The action (or the potential action/expression) of genes is figured as static – a ‘fact of nature’ to be discovered (‘unmasked’) – while the environment (‘nurture’) is, almost by definition, changeable. This presumed changeability of the environment renders it the obvious site of intervention in response to the ‘problems’ of both precocious puberty and childhood obesity (and obesity in general). In obesity, in the absence of any viable genetic intervention (apart from genetic tests to detect the presence or absence of a particular gene or gene sequence) the conclusions of papers habitually fall back on the significance of so-called ‘lifestyle’ changes in diet and exercise oriented both towards the achievement of weight loss, or especially in the case of children, the prevention of weight gain (see Gard and Wright, 2005: ch. 7). John Hewitt, for example, in his 1997 paper on obesity asking, ‘What have genetics studies told us about the environment?’ concludes that ‘Genetic studies are helping us to refine our understanding of environmental risks and to focus preventive or ameliorative efforts on those categories of risk that are most likely to be important’ (1997: 357). In another example, Friedman (2000: 634), while arguing that ‘tremendous scientific opportunities abound’, asserts that ‘for the moment, there is no panacea. So the final message is this … weight loss and exercise improve health’, although with the recognition that ‘a robust biological system makes it exceedingly difficult for most individuals to maintain weight loss for an extended period of time’ (Friedman, 2000: 634). Indeed, Speakman (2004: 2094S) suggests that ‘Addressing the genetic side of this interaction is potentially a far more tractable problem than addressing the environmental component by re-engineering society, because the level at which interventions might ultimately be made is the individual rather than society as a whole’. However, in the absence of a feasible, permanent intervention at the genetic level – ‘the ultimate ‘holy grail’ solution’ (p. 2102S) – he is forced to return to the more prosaic strategy of drugs (which are not yet available, but which we are ‘likely to see’) ‘at the same time as calorie restriction’ (p. 1021S).8

Similarly, in precocious puberty, the site of intervention has to be ‘the environment’, or at least the hormonal elements of the physiological processes involving genetic cascades; treatments for precocious puberty involve drugs which intervene in neuroendocrinological processes and prevent the cascading effects of hormones and gene pathways on developing bodies. Decisions to prescribe these medications are complex, because unlike obesity (which is never considered normal or acceptable and is therefore always open to intervention), ‘normal’ and ‘precocious’ puberty are slippery terms (puberty is ultimately desirable, but needs to happen at the ‘right’ time) and the decision to treat is made on a case-by-case basis. As for childhood obesity, however, genetic knowledge provides no direct avenue for intervention: the promise, as discussed in more detail below, is only to detect the children most at risk.

What kinds of intervention are possible?

Technoscientific claims about the proportion of influence accorded to genes versus other factors in obesity and early puberty are not simply a matter of intellectual argument, then, but constitute a debate that has serious implications for action. Although notions of genetic causes rarely lead to interventions at the level of the gene, the kinds of other interventions developed and offered are dictated by scientific beliefs about cause. If genes are implicated in both childhood obesity and pathological puberty, the focus of hope is to identify particular genes or gene sequences in order to find those individuals who are more likely to suffer from the condition, even before any evidence of ‘symptoms’. The hopes attendant on such early identification is that intervention could prevent the onset of conditions that are otherwise very difficult to treat, allow for scarce resources to be allocated most effectively, and produce individualized preventative strategies.

For children understood to be at genetic risk of precocious puberty, this dream is strong: in part because it is much easier to prevent puberty with hormonal drugs than it is to slow it down or reverse it once it has started. The clinical literature points to the benefits of this procedure for both children and parents, who would ideally be saved the trauma of unusually early development. Phillip and Lazar, for example, in their paper ‘Precocious Puberty: Growth and genetics’ (2005: 56), argue that ‘detection of these genes will provide a tool for identification of children at risk of developing CPP [central precocious puberty], ⁹ enabling early intervention with the aim of preventing its distressing outcomes.’ All of this work leaves aside the question of the psychological effects produced by the interventions themselves, figuring medical treatment as preventing distress, in contrast to the traumatic experience of precocious puberty. Whilst this is an empirical question yet to be explored (that is, the question of whether taking hormonal medication to prevent the onset of puberty creates psychological problems for children or simply alleviates them), there is evidence in related fields that such intervention may be harmful in both the short and long term. A recent study of adult women who were treated as children with hormones to prevent unusual (‘pathological’) height has shown that many of these women experienced this medication intervention (and its attendant examinations) as traumatic, and indeed, were more likely as adults to suffer from depression and anxiety (Rayner, 2009). The literature on surgical and hormonal interventions in babies and young children diagnosed with forms of intersexuality also shows that such interventions are rarely experienced as neutral, and can be viewed by children and by adults looking back to their childhoods as traumatic and psychologically destabilizing (Creighton and Minto, 2001). ¹⁰ The dream of early intervention, then, may ‘make sense’ biologically, but needs to be thought through in terms of its psychological consequences.

As argued above, childhood obesity is conceptualized as a problematic state rather than a process, and increasingly, as a state that is highly resistant to change. In this case, the potential of targeted intervention subsequent to genetic testing is more open-ended: children who are already overweight could be tested and targeted for clinical interventions, and those who are deemed genetically ‘at risk’ could become the focus of preventative strategies. As Perusse and Bouchard argue, ‘definition of these interaction effects for phenotypes related to obesity is therefore important because it will eventually allow the identification of individuals at risk of obesity, the development of complications associated with obesity, and the identification of those likely to be resistant to dietary interventions and hence requiring, perhaps, more drastic or better-adjusted dietary prescription’ (2000: 1285S). At the level of the individual, the psychosocial effects of this surveillance and intervention are not addressed within these recommendations. This is in spite of mounting evidence that schemes such as the routine weighing and measuring of schoolchildren (Colls, 2007; Evans 2007), and healthy eating / weight management educational programmes (Evans et al., 2003; Evans et al., 2004) disturb children's (and particularly girls') self-esteem and confidence in relation to their bodies and food, as well as producing physiological disruptions as children enter into the cycles of weight loss and regain that characterize the histories of many who struggle to maintain a ‘healthy weight’. It is also difficult to distinguish these risk-based interventions from those already in place for weight loss; indeed, one of the research team working on FTO commented in the Guardian, ‘When you have this risk factor or not, if you are overweight, you should eat less, exercise more’ (Randerson, 2007). However, this ‘rational prescription’ (Ogilvie and Hamlet, 2005: 1545) ignores the fact that over 80 per cent of all weight loss interventions are unsuccessful in the long term (Mann, Tomiyama et al., 2007). The recommendation to eat less and exercise more presumes the efficacy of this as a sustainable weight loss / management strategy, even in the face of overwhelming evidence that this is not the case. Indeed, radical dietician Lucy Aphramor (2005) argues that this is an unethical recommendation for doctors to make, since no reliable means currently exists to enable people to achieve it, exposing them to the risk of further failure and blame for their pathologized condition.

The persistence of normative elements: moral concerns around getting bigger

Discussions about the differing roles of genes and environments in childhood obesity and precocious puberty, at least currently, have little effect on the treatments offered to children. We are tempted to ask, with Keller (2008: 117), ‘why, given all the conceptual difficulties in attempting to sort nature from nurture, do we persist?’ Why do scientists expend so much effort attempting to make distinctions between genes and environments and quantifying their respective contributions to the production of embodied differences?

Keller's response in ‘Nature and the Natural’ (2008) is instructive. After describing the ways in which contemporary technoscientific discourses commonly take it ‘for granted that the organism can be regarded as a sum of genetic and environmental factors, with interactive effects constituting a persistent demarcation’ (Keller, 2008: 122), she suggests that the ubiquitous discursive aligning of ‘genetic’ with ‘natural’ (and therefore both ‘usual’ and ‘legitimate’) means that genetic investigation has a role in every attempt to produce knowledge about human (and other living entities') behaviour. Genes, it is assumed, can explain both the normal and the pathological: even whilst, as we have argued in our two case studies above, ‘environments’ are thought to matter (somehow), it is only insofar as they ‘interact’ with genes that they are considered powerful. Similarly, it is the presumption of the power of genetics to explain both normality and pathology that renders the possibility of successful intervention at the genetic level, to use Speakman's term, the ‘ultimate “holy grail” solution’ (2004: 2102S). It is this paradox – that genes will explain everything but yet are only ever part of the picture – Keller argues, that makes genetic science so powerful:

Thinking of our two case studies in these terms and within the current climate of intense concerns about obesity epidemics and the decrease in pubertal age across the western world, we suggest that the unremitting focus on genes is indeed part of an expansionist claim to know and potentially control unruly bodies. Claims to know obese and unnecessarily pubescent bodies are powerful in this climate because these bodies are matters of serious moral concern in medical, scientific and public discourses. These concerns take different, but related forms in our two case studies, with moral failure seen as causing and sustaining obesity, whilst being a likely consequence of precocious puberty. As Throsby (2009) argues elsewhere, public discourses on obesity constantly produce fat adult bodies as a threat to society: a moral failing of the will to health that is part of today's citizenship duties. Obese, or potentially obese children's bodies, are consequently harbingers of doom and sites of even more intense moral approbation: of course the child cannot be held responsible for his or her failure to maintain a ‘normal’ weight, but those in charge of his or her eating certainly can. Despite claims about genes, then, the parents of obese children are figured as morally reprehensible failures: unable to assimilate widely-available information about nutrition, unable to control their children's intake of food, or unable to understand the long-term consequences for the child and society more broadly of being overweight. The moral panic (Cohen, 2002) around obesity, particularly as it relates to children, has thrown up a predictable cast of contemporary folk devils – perhaps most famously in the United Kingdom, the ‘Burger Mum’, Julie Critchlow, who attempted to undermine celebrity chef Jamie Oliver's attempts to reform the school dinners being served at her children's school by passing burgers and chips through the fence and was subsequently labelled by Oliver himself a ‘big old scrubber’ (Martin, 2008). However, the singling out of individuals or groups (working mothers, McDonalds etc) is underpinned by a much more established and pernicious moral discourse around fatness – around bodies which are deemed to be out of control, disregarding of their own health, lacking in personal hygiene, lazy, unintelligent etc. (Cooper, 1998; Wann, 1998; Gard and Wright, 2005; Murray, 2005). These presumed characteristics of fatness are readily reproduced by children, who learn from an early age the unacceptability of the fat body, and by implication, the need to exercise discipline over it in order to exert change (Latner and Stunkard, 2003). In this light, the very presumption of the changeability of the environment/lifestyle (as opposed to the intractability of genes) reproduces these pejorative discourses of blame.

Significantly, several of the papers reviewed in preparation for this chapter address directly the issue of the stigmatization of obesity, arguing that genetic knowledge absolves individuals of blame for their condition. Friedman, for example, concludes that ‘We can only hope that advances in our understanding of the causes of this condition will lead to changes in the perception of what it means to be obese in a world of harsh judgments and facile conclusions that are not supported by a growing set of scientific facts. The stigma of obesity should be discarded, enabling this disease to join with other conditions that required that we look beneath the surface’ (2004: 568). ¹¹ But however good the intentions of these exhortations to abandon the stigmatization of obesity, these claims maintain the status of obesity as a problematic condition about which something can, and imperatively, should be done. It may not be an individual's fault that they are obese (because of their genes), but the obligation to act remains, even in the absence of a safe and effective intervention. This renders the failure to act morally problematic, regardless of genes or genetic predispositions. Furthermore, issues of body size, and the pressure to act in relation to those issues, are always profoundly gendered, with women confined by much narrower parameters for what constitutes acceptable body size and subject to far greater surveillance in relation to their weight than men; indeed, they already make up the vast majority of consumers of every weight loss intervention (Cooper, 1998; Stinson, 2001; Wann, 1998).

In precocious puberty, moral concerns focus on the presumed sequelae of early development: precocious sexual behaviour and higher risk of substance abuse. In the clinical and popular literature (in newspaper reports, for example), discussions of early puberty are often framed by worries about (female) sexual precocity and its potential consequences. A recent article in the Guardian's ‘Family’ weekend supplement, for example, summarized research that reportedly found that ‘entering puberty young (before 11) correlates with a host of problems, from teenage pregnancy to depression. Only 2 per cent of those who do so go on to enter higher education, regardless of their parents’ IQ and educational level’ (James, 2009: 2). Another headline makes the point more dramatically: ‘Girls With Early Puberty, Older Boyfriends At Greater Risk For Drugs, Sex, Alcohol’ (Anon, 2007), with the article arguing that girls who mature early are more likely (than boys who mature early) to have older romantic and sexual partners and to take illicit drugs. The risk posed to girls of entering puberty very young is, many claim, that they will be taken for sexual subjects before their time, and, even worse, may consider themselves sexual subjects and behave ‘inappropriately’ for their age: ‘Early maturing girls are more likely to be unprepared emotionally and cognitively for the changes of puberty itself and the social pressures that they may experience from boys and adults who perceive them to be more mature than their age and experience’ (Graber et al., 1999: 110, emphasis added).

Although in both the clinical and popular literatures, the dangers of such precocity tend to be assumed rather than spelled out or evidenced (see for example, Moshe and Lazar, 2005: 59; Posner, 2006: 319), these concerns are sometimes linked directly to recommendations for treatment. In an article on the use of drugs affecting the secretion of a particular hormone in the brain to treat precocious puberty, for example, Mul and Hughes (2008: 5) describe the problems caused by precocious puberty constituting indications for such treatment. After describing height ‘impairment’ (children who go through puberty early tend to end up shorter than their peers), and before advocating the use of hormones to ‘halt menses for hygienic reasons in some girls with severe cerebral palsy or developmental retardation’ (such girls tend to go through puberty early), they mention the ‘psychosocial outcome[s]’ associated with precocious puberty:

This statement is left hanging, and the reader can only assume that these ‘psychosocial outcomes’ of puberty are, like ‘hygienic problems’ and ‘height impairment’, considered suitable indicators for treatment.

Although their paper tends to indicate that the use of hormones to treat precocious puberty might prevent early sexual activities, Mul and Hughes (2008: 6) ultimately acknowledge that ‘embarking on treatment for the individual child with CPP [central precocious puberty] is not underpinned by any comprehensive evaluation of the long-term outcome data at a population level.’ Despite this lack of evidence, the emphasis on the potential of such treatments to produce psychosocial outcomes are highlighted in their conclusion. Such effects are seen to be ‘equally important’ to the physical effects pertaining to height and menstruation:

Psychologist Rachel Posner (2006) makes a similar claim about research into early puberty, arguing that it is imbued with assumptions associated with ‘societal anxiety about adolescent female sexuality [which] has inhibited young women's ability to healthfully and positively consolidate their sexuality into their identities’ (Posner, 2006: 316). Indeed, she asserts in her conclusion that ‘underlying the agenda of most of the research into the antecedents and consequences of early puberty has been a drive to control and limit adolescent girls’ sexuality. This line of research,’ she continues, ‘has aimed to determine how sexual behaviour among adolescents can be prevented’ (Posner, 2006: 320). Concerns about girls entering puberty early also focus on cultural judgements about the aesthetic qualities of bodies. Posner writes that,

There are obvious links here to concerns about obese girls, for whom failing to meet contemporary aesthetic standards is also seen as having serious psychological consequences. Indeed, Posner suggests that it is difficult to separate the effects of these conditions on girls' self esteem, as they are so often experienced together: ‘Further research is needed,’ she writes, ‘to disentangle the psychological effects of obesity from the effects of early puberty. That is, are girls concerned about their figures because they are early maturers or because they are overweight?’ (Posner, 2006: 319). ¹ 2

In both our case studies, gender is at stake in the moral framings of the conditions and their presumed consequences. Moral concerns focus on girls in particular, and, involve gendered assumptions about aesthetic qualities of bodies (thin is better), sexuality (girls should not express a sexual interest too early) and responsibility. In relation to the latter, in puberty discourses, if young girls excite inappropriate sexual interest, they – rather than those who display such interest – must be changed. Obese girls, similarly, must learn a gender-appropriate concern about their eating and exercise habits and learn to take responsibility for their size. In every case, we want to suggest that there could be alternative framings of the ‘problems’ caused by the conditions under discussion: framings that as we explain below, focus not on the individual, pathologized body, but on the social world constituting such pathologies

Conclusion: From interaction to relation?

In a careful analysis of the scientific literature on the genetics of ‘race’, Anne Fausto-Sterling (2004) argues that the contemporary focus on genetics as a means to explain the stratified epidemiology of particular health problems (heart disease, high blood pressure and diabetes, for example), assumes that diminishing such health inequalities cannot be done via other routes. What is papered over in the attempt to discover genetic bases for the racialized distribution of these conditions, she contends, is that basic health measures would have far greater impact on reducing morbidity and mortality (Fausto-Sterling, 2004: 24). Thus, ‘it is not what the new biology of race will produce,’ that constitutes the key risk of this field, ‘so much as it is that the new biology of race diverts our attention from solving problems using solutions we already have at hand’ (Fausto-Sterling, 2004: 30).

For Fausto-Sterling, this argument is linked to her proposal, made in a later paper on bone health and sex/gender, to retheorize the relationship between ‘genes’ and ‘environments’, understanding the actions of genes within complex systems. Rather than as causes, she suggests, ‘genes are best understood as mediators suspended in a network of signals (including their own) that induce them to synthesise new molecules’ (Fausto-Sterling, 2005: 1507). Genes, in this view, ‘do not build organisms from the bottom up; rather, their activities are sandwiched somewhere in the middle of chains and networks of events that integrate organisms with their environment’ (Fausto-Sterling, 2004: 26). In the case of racialized differences in health, she argues that ‘it is not that different biological processes underlie disease formation in different races, but that different life experience activates physiological processes common to all, but less provoked in some’ (Fausto-Sterling, 2004: 26). Our bodies, in other words, ‘naturally’ produce responses to our environments (Fausto-Sterling, 2004: 31): living as a black person in a racist US culture, for example, leads to increased blood pressure for many individuals.

In contrast to older scientific notions of homeostasis (where bodies are understood to be constantly trying to achieve a ‘normal’ state), this model, developed by physiologists in the late 1980s, argues that bodies are allostatic, functioning as a kind of ‘“smart” thermostat’ that regularly changes activity to meet anticipated demands (Fausto-Sterling, 2004: 27). This view of the body takes into account the social experiences of organisms (including humans) in understanding the development of particular disease conditions. In relation to high blood pressure, for example,

For us, what is compelling about this proposal is the challenge it poses to more prevalent models that attempt to separate and quantify the different contributions of ‘genes’ and ‘environments’, and, in particular, what this challenge means for health-related interventions. In the case of hypertension, Fausto-Sterling argues that the allostasis model emphasizes what she calls ‘higher-level’ interventions, adjusting entire systems of physiology and attending to the emotions and unmet needs that produce the negative physiological results of ‘hyper-vigilence’ around blood pressure (individuals' brains, she argues, can become extremely sensitive to changes in blood-related systems). The overall aim of interventions, she argues, should be to ‘reset the body's response systems, allowing function in a range less likely to produce excess morbidity and mortality’ (Fausto-Sterling, 2004: 29). In this case, citing neurophysiologist Sterling (2004), Fausto-Sterling approvingly lists ‘lifestyle interventions’ (dietary changes, weight loss, exercise and reduced alcohol consumption) as potentially leading to such systemic change, in part (in contrast to drug therapies) through acknowledging need and ‘enlarging positive social interactions and revivifying the sense of connectedness’ (Fausto-Sterling, 2004: 29–30).

While we are encouraged by her willingness to look to different kinds of interventions from the conventional drug therapies, we are less convinced that such interventions will have much effect in our examples. Indeed, part of what we have been critiquing here, particularly in the case of childhood obesity, is the fact that even in the wake of immense interest in genes and failure to demonstrate that ‘lifestyle’ interventions actually work in the long term, recommendations to ‘eat less and exercise more’ still retain most-favoured status in terms of intervention. Indeed, we have argued that such emphasis can have precisely the opposite effect to ‘enlarging positive social interactions and revivifying the sense of connectedness’: they can reinforce and increase experiences of discrimination and alienation. Although the case of early puberty (in which drug treatments are the normal recommendation) might hold closer parallels to hypertension (hence mirroring Fausto-Sterling's recommendation to move away from drugs to more ‘lifestyle’ oriented interventions), we would likewise suggest that what might be most important for children diagnosed with this condition is to simply avoid conceptualizing their puberty as a disaster.

To make effective interventions into the so-called epidemic of childhood obesity and the purportedly population-threatening trends towards early puberty, we need to think seriously about issues pertaining to economics, gender, class and ethnicity at both macro and embodied levels. Taking an allostatic model seriously means thinking about bodies in relation and consistently refusing a separation between genes and environments (even the separation assumed in most interactionist models). It also means challenging cultural norms around the moral freight and personal responsibility associated with particular bodily conditions. As our two case studies show, investigations into the genetic causes of conditions or attempts to delineate gene/environment interactions in producing them tend neither to reduce either the moral freight of conditions nor to shift already-existing attributions of responsibility. Parents of children diagnosed with obesity are still more likely to be held responsible for this than those whose children are early developers. Thinking of ourselves as ‘always 100 percent nature and 100 percent nurture’, as Fausto-Sterling suggests, means that moral questions cannot be separated off from physical ones, or left to be answered later. Cultural norms and practices associated with gender, class, ethnicity and other forms of body-based discrimination become sites of intervention that may well have lasting impacts at both individual and population levels.