x Introduction

fairly particular conceptions of a good life. Margaret Olivia Little investi- gates how cosmetic surgery can make purchasers and practitioners become complicit with what she calls "suspect" conceptions of normality, while from a vastly different perspective and in a very different mode, Susan Bordo articulates the same concern in "Braveheart, Babe, and the Body: Contemporary Images of the Body." Bordo presented an early version of that essay at our second project meeting; she then published it her book, Twilight Zone: The Hidden Lfe of CuItural Images from Plato to 0.1. (Berkeley: University of California Press, 1997). We reprint that essay here with permission of the author and the University of California Press.

The volume's final papers broach the daunting question, What sort of people do we want to become? Gerald McKenny reviews some of the dominant Western philosophical conceptions of how medicine should use technology to respond to the vulnerability of human bodies. In so doing he shows the often underappreciated ethlcal significance of those responses. Like McKenny, Mary Wlnkler also undertakes her explora- tion in a rich historical context. Whlle Winkler has concerns (as does McKenny) about using biotechnologies to enhance our capacities and

traits, she does not argue for the blanket of any pven enhancement purpose. Rather, like the project as a whole, she tries to articulate as clearly as possible what is ethically at stake when we use new biote~hnolo~ies to achieve such purposes. She helps us ask the right questions.

If it weren't for LeRoy Walters's talk on enhancement in 1993 at The Hastings Center, the energy necessary to drive thinking about a project might not have been generated. If it weren't for Daniel Callahan's guidance and nudging, the grant application to the NEH would not have been finished. And if it weren't for the careful administrative work of Nicole Rozanski, this volume would not have made it to John Samples's office at the Georgetown University Press. I am indebted to those individuals, to the contributors to this volume, and to the following individuals who also made presentations and/or attended project meet- ings: W. French Anderson, Adrienne Asch, Erika Blacksher, Bette Crigger, Eve DeVaro, Lawrence Diller, Strachan Donnelley, Harold Edgar, Mark Hanson, Bruce Jennings, Peter Kramer, Sheldon Krimsky, Tracy Macdonald, Glen McGee, Ellen Moskowitz, Leigh Turner, Law- rence Vogel, LeRoy Walters, and Peter Whitehouse.

Is Better Always Good?

The Enhancement Project

Worry about enhancement? Why not worry instead about apple pie? Enhancement, after all, is somedung we seek for ourselves and think others should too. We praise individuals who exercise so that they will live longer, be thinner, and if not richer, at least happier. We applaud individuals who seek excellent schools to enhance their intellectual development. We praise parents who do everythmg they can to enhance their children's moral development. So why would anyone worry about , a new cosmetic surgery technique that promised to make us thinner? Why worry about a new psychopharmacological agent that promised to enhance concentration and performance in school? What about a new psychopharmacological or genetic technology that promised to make us kinder and gentler?'

The following essay begins to say why and when it will sometimes make sense to worry about the prospect of aiming new biotechnologies at the enhancement of human capacities and traits. When we began our two-year project funded by the National Endowment for the Hu- manit ie~,~ we hoped to articulate for policymakers what we called "a continuum of uses of 'enhancement technologies,' from those that promote shared values, to those that seem neither to promote nor threaten shared values, to those that threaten such values." That hope was misguided in a couple of ways. First, it failed to appreciate that the heterogeneity of the technologies and the number of problems surrounding their regulation make the idea of "a continuum" unrealistic. Second, the phrase "enhancement technologies" itself is potentially mis- leading. The phrase could be read to suggest that "enhancement techno- logies" are in a class different from, say, the class of "health technologies." But of course they are not. The same technology can be aimed a t

different purposes. A genetic technology that could increase muscle

2 Erik Parens

mass for the purpose of treating a patient with a degenerative muscle disease could also be used to enhance the ability of an athlete to compete at lifting weights.

In a word, we quickly discovered that our project's primary aim should be to help clear some of the conceptual ground. This purpose entailed not only trying to clarify the different ways in which the term enhancement is used, but trying to clarify some good reasons why anyone might worry about aiming new biotechnologies at the enhancement of human capacities and traits.

One of the things we learned is that to understand womes about enhancement, one needs to notice that the term enhancement is used in at least two different, albeit sometimes overlapping, sorts of conversa- tions-and for dfferent reasons. In the first sort of conversation, en- hancement is one pole of the treatment/enhancement distinction. It is used in conversations by people attempting to say what doctors, as doctors, should and shouldn't do or by people attempting to say what a just system of health insurance should and shouldn't provide. Ths conversation is often conducted, explicitly or implicitly, in terms of the proper goals of medicine.

In the second sort of conversation the concern is not primarily that doctors might provide an intervention that would undermine the proper goals of the profession. Rather, the concern is that anyone who provided the intervention would be undermining extramedical, social goals or would be exacerbating already existing social problems. The first half of this essay is devoted to enhancement as it appears in conversations about the goals of medicine; the second half is devoted to enhancement as it appears in conversations about what might be called the goals of society.

In the essay that follows I draw heavily on the work of the project participants, but do not claim that all would share my conclusions. In particular, some participants think the term enhancement is so freighted with erroneous assumptions and so ripe for abuse that we ought not even to use it. My sense is that if we didn't use enhancement, we would end up with another term with similar problems. Rather than attempt to come up with a term that is free of such problems, it is my view that we ought to begin with what we've got, and try to articulate as clearly as possible what the dangerous and problematic uses are. I elaborate such uses below, but invite the reader to consult the essays that follow. Indeed nearly every aspect of my overview is elaborated in at least one of these essays.

Is Better Always Good? The Enhancement Project 3

Enhancement and the Goals of Medicine

As mentioned above, the treatment/enhancement distinction is often used in the context of conversations about what falls within and what falls outside the proper goals of medicine. But as anyone who has participated in or observed such a conversation knows, there is no one universally accepted conception of the goals of medcine. The lack of such a consensus has much to do with the fact that there is no one universally accepted conception of what health is. And thus neither is there a universally accepted definition of what "going beyond health to enhancement" means.

Within the goals of medicine conversation, there is, in the starkest terms, a long-standing debate between those who view health as freedom from disease and those who, like the authors of the famous World Health Organization definition, view health as "a state of complete physical, mental, and social well-being." In Norman Daniels and James Sabin's terms, there is a long-standing debate between "hard-line" and "expansive" conceptions of health, and thus between "hard-line" and "expansive" conceptions of the goals of medicine in particular and of health care more generally.

The Treatment/Enhancement Distinction and the Normal Function Model Perhaps the most persuasive defender of the "hard-linen-or "normal

function'-view is Norman Daniels. On this view, "disease and disability are seen as departures from species-typical normal functional organiza- tion or fun~t ion in~ ."~ As Daniels puts it, "According to the normal function model, the central purpose of health care is to maintain, restore, or compensate for the restricted opportunity and loss of function caused by disease and disability. Successful health care restores people to the range of opportunities they would have had without the pathological condition or prevents further deterioration."' One of the roots of this view is the conviction that the primary aim of health care is to provide people with normal function so that they can have an "equal opportunityn to pursue their life plans.

The terms "normal" and "equal" can be a bit confusing here. At the heart of the normal function model is the view that health care ought to help people become "normaln-which is not to say uequal"-competi- tors. Crudely put, the normal function model accepts that people are unequally endowed with respect to traits and talents; it accepts that

4 Erik Parens

"by nature" individuals are not equal competitors. The normal function model insists, rather, that those unequal competitors are entitled to an equal opportunity to pursue their life plans within the limits set by those natural endowments. On this view, medicine's primary goal is to restore people to the normal function that disease and disability diminish and whlch is the necessary condtion for them to pursue their life plans.

Proceeding from such a conception of health, disease, and the goals of medicine (and health care), Daniels writes: "Characterizing medical need [as what has to be done to restore species-typical functioning] implies a contrast between medcal services that treat disease (or disabil- ity) conditions and uses that merely enhance human performance or appearancen6 (emphasis added).

There are at least two uses of the distinction between interventions that aim at treatment and interventions that aim at enhancement, be- tween interventions that aim at the restoration of species-typical function and enhancements that aim at something more. The primary use, and the one that motivates Daniels, is as a tool to articulate what just health care entails. On his account, a just and basic package of care would include treatments but exclude enhancements. A just system of national health care insurance, for example, would cover the former but not the latter.

The second use is as a tool in the fight against medicalization. That is, the normal function model helps to identify the proper domain of medicine s u c h that some forms of disease are beyond its proper reach. Daniels and Sabin introduce an example of what I mean in their essay, "Determining 'Medical Necessity' in Mental Health Practice."' They point out that many different kinds of shyness can produce dis-ease in this society. The normal function conception enables us to distinguish among such kinds: to distinguish, for example, between shyness that is caused by "illness" and hence deserves treatment, and shyness that is caused by "life" and which, while worthy of response, does not deserve the services of a health care system with limited resources. As Daniels and Sabin point out, in contrast to expansive models of the goals of medicine, the normal function model enables us to make a "moral distinction between [the] treatment of illness and [the] enhancement of disadvantageous personal capabilitiesn (p. 10).

According to the normal function model, "complete physical, men- tal, and social well-being" is beyond the proper domain of medicine. The ability to identify what is beyond medicine's proper domain is

Is Better Always Good? The Enhancement Project 5

enormously appealing to people who worry that too much is being brought within it. Insofar as the normal function model accepts that people are thrown into the world with different endowments, it can be a tool to fight medicalization; it can help us to remember that there are natural differences and characteristics that medicine ought not to be used to erase.

Problems with the Treatment/Enhancernent Distinction As do all distinctions and models, however, this versions of the

treatment/enhancement distinction and the normal function model has several problems-having to do with the intelligibility of the distinction and with the assumptions embedded in it, as well as with the uses to which it might be put by unreflective policymakers.

One of the first problems with Daniels's version of the treatment/ enhancement distinction is that it can be confusing: both interventions aimed at treating disease and ones aimed at enhancing human perfor- mance are improvements. That may be one reason why LeRoy Walters and Julie Palmer have chosen, instead of distinguishing between treat- ments and enhancements, to distinguish between health-related enhance- ments and nonhealth-related enhancement^.^ There are at least a couple of virtues to this approach. First, Walters and Palmer's distinction conveys the sense that both sorts of intervention are improvements over an existing condition: one is health related and the other isn't. Second, the category of health-related enhancement is large enough to accommodate treatment and prevention-a virtue for those who worry that "enhancements" aimed at preventing disease (such as vaccines) will be pointed to as a way to undermine altogether the notion that enhancement is a class worthy of special attention. The downside is that the new version of the treatment/enhancement distinction may obscure the fact that the health-related enhancement/nonhealth-related enhancement distinction carries very similar difficulties (for example, what is the difierence between health- and nonhealth-related traits?) and thus just postpones having to deal with them. While our group did not reach any consensus about this matter, it may be that rather than try to craft a single term such as health-related enhancement to encompass treatment and prevention, we should just concede that we need to add to the categories treatment (of disease) and enhancement a third: prevention (of disease)."

A second, widely discussed ~roblem with Daniels's account of the treatment/enhancement distinction is that it can appear to be arbitrary.

6 Erik Parens

To make that point, David B. Allen and Norm Fost offer the follow- ing scenario:

Johnny is a short eleven-year-old boy with documented growth-hormone deficiency resulting from a brain tumor. His parents are of average height. His predicted adult height without growth hormone (GH) treatment is approximately 160 cm (5 feet 3 inches). Billy is a short eleven-year-old boy with normal GH secretion according to current testing methods. However, his parents are extremely short, and he has a predicted adult height of 160 cm (5 feet 3 inches)."

Johnny's shortness is a function of disease and thus, on Daniels's account, descrvcs treatment. Billy, however, has a normal genotype, one that produces normal levels of GH. Thus Billy's shortness is not a function of disease, and on Daniels' account does not deserve treatment.

Whle Johnny and Billy are different with respect to GH secretion, they are similar in that both will suffer equally from being short in a culture that values tall stature. Thus one might ask, does the treatment/ enhancement distinction obscure our res~onsibilitv to res~ond to the

1 J 1

suffering of both-regardless of the fact that one has a disease and one is healthy? Assuming for the purposes of argument that GH would be equally effective in both cases, would we make a mistake if we said that giving GH to Johnny would be a treatment, but giving it to Billy would be an enhancement?

Whle Daniels acknowledges that this is a hard case, he argues that his normal function model remains the best alternative for those trying to articulate a basic package of health care. In the end, he reminds us-tha; his model assumes that different individuals have different capabilities and traits. The purpose of medicine is not to eliminate all differences. Rather, it is to restore people "to the range of capabilities they could be expected to have had without disease or disability" (p. 124), given their draw in the so-called natural lottery. Thus whle Johnny and Billy are a hard case for those who in general are committed to responding to suffering, treating Johnny and Billy as the same would produce a still larger problem. Treating them the same would entail undermining our fundamental commitment to preserving differences, to promoting the health of populations made up of people whose normal function takes different shapes. If we abandoned Daniels's account of the proper purposes of health care, he argues that we would have to accept the still more problematic aspiration to level all differences to the extent

Is Better Always Good? The Enhancement Project 7

that we can. At least for many who reject the aspiration to level such differences, Daniels's argument is persuasive.

There is another problem with the normal function version of the treatment/enhancement distinction, which Eric Juengst raises,'> for which there may not be as clear a response as there was to Allen and Fost's. The normal function account runs into conceptual trouble when it is applied to "a limitlessly beneficial personal enhancement like moral sensitivity, intellectual acumen, or social grace." Juengst points out that on Daniels's own account, the notion of species-typical functioning is not merely a statistical notion, but implies a theoretical account of the design of the organism (that describes the "natural functional organization of a typical member of the species"). Juengst suggests that-statisti- calh-it may be possible to draw out a spectrum of human psychosocial capacities, with an average middle term. Theoretically, however, i t is very difficult to know what species-typical moral sensitivity, intellectual acumen, or social grace is. Thus, the species-typical functioning account doesn't provide definitive guidance in those cases where we are talking about the prospect of enhancing such capacities. And thus even if one accepts that the treatment/enhancement distinction is not arbitrary when it comes to some physiological functions like heart rate or growth hormone secretion rate, it is not easy to know how far the normal function model can get us with psychosocial functions like moral sensitiv- ity or social grace.

There are not only problems with the intelligibility of the distinction itself. According to Anita Silvers, another problem with the distinction between treatment and enhancement is that it presupposes a notion of, and inadvertently valorizes, "the normal."13 She suggests that the usual deployment of this distinction presupposes that to "promote equality of opportunity we must create a system that restores inferior individuals to average competence." On her view, a commitment to equalizing opportu- nity through "normalizing the functionality of those who have hsabilities" invites coercive and costly practices. One can read Norman Daniels's account of the importance of species-typical functioning upon which her argument depends differently from the way she does, and still accept the seriousness of the concern about the inadvertent valorization of "the normal." Indeed, in fairness to Daniels, it should be said that he is committed to trying to secure for individuals a range of opportunities, not to replicating specific forms of function-and not instead of securing better compensatory measures.

8 Erik Parens

Nonetheless, it is important to grant the possibility that the treat- ment/enhancement distinction and the conception of normality upon which it depends could be used for coercive purposes. That conception of normality was in fact used, for example, when people who were post-polio were forced to use braces so that they could approximate "normal function"–rather than allowed to use (what in most cases would have been far more helpful) wheelchairs. But to grant h s possibility suggests that we need to be on guard against this pernicious use of the distinction; it does not foreclose using it altogether.

There are at least two more important that will attend any attempt to employ the treatment/enhancement distinction to de- scribe the proper domain of medical practice and/or insurance reim- bursement. The first is that any individual's or any group of individuals' attempt to articulate a distinction like the treatment/enhancement one will, like all distinctions, take on a life of its own-regardless of the care with whlch someone like Daniels lays it out. Whereas Daniels employs the distinction with a view to providing people with what they need (a basic package of care), David Frankford's fundamental wony is that it will be used to keep people from getting what they need.

Frankford suggests that "students of public policy have long known that policy is rarely implemented as f~rmulated."'~ The danger of a group such as ours malung a policy statement about any version of the treatment/enhancement distinction is that, to begin with, "a statement that formulates a treatment/enhancement distinction potentially makes that distinction 'real.' " The problem with such a distinction becoming "realn is that it will mesh all too well with our current discourses about health policy, which "stress technical efficiency, and technical efficiency as a means to increase the size of the overall pie: the greatest goodies for the greatest number-utilitarianism (but nodung like sophisticated hybrid consequentialist models that attempt to account for distributive concerns)." That is, a version of the treatment/enhancement distinction like Daniels articulated in the hopes that it might do work in a sophisti- cated consequentialist scheme might in fact be appropriated by others and used for purposes very different from those intended by him. No matter how much we hope that we can specify our concerns about distributive justice, and no matter how much we specify our understand- ing of the tentative and problematic nature of the terms treatment and enhancement, they will be wrenched from the context in which we have articulated them.

Is Better Always Good? The Enhancement Project 9

In spite of his profound reservations about the uses to which the distinction will be put, Frankford acknowledges that it is already part of our "intersubjective use." This notion leads him to suggest that the way to minimize the potential for abuse is to try to limit our employment of the distinction to our conversations with medical professionals. Differ- ent from insurers, who on his view will surely tear the distinction from its "ethical mooringsn and sweep it "into a sea of cost containment," he hopes that the distinction might be more thoughtfully used by medical professionals, who, as practitioners of "the art of the particular," attend to particularity and context. He concludes, "administration of a treat- ment/enhancement distinction in professional practice stands a much greater chance of being highly contextualized and incorporating all of cognitive, aesthetic, and ethical practical knowledge, than would administration of the distinction in the 'policy' world of contracts and health insurance."

Even if, in the age of managed care, one is skeptical about how clear the distinction is between insurers and medical professionals, Frankford's fundamental worry about the distinction being used to keep people from getting what they need will not go away. Whoever wants to wield the distinction needs to be committed to fighting the sorts of abuses he fears.

Last but not least, there is another practical difficulty with Daniels's view that the normal function version of the treatment/enhancement distinction can be used to help articulate a basic package of care. Assuming that one of the reasons we like the treatment/enhancement distinction is that it helps us to articulate such a package, it will be important to remember the following problem. The distinction does not square perfectly with current insurance practices that many of us take to be just nor will it square perfectly with what many of us would take to be a basic package of care. As Dan Brock points out, some treatments (for example, autologous bone marrow transplants for meta- static breast cancer) are not now covered by some insurance plans because they are deemed experimental and not cost effective." Further, much insurance does cover some services that are not treatmenb, such as abortion. Finally, on Brock's account, some enhancements are covered by insurance because they prevent disease (for example, vaccination "enhances" normal immune system function). Parenthetically, as I men- tioned above, it seems to me that it is not helpful to refer (as Brock does here) to vaccinations as enhancements; such interventions would

10 Erik Parens

fall directly into the prevention category if we could agree that we need a h r d category in addition to treatment and enhancement.

Regardless of where one comes out on the prevention question, it is clear that there is neither a perfect match between treatment and what insurance does or should pay for, nor between enhancement and what insurance does not and should not pay for. The absence of such a perfect match is a limitation for those who want to use the distinction as one part of their attempt to define a basic package of care.

In sum, like most distinctions, the treatment/enhancement distinc- tion is fraught with problems. If, however, we recognize these problems, then we can use the distinction as one way to begin conversations about what doctors should and shouldn't do and what just systems of health care should and shouldn't reimburse. But as the foregoing discussion of the conceptual and practical problems suggests, it would be a mistake to think that it will be possible in some straightforward manner to read off the distinction itself what we should and shouldn't do. There is a big difference between hoping that a given distinction can begin conversation, and thinlung it can end one.

In the current context, where there are financial incentives to provide fewer services, recopzing these problems should make us wary of decisions to refuse services based on the distinction. At the same time, critics of such cost savings should realize, as does Kathy Davis in her contribution to the collection,16 that a nuanced conception of the distinction could in principle be used to help us begin to identlfy what we owe to each other. It could be one tool used to say what is and isn't included in a basic package of care.

The Schmocter Problem So far I have said how the treatment/enhancement distinction tends

to be used by those who want to identify the proper goals of medicine- and ultimately the proper constituents of a basic package of health care. I have also identified several problems associated with that goals of medicine approach and the normal function model upon which it de- pends. But even if the normal function model and the treatment/ enhancement distinction were without limitations, even if we could clearly identify the important exceptions and caveats to the rule that medicine is only for the sake of treatment, we still would not yet have the theoretical resources we need to deal with concerns about what I have previously called the goals of society.

Is Better Always Good? The Enhancement Project 1 1

To appreciate this point, i m a p e for a moment a group of people who call themselves schrnocters, a term coined by my friend and former colleague James Lindemann Nelson. l 7 Schrnocters don't claim to practice medicine. They widely advertise that they practice schrnedicine. That is, they are expert in using new biotechnologies to enhance human capacities and traits, and they sell their expertise to willing, indeed, enthusiastic purchasers. Like some plastic surgeons today, these schrnocters of the future don't rely on insurance reimbursement to make a living. More than enough people are eager to buy their services. Thus, even if talk about the "goals of medicine" could dissuade doctors from providing some services on the grounds that they are enhancements, and even if insurers refused to reimburse "enhancement" services, there is no good reason to think that schrnocters would be dissuaded from providing those services. By definition, schmocters don't care about the goals of medi- cine, they care about the goals of schmedicine. The argument that appeals to the goals of medicine to shore up a prohibition of doctors pursuing enhancement would not suffice to prohibit schrnocters from pursuing the same.

Enhancement and the Goals of Society

Thus while the treatment/enhancement distinction makes sense in the context of the goals of medicine conversation, it does not make sense in another sort of conversation. And that is the sort of conversation that we will increasingly have to have: one about what we might call the goals of society. In this sort of conversation, one can't argue against a particular intervention on the grounds that it is not a treatment or not consistent with reasonable insurance practice. The problem is harder. In this sort of conversation, if one wants to claim that a given "improve- ment" will in