CHAPTER THREE: GENES, JUSTICE, AND HUMAN NATURE

I. Distributive Justice Issues Raised by Genetic Intervention

The genetic revolution in molecular biology will not benefit all equally, and some may in fact be greatly disadvantaged by particular applications of genetic science. The fairness of the distribution of benefits and burdens is a matter of social justice, and has appropriately received considerable attention in bioethical writing on the Human Genome Project. Many of the distributive issues are near-term and quite tangible. They include:

! whether it is just to exclude individuals from employment or from health, life, or disability insurance if they are known to have genetic diseases or genetic factors that predispose them to diseases;

! whether it is just for only those who can afford genetic services to have access to them, especially since much of the initial research that led to these services was publicly funded;

! whether the right to health care includes entitlements to genetic enhancements as well as treatment and prevention of diseases;

! whether international distributive justice requires that the fruits of genetic science (in medicine and agriculture in particular) be shared with those in poorer countries that cannot afford to develop the technology, as opposed to leaving the distribution of these benefits to the global market; and

! whether the direction of genetic research and development should be shaped by expected market demand, as opposed to having ethical principles determine priorities (for example, channeling funds to cure devastating diseases that afflict large numbers of people rather than to the enhancement of normal traits for the rich or the discovery of treatments for less serious conditions, such as male baldness) (Murphy and Lappe, 1994).

At a deeper level, some writers have explored a sixth issue: puzzles concerning the use of genetic interventions to prevent serious disabilities, pondering whether the failure to do so can be morally wrong when the only way to avoid the disability is to avoid the birth (or conception) of the individual who would have it.[1]

The first five concerns are all clearly questions of distributive justice. All have already been discussed in the burgeoning literature on the ethics of the genetic revolution. The sixth, which has also been extensively explored, bears on the theory of distributive justice, at least so far as the scope and limits of the rights of future generations are concerned, even though those who consider it have generally spoken only of obligations to prevent harm rather than those of justice. But the broader implications of the possibility of identity-altering interventions for theorizing about justice have not been drawn by those who have explored the sixth concern.

Resolving these issues, with the possible exception of the sixth, does not require a fundamental reorientation in our thinking about distributive justice. They are old questions in new guises.

There is, however, another set of distributive justice issues raised by rapid advances in genetic science that have not yet even been systematically articulated, much less resolved. As the possibilities for significant and large-scale genetic interventions on human beings come closer to being actualized, we may be forced to expand radically our conception of the domain of justice by including natural as well as social assets among the goods whose distribution just institutions are supposed to regulate, to abandon the simple picture of justice being about distributing goods among individuals whose identities are given independently of the process of distribution, and to revise certain basic assumptions about the relationships between justice, human nature, and moral progress. In this chapter, we are concerned about these more fundamental and less discussed issues of distributive justice.

II. Including the Distribution of Natural Assets in the Domain of Justice

The Traditional View: Natural Inequalities Are Not a Concern of Justice

At least in the West, systematic thinking about justice begins with Plato's Republic. Plato assumed that there are signficant natural inequalities--what we would call genetic differences--among human beings. He also believed that differences in the capacity to develop higher forms of intelligence and to cultivate certain virtues are among these natural inequalities. Furthermore, he assumed, without argument, that these natural inequalities (and perhaps all natural inequalities) are not themselves within the domain of justice. For Plato, the (alleged) fact that some are born with a lesser capacity for developing intelligence, or wisdom, or courage, or temperance is not itself something that justice requires efforts to correct. Those who happen to be born with lesser capacities have no claim of justice.

Later theorists have tended to share this assumption about the boundaries of the domain of justice without questioning it--perhaps without even noticing it. Recently, however, some theorists, including Ronald Dworkin and John Roemer, have suggested that justice requires redistributing social goods in order to compensate those with less desirable natural assets. But they have not considered the possibility that justice might sometimes require altering the natural assets themselves, perhaps for the simple reason that until very recently this has been unthinkable.[2]

Challenging the Traditional View

The new molecular genetics challenges the assumption that justice does not require interventions to alter peoples' natural assets because it offers the possibility of selectively, rapidly, and accurately modifying or replacing individuals' genes. Moreover, even if it should never become feasible to achieve large-scale direct genetic intervention to alter or replace genes that significantly influence life prospects, it may well become possible to exert much greater control over a wider range of phenotypic characteristics than ever before through genetic pharmacology.

If precise and safe control over the distribution of natural assets becomes feasible, then those who believe that justice is concerned with the effects of natural assets on individuals' life prospects will no longer be able to assume that justice requires only that we compensate for bad luck in the natural lottery by intervening in the social lottery, rather than by attacking natural inequalities directly. As we shall see, there may be moral reasons for not doing what we would then be able to do, but whether such reasons are available depends on the theory of justice followed.

Equality of Opportunity

Many contemporary theorists of justice follow Rawls, who holds that the principles of justice include a principle of equal opportunity. There are disagreements, however, about how equal opportunity is to be understood. Three major alternative interpretations of equal opportunity may be distinguished in the historical and contemporary literature (Buchanan 1995):

A. Equal opportunity requires only the elimination of legal barriers to similar prospects for persons of similar talents and abilities (sometimes called "Formal Equality of Opportunity, or "Careers Open to Talents")

B. Equal opportunity requires the elimination of legal and informal barriers of discrimination for persons of similar talents and abilities ("informal barriers" includes extra-legal discrimination based on race, gender, sexual preference, ethnicity, class, religion, etc.)

C. Equal opportunity requires not only the elimination of legal and informal barriers of discrimination, but also efforts to eliminate the effects of bad luck in the social lottery on the opportunities of those with similar talents and abilities. (The "social lottery" here refers to the ways in which one's initial social starting place--family, social class, etc.--affect one's opportunities. Hence, one of the most important measures required by this third conception of equal opportunity is free basic education).

The first and second interpretations are both instances of what John Roemer has called the nondiscrimination conception of equal opportunity (Roemer 1996). According to these, it is only the limitations on opportunities that result from legal or informal discrimination that are unjust and hence, as a matter of justice, call for elimination. Once the first interpretation is accepted, it is difficult if not impossible, on pain of inconsistency, to refrain from replacing it with the second: legal discrimination is clearly unjust, but if that is true, then private or informal discrimination appears to be unjust as well. Following John Roemer's terminology, we can call the third interpretation the level playing field concept of equal opportunity.

It is probably fair to say, as a broad generalization, that the dominant view in contemporary liberal political philosophy is that equal opportunity requires at least the elimination of both legal and informal discrimination. Some who would describe themselves as liberals, however, would resist any effort to expand the requirements of equal opportunity further, to encompass the third interpretation. On their view, the fact that some individuals opportunities are seriously limited by the poverty or lack of education of their families of origin is unfortunate, but calls for no remedy as a matter of justice.

Nevertheless, the third interpretation of equal opportunity has considerable appeal, once the expansion from the first to the second interpretation is granted. Its core idea is that the fact that someone is born into a family with low socioeconomic or educational status should not in itself lead to that person's having lower life prospects than other persons of similar talents and abilities who were born into more fortunate social circumstances.

Two Variants of the Level Playing Field Conception

Two quite different reasons can be invoked to support the idea that the opportunities of people with similar talents and abilities should not be disparate due to the effects of the social lottery. Each gives rise to a different form of the level playing field concept of equal opportunity.

The first is that, at least in societies like the United States, how someone fares in the social lottery is significantly influenced by the ongoing effects of unjust social structures. Perhaps the most obvious example here is the continuing effects that past discrimination have on the opportunities of African Americans. The distribution of initial social assets would also be influenced, however, by the present effects of unjust inequalities in the distribution of wealth and income that are not due to racial discrimination and that would persist for sometime even if appropriate principles of distributive justice were now implemented. Not all past injustices were a matter of racial discrimination; there was also economic exploitation, facilitated by the abuse of governmental power and in some cases lawless coercion.

The first version of the level playing field requires that something be done to counteract the opportunity-limiting effects of bad luck in the social lottery so far as these limitations result from the ongoing effects of unjust social structures. It can be argued that Rawls is endorsing this social structural view when he says that "those who are at the same level of talent and ability, and have the same willingness to use them, should have the same prospects of success regardless of their initial place in the social system" (Rawls 1971). This view does not limit efforts to achieve equality of opportunity to countering the lingering effects of discrimination; it also requires efforts to counter the ongoing effects of other forms of past institutional injustice, including the unjust distribution of wealth. But the emphasis, to repeat, is on limitations on opportunity that originate in unjust institutions, not in natural differences among persons.

The second variant of the level playing field concept is based on a different assumption: the moral intuition or considered judgment that persons should not have lesser opportunities as a result of factors beyond their control, in the sense of being unchosen. Thomas Scanlon has labelled this the brute luck view of equal opportunity--the contrast being between matters of brute luck, which are not within one's control at all, and misfortunes that depend on a person's choices (Scanlon 1989). According to this view, persons should not have lesser opportunities due to how they fare in the social lottery--whether born into a poor, uneducated family, and so on, regardless of whether the limitations on their opportunities originate in unjust institutions.

When it comes to the social lottery, the implications of the two variants of the level playing field conception are closely congruent, at least in a society with a history of unjust social institutions. In that case, many of the inequalities in initial social assets (all of which are beyond the control of the individual) will be the result of unjust social structures. But when it comes to the natural lottery, the social structural view and the brute luck view have quite different implications. The former has no direct implications for inequalities in opportunity resulting from the natural lottery--the distribution of natural assets or endowments. The latter does: equal opportunity, on the brute luck view, requires efforts to counteract the effects of all factors beyond an individual's control. And if anything is beyond a person's control, it is how the individual fares in the natural lottery.

The difference between these two variants can hardly be overemphasized. The social structural view, like the discrimination conception of equal opportunity, limits the domain of equal opportunity, at least in the first instance, to social inequalities, because it is concerned only with how social structures, and more specifically, unjust social institutions, influence a person's success in competing for desirable offices and positions in society. The brute luck view is much more expansive: it enlarges the domain of equal opportunity to include natural inequalities.

It is worth noting that there are passages in Rawls's much-cited discussion in his book A Theory of Justice of how his Principle of Fair Equality of Opportunity and his Difference Principle fit together that might be interpreted as endorsing the brute luck view. For example, in explaining why he rejects the nondiscrimination conception of equal opportunity, Rawls notes that if equal opportunity were construed in this narrow way, this would be inadequate because "the initial distribution of assets for any period of time [would be]...strongly influenced by natural and social contingencies." Similarly, he also says that a theory of justice that allows "the distribution of wealth and income to be determined by the natural distribution of abilities and talents" appears to be "defective" [ATOJ, pp. 73-74]. Both these passages might at first appear to go beyond the social structural view toward the brute luck view (Rawls 1971).

It is true that the passages do seem to commit Rawls to the view that justice is concerned with natural as well as social inequalities. However, a closer reading of the text suggests that Rawls does not seek to address natural inequalities under the heading of equality of opportunity. Instead, he appears to restrict equal opportunity to efforts to counteract the opportunity-limiting effects of unjust social institutions (that is, the social structural version), while noting that the operation of a distinct principle of justice, the Difference Principle, will do something to mitigate the effects of natural inequalities. (The Difference Principle requires that inequalities in wealth broadly construed work to the greatest advantage of the worst off). In the second passage cited above, Rawls may be merely saying that it would be impermissible to base a person's entitlement to a share of social goods on the mere fact that he happens to have been more fortunate in the genetic lottery. That view does not commit him to the brute luck thesis that all natural inequalities require redress or compensation as a matter of justice.

If this interpretation is correct, then Rawls's remarks about the moral arbitrariness of natural inequalities do not signal that he has moved to the more radical brute luck conception. However, they do mean that in some sense he regards natural inequalities as falling within the domain of justice. Other passages in the same discussion lend further support to the hypothesis that Rawls's conception of equal opportunity is the social structural view (for example, his discussion of how the family influences peoples' opportunities) (Rawls 1971).

Our suggestion, then, is that in spite of his recognition that in some sense it is contrary to justice to allow natural inequalities to affect peoples'life prospects by basing their distributive shares on their natural endowments, Rawls opts for a combination of a more restrictive conception of equal opportunity with the Difference Principle, rather than a more radical interpretation of equal opportunity that would require counteracting the effects of natural inequalities or eliminating natural inequalities altogether, if this were possible through genetic intervention.

A proponent of the brute luck view of equal opportunity might at this point say that Rawls's response to natural inequalities is inadequate. If equal opportunity is so important as to enjoy the priority that Rawls accords to it, then the Difference Principle cannot adequately mitigate the opportunity-limiting effects of natural inequalities.

As we discuss in Chapter Four, the matter may not be so simple. Rawls's apparent choice to use the more limited, social structural conception of equal opportunity and to rely on a kind of division of labor between that and the Difference Principle represents a more complex view of how a theory of justice should be structured. One virtue of this approach is that it preserves the historical roots of contemporary notions of equal opportunity in the concept of formal equality or "careers open to talents," which focuses only on limits on opportunity due to the influence of social institutions--more specifically, legal rules that entrench inherited privilege to the detriment of meritocratic competition. However, one can only determine which is the correct choice in constructing a theory of justice--the narrower social structural conception of equal opportunity plus a subordinate principle for the distribution of wealth, or the more radical brute luck view--by evaluating the overall strengths and weaknesses of the rival theories in which these options are situated.

Our purpose here is not to engage in Rawls exegesis. We have suggested that the various passages in A Theory of Justice just cited can be interpreted as an endorsement of the social structural view. We acknowledge, however, that others have interpreted Rawls's concern with natural inequalities as pointing toward the more radical brute luck view. In that sense, Rawls's classic discussion can be viewed as a matrix out of which two dominant strains of contemporary liberal thought on equal opportunity can be generated.

The brute luck view (with various twists) is rather unambiguously embraced by a number of other prominent contemporary philosophers, including John Roemer, Richard Arneson, and G.A. Cohen (Roemer 1996, Arneson 1989, Cohen 1993). Our chief aim is to trace the implications of these two conceptions of equal opportunity for the uses of genetic intervention technology. We shall argue that in spite of the profound difference between these two conceptions as to the domain of equal opportunity, there is a surprising overlap in their implications for how genetic intervention technology ought to be deployed, at least for the foreseeable future.

Each of the two level playing field conceptions has some appeal. The strength of the social structural variant, as noted earlier, is that it preserves a clear connection with the historical roots of the notion of equal opportunity: the idea of careers open to talents, or formal equal opportunity. Like it, the social structural view focuses on the very plausible idea that equal opportunity has to do not with just any inequalities, but with inequalities due to defective social institutions.

Yet the appeal of the brute luck view is simple and straightforward: to many people it seems unfair that some should have lesser opportunities as a result of factors over which they have no control--circumstances that did not result from their choices. Not only are natural deficits as much beyond the individual's control as initial social assets; in some cases they may impose even more severe constraints on opportunities. This is clear enough in the case of a straightforward genetic disorder such as phenylketonuria (PKU), which, if untreated, produces mental retardation, or Tay-Sachs Disease, which causes much suffering and inevitably results in death in early childhood. Whether an individual has certain abilities and talents may depend upon his or her luck in the natural lottery. Restricting our concern about equal opportunity to differences among those with the same talents and abilities seems arbitrary.

Most industrialized societies recognize a societal obligation to use medical interventions to cure seriously disabling congenital disorders, whether their source is genetic or is due to environmental or developmental factors during pregnancy or birth conditions. If a baby is born with a hip deformity, for example, an effort will be made to marshall social resources to pay for surgical repair of this condition if the parents lack health insurance and cannot afford to pay the surgical bill. One obvious and compelling justification for subsidizing this procedure is that it is necessary in order to remove a serious obstacle to opportunity. In this sense, some of our most basic social institutions reflect a commitment to intervening in the natural lottery for the sake of equal opportunity, at least when it is a hereditary or congenital disease that threatens opportunity. And presumably any philosophical account that accords an important place to equal opportunity as a principle of justice must acknowledge the necessity of such interventions.

But an individual's place in the distribution of natural assets can severely limit her opportunities even in cases in which she does not suffer from anything that would uncontroversially count as a genetic disorder or a disease. For instance, suppose that only those whose genetic assets fall within certain parameters tend to develop certain cognitive abilities beyond a certain level. Suppose also that, in general, only those who develop these abilities beyond this level are able to learn the mathematics needed to succeed in all but the very least desirable jobs in a technically advanced society. Under such conditions, those whose genetic constitutions prevent them from reaching the needed threshold of abilities will experience significant limitations on their opportunities unless something is done to overcome this impairment.

Notice that the preceding example of a significant natural inequality that is not a disease is presented as hypothetical. At this point in the infancy of genetic science, no one can say whether there will turn out to be a significant number of genetic conditions that do not qualify as diseases but that seriously limit peoples' opportunities. Notice also that the hypothetical does not buy into anything so gross and problematic as the view that IQ is genetically determined. Instead, it only suggests the possibility that some particular aspect of cognitive functioning might turn out to have two features: first, there are significant inequalities among individuals in its distribution within the range of normal functioning for our species (hence those with lower levels of the skill do not have a disease), and second, because of particular features of the society in question, those with lower levels of the skills experience significant limitations on their opportunities. Our main concern is not whether there are such differences, but rather what the possibility that there are shows us about how we should think about justice.

Roemer provides an example (similar to our Scenario 4 in the Introduction) that is less hypothetical and yet also shows the apparent arbitrariness of restricting equal opportunity concerns to the genetic disadvantages that count as diseases. He considers the case of a child who "is not slow to learn but has an emotional cyclicity, clearly inborn, which makes it harder for him to carry out plans, or succeed in school" (Roemer 1989). His point is that because the child's psychological condition is beyond his control (being inborn) but limits his opportunities, equal opportunity requires that something be done. Notice that Roemer does not say that the psychological condition is a disease, though of course "emotional cyclicity" is broad enough to cover cases of the mental illness known as bipolar disorder.

The point is that there is a continuum of psychological conditions--from mild mood swings to bipolar disorder--and that what really matters is whether a condition limits opportunity and is beyond a person's control (Siever 1996). Whether there are a significant number of cases in which a genetically based condition (whether psychological or physical) that is not a disease substantially limits persons' opportunities is an empirical matter. The facts are in dispute, but genetic research may well document a number of such cases, with the result that we will be less confident about drawing moral lines between opportunity-limiting conditions that are diseases and those that are not.

The message of these example is not that the concept of disease is so fuzzy as to be unusable. Throughout this book we use a conception of disease proposed by Christopher Boorse and developed and used by Norman Daniels in his work on just health care. Disease, according to this view, consists of conditions that are adverse departures from normal species functioning. So genetic diseases, roughly, would be genetically based conditions that are adverse departures from normal species functioning. What Roemer's example shows is that there can be genetically based conditions that limit people's opportunities, and that what matters, from the standpoint of a general account of equal opportunity, is not whether they are diseases, but whether they limit opportunity.

In some cases, of course, a genetically based disadvantage can be remedied by nongenetic means, and the question of genetic intervention will not arise. For example, although PKU is a genetic disease, the mental retardation that it produces can be avoided by administering a special diet to the individual for the first few years of life. Similarly, as noted in Chapter One, although nonacquired hemochromatosis is caused by a defective gene, it is now successfully treated by nongenetic means (regular phlebotomies--bloodlettings).

Nevertheless, there are some genetically based conditions for which there is no known environmental treatment or nongenetic remedy, and more are likely to be discovered. The point is that from the standpoint of the brute luck view, the distinction between natural inequalities that are diseases and those that are not is insignificant, at least in principle. The real issue is whether inequalities in natural assets, as one form of brute luck, limit opportunities. If so, then whether they constitute diseases or not, they are a concern of justice.

The Social Structural View of Equal Opportunity and the Right to Health Care

As Norman Daniels has argued, the case for a moral right to health care relies, at least in part, on the fact that health care promotes equal opportunity by preventing and curing disease (Daniels 1985).

Daniels has developed a sophisticated theory of just health care based on Rawls's principle of equal opportunity. Rawls himself offers no theory of just health care, though he has endorsed Daniels's project. In A Theory of Justice, Rawls notes that he makes a simplifying assumption, which explains why he offers no account of the right to health care: the idealized parties who are to choose principles of justice are to proceed on the assumption that they are normally functioning, full participants in social cooperation. Daniels's Rawlsian theory of just health care relaxes this simplifying assumption, arguing that equal opportunity provides a basis for the right to health care.

Daniels does not endorse the brute luck conception, however, as will become clear in the next chapter. His view of equal opportunity is closer to what we have called the social structural variant of the level playing field conception. (Moreover, he maintains that it is the latter rather than the former that provides the most consistent interpretation of the totality of Rawls's writings that bear on equal opportunity).

At first blush, it might seem that it is problematic to take the tack Daniels does--to reject the brute luck view and embrace the claim that equal opportunity requires efforts to cure and prevent disease. At least in some cases, perhaps many, disease is a result of bad luck in the natural lottery, not the effect of the social structure. So if equal opportunity, as the social structural view holds, is not concerned with natural inequalities as such but only with the opportunity-limiting effects of social structures, how can it serve as the foundation for a right to health care as a response to the natural inequality of disease?

We noted earlier that the social structural view has no direct implications for counteracting natural inequalities. This qualification is important. One way of understanding how a Rawlsian theory of equal opportunity can consistently reject the brute luck view and yet base the right to health care on equal opportunity is to focus on an idea we encountered earlier in Rawls's simplifying assumption about heath care needs: the idea of being a normally functioning, full participant in social cooperation.

In the context of concerns about equal opportunity, we can think of being a normally functioning, fully participating member of society as having the characteristics necessary to be a "normal competitor" for desirable social positions. Clearly, diseases--as adverse departures from normal species functioning--can prevent an individual from being a normal competitor. We can think of equal opportunity, then, as being concerned not only to counteract the opportunity-limiting effects of social institutions but also to cure or prevent diseases, insofar as these preclude an individual from being a normal competitor in social cooperation. In other words, equal opportunity has to do with ensuring fair competition for those who are able to compete, but also with preventing or curing disease that hinders people from developing the abilities that would allow them to compete.

In this view, equal opportunity not only requires that competition be fair; it also requires efforts to bring people up to the threshold normal functioning that enables them to compete under conditions of fairness. This allows a consistent appeal to equal opportunity as a moral foundation for the right to health care without embracing the brute luck view, and with it the thesis that equal opportunity must somehow counteract all natural inequalities, not just those that constitute diseases.

It would be a mistake, however, to conclude that the social structural view as developed by Daniels to include a Rawlsian theory of just health care strictly limits genetic interventions to cure and prevent disease. The significance of disease is that it limits opportunity in the most serious cases, at least, by preventing persons from developing the threshold of abilities necessary for being "normal competitors" in social cooperation. It is possible, however, that there are some natural inequalities that are not adverse departures from normal species functioning but that nonetheless so seriously limit an individual's opportunities that he or she is precluded from reaching the threshold of normal competition. In such cases, genetic intervention might be required if it were necessary to remove this barrier to opportunity.

Whether there are such natural inequalities will depend in part upon the nature of the framework for social cooperation. It is not at all implausible to think that in some cooperative frameworks, persons with certain abilities at the low end of the normal distribution will be prevented from being effective competitors for desirable social positions, and even from participating in the mainstream economic activities of the society. Two possible examples of conditions that fit this description have already been considered: emotional cyclicity that impairs work and personal relationships but is not so severe as to count as the disease of bipolar affective disorder; being deficient in certain genetically based cognitive skills that are required for higher mathematics in a society in which sophisticated quantitative skills are needed for a broad range of jobs.

In Chapter Seven, we explore the application of conceptions of justice to the choice of a basic cooperative scheme, insofar as the nature of the cooperative scheme determines who is able to be a full participant or normal competitor in social cooperation. At this point, we only wish to emphasize that while the Rawls-Daniels view avoids the kind of wholesale commitment to genetic intervention--or to "genetic equality"--to which the brute luck view seems wedded, it may require some interventions that go beyond the cure and prevention of disease. Whether it does will depend on what the normal distribution of various characteristics is and how that relates to the most fundamental requirements for successful participation in social cooperation in a given society.

The Brute Luck View and the Scope of Intervention in the Natural Lottery

We have argued that the social structural view, at least when developed along the lines of Daniels's Rawlsian account, implies no commitment to regard all natural inequalities as subject to the dictates of equal opportunity. The brute luck view, however, seems to require just such a profound expansion of the domain of equal opportunity. For some, this expansion will seem implausible, if only because it seems to sever the notion of equal opportunity from its historical roots in the idea of "careers open to talents". For as we have seen that notion--like the social structural view, but unlike the brute luck view--anchors equality of opportunity to a concern for the opportunity-limiting effects of defective social institutions and is nothing so ambitious as an attempt to free human beings from opportunity-limiting effects of misfortune generally.

It would be a mistake, however, to make an assessment of the basic moral intuition underlying the brute luck view depend solely on whether it supports an adequate conception of equal opportunity. A number of the most prominent theorists who believe that people's prospects should not be limited by factors beyond their control have attempted to work out the implications of this conviction for a general theory of justice. They would be unimpressed by the criticism that their view does not comfortably fit within the confines of what has traditionally been regarded as equal opportunity.

Resource Egalitarianism and the Domain of Justice

The brute luck view has been advanced in some of the most interesting and rigorous work in the theory of distributive justice in recent years--the so-called "Equality of What?" debate. In initiating this controversy, Amartya Sen asked: Given that a sound theory of distributive justice will include an egalitarian element, what is it that is to be distributed equally? One answer that has been explored in some detail and has attracted the support of several formidable thinkers is resources (Sen in McMurrin 1980; Dworkin 1981a,b; Cohen 1989). Stated as a principle of justice, the resource egalitarian view is simply this: Resources ought to be distributed equally among persons.

According to John Roemer, the chief task for resource egalitarians is to devise a "resource allocation mechanism" to achieve an overall equal distribution of resources among persons by compensating for inequalities in the distribution of natural resources through special redistributions of social resources (Roemer 1985). This requirement, that those with lesser natural resources (genetic endowments) ought to be compensated by redistributing social resources to them, may be called the Resource Compensation Principle.

What is striking is that resource egalitarians have focused exclusively on the Resource Compensation Principle, not on the Equal Resources Principle itself, to the extent that they have speculated about how institutions should function if they are to achieve justice. They have not explored the possibility of achieving greater equality by intervening directly in the natural lottery by which genetic endowments are currently distributed. Yet from the standpoint of a theory that emphasizes the primacy of resources for distributive justice, and that appears to have no room for any fundamental distinction between social and natural resources, there is no obvious reason to restrict the operation of the institutions of justice to the redistribution of social resources. If resources ought to be distributed equally and natural endowments are resources, then we ought to intervene in the natural lottery whenever doing so would be the best way of equalizing resources. In contrast, as we have seen, thee is one way of interpreting Rawls's theory--namely, as endorsing the social structural view of equal opportunity rather than the brute luck view--that avoids the implication that justice requires interventions to counteract all natural inequalities.

Individual Liberty and Genetic Intervention

It might be argued that even if resource egalitarians took seriously the feasibility of interventions in the natural lottery, they would have moral reasons for not endorsing them. Most resource egalitarians are liberals; they might argue that respect for basic individual liberties precludes interventions in the natural lottery, requiring instead that social resources be distributed so as to compensate those who fared badly.

This response is unpromising for several reasons. First, resource egalitarians have generally been reluctant to assume that the particular resources Rawls calls the basic liberties are of such extraordinary importance relative to other resources as to enjoy an absolute priority. Consequently, they cannot simply dismiss the possibility of intervening directly in the natural lottery on the grounds that doing so would clearly be incompatible with respecting basic liberties.

More important, if proper regard for individual liberty is to provide a reason for not concluding that resource egalitarianism is committed to a profound expansion of the domain of justice into the natural lottery, it will have to be shown that genetic interventions violate the liberty of either prospective parents or their offspring or both.

Consider the claim that genetic interventions directed toward prospective parents (such as the insertion of genes into embryos or gametes) are precluded by respect for the liberty of the prospective parents. First of all, not all genetic interventions will require treatments to be performed on the parents. Some may be applied directly on the individual who would otherwise have lesser natural assets. In such cases, the priority of the parents' liberty could not be appealed to in order to block the intervention.

Second, even where intervening in the natural lottery requires performing procedures on the prospective parents, these will in many cases be welcomed, since they are for the benefit of their offspring. Thus it is reasonable to assume that in most cases prospective parents would freely consent to genetic interventions to increase their children's opportunities. But if this is so, then it is hard to see how their basic liberties could be an obstacle to extending the operation of equal opportunity to the natural lottery, to the extent that they would willingly cooperate in improving their children's prospects. For if the basic liberties include the right to refuse invasive interventions, it is accompanied by the right to accept them as well (as it is in the case of ordinary medical interventions). Finally, even in cases in which the parents for some reason or other did not consent to interventions on themselves for the sake of their offspring, it cannot be assumed that their interest in avoiding nonconsensual medical treatment always outweighs the fundamental interest that all individuals are supposed to have in equality according to resource egalitarianism.

Consider next the implications for individual liberty of intervening only on the individual for whose sake it is undertaken. If the intervention were early enough in life, it would be implausible to argue that the individual's interest in liberty--in being free from nonconsensual medical treatment--outweighed his or her interest in avoiding significant limitations on opportunity.

This conclusion is buttressed, once we recognize that genetic interventions might in fact be less threatening to individual autonomy and privacy and more likely to succeed at removing barriers to opportunity than the kinds of large-scale changes in family life needed to counteract the opportunity-limiting effects of disadvantages in family culture and early childhood experiences. In some cases, intervening in the genetic lottery might well prove more efficacious and less morally problematic than intervening in the family to correct for disadvantages resulting from the social lottery. A genetic intervention to correct or prevent a disabling metabolic disorder--or perhaps even to improve certain dimensions of memory that would enhance learning--might well be more effective in expanding individuals' opportunities as well as less intrusive than some existing social welfare programs are. It seems, then, that considerations of individual liberty pose no general bar to genetic interventions in the name of resource equality.

To summarize: neither Rawls nor the resource egalitarians consider the possibility of direct interventions in the natural lottery for the sake of justice. In Rawls's case, there is a plausible interpretation of his theory of equal opportunity, which, when combined with Daniels's theory of just health care, explains why justice requires intervention to prevent or cure disease but does not require any general effort to counteract or eliminate natural inequalities as such. Resource egalitarians, in contrast, appear committed to the thesis that justice requires direct interventions in the natural lottery, and not just to prevent or cure diseases, whenever doing so is the best way to achieve an overall equality of resources. Absent a principled distinction between natural and social resources, it appears that resource egalitarians must either abandon the commitment to equalizing social resources or abandon the view that it is only necessary to compensate for, rather than intervene directly in, the distribution of natural resources.

Genetic Equality?

So far we have argued that, at least from the standpoint of the brute luck view of equal opportunity and resource egalitarianism, the feasibility of genetic intervention requires a profound expansion of the domain of justice. But nothing has been said about which sorts of genetic interventions, under what conditions, are required. In particular, no conclusion has been drawn about the necessity of strictly equalizing the distribution of natural assets.

There are at least two important reasons for avoiding the leap from the thesis that in principle justice requires intervention in the natural lottery to the conclusion that justice requires an equal distribution of natural assets.

First, even if justice ought to be concerned with inequalities in natural assets, what counts as such an asset (or deficit) is at least partly determined by the social structure--and preeminently by which sorts of traits the dominant cooperative framework of a given society favors. (By the dominant cooperative framework we mean the set of basic institutions and practices that enable individuals and groups in a given society to engage in on-going mutually beneficial cooperation.) To take an obvious example: in a pre-literate hunting and gathering society, having a neurological condition that impaired the development of reading skills but that did not interfere with hand-eye coordination, motor skills, and normal oral communication would not be a deficit.

Because the dominant cooperative framework changes over time, the value of various traits changes also. In addition, the value of particular traits may depend not only only their scarcity but also what may be called complementarity--that is, the existence, in sufficient numbers, of persons with other traits with which they can be combined in cooperative interaction.

Thus, there is no such thing as a resource per se. Different traits will be resources in different social environments. Recognition of this simple fact imposes a fundamental constraint on any attempt to intervene in the natural lottery in the name of equality of opportunity or of resources.

Nevertheless, there are presumably some very basic characteristics that are what Rawls calls primary goods--maximally flexible assets, characteristics conducive to the successful pursuit of a broad range of human projects in a diversity of social environments. To the extent that the genetic factors that contribute to these can be accurately identified and subjected to safe and effective human control, there is a strong prima facie case for undertaking efforts to reduce the impact of inequalities in their distribution. Nevertheless, the fact that what counts as a resource will sometimes vary across social environments makes it misleading to talk glibly about "genetic equality."

The second reason not to leap to the conclusion that justice requires an equal distribution of natural assets is that any thought of literally equalizing such assets would almost certainly betray a failure to appreciate what might be called the fact of value pluralism (or the diversity of the good). What is regarded as a natural asset as opposed to a natural deficit, and which natural assets are regarded as most valuable, depends in part on what we assume to be a good human life. And that is typically much more controversial, and less capable of being determined by anything like objective, universal criteria, than most of us like to admit.

In addition, those who are wary of genetic determinism or reductionism rightly observe that it is extremely unlikely that desirable traits such as "altruism" or "cooperativeness" or "initiative" will be strongly determined by one gene or even by a complex of genes rather than by a complicated interaction of environment and many genes. This important point does not go far enough: it is also a mistake to assume that we all desire the same thing when we say that such traits are desirable.

In some cases, or perhaps many, superficial agreement as to what counts as desirable traits masks deeper disagreement in values. For example, it might be thought that there are some traits, such as "initiative," "altruism," and "cooperativeness," that all or most of us agree are desirable. But serious differences of opinion surface once we begin to think about just what sorts of psychological dispositions we would be seeking to foster. What some consider to be a laudable disposition to take the initiative, others will regard as excessive forwardness or even aggression. What some believe to be the right proportion of altruism in a person's motivational structure, others will condemn as weakness, as a failure to stand up for oneself. And what one person sees as admirable cooperativeness, another may scorn as lack of independence or an excessive willingness to "fit in" at the expense of individual judgment and autonomy.

The point is that the traits we find most desirable are often complex dispositions involving the exercise of evaluative judgments--what Aristotle called virtues. (How cooperative should I be [with these people, under these circumstances]? Should I take the initiative now or wait for a consensus in the group? Should I subordinate my interests to those of my friend in this instance?) Because these traits themselves involve the exercise of complex evaluative judgments, different individuals and communities may have different views about which traits, in which proportions in a person's motivational structure, really are desirable. Thus any sane and responsible attempt to harness the powers of society for large-scale genetic interventions in the name of resource egalitarianism or equal opportunity must recognize the limitations imposed by the fact of value pluralism, which includes disagreements rooted in the complex evaluative nature of some of the traits we find most desirable.

Finally, the proposal to equalize natural assets becomes even less plausible once it is seen that what would have to be equalized, presumably, would be overall packages of natural assets, rather than each natural asset, however the latter are understood. The fact of value pluralism would impose even more serious limits on attempts to compare overall packages of natural assets, making it unclear to what degree there is inequality among packages or even whether they are unequal.

A "Genetic Decent Minimum"?

Taken together, the fact of value pluralism and the fact that the value of traits is relevant to social conditions call for caution about any commitment to genetic equality--and perhaps point toward a more modest goal. At least for the foreseeable future (if not forever), the appropriate objective, from the standpoint of both the brute luck conception of equal opportunity and resource egalitarianism, may be something more like the attainment of a "genetic decent minimum"--to the extent that this can be identified with a reasonable degree of consensus--than the elimination of all inequalities in natural assets.

In practice, this would mean a strong societal commitment to use advances in genetic intervention to prevent or ameliorate the most serious disabilities that limit individuals' opportunities across a wide range of cooperative frameworks. Whether or to what extent such efforts would go beyond attempts to prevent or cure genetically based diseases is largely an unanswerable question at this point. To answer it one would have to know whether there are genetic conditions that significantly limit opportunities but that do not constitute diseases and that we will be able to prevent or ameliorate through the application of genetic science.

Points of Convergence

Thus despite deep theoretical differences, the brute luck conception of equal opportunity and resource egalitarianism, on the one hand, and the social structural view of equal opportunity augmented by Daniels's account of just health care, on the other hand, may have largely similar implications for social policy. Both seem to require that efforts should be directed, for the most part and at least for the foreseeable future, toward the use of genetic interventions to prevent or cure disease. Both views, however, allow for the possibility that justice may require genetic interventions that go beyond that, though the resource egalitarian view allows much greater scope, in principle, for such interventions.

III. The Colonization of the Natural by the Just

We now wish to explore a simple but rather radical way of reinterpreting the distinction that theorists of distributive justice have drawn between natural and social inequalities. We do not offer it as an exegesis of any particular view. Instead, we advance it as a way of avoiding an inconsistency on the part of the resource egalitarians.

The inconsistency, noted earlier, is that those who endorse the thesis that resources are to be equalized seem to be committed to direct and wholesale intervention in the natural lottery, yet opt instead for using social resources to compensate for natural inequalities social goods rather than attacking them directly. The problem is that the brute luck theorists and resource egalitarians give no good reason to opt for this principle and to eschew direct interventions in the natural lottery (whether for the sake of equal opportunity or for equality of resources, respectively). The apparent inconsistency disappears, however, if we understand the distinction between the social and the natural as that between what is subject to human control and what is not.

Nature, or the natural, is often thought to be not only that which is given but also that which must be accepted, as beyond human control. In that sense, to say that something is due to nature is to relegate it to the realm of fortune or misfortune, rather than justice or injustice. (Hence the term "natural lottery," a lottery being a game of chance.) It is not surprising, then, that to a large extent traditional thinking about justice has associated natural disadvantages with misfortune rather than injustice, since there was little or nothing that could be done to prevent them.

In contrast, nature subdued--nature mastered by human intelligence and directed to human purposes--is no longer the given, no longer that which must be accepted, and hence no longer the domain of fortune and misfortune. Paradoxically, nature brought within human control is no longer nature.

The boundary between the natural and the social, and between the realm of fortune and that of justice, is not static. What we have taken to be moral progress has often consisted in pushing back the frontiers of the natural, in bringing within the sphere of social control, and thereby within the domain of justice, what was previously regarded as the natural, and as merely a matter of good or ill fortune.

Sometimes this is accomplished by simply coming to see that what we took to be natural was all along social, as when the rising bourgeoisie demanded the end of aristocratic privileges that had, for so long, paraded as natural inequalities. In other cases, we actually extend the social by expanding our control. (For example, the annual flooding of a great river is transformed from a fact of nature to an object of administration.)

Until now, human beings have achieved this expansion of control mainly through developing technologies for controlling the nonhuman parts of nature. Now we stand on the threshold of a great expansion of the domain of the social into us. If it becomes within our power to prevent what we now regard as the misfortune of a sickly constitution (a weak immune system) or the catastrophe (the natural disaster) of a degenerative disease such as Alzheimer's dementia, then we may no longer be able to regard it as a misfortune. Instead, we may come to view the person who suffers these disabilities as a victim of injustice. As our powers increase, the territory of the natural is annexed to the social realm, and the new-won territory is colonized by ideas of justice.

Thus we might interpret the resourcists' distinction between social and natural resources as that between resources whose distribution we can control and those we cannot. Then reliance on the Resource Compensation Principle would make perfectly good sense: a commitment to equalizing resources would require that social resources, those resources whose distribution we can control, be distributed so as to achieve an equal distribution of all resources, social and natural, by using social resources themselves to compensate for differences in resources that we cannot control--natural ones.

This reinterpretation would radically transform our understanding of resource egalitarianism. The Resource Compensation Principle would only operate where direct intervention to counteract natural inequalities was either not feasible or for some reason less efficient than compensation. Understood in this way, resource egalitarianism would require that the domain of justice expands as our powers of genetic intervention develop. So even if the practical implications of the Rawls-Daniels social structural view and the resource egalitarian view would tend to converge for a time, they might yield profoundly different implications in the future if mankind's genetic powers continue to develop.

IV. Blurring the Distinction Between the Subjects and Objects of Justice

So far we have seen that both the brute luck conception of equal opportunity and resource egalitarianism appear to yield a profoundly expanded conception of the domain of distributive justice when applied to the possibility of genetic intervention. In both instances, the distribution of what we now call natural assets in principle falls squarely into the domain of justice, at least at the deepest level of theorizing. How much of what we now regard as natural assets actually becomes subject to intervention in the name of justice will depend on the extent to which our powers grow. Now we wish to argue that as the possibilities of what may be called radical genetic intervention come closer to realization, the most fundamental single framing assumption of our ordinary ways of thinking about justice, both in theory and in practice, will be shattered.

The basic problem of distributive justice, as it has hitherto been conceived, is how goods ought to be distributed among persons when their identities, at least for purposes of justice, are given independently of the distribution of goods (Roemer 1985). But if it becomes possible to distribute the genetic bases of all "natural" human characteristics, including those that are partly constitutive of the identity of persons, then this fundamental assumption--this picture of subjects waiting to receive objects through the workings of some distribution mechanism--will no longer be applicable. Yet it is not clear what alternative picture will replace it.

To focus on this problem, we will reserve the phrase "radical genetic intervention" for the use of genetic technology to determine the genetic bases of identity-constituting characteristics in human beings. The basic point is this: if radical genetic intervention becomes feasible, we shall need to reconceive the fundamental problem of distributive justice.

Note that to make this point, a person need not be identified with his or her genes. No variety of genetic reductionism need be assumed. Instead, we can acknowledge that, for fairly obvious reasons, the conception (or conceptions) of personal identity human beings have developed do not focus on genes but rather on phenotypic characteristics, whether psychological or physical or both. But to the extent that these characteristics turn out to be expressions of genes (interacting with environmental factors, of course), the ability to design the genomes of individuals is the ability to design and create individuals with particular identities.

If this is the case, then intervening in the natural lottery by genetic means will not be limited to distributing goods among persons whose identities are fixed prior to the act of distribution. Instead, the distribution of genes itself may in part determine the identities of the persons. (And this is true even if, as some conceptions of identity maintain, a person's identity is also determined in part by factors in the physical or social environment as they shape the individual's development over time.)

To put the same point in a slightly different way, we think of justice as justice to persons. But we may soon have to contemplate the idea of justice in the designed creation of persons. At present we are not well equipped to do this, in part because even the most sophisticated theories of justice have proceeded on two increasingly obsolete assumptions--that the distribution of the characteristics with which people are born is a given, beyond human control, so that any resulting inequalities must at best be compensated for rather than attacked directly; and that the fundamental problem of justice is that of distributing goods among antecedently existing particular persons.

The breakdown of the distinction between the subjects of distribution and the objects of distribution (between persons and goods) seems inescapable once we are committed either to resource egalitarianism or to a brute luck conception of equal opportunity (according to which all unchosen and undeserved significant limitations on persons' opportunities are suspect). However, the breakdown is not inevitable from the perspective of the narrower, social structural conception of equal opportunity, according to which the only suspect limitations are those that are due to unjust social structures or that prevent a person from reaching the threshold of abilities needed to be a normal competitor in social cooperation.

No doubt some will take the fact that resource egalitarianism and the brute luck conception of equal opportunity lead to a blurring of the distinction between the subjects and objects of justice as a reductio of these theories. Others may bite the bullet and embrace the conclusion that we must learn to think of justice not simply or primarily as the distribution of goods among persons, but as including the distribution of person-constituting characteristics in order to ensure that whichever persons do exist have equal opportunities (or equal resources). Those who take the latter, bolder path have yet to develop their own, alternative explanatory metaphors, much less a coherent theoretical apparatus to replace the traditional ways of framing the fundamental problem of distributive justice. It would be premature to conclude that they will not be able to do so, but equally unwarranted to assume that the task will be completed successfully.

V. Justice, Human Nature, and the Natural Bases of Inequality

Thus far we have discussed bringing the distribution of what have been regarded as natural assets within the domain of justice and challenging the assumption that the problem of justice is about distributing goods among persons whose identities are given independently of the distribution. There is yet another way in which the prospects of genetic intervention require a transformation of how we think about justice. The possibilities of genetic intervention threaten to sever the fundamental connections between justice and human nature that have been assumed in many if not most traditional theories of justice.

It is with some hesitation that we enter this terrain, because the profound sorts of genetic interventions that would call into question traditional assumptions about human nature and its relationship to justice may never become possible. The discussion that follows is therefore even more speculative than most of our explorations in this book, attempting as it does, to peer into an even more distant possible future. We make no assumptions about the degree to which changes in the genetic constitutions of humans will become possible.

In some respects, the most far-reaching insight of the genetic revolution in molecular biology is the realization of the common double-helix structure of all DNA across all species. The basic techniques of manipulating DNA render the so-called barriers between species breachable. More to the point, the very idea of a species becomes less important. Once genetic material from one species can be introduced into individuals in another species, the old definition of species as groups of individuals that can reproduce fertile offspring seems quaint, given the possibilities for creating reproducible, novel kinds of creatures.

Given the common structure of DNA and its ubiquity and fungibility in virtually all living things, we face the prospect of literally changing human nature through the introduction into human beings of genes from other species or even synthesized, novel genes that have never occurred anywhere in nature before. (Once it becomes feasible to construct novel sequences of base pairs of nucleotides, it will be appropriate to speak of creating genes, not just discovering them.)

Human nature has traditionally been regarded not only as that which unites us with one another and distinguishes us from other kinds of beings, but also as unchanging. So the possibility of changing even our "essential" characteristics would seem to render the very concept of human nature obsolete, so far as it includes the idea of an unchanging core of essential characteristics. To the extent that our theorizing about justice has been based on assumptions about human nature, and in particular on the assumption that there is (and will remain) one human nature that provides the basis for the moral equality of persons, the radical malleability of life through the application of genetic science presents yet another basic challenge to our thinking about justice.

Theories of justice (and of morality generally) have been thought of as based on conceptions of human nature. Indeed, advocates of rival theories of justice have typically criticized one another by trying to show that their opponents assumed an inaccurate account of human nature. In some instances, justice has been explicitly identified with acting according to nature or, more specifically, according to our (unchanging) human nature.

According to some theorists, including most famously Aristotle, human nature is simply rationality. According to others, there are other features besides rationality included in human nature. For example, it has been said by Hume among others, that it is human nature to be capable of only limited altruism--that according special priority to our own interests and those of our loved ones and close associates is a characteristic of all human beings, regardless of their culture or location in history. Indeed, proposals for how society ought to be structured are often criticized on the grounds that they fail to take seriously this alleged feature of human nature. Thus a standard criticism of more radical socialist visions of the good society is that they attribute too much altruism to human beings.

The assumption behind all such criticisms is that a conception of justice for human beings must at the very least reflect the motivational and cognitive limitations of human nature. Human nature is thus understood as a constraint on theorizing about justice. But if theorizing about justice begins with a conception of human nature as given, then it is hard to see how it can provide answers to such questions as: Ought we preserve human nature (as we have understood it thus far)? Or, if it is permissible to change "human nature," how should we change it?

Three Conceptions of the Relation of Human Nature to Ethics

The relevance of human nature to theorizing about justice and morality generally is a matter of dispute. There are three quite distinct, major views. Each gives a different sense to the common idea that morality must be based on human nature and that moral theory must somehow reflect this dependence.

First, there is the rather minimal and highly plausible view that human nature has to satisfy certain conditions if morality, and hence theorizing about morality, is to be possible at all. For example, for Aristotle and Kant, human beings must be free and rational if there is to be such a thing as morality (for human beings), and if theorizing about what morality requires is to have any point at all.

Second, there is the view, already sketched above and attributed to Hume, that even if some features of human nature make morality possible, that nature also constrains the content of morality by limiting the demands it can make on beings like us. (Thus, for example, if we are by nature creatures of only limited altruism, then morality cannot demand that we be wholly self-abnegating.)

A third, much more ambitious--and much less plausible--conception of the relevance of human nature to morality and moral theory is that we can derive the substance or content of morality from a proper understanding of human nature. Some readers of Aristotle have thought that human nature plays this role in his moral theory because they have assumed--probably mistakenly--that his notion of natural function is intended to yield substantive ethical principles. None of the three views takes seriously the possibility that the features of human nature in question might change, much less that human beings might come to have the power to change them.

The appeal of the first view is most evident: unless human beings are sufficiently rational that they can have a conception of what they ought to do as distinct from what they desire to do and unless they are motivationally capable of acting on this, it is hard to see how they can be thought of as subject to morality at all. The second and third views, however, are much more difficult to justify, though they have too often been assumed to be correct.

The great difficulty with the second view, of course, is that it runs the risk of mistaking limitations that are due to environmental and social influences for the constraints of our nature. As Marx frequently emphasized, even the most powerful theorists have tended to mistake the distinctive character of human beings in their particular social milieu for human nature (Marx (1844) in McLellan 1984). Indeed, the controversy between those who think that human beings by nature are egoistic and those who attribute more altruism to our species has been simply a matter of assertion and counterassertion, with neither side marshalling adequate empirical evidence to support its case. So even if it is true that human nature constrains morality, the question of what constraints it imposes is no more tractable than other questions concerning the relevant contributions of genes and the environment.

The third, most ambitious view of the relevance of human nature to morality and moral theory is even less supported than the second. Although we cannot hope to canvass the objections to the many versions of this view, it is fair to say that the dominant conclusion in ethical theory for at least the past century is that attempts to derive a comprehensive, substantive morality from human nature have failed. Let us then set aside the third, unpromising view of the relationship between human nature and morality, and concentrate on the first and second.

Genetic Causation, Freedom, and the Possibility of Morality