Luck and the responsibilities to protect one’s epigenome

ABSTRACT

This paper builds upon the debate on ‘moral luck’ – i.e. the import that factors beyond one's control have on the cogency of normative claims such as responsibilities – to criticise claims towards backward- and forward-looking responsibilities to protect one's epigenome for the sake of personal or future generations' health. Luck, I argue, is part and parcel with the actions required to protect our epigenomes, and points to the need of dismissing the ensuing individual responsibility claims. But what about the consensual alternative of appeals to collective responsibilities? If a consideration of luck reveals the vulnerabilities, circumstances and uncertainties that call into question individual responsibilities, the same kind of reasoning could apply analogously to collective agencies. Luck is no less of a challenge to our moral intuitions in the case of collective epigenetic responsibilities, and demands abandoning dominant atomistic framings parsing out individuals and collectives in the societal uptake of epigenetics.

KEYWORDS:

• Individual responsibility
• collective responsibility
• moral luck
• epigenetics
• epigenetic responsibilities

Introduction

Luck is said to play a vital role in scientific discoveries. Some compelling, influential and/or far-reaching findings can be attributed to luck, or serendipity – a mixture of scientific wisdom and a lucky course of events driving scientists towards the realisation of an epistemic innovation (Merton 2006; Yaqub 2018; Copeland 2019; Sand 2020). Little, however, has been written about the role of luck for the normative dimensions of scientific knowledge in contemporary societies. In this paper, I focus on the role of luck in public discourses and representations of epigenetic knowledge. In particular, I use the concept of luck to contribute to the debate around responsibilities to protect one’s epigenome for the sake of personal health, or the health of future generations (Hedlund 2012; Dupras and Ravitsky 2016; Chiapperino 2018; Meloni and Müller 2018). First, I show that bioethics and Science and Technology Studies (STS) scholarships have given extensive treatment to the limitations of claims towards individual responsibilities. Specifically, these critique have clustered around the limitations of two intertwining concepts of responsibility (Vincent 2011): one that is synonymous with accountability for damaging one’s epigenome (liability or backward-looking responsibility); and another one that highlights the prospective duty to protect it (remedial or forward-looking responsibility). This, I argue, resulted in a consensual emphasis put on collective epigenetic responsibilities as an ethically and politically sound uptake of epigenetic evidence. With regard to this debate, I show that a consideration of ‘moral luck’ (Williams 1982; Statman 1993) – understood as the import that factors beyond one’s control have on the justification and cogency of normative claims such as responsibilities – provides further ground to dismiss the attribution of individual epigenetic responsibilities. Yet, I argue, the role of luck for our morality also points to some fundamental flaws in the critiques moved to such individual responsibilities. If luck reveals the vulnerabilities, circumstances and uncertainties that call into question the cogency of responsibility claims for individual agencies, the same types of considerations apply analogously to collectives, which are already a far more contested source of responsibility (Lewis 1948; Feinberg 1968; French 1984; Arendt 1987; Smiley 2017). ¹ In my view, collective epigenetic responsibilities fail to be an obvious alternative normative construct to their individualistic counterparts. Following these considerations, I conclude that taking luck seriously in the case of (both backward- and forward-looking versions of) responsibilities for the stewardship of one’s epigenome demands a shift in the critics’ focus. One that does not consider responsibility as a ‘placeholder’ (Pettit 2007, 198), and turns away from considerations of desert, capacity, control and cognisance in the societal appraisal of epigenetics. Such an approach – capable of considering the complex circumstances, social processes, and transformative experiences characterising the collective and individual moral life of epigenetic knowledge – is still needed.

Epigenetic responsibilities: backward- and forward-looking

Epigenetic modifications – which include DNA methylation, histone modifications and the regulation of gene expression by RNAs – have been presented as potential bio-dosimeters of lifestyle and environmental exposures (Guerrero-Preston, Herbstman, and Goldman 2011). Global and gene-specific methylation patterns have been shown to account for the biological effects of different individual behaviours and exposures (Cooney 2007; Alegría-Torres, Baccarelli, and Bollati 2011), such as psychotropics consumption (Wanner, Colwell, and Faulk 2019; Schrott and Murphy 2020), smoking (Hillemacher et al. 2008), nutrition (Liu et al. 2003; Lillycrop and Burdge 2015) and physical activity (Pareja-Galeano, Sanchis-Gomar, and García-Giménez 2014; Grazioli et al. 2017). Epigenetic modifications are regarded as a footprint of health trajectories, since they constitute a chief mechanism through which lifestyles and environmental exposures affect gene expression, organismic development, late life diseases and potentially future generations’ health (Cavalli and Heard 2019). To this purpose, several epigenetic tests are currently being developed and sold directly to consumers (Dupras, Beauchamp, and Joly 2020). These tests promise to precisely follow the footprints of one’s health, by quantifying the biological traces of past behaviours or environmental exposures. For instance, tests based on quantitative DNA methylation analysis currently hold to detect smoking behaviours for several weeks and are recommended by producers for uses which include ‘establish[ing] the smoking status of those seeking life and health insurance’, or ‘prioritization for expensive medical procedures’ such as transplants (Behavioral Diagnostics 2020). If not for the purpose of establishing and measuring the effects of past imprudent behaviours, epigenetic tests are also presented as a way to provide consumers with an upshot of their biological health status. Often summarised as ‘a true measure of [customers’] biological age’, some epigenetic tests are sold under the assumption that knowledge of one’s predispositions to disease provides the relevant ‘insights’ to ‘turn back the [aging] clock’ or ‘make smarter lifestyle choices to avoid illness’ (Chronomics.Com n.d.). ²

These uses of epigenetic analyses illustrate a very predictable usage of epigenetics at the juncture with the long-standing debate on responsibilities for health (Buyx 2008; Feiring 2008; Albertsen 2015). Even before these tests were commercially available, critics warned us that epigenetics may become an enabling platform for normative options pivoting around desert, accountability and responsibility in healthcare (Hedlund 2012; Rothstein 2013; Loi, Del Savio, and Stupka 2013; Chiapperino and Testa 2016; Pentecost and Meloni 2018; Meloni and Müller 2018). Reflections on responsibility and epigenetics have been particularly fruitful in that this type of knowledge seems connected with some aspects of standard models ³ of responsibility in moral philosophy (Pettit 2007; Aristotle 2009, III 1–5; Talbert 2019) First, it is a way to causally connect a particular agent, or a specific set of actions (e.g. lifestyles, environmental exposures) to the circumstances for which responsibility is sought (e.g. one’s health condition). For what is widely regarded as a necessary condition for moral attributions of responsibility, epigenetic marks of past behaviours and exposures provide a crucial – although far from uncontroversial (see below) – factual resource to make causality claims. Second, epigenetic information fuels the intuition that the more we are knowledgeable about the consequences of our actions, the more we are in a position to see what is at stake, the more we are responsible for those same actions. Also known as the epistemic condition for responsibility (Pettit 2007; Aristotle 2009), this idea has been particularly cogent in the scholarly attention given to epigenetics (Hedlund 2012). Epigenetic knowledge may allegedly ‘make a change in degree’ (Hedlund 2012, 178) as to responsibilities to protect one’s health, since it supposedly provides the agent with the access to evidence making her able to estimate much better the relative value of her actions (e.g. lifestyles) on health.

Starting from these considerations, different scholars have focused on distinct responsibility claims relating to epigenetics. Some have looked at the possibility to measure past unhealthy behaviours through epigenetic marks and how this lends legitimacy to backward-looking responsibility claims. This notion of responsibility can be qualified as cognate to ideas of accountability and liability (Vincent 2011). Specifically, epigenetic tests – such as the one detecting past smoking behaviours mentioned above – are a concrete example of how epigenetics brings us beyond the epidemiological evidence of at-risk groups, and provides knowledge of the molecular traces of one’s risky lifestyles. What several scholars considered to be only a (problematic) possibility of epigenetics (Hedlund 2012; Chiapperino and Testa 2016; Dupras and Ravitsky 2016; Pentecost and Meloni 2018) is currently a commercially available means to provide factual support to the widely shared policy and moral emphasis on individual liability in healthcare. Furthermore, diagnostic applications of epigenetic knowledge also seem to instantiate another interpretation of epigenetic responsibilities for health; namely, the one compelling individuals ‘to do something’ for protecting one’s epigenome (Chadwick and O’Connor 2013, 464). This type of remedial, prospective or forward-looking responsibility (Vincent 2011) rests upon the possibilities offered by epigenetics to guide future actions (e.g. lifestyle changes) directed at health improvement, and has attracted no less critical scrutiny (Dupras and Ravitsky 2016). Like in the abovementioned case of the epigenetic test for biological aging, test providers advertise these tools as an empowering technology providing citizens with actionable knowledge to improve their own health (cf. Chiapperino and Testa 2016).

This body of scholarship provides extensive ground to dismiss claims towards both backward- and forward-looking individual responsibilities for protecting one’s epigenome (reviewed in Chiapperino 2018; Dupras, Saulnier, and Joly 2019). Primarily, the reasons provided have to do with the nature of the epigenome and some epigenetic modifications. As I argued elsewhere (2018), the debate internal to epigenetic epidemiology (Heijmans and Mill 2012; Mill and Heijmans 2013; Mitchell 2018) provides epistemic ground to criticise claims in favour of backward-looking responsibilities. A closer look at the informational and actionable value of epigenetic marks shows in fact that the characterisation of epigenetic effects of past individual habits on health – but also exposures, life conditions, psychosocial factors – is limited in multiple ways. First, little proof exists today as to the causal role of epigenetic modifications in pathogenesis. Second, no possibility exists to adjudicate whether an epigenetic modification relating to a health condition is due solely to lifestyles, or environmental stimuli, or genetic differences. As several epigeneticists point out, health trajectories are likely the result of a combination of the three (Cavalli and Heard 2019): it may thus be a hopeless task to try to determine the contributory effect of lifestyles vis à vis genetic predispositions and/or environmental factors in pathogenesis. Dupras and Ravitsky (2016) took instead issue with claims of forward-looking responsibilities, and argued that it would be very tricky to enact such responsibilities since no definition of ‘epigenetic normality’ (536) exists. The epigenome is highly contextual and its effects on organism functioning and pathogenesis very much situated in time and space. If, they argue, our epigenome is highly sensitive to temporalities of programming, individual genetic variation, and stochasticity (cf. Feinberg 2014), what healthy behaviours should individuals strive for, in order to act as responsible stewards of their epigenome? If we follow Hedlund (2012), instead, countering claims towards individual responsibility to protect one’s epigenome does not require assessing the actionability of epigenetic knowledge (or lack thereof). Echoing long-standing arguments against individual responsibility for health (Minkler 1999; Resnik 2007), Hedlund points to the ‘circumstances that to varying extent constrain individual choice’ (179). Intentionality, voluntariness and capacity to act – all generally thought to be necessary conditions for responsibility attributions – rarely apply anyway to actors constrained by unequal social and economic structures. Lifestyle behaviours are seldom the result of individual deliberate choices, but rather stem from an intricate web of socio-political factors that ‘strike unevenly’ (179) in our societies, and thus unevenly hamper individual capacities to take up full responsibility. This is what makes Hedlund suggest that the normative burden of backward- and forward-looking responsibilities should largely fall onto collective agencies (e.g. the state, corporations, public health agencies):

Epigenetic research advancements pointing at the role of individual life style for the development of disease might contribute to an emphasis on individuals’ responsibility for their own health and the health of their future descendants, potentially leading to expectations that individuals who do not avoid injurious behaviour should pay for health care, and the risk of stigmatisation of those who do not take appropriate responsibility for their own health. Such a development is, however, not self-evident. The emerging knowledge about epigenetic mechanisms also calls attention to the role of structural conditions, which as well could give rise to a focus on the role of society and the state to protect and care for health and wellbeing of individuals, present and in the future. (Hedlund 2012, 181; emphasis added).

Hedlund’s proposal has had a large influence on bioethical and social sciences scrutiny of epigenetics. Needless to say, some critical voices exist in this debate which do not share the idea of epigenetic responsibilities for health being largely a collective matter. Huang and King (2018, 73) point to the fallacy of discounting the role of ‘individual agency in mitigating environmental exposures’ and claim that ‘it does not necessary follow’ (original emphasis) that identifying larger structural configurations producing epigenetic effects on individual health requires collective action rather than individual ones. Similarly, Dupras and Ravitsky (2016) nuance their dismissal of individual responsibility claims by considering the attribution of responsibility only to states and collective agencies to be too simplistic. Provided that strict conditions of capacity and timeliness to act are met, they argue, it is conceivable that several consequences of aberrant epigenetic effects may legitimately fall under the scope of individual agency. Otherwise stated, the question of ‘balance of individual and collective responsibility is as of yet unsolved’ (Hens 2017b). Amidst this diversity of positions, however, several scholars have supported Hedlund’s call for a normative translation of epigenetics promoting ‘a forward-looking approach that calls for collective responsibility’ (Pentecost and Meloni 2018, 62). They have done so by raising the flag against the metaphors (Stelmach and Nerlich 2015), the deterministic (Waggoner and Uller 2015), reductionist (Müller et al. 2017; Meloni and Müller 2018) and gendered translations of epigenetic knowledge (Warin et al. 2011; Hanson and Müller 2017; Kenney and Müller 2017; Chiapperino and Panese 2018; Sharp, Lawlor, and Richardson 2018), which lend legitimacy to individualising discourses of public health policy. These authors have criticised the multiple ways knowledge-production in epigenetics fuels the claim that we are morally accountable and responsible for epigenetic predispositions since these depend allegedly on factors under our control. Pointing to ‘the social and political conditions in which we live – not just individual decisions – as critical factors in shaping health outcomes’ (Lappé 2016, 94), many contributors to this literature suggest shifting the emphasis away from individuals and towards the importance of addressing epigenetic responsibilities as a collective matter. Let us turn now to a consideration of how moral luck contributes to the debate on backward- and forward-looking responsibilities to protect one’s epigenome.

Constitutive, circumstantial and resultant luck: further restricting individual responsibilities

In the late 1970s luck surged as the subject of an intricate philosophical debate, pivoting around its relevance in the area of morality. The debate took off from the recognition of a prima facie irrelevance of luck to the distribution and exercise of moral attributions. One’s moral worth being subject to chance and factors beyond one’s control is in fact a rather oxymoronic statement for most moral theories. Freedom and rationality are often the highlighted sources of agent’s dispositions, decisions and actions (Williams 1982). Yet, a closer scrutiny of these dispositions, decisions and actions reveals that, though agent-directed, they are substantively open to luck. This recognition has led several scholars to highlight a paradox raised by so-called moral luck. Either we hold the view that we cannot be held morally responsible for things we do not control (or we could not anticipate as a result of our actions), and hence we admit that the pervasiveness of luck in our lives dramatically prevents our judgements of responsibility. However, this seems counterintuitive in light of our ordinary moral intuitions and practices in which we do hold other people morally responsible. Or, we try to shield our judgements of someone’s moral worth/responsibility from luck, but then we end up dramatically restricting the ground for attributability and ownership of actions. Luck is in fact ubiquitous in our lives, and very little would be left for responsibility to apply.

On this matter, the seminal contributions of philosophers Bernard Williams and Thomas Nagel – among others ⁴ – may help us better frame the issue, and illustrate how taking seriously the role of luck questions also those moral intuitions about responsibility populating the societal uptake of epigenetics. Nagel’s contribution to this debate consists of a very useful taxonomy of the different ways in which luck affects some central tenets of modern conceptions of responsibility. For present purposes, I will focus on how Nagel explicitly ties the idea of luck to the intuition that we can be held responsible for our actions (and their outcomes) because we cause them and we perform them voluntarily. Luck undermines these moral assumptions in three different ways. ⁵

First, moral thinking rests upon the systematic attempt to obliterate what Nagel calls constitutive luck. The qualities of one’s character, one’s attitudes, one’s capacity to make (good or bad) decisions, one’s inclinations, one’s instincts cannot be described merely as deliberately chosen properties of agents and their characters. These rather depend very much on factors beyond individual control, they are induced from external factors, which may be arguably qualified as luck. So does, for instance, the capacity to consider health and wellbeing as valuable objectives to pursue in one’s life, which is largely the product of social (e.g. upbringing), cultural (e.g. the value of health in historically and geographically distant social groups), or even biological (e.g. impulsivity) factors (cf. van Hooft 1997; Thomas, Fine, and Ibrahim 2004; Crawford 2006; Broom et al. 2019). Similarly, the specific value assigned to prevention of epigenetically-mediated health risks both for one’s own sake, and for the sake of future generations is the result of local configurations of health promotion, global discourses of policy-making, as well as historical circumstances of health inequalities and affordances of healthcare systems (Pentecost and Cousins 2017; Pentecost and Meloni 2018). Recognising how luck intervenes into the constitution of individual conceptions of a good life wears out the element of voluntariness in attributions of responsibility. The states of character and will that influence preferences, values and choices – including, as prime example, those related to prevent aberrant epigenetic effects of one’s behaviour and lifestyles (cf. Dupras and Ravitsky 2016) – are aspects of the individual moral character that simply are not just voluntary deliberations.

Second, luck undermines some commonly held intuitions about responsibility for what Nagel, following Williams (1982), calls circumstantial luck. Here, the bearing of luck is on the difficult choices, obligations, tests, and exercises of responsibility we face during our lifetime. The way agents respond to these challenging situations is a fundamental part of the worth assigned to characters, identities and biographies through both self- and other-regarding judgements. Some people may be confronted to moral quandaries of the highest complexity, and certainly several of them may fail to meet these challenges in a praiseworthy manner. These situations, Nagel argues, exemplify how substantive matters that partake to the definition of one’s character, or others’ judgement of one’s action are determined by factors falling beyond individual control. Whether agents fail or succeed in acting responsibly, for instance, with regard to the correction of the aberrant epigenetic effects of major traumatic events (Yehuda and Bierer 2009; Lehrner and Yehuda 2018), humanitarian crises (Heijmans et al. 2008; Roseboom et al. 2011), or the omnipresence of endocrine disruptors in the material environments of industrial societies (Snyder et al. 2003; Skinner 2015), cannot be simply taken as a matter of choice and responsibility of these agents vis à vis the available knowledge of a preventable health risk. These situations are a substantial aspect of the moral lives of these individuals that, incidentally, also largely escapes their own control. Starting from a consideration of what Williams, Nagel and others (Statman 1993) called circumstantial luck, it is therefore questionable whether one should be held accountable for failing to prevent the aberrant health effects of specific nutrients, toxic exposures, or stressful experiences that largely depend on situational factors beyond individual control. At the same time, similar considerations could be extended to claims towards remedial (forward-looking) epigenetic responsibility. How fair is it to hold someone responsible for failing to meet expectations of praiseworthiness when confronted with a complex state of affairs that is largely due to unlucky circumstances? Arguably, other people simply happened to avoid these fundamental and difficult tests in their (luckier) life (Hartman 2016) ⁶ : should this be relevant for an evaluation of people’s failures in these contexts? Circumstantial luck highlights how the result of what people do or fail to do appears, upon reflection, to be also the product of circumstances beyond one’s control. So, one might ask, how fair are the judgements of moral attitudes expressed in these circumstances, if we can arguably maintain that one is ‘morally at the mercy of fate’ when being confronted with them (Statman 1993, 65)?

Third, the last way in which luck wears out the moral concept of responsibility is with regard to the resulting effects of one’s action. This is the luck that intervenes, Nagel argues, in ‘the way things turn out’ (Statman 1993, 61), and thus relates to the existing hiatus between one’s decision, actions, behaviours and their consequences. The case of the alleged responsibilities to protect one’s epigenome is once again useful to illustrate how these moral claims rest upon the obliteration of (resulting) luck. Our epigenome is thoroughly characterised by environmental plasticity, individual variability and a general indeterminacy of change-effect mechanisms. Whether or not early-life epigenetic metabolic programming is a health risk depends, for instance, on the degree of mismatch between the resulting phenotype and the future environment (Godfrey et al. 2007). Similarly, stressful experiences may lead to psychopathology in ‘vulnerable’ individuals, as opposed to ‘resilient’ ones, on the basis of an interplay of genetic predispositions and epigenetically mediated socio-environmental effects (Champagne 2010; Dudley et al. 2011). Finally, actions undertaken to remedy to epigenetic predisposition may be effective only in specific time windows of development, while being ineffective at later stages of life (Zhang and Ho 2011; Wang et al. 2013). This means that the value of ‘acting responsibly’ to correct, or prevent aberrant epigenomic changes depends on several intertwining factors outside of individual control: for instance, the temporal windows of development in which these actions take place, ⁷ the genetic differences among individuals affecting malleability of specific epialleles, or, also, stochasticity (Feinberg 2014). In other words, the same undertakings (damaging or protecting one’s epigenome) may result in different (successful or not) outcomes. So, in what ways can we take such outcomes to be an essential justification for both backward- and forward-looking responsibility attributions about them? By putting resultant luck into the picture, we are left with a very different uptake of the moral cogency of claims towards (epigenetic) responsibility. First, the success – and, perhaps, also the praise or blame – attached to these exercises of responsibility seems to be the result of much more than behaviours, choices and actions of the concerned agents. In the illustrative case of choices directed at correcting or preventing aberrant epigenetic predispositions, the outcome of these actions is conditional to factors that are not affected by the actions themselves (e.g. temporalities, variability, stochasticity). Second, and relatedly, the fact that epigenomes change over temporalities of development, over individual genetic differences, and due to stochasticity, also seems to run counter a precise definition of the healthy behaviours individuals should strive for to protect their epigenomes. Whether an action is responsible or irresponsible, in a morally relevant sense, rather seems to be partly determined by the actual result of the actions (i.e. its actual effect on health), and not by a priori good or bad deeds.

Let us try to take stock of the argument here. Luck is not often part of the ways our moral intuitions about responsibilities are formalised, attributed or exercised. The concept of moral responsibility rather rests upon a series of assumptions about agent’s constitution, authorship of resulting consequences, causality, control and circumstances (Talbert 2019) for which luck should, at least in principle, be irrelevant (Zimmerman 1987). A closer scrutiny of the ways luck intervenes in the constitution, circumstances and resultant outcomes of the actions for which we are held responsible, shows instead that intuitions about responsibility are at odds with the fact that what we do, whether we succeed or fail, depends to a large extent from factors beyond our control. Taking seriously how luck intervenes in exercises of responsibility brings about a philosophical paradox that is deeply embedded in such concept:

A person can be morally responsible only for what he does; but what he does results from a great deal that he does not do; therefore he is not morally responsible for what he is and is not responsible for (Williams and Nagel 1976; reprinted in Statman 1993, 66).

These philosophical reflections on luck and responsibility have certainly a larger bearing on the moral concept of responsibility than the one relating to the normative claims coming out of epigenetics. The seminal contributions of Williams and Nagel have given rise to a heated debate that goes from questioning the very existence of anything such as moral luck (Rosebury 1995; Kneer and Machery 2019) to exploring and debunking its major implications for morality (Zimmerman 1987, 2002), or reconciling its import on morality with a coherent version of responsibility (Hartman 2016). Other, more moderate receptions have instead toned down these views, by pointing out that luck may not be as disruptive as Nagel and Williams conclude, since our conceptions of moral agency and responsibility can and do function also under the assumption that what we do is not completely under our control (Walker 1991). ⁸ Finally, addressing the problem of moral luck, some have argued, is far from being a disruption of our moral systems: it rather demands recognising that ‘one’s degree of responsibility is a function of various factors, including one’s exercises of will and perhaps one’s character’, and that ‘[t]hese factors can be affected by luck’ (Hanna 2014, 693).

Nonetheless, an analysis of luck for the normative scrutiny of epigenetics provides further reasons, or specification to the available repertoire of arguments against claims towards individual responsibilities stemming from this knowledge. Starting from a consideration of how luck affects the circumstances (circumstantial luck), the conditions (constitutive luck), the processes and the results (resultant luck) of one’s actions, the validity of arguments in favour of individual responsibilities to protect one’s epigenome gets restricted on further grounds. A consideration of moral luck provides an alternative language to the moral emphasis scholars have put on ‘the role of structural conditions’ (Hedlund 2012, 181) that limit, if not undermine, individual responsibilities to protect one’s epigenome. This individualised reasoning in the societal uptake of epigenetics is fundamentally defective to capture the indeterminacy and constrained nature of choices and actions relating to the protection of our epigenomes, as well as the contingency of these actions’ results and outcomes.

Taking full stock of ‘luck’: contra collective responsibilities

But what is the import of these considerations on the shared tendency to ascribe epigenetic responsibilities to collective agencies? What can a consideration of luck teach us on the consensual calls for collective epigenetic responsibilities? Are these responsibilities an appropriate alternative to assess this issue? Those who champion collective responsibilities may welcome the critique stemming from a consideration of luck, which points to the vagaries of individual agency and the limitations these put on attributions of responsibility. Concerning this point, however, I wish to underline how, in these critiques, we seem to be dealing with intuitions about the coherence of collective agencies that are too generous compared to those granted to individual agencies. From environmental factors producing epigenetic harms through exposures, to the structural differences in society that produce health-threatening lifestyles with putative epigenetic effects on the worse-off, analysts of epigenetics seem to consider collective (e.g. the state, corporations, or public health agencies) more than individual agencies as morally assessable for these situations. Presumably, this is because what is assessed can be attributed to such collectives and strictly depends on factors under their own control. ⁹ However, I shall argue, the pertinence of these attributions to collectives is a questionable matter as much as it is for individual agents.

The existence of collective responsibilities – as a judgement about collectives’ capacities to qualify as agents and being morally charged with responding for/bringing about a state of affairs – is far from self-evident. Certainly, there are a number of ways in which we think of collectives as holders of responsibility. We legally set collectives with goals, we delegate power to institutions, corporations or organisations in ways that are delineated in policies and transformed into actions through the activities involving them. Our ordinary language confers, in other words, an important social and pragmatic function to collectives, which is often expressed in the language of responsibility (Werhane and Freeman 2009). Yet, as shown above, being an appropriate addressee of moral responsibility involves the attribution of certain capacities and powers – like intentionality, knowledgeability and causality over the course of events for which we are judged. Whether a collective meets these conditions is an important matter for those arguing, for instance, that epigenetics ‘requires greater attention to collective responsibilities’ (Guthman and Mansfield 2013, 500). Specifically, it is a morally salient matter in the sense that it could illuminate on what grounds collective agencies should be blamed for failing to meet these expectations.

Collective responsibilities get precisely questioned on the grounds of common conditions for moral responsibility. For instance, the idea that there exist distinct collective intentions and actions as opposed to the contributory intentions and actions of individuals members of a group is a long-standing source of philosophical scepticism (Lewis 1948). Talks about collective responsibility can be coherent, according to some, only because they are often of a metaphorical form: we do not really consider collectives as bearers of agency and capacity for blameworthiness, but rather we direct such responsibility claims to the individual(s) delegated to make decisions on behalf of the collective (Narveson 2002). In other words, we do employ the language of collective responsibilities simply as a ‘pragmatic foreshortening’ for people who hold positions of power that could remedy to a certain situation, or for the lack of knowledge with regard to actual decision-making processes and the particular individual responsibilities behind a course of action (Sand 2018, 234). Such attributions of collective responsibilities raise the problem of determining the contributory liability (backward-looking responsibility) of individual members – as well as a distinction between the liabilities of past and present members (Feinberg 1968; Arendt 1987). What are, for instance, the causal connections that tether past wrongdoings leading to the epigenetic effects of trauma (Yehuda and Bierer 2009; Lehrner and Yehuda 2018), humanitarian crises (Heijmans et al. 2008; Roseboom et al. 2011), or pollution (Snyder et al. 2003; Skinner 2015) with the political responsibilities of present public health agencies (cf. Darby and Branscombe 2014)? This suggests that collective responsibilities are as much open to constitutive luck as individual ones. The moral standing of the collective is not just determined by its deliberations and choices of conduct (if any). Rather, it is also open to the independent contributions of its (past and present) members, in ways that cast a doubt on the blameworthiness of the whole collective vis à vis a state of affairs. Does a ‘handful of dishonest’ (present/past) people suffice to jeopardise collective conduct ‘despite the best efforts of everyone else’ to act in ‘a morally laudatory manner’ (Whitman 2008, 38; see also Hartman 2019)?

Things are no less problematic when it comes to forward-looking collective responsibilities. First, scepticism about collectives meeting the conditions of agency necessary to attributions of responsibility applies no less to prospective, forward-looking reasoning. Here, instead of questioning whether collectives can be held accountable for a morally problematic state of affairs, the matter of dispute lies in the difficulty to coherently attribute agency, or to morally charge the collective with the responsibility to improve upon a certain situation (French and Wettstein 2014). One alternative to this strong version of forward-looking collective responsibility is to claim that collectives are the addressee of conventional or pragmatic attributions of responsible agency. But then, one might wonder, are these responsibilities sufficient grounds for any moral evaluation of collective action (French 1984)? Some authors equate in fact remedial collective responsibility to the capacity – obtaining through conventions that regulate the organisation of our societies – of taking informed action about a given situation (Pettit 2007; Hess 2014). For instance, public health agencies ought to do something about the epigenetic effects of structural configurations of society because they are ‘best placed to do [so]’ (Hedlund 2012, 171). Yet, it is unclear why this is enough of a basis for a moral evaluation of collectives. In fact, one could argue that this type of ‘role responsibility’ provides little information on a moral assessment of collective action (O’Connor and Sandis 2010, chap. 39). If collectives are really compelled to assume the role of correcting these hazards (meaning that there are no circumstances under which an alternative agent could take over this role), one could wonder to what extent this does not mitigate their blameworthiness for not fulfilling their role-responsibility. Much like in the case of individuals (Schwan n.d.), having one’s available options constrained, or being confronted to the circumstances of an unwanted moral challenge, may be regarded here as being responsibility-undermining. Or, one would need to incorporate an additional justification than mere ‘capacity for solving a problem’, in order to show how these considerations apply differently to collectives (see Smiley 2017, chap. 7 for an overview).

Second, and relatedly, appeals to forward-looking collective responsibilities do not automatically support the idea that action should tackle the structural configurations of society producing epigenetic hazards. As argued also by Huang and King (2018), epigenetics may only be a powerful case (among others) for the vulnerabilities to environmental exposures of particular individuals, but it does not necessarily ‘serve as sui generis evidence for moral or ethical obligations’ to correct them (74). Understood as role responsibility, this remedial attribution is no more than an agent’s mandate, or legal/institutional oversight of a state of affairs, and its exercise depends on considerations that are external to the mere capacity and sovereignty over change (i.e. social policy). For example, remedial collective responsibilities may still weigh whether action on the arrangements producing vulnerabilities is more effective than action on more proximate causes through medication (Szyf 2009). The role of epigenetic signatures in pathogenesis may be different from individual to individual – due to a unique combination of genetic predispositions and environmental exposures (Cavalli and Heard 2019) – thus requiring personalised rather than one-size-fits-all, or structural interventions. In sum, the capacity of collective agencies to remove the epigenetic signatures of social injustices and environmental exposures is, again, not a direct justification for distribution of moral work.

Third, even if we hold a coherent view of collective agencies as the bearers of remedial responsibility, these are no less exposed than individual ones to contingencies and circumstances of agency or the stochastic and highly contextual dependency of epigenetic predispositions to disease (Feinberg 2014; Cavalli and Heard 2019). Recall here the importance that ‘complex causality relations’ and ‘structural conditions constraining […] choice’ play for a critique of appeals to individual epigenetic responsibilities (Hedlund 2012, 211). One may suggest that collective epigenetic responsibilities may be deflected by the same complexities, luck and vagaries that intervene upon individual agencies. Collectives are no less confronted also to these intricate circumstances that result from historically distant actions, practices and schemas that sustain and reinforce aberrant developmental-epigenetic effects (Jablonka 2016). It may be a challenging moral test to break these negative loops just with social policy, and one that would (as shown above) presumably mitigate moral blameworthiness of collectives for not fulfilling this role. Furthermore, addressing persistent, socially induced deleterious epigenetic effects through collective action may also fail to prevent, neutralise or reverse those effects due to a series of intertwining factors outside of policy control. As shown above, the temporalities of intervention, the genetic differences affecting the malleability of specific epialleles, or stochasticity (Feinberg 2014), are all factors related to resultant luck that would temper attributions of responsibility also at the collective level.

Political-social action may be, in other words, far from being an effective or, at least, uncontroversial ‘responsibility placeholder’ (Pettit 2007, 198) to break away from the limitations of addressing aberrant cycles of social/environmental conditions and epigenetic modifications at the individual level. Rather, it may seem unwarranted, upon reflection, to exempt also collective agencies from a consideration of how intrinsic limitations and deficiencies, trying and unwanted circumstances, as well as imperfectly predictable results temper their blameworthiness for failing to act responsibly to protect our epigenomes and health.

Conclusions

Like many critics have pointed out (Hedlund 2012; Dupras and Ravitsky 2016; Chiapperino 2018; Meloni and Müller 2018), appeals to individual responsibilities to protect one’s epigenome overestimate individual capacities of bearing such backward- and forward-looking types of responsibilities. Here, I have provided another version of the same critique based on the influence of luck on individual agency. In brief, factors beyond one’s control intervene in the constitution, the circumstances and the results of actions affecting one’s epigenome, thus suggesting a further line of argument to dismiss claims towards individual epigenetic responsibilities. Yet, taking full stock of luck with regard to such responsibility claims suggests that appeals to collective epigenetic responsibilities may neither be a sound version of these claims. Several shortcomings of notions of individual epigenetic responsibilities seem to apply as well to the collective agents (e.g. the state, corporations or public health agencies) supposedly bearing a moral duty to protect our epigenomes. If, as seen above, collective responsibilities are intended as accountability (backward-looking responsibility) of collective agencies for a state of affairs, then it could be questioned whether these claims can actually build on a coherent view of collective intentions, voluntariness, or causality as conditions for their blameworthiness. If claims towards collective responsibilities are instead of the remedial kind (forward-looking responsibility), then it is far from clear whether and how these consensual appeals are in any way morally binding. The reasons for collective agencies to do something about the epigenetic effects of socio-environmental configurations go beyond their mere capacity to produce change through social policy. These considerations touch upon the web of circumstances and uncertainties entangling collective agencies, and mitigating their blameworthiness in the case of failure to exercise such responsibilities. Luck is, in a nutshell, no less of a challenge to our moral intuitions in the case of collective responsibilities.

Does this mean that taking into account luck leaves us without any notion of epigenetic responsibility altogether? A critique of responsibility grounded on the moral relevance of luck does not entail ruling out the existence of motives, will, decisions, and assessments of individual and collective moral lives. Luck does not exclude the possibility that favourable conditions of actionability and reversibility of a given epigenetic mark may compel individuals to take responsibility to protect their own health (cf. Dupras and Ravitsky 2016). Nor does it prevent us from delegating to collective agencies the reparation of far more complex ‘historical circumstances that have structured inequalities’ (Pentecost and Meloni 2018, 62) in socio-environmental exposures with aberrant epigenetic consequences. We/I have plenty of reasons, luck notwithstanding, to advocate for such collective actions. Epigenetic findings could be framed as a mode of attention towards the need of solidarity-based policies addressing the uneven effects of socio-environmental determinants of health (Hendrickx and Van Hoyweghen 2018). Collective agencies can act on the epigenetic effects of structural inequalities in ways that are uniquely capable of promoting a moral optimum of society by bringing it closer to a full appreciation of universal human rights and flourishing (Dupras, Joly, and Rial-Sebbag 2020). Regardless of the role of luck in potentially mitigating blameworthiness for failing to produce, or bring about a certain outcome, these lines of actions may be worth pursuing for the sake of, for instance, a collective’s cultivation of its virtues. If not on value-laden grounds, an argument for collective epigenetic responsibilities could be made on epistemic grounds; that is, based on the efficacy of collective action in removing the fundamental causes of several epigenetically-mediated risk factors at once thus affecting multiple disease outcomes (e.g. poverty leads to stress, but also unhealthy eating, thus mitigating poverty may prevent the health consequences of both stress and unhealthy eating; cf. Link and Phelan 1995). Such justifications of collective action would have the merit of shifting the debate around epigenetics, and more generally the social determinants of health (Schwan 2020), away from considerations of desert, capacity, control and cognisance that are at the core of the moral language of responsibility. If anything, the above uptake of luck may in fact raise substantive scepticism as to whether these considerations are the good criteria to distribute moral labour around knowledge of the mutual entanglements of our biological and social lives. A luck-informed view of agency and morality questions whether these a priori criteria for responsibility are needed for an assessment of blame/praise with regard to (existing and future) epigenetically-mediated effects of socio-environmental determinants of health. Rather, a luck-informed view points to the need of delineating pragmatic, conventional or role collective responsibilities, based on distributive theories of agency, on accessory justifications of autonomy, solidarity, vulnerability and human flourishing, or other norms of our moral and political life that could help ‘actors to collectively embrace their responsibility for future equitable conditions’ (Neuhäuser 2014, 233). In a nutshell, the critique from luck may perhaps boil down to the requirement of a conceptual sharpening of the directions in which epigenetics may steer our moral lives.

The kind of societies that are set to emerge from the use of epigenetic knowledge is still far from clear. Yet, our biological interconnectedness, social interdependence and biosocial entanglements have never been more evident thanks also to this knowledge. The moral exercises required by epigenetic knowledge may thus also need to be inextricably relational and complex, much like being capable to take into account how communal/collective and individual actions fall prey to uncertainties and luck. Dominant atomistic framings in which either (categories of) individuals or collectives are the isolated agents compelled to act responsibly with regard to this knowledge seem, in the face of these considerations, to be rather the reduction of thick moral experiences, complex circumstances, probability estimates and biological stochasticities to an abstract analytical modelling of ethical concerns. The problem luck raises for arguments in favour as well as critiques against individual moral responsibilities to protect one’s epigenome is, therefore, that the category of responsibility is ‘thought, as it were, to apply itself’ (Williams 1997, 97). The use of language and moral attitudes by critical scholars may rather be more accurate and plausible – if not valuable – insofar as it acknowledges the radical contingency of these situations, and champions their intricacies also for its own moral ambitions. An individual responsibility to correct aberrant epigenetic predispositions, I argued, is an effect of socio-environmental and collective vagaries wrongly considered. But the alternative appeal to collective responsibilities, I reckoned, may also be a fairly limited way to address the complexities of ‘biological and social forces, organism and environment, agent and milieu’ that make up our epigenomes (Meloni, Cromby, and Lloyd 2018, 5).

Acknowledgments

Support for this research came also from the facilities, infrastructures and resources made available to visiting fellows at the Brocher Foundation (brocher.ch). A substantial part of this paper was written during a month’s residency at the Brocher Foundation (brocher.ch). Thanks to Anyck, Marie, Philippe, Eric, Sabine, Roland (the foundations’ staffers) and to my fellow residents for their support throughout the residency. My gratitude for a thorough and constructive feedback on this manuscript goes to Kristien Hens, Emma Moormann, Jo Bervoets, Martin Sand and the two anonymous reviewers.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes on contributor

Luca Chiapperino is SNSF Ambizione Lecturer at the STSLab, Institute of Social Sciences, University of Lausanne. His research interests lie at the intersection of STS and Public Health Ethics, with a specific focus on the socio-political and policy dimensions of biosocial themes in post-genomic biosciences.

Additional information

Funding

This work is part of the Swiss National Science Foundation (SNSF – Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung) Ambizione project ‘Constructing the Biosocial: an engaged inquiry into epigenetics and post-genomic biosciences’ (PZ00P1_185822). Partial support

Notes

1 My focus on the role of luck for the foundations and value of individual and collective responsibilities sets aside an equally important and developed literature on distributive questions of responsibility in the societal appraisal of epigenetics. Several authors have employed luck egalitarian and Rawlsian perspectives to adjudicate this matter, with a distinct focus on how the social determination of epigenetic effects blurs the boundaries between natural and social health disadvantages, as well as factors beyond personal control (i.e. brute luck) and ‘socially generated’ endowments – all distinctions that matter to equality of opportunity and theories of justice in healthcare (Loi, Del Savio, and Stupka 2013; Stapleton, Schröder-Bäck, and Townend 2013; Kollar and Loi 2015). These contributions often build upon the assumption of a coherent attribution of responsibility to individual and/or collective agencies. The present paper matters to this debate only to the extent of problematising this assumption.

2 The validity and utility of epigenetic tests, especially those commercialised direct-to-consumers, is at the centre of a substantive methodological controversy that informs also their ethical scrutiny. Charles Dupras and his colleagues have provided some timely and compelling analyses of how the epistemic value of these tests (specifically, epigenetic age tests) informs our ethical reasoning on them (Dupras et al. 2019; Dupras, Beauchamp, and Joly 2020).

3 I am leaving aside here one dimension of moral conceptions of responsibility, which is implicit in most examinations of this matter around epigenetics. Namely, the idea that one is an apt candidate for praise or blame, for being responsible or irresponsible if and only if the action under scrutiny is voluntary; see Talbert (2019) for an introductory overview of different approaches to the voluntariness condition in competing conceptions of responsibility.

4 Daniel Statman (see Statman 1993) has published a seminal collection of the most relevant philosophical contributions to this debate, including Nagel’s and Williams’ articles that inaugurated it. These articles originally appeared in Williams and Nagel (1976). All citations in the present text refer to the revised versions of their contributions in Statman (1993).

5 Nagel’s taxonomy points also to the ways luck decisively affects the conditions of causality on which claims towards responsibility rest upon. This causal luck is the sole questioning the existence of a free will tout court and applies less to the debate around epigenetics.

6 See Zimmerman (2002) for a critique of this version of circumstantial luck, pointing to a luck-insensitive interpretation of the role of circumstances in exercises of responsibility (pp. 563–565).

7 It should be made clear that developmental possibilities for shifting epigenetic predispositions do not provide any justification to a normative emphasis put on parental, and especially maternal responsibilities around epigenetics. There exist several other reasons that go against such a normative construct. These include broader discourses and figurations of parenthood in the representations of epigenetic knowledge (Kenney and Müller 2017; Hens 2017a, 2017b), or established biases and epistemic limitations in the experimental repertoire of epigenetic sciences (Chiapperino and Panese 2018; Sharp, Lawlor, and Richardson 2018).

8 Besides Walker’s response to the problem, based on a defense of impure moral agency, the relation of luck and control to accountability is at the core of this philosophical debate. Rejection of moral luck is often motivated by the so-called Control Principle (i.e. the idea that we are morally assessable only to the extent that what we are assessed for depends on factors under our control), and its corollary (two people ought not to be morally assessed differently if the differences between them are due only to factors beyond their control). I ignore the full complexities of this debate, but besides Dana Nelkin’s introductory analysis of this matter (Nelkin 2013), the reader may find useful: the original formulation of the problem of control in Nagel’s contribution (Statman 1993, 58 and ff.); a rejection of the idea that luck poses a paradox for coherently supporting the Control Principle and its corollary (Zimmerman 2002, 559); a critique of the argumentative strategy (i.e. the counterfactual gambit) scholars have used to delimit the impact of luck on the Control Principle and its corollary (Hanna 2014).

9 Arguments in favour of collective epigenetic responsibilities are often presented in a truncated form, and seldom present a structured taxonomy of the different concepts of responsibility they employ (see Hedlund 201 2 for a structured example). The task of disambiguating all these different versions of the same argument goes beyond the scope of the present paper. In this section, I have tried to address appeals to collective epigenetic responsibilities, as if these were based on the standard model of responsibility (grounded on causality, voluntariness and knowledgeability) structuring the debate around individual claims. I have signalled in this section where references to these dimensions of responsibilities for collectives could be found into any concrete version of the argument in the literature. By doing so, however, I do not mean to misrepresent any of these colleagues’ positions. These appeals to collective epigenetic responsibilities may simply be a polemic and political tool against individual responsibilities, which is open to further refinement that may well escape the challenges raised by luck.