Innovations in science and technology are often the source of public concern, but few have generated debates as intense and at the same time with such a popular fascination as those surrounding genetic technologies. Unequal access to preimplantation diagnosis could give some individuals the opportunity to select children with more advantageous predispositions.Footnote 1 Genetic therapies and enhancements (improvements of healthy characteristics) involving changes in germ-line cells (egg or sperm cells, or their precursors) could lead to inheritable genetic advantage and then to an accumulation of natural advantage along lineages.Footnote 2 Unequal access to genetic technology could contribute to an increase in existing social inequalities, by intensifying congenital inequalities, which would lead to significant competitive disadvantages for the socially and economically disadvantaged members of society.
However, more inequality is not necessarily unjust. As many contemporary egalitarians are happy to admit, some inequalities might not be objectionable if, through trickle-down or incentive effects, they allow further improvements of the conditions of the worst off. This general consideration is relevant because it is misleading to frame access to therapies and enhancements as a zero-sum game, in which one person’s win is another’s loss.Footnote 3 That would be the case if the only goods produced by cooperation were positional: goods like being tall, whose value increases the higher one is in the hierarchy of people having access to that good.Footnote 4 On the contrary, both therapies and enhancements, like education, contribute to creating resources that governments could redistribute.Footnote 5 To judge whether an unequal distribution of a good is unjust, one must know where the inequality comes from. For instance, is it produced by a scheme of cooperation that is in the long-term benefits of the least advantaged, or by a scheme with a tendency to worsen the expectations of the least favored group?
An additional complication is that different conceptions of justice measure injustice by focusing on different types of inequalities. When justice is said to obtain in virtue of how the good X is distributed, X may be called the currency or distribuenda of the conception of justice in question. It is useful to distinguish theories in terms of the currency in which social institutions operate. John Rawls’s theory of justice as fairness, for example, evaluates justice on the basis of how social institutions distribute social primary goods. These are, by definition, all-purpose instrumental goods whose distribution is directly controlled by social institutions.Footnote 6 (Revisions of this formula in more recent formulations of the idea of justice as fairness change nothing for our discussion.)Footnote 7 Natural primary goods are also, by definition, general purpose goods; one might think that health, vigor, intelligence, and imagination are such goods. They are not regarded to be the currency of distributive justice, because, according to Rawls, they are only influenced, not directly controlled, by them.Footnote 8
Rawls’s choice of social primary goods as the sole currency of justice has been criticized for several reasons. It has been argued, for instance, that his theory does not directly justify special compensations for disabled and chronically ill individuals. If one considers natural primary goods, welfare,Footnote 9 or Sen’s capabilitiesFootnote 10 instead, one can justify them straightforwardly, on the basis of the observation that most disabled individuals obtain fewer primary goods than most healthy people, or that they are less able to convert external resources into welfare or capabilities. For that reason, some philosophers claim that conceiving justice as a function of the distribution of social primary goods currency cannot account for widespread moral convictions, such as the alleged one concerning special compensation for disabilities.Footnote 11, Footnote 12, Footnote 13 Other philosophers think that justice involves the rectification of inequalities that are due to brute luck, that is to say, to factors for which individuals cannot be held responsible.Footnote 14 They are suspicious of the distinction between natural and social primary goods, because it is orthogonal to the distinction between brute luck and responsibility. A person’s initial social circumstances, for instance, are socially produced and are certainly a matter of brute luck for the person who benefits from them. Health is perhaps describable as a natural (biological) good, but bad health can result from either brute luck, for instance, genetic predispositions, or from choices analogous to deliberate gambles, for instance, participation in winter sports, or from both.
The debate about access to gene therapy and genetic screening suggests a different criticism of the social primary good currency, namely that the boundary between what counts as natural and what counts as social is blurred.Footnote 15 This critique is made more vivid by recent advances in biotechnology that suggest that one might be able, in the future, to alter the distribution of initial natural characteristics to an unprecedented degree. Among the philosophers who have argued in this guise, Colin Farrelly’s argument deserves special attention, because he is among the few to draw a precise normative implication from it. He has argued that
Rawls’s general conception of justice stipulates that only the SPG [social primary goods] are to be equally distributed, unless an unequal distribution of any, or all, of these values is to everyone’s advantage. But what about the NPG [natural primary goods]? . . . [Genetic] technologies will make it possible for the distribution of NPG to be directly (though not totally) under our control. Advances in screening and diagnostic technology, gene therapy, and genetic enhancement mean that the goods to be distributed by principles of social justice will be radically changed.Footnote 16
As we shall see, Farrelly concludes that access to genetic technology ought to be regulated by a principle governing the distribution of genes relevant to the distribution of natural primary goods: the genetic difference principle.Footnote 17 This argument appears to be independent from the arguments coming from utilitarians, capability theorists, and luck egalitarians, cited previously, because it does not rely on the same premises and moral intuitions.Footnote 18, Footnote 19
This article has the following structure: in the first section I explain why Farrelly correctly avoids turning justice into a function of the actual distribution of natural primary goods. In the second section, I criticize the lax genetic difference principle, which turns justice into a function of the distribution of genetic endowments (which affect expectations of natural primary goods). I argue that the second approach provides a poor guide for policy when interactions among genetic and environmental factors are significantly complex.
Natural Primary Goods
Rawls rejects the inclusion of natural primary goods in the currency of justice by arguing that they are not controlled directly by social institutions. Does the possibility of controlling genes make any difference? To answer this question, one must specify what direct control involves and why it is morally relevant.
To clarify matters, it is useful to introduce the classical biological distinction between genotype and phenotype. Following Lewontin, we can define the genotype as the inherited set of DNA molecules present in the fertilized egg and the phenotype as that “set of traits that characterizes the whole organism at each stage of its life,” which develops in virtue of “complex . . . processes that occur not in a vacuum, but in an impinging world of environmental circumstances and developmental accidents.”Footnote 20
Consider the natural primary goods listed previously. They are phenotypical traits and, as such, influenced by both genetic and environmental circumstances. The distribution of health, vigor, intelligence, and imagination is not directly controlled, but is only influenced, by the initial genetic distribution. Even if individuals begin life with identical genomes, they will be influenced by different developmental accidents. Even if a population consists of identical twins, they will sooner or later become free to choose (to some degree) the environmental influences affecting their development.
The problem with directly controlling distributive shares of health, vigor, intelligence, and imagination is that, in order to realize a specific pattern of distribution, the state must engage in continuous redistribution, thus interfering with people’s lives in an especially burdensome way. Developmental accidents and personal choices cause a departure from the distributive target specified by the principle in question. For instance, if the principle prescribes equality of health, vigor, intelligence, and imagination, inequalities due to accidents or individual choices must be corrected or compensated for, in order to reestablish equality. It seems more difficult and morally problematic for the state to control natural, rather than social, primary goods in this way, as I shall explain shortly.
To achieve a target distribution of wealth and income, the state taxes individual assets; if taxes are not delivered, it may expropriate some properties. These assets are things, features of the environment, external to the person. On the contrary, qualities like health, intelligence, vigor, and imagination are essentially connected to physical and mental properties intrinsic to the person. Controlling (rather than influencing) a person’s assets of these goods might require controlling amounts of physical exercise, diet, and rest, or even modifying the body through biological interventions, in the most challenging cases. Few would accept extending the authority of the state over the individual in this way.
The point is that, in contrast to wealth, health, vigor, intelligence, and imagination cannot be transferred from those who have more of them to those who have less of them, without violating the right to physical and mental integrity of the person, by which I mean the (moral) right of self-ownership of mind and body, one of the basic liberties recognized by any state calling itself liberal.Footnote 21 These goods are, in that sense, what economists call indivisible goods. These assets cannot be divided, but their distribution can be affected: taxes on cigarettes and subsidies on sports can affect a person’s tendency to build assets of such goods. Legitimate social institutions can influence the distribution of these assets but cannot control them. If justice, on the contrary, required realizing patterns of natural primary goods, it would require direct control over the distribution of such goods, leading to excessively intrusive governmental policies.
This leads to moving from distributing natural primary goods such as health or intelligence to distributing opportunities for these goods; that is to say, ensuring that the playing field be appropriately leveled in relation to chances of obtaining such goods, by setting up institutions accordingly. Many social goods, in particular educational opportunities and access to healthcare, contribute to determining a person’s lifetime opportunities not only for social but also for natural primary goods, like health and intelligence. It follows that social opportunities for natural primary goods—social factors affecting the expectations of natural primary goods—can be controlled (modified, redistributed, etc.) by controlling the distribution of external goods, like wealth and legal entitlements. This strategy avoids the problem of personal integrity.
Are there natural goods that, like wealth and unlike intelligence, can be directly controlled (e.g., added, reduced, transferred) without violating personal integrity? Genetic endowments may satisfy this description in certain conditions. If a very young embryo in a test tube is not regarded as a person, intervening on the genes of this embryo (or selecting a different embryo for implantation) cannot be regarded as a violation of that future person’s personal integrity. Let us concede this possibility, for the sake of argument.
This reasoning leads to and justifies Farrelly’s approach. He explicitly acknowledges Rawls’s point that society cannot directly control the distribution of natural primary goods and concludes that institutions should seek to “fairly distribute those genes that are most important for the [natural primary goods].”Footnote 22 In the next section, I shall argue that this—more promising—strategy is unsatisfactory, too.
Genes
Farrelly’s genetic difference principle, henceforth GDP, states that “inequalities in the distribution of genes important to the NPG [natural primary goods] are to be arranged so that they are to the greatest benefit of the least advantaged.”Footnote 23 To apply this principle, one has to be able to identify the worst-off group and compare different distributions of genetic endowments in terms of the expected benefit to its members. A just distribution of genes is one (rationally) expected to maximize the natural primary goods holdings of those with lesser primary goods. The principle does not assume genetic determinism; genes do not determine individual levels of health, vigor, intelligence, and imagination but affect the probabilities that specific levels will be reached. Farrelly’s view is not egalitarian but prioritarian in inspiration. Prioritarianism is characterized by Farrelly as the idea that “it is morally more important to benefit the people who are worse off.”Footnote 24
It is more instructive (and intuitive) to imagine the principle applied to single-gene traits. Farrelly points out that “there are over 6000 known single-gene disorders, which occur in 1 out of every 200 births. A single-gene disease is a disease caused by a single malfunctioning allele. Such disease typically can develop in practically all usual environments.”Footnote 25 Suppose that we are dealing with a genetic disease for which no easy and inexpensive environmental fix exists. Huntington’s disease, for instance, leads to severe cognitive impairment and paralysis after age 40. The mutation that is responsible for it obviously reduces lifetime expectations of natural primary goods. In a world in which no single-gene condition worse than Huntington’s disease exists, the GDP would prescribe adopting institutions that increase the chance that individuals with Huntington’s disease will obtain gene therapy. Yet even in this hypothetical scenario, the principle is problematic, because it would justify channeling too many social resources into genetic technology. For genetic inequalities are not the only inequalities that deserve attention from the point of view of justice. As Farrelly points out, “many of those who are worse off in terms of their expected lifetime acquisition of natural primary good are worse off, not because of their genetic endowments, but because they lack access to adequate housing, basic healthcare, or long-term economic security.Footnote “26 Farrelly imagines a government that must choose between improving public education and investing in public genetic enhancements. He imagines a case in which “if the government had pursued improving education they would have improved the life prospects of many other people who are not genetically disadvantaged. . . . Even those who have both favourable genes and social environment could benefit from quality public education.”Footnote 27 As Farrelly acknowledges, the stringent interpretation of the genetic difference principle leads to unreasonable policies, based on the implausible idea that “a small gain to the genetically disadvantaged is more important than a great loss to both the socially disadvantaged and [the overall] advantaged.”Footnote 28 This leads to what Farrelly calls the “problem of weight.” It is not a specific problem for prioritarian theories of genetic justice. It is equally a problem for egalitarian theories, and, indeed, for any conception of justice in which the distribution of genes is regarded as one of justice’s distinctive goals. In the context of egalitarianism, it can be defined as the problem of “balancing the desire for achieving genetic equality with the desire for achieving other kinds of equality (e.g., wealth and income) and other values (e.g., utility and freedom).”Footnote 29 Farrelly claims to endorse a form of pluralistic prioritarianism. This is, as defined by Farrelly himself, a kind of prioritarianism that takes into due account the two complications cited previously: the importance of values other than justice and the existence of nongenetic forms of advantage.Footnote 30
Farrelly claims that the problem of weight can be solved by moving away from the strict interpretation of the GDP. He endorses the lax genetic difference principle, which says that when selecting institutions affecting the distribution of genes, society should aim to ensure the “greatest reasonable benefit to the least advantaged.”Footnote 31 The qualification “reasonable” “permits a number of diverse considerations”—such as general utility and freedom—“to come to the fore.”Footnote 32
Moreover, pluralistic prioritarianism must take into account the so-called problem of genetic complexity. Farrelly notices that
the most prevalent diseases—like heart disease, cancer, and diabetes—are not caused by a single malfunctioning allele. These more common diseases are multifactorial diseases. Their development depends on a variety of factors beyond our genetic constitutions. Environmental factors like diet and lifestyle (e.g. exercise, smoking, stress levels, etc.) often play a more important role in determining our risk of developing multifactorial diseases. . . . Our genes do play a role. . . . But . . . environmental factors play a much greater role.Footnote 33
Farrelly claims that “to take genetic complexity seriously, justice theorists ought to embrace pluralistic prioritarianism.”Footnote 34 Yet Farrelly’s pluralistic prioritarianism has another feature on which Farrelly does not elaborate: it is conceived as the result of an intuitive weighing of different particularistic claims of justice. To understand this particularistic aspect of pluralistic prioritarianism, it is useful to contrast it with an instance of what I shall label “monistic” prioritarianism: Arneson’s “responsibility catering” welfare prioritarianism.Footnote 35 Like pluralistic prioritarianism, monistic prioritarianism “is egalitarian in tilting in favor of those who are badly off.”Footnote 36 But in contrast to the pluralistic version, “priority is assigned to aiding an individual in virtue of how badly his life is going, as measured by an objective scale of well-being.”Footnote 37 “Well-being” here provides the currency of justice, in the sense that the moral weight of removing genetic or environmental forms of deprivation is established by converting all the relevant factors into a unified welfare metrics. Suppose that the well-being of people is affected only by the distribution of four goods: germ-line gene therapy, healthcare, access to education, and wealth. Although each factor adds something to well-being, there is no need to invoke distinct principles for dealing with each factor. Rather, the weight and importance of each depends on its contribution to distributing well-being. It is not necessary to calculate separately which educational arrangement, for example, would maximize well-being for the worst off. Rather, entire social assets are evaluated to see which combination of genetic modification, healthcare, access to education, and wealth is most likely to benefit the worst off.Footnote 38
By contrast, Farrelly’s prioritarianism includes (explicitly) a principle of genetic justice and (implicitly) norms of nongenetic justice; that is to say, different norms exert competing demands on the social system, with no preestablished order of priority.Footnote 39 No general formula is provided for weighing them, although it is recognized that, for philosophy to be relevant for policy, it must achieve some weighing.Footnote 40 This analysis is supported by analyzing three features of Farrelly’s approach: First, it analyzes the claim that we need “an account of genetic justice, that is, an account of what constitutes a fair distribution of genetic endowments that influence our expected lifetime acquisition of natural primary goods.”Footnote 41 Second, notice that according to both the strict and the lax version of the genetic difference principle, genetic justice can be realized without modifying the distribution of other goods; the genetic difference principle never requires distributing goods beside genetic endowments. Third, one finds the idea of overall justice as the outcome of a reasonable weighing of heterogeneous moral claims; justice all things considered results from finding the appropriate balance between particularistic claims of justice—claims of justice involving different particular goods. If that is true, it makes sense to distinguish all-things-considered justice, which is what follows from such weighing, and genetic justice in the strict sense, which is what the genetic difference principle would require, if it didn’t have to be weighed and compromised against equally legitimate moral claims of justice and other values. When Farrelly claims that we need an account of genetic justice, I take him to mean that we need an account of what a fair distribution of genetic endowments would have to be apart from the requirements of other particular aspects of justice. Otherwise, what justifies invoking the genetic difference principle, a principle for dealing specifically with genetic endowments?
It might be objected that, in the first section, I have already provided the answer to the question I’m asking. I argued that principles dealing directly with the distribution of natural primary goods involve a violation of personal integrity. This would be correct if the only available options were (1) distributing actual natural primary goods or (2) distributing genetic endowments “that influence our expected lifetime acquisition of natural primary goods.”Footnote 42 There are, however, many other options, for instance, (3) distributing opportunities, by modifying social factors affecting (statistically definable) expectations of natural primary goods. This is not the same as distributing genetic endowments because the distribution of opportunities is affected not only by genetic endowments but also by a person’s initial position in the social system. For instance, in most existing societies, parental socioeconomic status is a strong predictor of a person’s future educational level and income,Footnote 43 which are, in turn, strong predictors of health outcomes.Footnote 44, Footnote 45
In light of this analysis, one could object that Farrelly provides no justification for the claim that we need an account of genetic justice understood in this way. What, in his theory, explains the special moral importance attached to genetic endowments? What justifies appealing to a prioritarian principle whose scope is so specific? I shall not press such questions further. Rather, I shall point out that, when there are significant interactions between genetic and environmental factors that can be changed politically, Farrelly’s particularistic approach fails to justify searching for better policies (better in prioritarian perspective), thus contradicting its own prioritarian rationale.
Let us consider the distribution of the genes responsible for predispositions to diseases that respond to both environmental and genetic factors. One might be convinced on the basis of epidemiological evidence that “the major preventable environmental causes of illness and death are tobacco use, unhealthy diet, physical inactivity, excess alcohol use, infections, trauma, and exposure to environmental toxins.”Footnote 46 Merikangas and Risch conclude that, for multifactorial disorders amenable to environmental modifications, “gene hunting for disorders that appear to be highly amenable to environmental modification, such as type 2 diabetes, AIDS, alcohol dependence, and nicotine dependence, would have lower [public health] priority, even though genes may be involved in their etiology.”Footnote 47 Khoury and colleagues object that this way of thinking “could perpetuate the false competition between nature and nurture.”Footnote 48 Their point is not merely speculative. Some known environmental causes of disease, like smoking and drugs, also cause addiction, which affects the success rate of interventions at the population level, like information campaigns. Addiction is, in turn, influenced by genetic predispositions, which could perhaps be better dealt with through advances in pharmacologic and behavioral interventions, derived from genomics research.Footnote 49
A similar critique applies to Farrelly’s approach. His concrete proposal involves “defin[ing] the category of the genetically ‘least advantaged’ as those individuals whose genetic constitutions place them below half of the median for the expected lifetime acquisition of natural primary goods.”Footnote 50 On the basis of this criterion, people with a genetic predisposition that makes it especially difficult to overcome drug addiction could never be among the genetically worst off in the following hypothetical scenario. Let us suppose that the gene in question is spread uniformly across different social groups, so that 80 percent of the people with that genotype have, for social reasons, little chance of developing an addiction to begin with. However, 20 percent of the population with the predisposition in question belong to a socially disadvantaged group with a high chance of drug abuse. Among the members of this social group, those with the unfavorable genetic predisposition could fall below half of the median for the expected lifetime acquisition of natural primary goods within that (already disadvantaged) group. People from unfavorable social backgrounds with a genetic predisposition to hold on to their addictions could represent, within society, the sociogenetic group with the lowest expectations of natural primary goods. (Even if having the genetic predisposition in question, in and by itself, does not place the individual below half of the median for the expected lifetime acquisition of natural primary goods.) So if it were the case that gene therapy could remove the genetic cause of disadvantage among those from a disadvantageous social background, by increasing their chances of quitting addiction, committed prioritarians should perhaps favor the policy in question. Notice that this is a policy addressing the distribution of genetic endowments, yet Farrelly’s principle cannot be used to justify it.
I shall conclude the argument against Farrelly’s particularistic idea of genetic justice by arguing that the lax genetic difference principle fails to deal adequately with cases in which interactions between genes and environment involve what have been called interactive predispositions. These are cases in which, by definition, “the presence of a genetic difference between various groups both increases and decreases the probability of individuals from one group, in comparison to individuals from the other group(s), developing a particular phenotypic trait depending on the environmental conditions experienced.”Footnote 51 What is in question here is a particular type of interaction between genes and environments (GxE interaction) in which features of the environment not only affect the probability for individuals with a certain genotype to develop the phenotype in question, but also involve a change of rank: a case in which “different genetic groups respond differently to the same array of environments and that difference in phenotypic response is so extreme that the higher ranking group in one environment becomes the lower ranking group in a different environment.”Footnote 52
Again, this is not a mere theoretical possibility, but it has already been found in a genetic predisposition to sociopathy,Footnote 53 as reported by James Tabery:
In a widely acclaimed study from 2002, Avshalom Caspi, Terrie Moffitt, and their colleagues found a case of GXE for a gene controlling neuroenzymatic activity (low- vs. high) monoamine oxidase A [MAOA] and the development of antisocial personality disorder (ASPD). ASP is the clinical term for sociopathy, defined by a pervasive pattern of disregard for and violation of the rights of others. MAOA is a metabolic enzyme that inactivates neurotransmitters, and deficiencies in the enzyme have been associated with aggression. In the Capsi-Moffitt study, childhood maltreatment consisted in physical and sexual abuse. However, other experiences also counted as childhood maltreatment: harsh discipline, neglect, multiple changes in the primary caregiver, and poor mother-child interactions.Footnote 54
The special ethical and philosophical significance of this case, which Tabery’s article elucidates, is that “the MAOA study by Caspi and Moffitt is . . . an instance of GxE resulting in a change of rank.”Footnote 55 In other words, one allele (version of the gene) correlates to better outcomes if the degree of childhood maltreatment is high; the other correlates to better outcomes if the degree of childhood maltreatment is low. Cases of GxE interaction involving a change of rank create a special problem for Farrelly’s approach, as I shall show by discussing a hypothetical example.
In order to simplify the analysis of the case at hand, I shall imagine a society made up by genetically identical individuals, except for the two genes, P and R,Footnote 56 which are roughly equally distributed across the population. The two genes confer an interactive predisposition with an inversion of rank: gene P is a predictor of higher IQ in poor learning environments and gene R is a predictor of higher IQ in rich learning environments. Suppose that parents could buy gene R in an unregulated genetic “supermarket” for their children. It would be rational for benevolent wealthy parents to confer that genetic endowment on their children, because they could safely predict a rich learning environment for their children. Assume, for the sake of argument, that this group of individuals with higher IQ contributes little to raising the expectations of the natural or social primary good of those who are worst off in social and economic terms. That might be the case if, for instance, a significant proportion of high IQ individuals undertakes careers in art, philosophy, and social science that have many intrinsic merits but cause little, if any, trickle-down effects.
Let us also assume that the above-described market for genetic enhancements contributes to raising the expectations of natural primary goods (intelligence, above all) of people with a favorable starting position in society, while making it increasingly difficult for the unenhanced to compete for social positions that contribute to accumulating assets of goods, such as, let us suppose, positions in art or education. If so, the enhancement market reduces the average expectations of intelligence for citizens from disadvantaged social backgrounds, in absolute terms. By raising inequality in the distribution of a natural primary good (intelligence), said market exacerbates the sort of inequalities with which Farrelly’s principle is concerned, without any corresponding benefit for those who are worst off socially and economically.
Farrelly’s approach has problems dealing with cases like this one. I shall consider two cases: one in which roughly half of the population is from a socially and economically privileged background and obtains a rich educational environment, and one in which only a minority of the population enjoys this privileged condition.
In the first scenario, 50 percent of children enjoy a rich educational environment. In this case, the lax genetic difference principle does not single out P or R as, all things considered, better genes, in terms of expectations of natural primary goods. After all, R is not correlated with better outcomes than P overall, given that 50 percent of the R-gene population grows up in disadvantaged scenarios, where R is a predictor of IQ disadvantage. It follows that citizens with the R gene cannot be classified as the genetically advantaged and that the reverse is also true. Although the genetic supermarket affects the distribution of expectations of natural primary goods across differently advantaged social classes, it does not affect the distribution of expectations across differently advantaged genetic classes, for in this scenario all genetic classes are equally well off. Hence it could not be prohibited by invoking the lax genetic difference principle. Nor can this principle be invoked to justify a tax-subsidy scheme, in which taxes on genetic enhancements are used to boost education for the socially least well-off citizens. This follows from at least two considerations: first, the redistributive policy in question cannot be said to improve the expectations of the genetically worst off; socially disadvantaged citizens are not ipso facto worst off genetically. Second, principles of genetic justice are not intended for educational policy; in Farrelly’s pluralistic framework the requirements of genetic justice are meant to be balanced against those of educational justice, which means that they cannot possibly be meant to include them.Footnote 57
Let us now consider the second scenario, in which only a minority of the population (say, 10 percent) grows up in a rich and stimulating learning environment. There are two ways of analyzing the condition of being born with the R gene in this scenario. One option is the quasi-utilitarian one of regarding the R gene as the “worst” genetic predisposition, because it correlates with the worst outcome in the most widespread environmental circumstances (those involving poor learning conditions). The number of cases in which R correlates with better expectations is significantly lower than the number of cases in which R correlates with worse expectations, because a substantial majority of the population enjoys poor learning environments, in which the P gene is advantageous.
Under this (objectionable, as we shall see) interpretation, individuals who have inherited the R gene represent the genetically worst-off group. It turns out that when privileged parents buy gene therapy interventions for their socially privileged children, they maximize the natural primary goods expectations of the genetically worst-off group. At the individual level, every wealthy couple who, through germ-line gene therapy, replaces the P gene with the R gene in their child improves the lot of R-gene children as group: thanks to the enhancement market, a larger proportion of children with the R gene will be born in socially and economically advantaged circumstances and have better expectations of IQ. The market therefore improves the (average) expectations of the natural primary goods of the genetically worst-off group, and therefore the genetic difference principle is satisfied.
It is remarkable that, in this case, prioritarian genetic justice can be achieved by changing the genes of socially advantaged children, but not by improving the social conditions of the socially disadvantaged children with the R gene. Changing the genes of socially advantaged children from the P to the R allele raises the expectation of individuals who are born in favorable social circumstances and, because in those environmental conditions the R gene becomes relatively advantageous, in favorable natural circumstances too. The set of individuals whose expectations are thus improved, defined by both genetic and social characteristics, has the highest expectations of natural primary goods, overall. What is questionable, therefore, is that the prioritarian policy justified by the genetic difference principle justifies improving the conditions of a group that is better off in terms of expectations of natural primary goods, under such (objectionable) classifications of advantageous and disadvantageous genetic predispositions.
It could be objected that it is a mistake, conceptually speaking, to treat an interactive predisposition as an ordinary one. When a genetic predisposition is interactive, neither of the two traits should be considered better or worse as such, that is, as the predisposition attached to the worst expectations, or the one to modify or screen away.Footnote 58 If this point is granted, the problem with genetic justice is that it fails to address inequalities arising in virtue of combinations of genetic and environmental factors in which the two interact with inversions of rank. In the above-described case, a market of genetic enhancements leads to a distribution of the R allele, which contributes to an increase in inequality of natural primary goods. Yet, from the standpoint of the genetic difference principle, this inequality neither augments nor reduces genetic inequality, which is by definition an inequality among groups defined by their genetic endowments (not by membership in a social group).Footnote 59 For all we know, many genetic predispositions could involve inversions of rank. If principles of genetic justice cannot be applied to such cases, what are they useful for?
Although it promotes a more unequal society in terms of natural primary goods, a free market of R-gene enhancements in the second scenario is not unjust from the point of view of genetic justice. Saying that an unregulated market of genetic enhancements is a worse policy, for the worst off, than one in which the enhancement market is taxed and the tax used for improving education for members of socially and economically disadvantaged classes represents a piece of sound moral reasoning that cannot be framed in terms of the weighing idea. It can only be applied to the choice of whether the worst group in terms of genetic determinants of natural primary goods or the worst group in terms of social determinants of social primary goods is to be privileged by policy, on a tight budget constraint. The framework can avoid such counterintuitive consequences only if one claims that interactive predispositions fall outside its intended scope, but (depending on how the biological facts turn out) that might amount to conceding the practical irrelevance of the enterprise.
Conclusions
It should be clear at this point that the problem of genetic complexity challenges an essential feature of Farrelly’s idea of pluralistic prioritarianism: taking genetic endowments as the currency of a distinct principle of justice, despite the fact that the value of genetic endowments cannot be assessed independently of environmental circumstances. In this perspective, a redistribution of the social determinants of advantage (education) cannot be considered a requirement of genetic justice. What appears as an intuitive and elegant framework, however, makes no sense in the light of the way in which genes and environments interact. The objection against pluralistic prioritarianism is that it does not require the ministry of genetic justice and the ministry of education to talk to each other. It is plausible to think that this result can be extended to non-prioritarian frameworks that require balancing the claims of different particularistic principles having to do with genes and other goods.
The conclusion is not that a conception of social justice ought to be monistic, that is to say, a function of the realization of a single principle or of the distribution of a single good, such as Arneson’s responsibility catering (welfarist) prioritarianism. Rather, the point is that the division between genetic goods and other goods makes no sense in the light of the importance of potential interactions between them. Farrelly’s framework of pluralistic prioritarianism is unable to cope with the fact of genetic complexity. One ought to abandon not only Farrelly’s genetic difference principle but also the very idea that regulating access to genomics involves an account of genetic justice, that is, an account of what would constitute a fair distribution of genetic endowments, describable as a distinct ideal, independent from the social bases of inequality.
