Personalized medicine (PM) holds great promise for improving the quality of healthcare. Promises, however, come with caveats. The successful development and practice of PM requires the establishment of large-scale genomic databases and the relevant biobanks (Hamburg & Collins 2010; Hewitt 2011). The construction and operation of these databases give rise to one of the central ethical worries that have been associated with PM, namely, the dangers of loss of “genetic privacy” – or in effect, “genomic privacy” (Newcombe 1994; Ursin 2008; Guttman & Wagner 2013; Chadwick 2013; Gostin 2017). The practice of PM hence gives rise to the dilemma of balancing the benefits to the public against (among other things) the risks to the individual patient of breaches of genomic privacy.
To solve dilemmas successfully, we need to be able to assign relative weights to the different considerations that make them up. To assess the significance (i.e. the importance, not the probability) of the risk of breach of genomic privacy, when making policy decisions regarding PM, we need a good understanding of what the feared potential drawbacks precisely are. Only thus can we understand the force of the constraint. Here, however, we face a hurdle, in the form of a persistent lack of clarity attached to the alleged “specialness” of genomic information, based on the unique status and characteristics of the genome, and in turn of a certain vagueness regarding the ethical considerations that accrue to genomic information. This alleged specialness has invoked a wave of discussions on “genetic exceptionalism”, to be followed by a backlash against that idea (Bernhardt et al. 1997; Murray 1997; Green & Botkin 2003; Everett 2004; Rothstein 2005; Goodman 2016). Although that debate has lost momentum, it has not been resolved – the latest evidence being a forthcoming target article in the American Journal of Bioethics advocating a middle ground in the form of what the authors term “genomic contextualism” (Garrison et al. forthcoming).
Although the alleged “specialness” or “exceptional status” can to a large extent be analyzed into specific parameters (unusual wealth of information, involvement of biological relatives, etc.) whose claim to specialness can then be assessed rationally, there remains a more elusive element that transcends those, and which has contributed to the perception of a “DNA mystique” (Nelkin & Lindee 1995). To overcome unnecessary mystifications and to assess clearly the nature and force of the ethical constraint the requirement of “genomic privacy” poses to the practice of PM, I will focus here on that elusive element; once diagnosed, I will propose a method to evaluate it.
To get there, however, we should first take a step back and briefly state what the right to informational privacy in general is meant and devised to protect. The moral reasons for this right can be divided into two general ones: (1) protecting subjects from harms secondary to misuse of divulged personal information about them, and (2) protecting the private sphere from certain forms of scrutiny, which as such (i.e. without any consequent harm to personal interests) threaten human dignity. Keeping this double function in mind can help us track – and then dispel – any “mystical” aspects of the genomic privacy requirement. Specifically, distinguishing considerations of harm, which are more tangibly definable, from considerations of human dignity, which are notoriously more arcane, and then asking the right questions to clarify the latter component, can help us think more clearly about ethical dilemmas concerning (the adequate protection of) genomic privacy.
In this brief exposition, I will forgo both the discussion of harms and of the elements of dignity that are not or are less unique or obscure, and that I want to sift out. I shall but mention that one of those latter elements is the fear of stigmatization consequent to exposed genetic information. Although stigmatization indeed violates dignity, it is best classified as a harmful consequence; in addition, it is contingent on having a stigmatized genetic endowment. In contrast, we will here focus on the injury to human dignity via breach of informational privacy as such, i.e. regardless of the content of the information divulged or of any harmful consequences. The claim I will advance is this: protecting informational privacy as such is protecting human dignity by preventing a sense of personal violation, as a function of exposure of what is intimate (what is intimate need not involve stigma; it may even be admired, only unfitting for exposure.) Here we finally arrive at what I take to be the deepest source of elusiveness surrounding the special or exceptional status to which the genomic privacy protection should supposedly respond. We now ask: to what extent, if at all, are breaches of genomic privacy uniquely prone to generating a sense of personal violation? And we similarly ask: is genomic information inherently intimate?
While the debate has not been framed in this precise terminology, I believe that it captures a deep element of the genetic exceptionalism spirit. That exceptionalist intuition involves the idea that DNA determines the essence of personal identity.
Consider the following illustrations: Walter Gilbert, the prominent Harvard biologist, opined that understanding human genetic structure is “the ultimate answer to the commandment ‘Know thyself!’” (Miller 1998: 180); the title of Carl Cranor’s (1994) important anthology is telling in this respect: Are Genes Us? Dorothy Nelkin and Susan Lindee call DNA “the ultimate identifier” (1995: 47) and opine that DNA has taken on in our culture the role of “the Biblical soul.” This last view is suggested by no other than Nobel laureate Francis Crick in his book The Astonishing Hypothesis: The Scientific Search for the Soul (1994). And Crick’s partner, Nobel laureate James Watson asserts: “Now we know, in large measure, our fate is in our genes” (Weiner 1994: 31).
The idea of the genome as the essence of selfhood, alongside the ensuing idea that “genetic information is both potentially embarrassing and uniquely personal” (Annas 1993: 106) naturally entails an adamant requirement of genomic privacy, so as to protect “the most intimate and personal information that will ever exist” (Goerl, Hyer, and Farkas 1997: 86).
Can this view legitimately ground a strong genomic privacy requirement? A negative answer, I will argue, relies on three claims: (1) most obviously: it is not true that genome equals personal identity; (2) even if it did: personal identity per se is not the parameter directly relevant to moral considerations of privacy; rather, the sense of identity is what counts; (3) not even every exposure involving elements of one’s sense of identity correlates significantly with a sense of personal violation – which is what privacy requirements seek to avoid. After explaining (1)-(3), I will proceed to provide a sketch of how the question regarding a “sense of personal violation” is to be approached and investigated directly.
- The identification of one’s genome with one’s identity both clashes with commonsensical intuitions, such as that identical twins are not identical persons, and is untenable given our knowledge of the effects of epigenetic, environmental, and cultural factors on one’s character and traits. It might be countered that the idea is not that genes as such are literally “us” but rather that the genome is symbolically expressive of who we are (Hellman 2003). Whether true or not, this more nuanced idea shows, however, that personal identity is constructed via the hermeneutics of identity, which shifts the weight to the next parameter, focusing on the sense of identity.
- Given the realization that personal identity is partly a function of a sense of identity (metaphysical disagreements about the “true” determinants of personal identity are in an important sense wrongheaded), any imperative to protect personal identity must respond to the sense of identity. One’s sense of identity, however, never fully overlaps with a more objective, 3rd-person perspective on personal identity, yet a person’s privacy interests – on which we focus – correlate with the former, not the latter.
- Not every element in one’s first-person sense of identity is something one cares much or even at all to protect from scrutiny; in fact, there are features of one’s sense of identity with respect to which one seeks exposure (one who self-identifies as a singer, for instance, may be happy that video clips of her musical performances are watched by as many as possible). Hence the protection sought in privacy rights does not correlate well enough with exposure of elements of one’s sense of personal identity but rather with those elements whose exposure risks creating a sense of personal violation.
We must conclude, then, that to assess the protection-of-identity rationale for genomic privacy, we must probe the relevant “sense of personal violation” directly.
How can such an investigation be carried out fruitfully? Importantly, injury to human dignity by means of evoking a sense of personal violation refers not primarily to a theoretical construct but to an authentic human experience. Hence, investigating it requires turning to experimental moral philosophy (experimental philosophy has to date not examined the basic moral concept of human dignity.) Accordingly, with the sense of personal violation in mind, I will briefly sketch what form an experimental approach to investigating the significance of genomic privacy might take.[1]
The basic approach, I believe, should be a traditional questionnaire one, where people report their sentiments regarding the possibility of public exposure of their de-identified genome, as compared to their sentiments toward exposure of benchmark paradigmatic (de-identified) elements of personal intimacy. Revealing questions will compare the sense of personal violation due to exposure of (de-identified) genomic information to the sense of personal violation with respect to the following.
- Publication of de-identified personal narrative (this often happens when therapists write books and papers describing de-identified stories of their patients).
- Exposure of de-identified personal thoughts (think for instance of the use of fMRI lie-detectors).
- Publication of de-identified pictures of one’s exposed body (think for instance of the use of patient’s pictures in dermatology).
- Public imitations of one’s de-identified voice, behavioral patterns, and the like (as e.g. sometimes happens for advertising purposes).
- Public exposure of de-identified intimate personal objects (e.g. toiletry, undergarments, even fridge contents).
Such questions should address the projected sense of personal violation also irrespective of the risk of (re)-identification, i.e. also in case de-identification or anonymity are maintained (a person may understand that no one knows it is him exposed out there, yet his knowing this may be enough for a sense of personal violation).
The kind of research proposed here relies on a philosophical perspective that can be but hinted at here without elaboration, viz. that the idea of human dignity is neither strictly theoretical nor metaphysical – it is rather a function of personal attitudes and sentiments, and should be constructed bottom-up, through (among other things) people’s experiences of a sense of personal violation.
I should stress that the suggested empirical investigation cannot practically-speaking be circumvented, by simply procuring informed consent for participation in large-scale genomic biobanks, since requiring real informed consent for such participation may well undermine the entire project (Knoppers 2005; Manson & O’Neill 2007). Furthermore, the empirical research suggested will itself indicate the required level of stringency of any suggested policy of informed consent for genomic biobanks.
Our discussion was motivated by the fact that genomic biobanks are crucial for the successful implementation of PM. The analysis undertaken and suggested here should help determine the relative moral weight of the genomic privacy constraint in devising policies regarding participation in genomic biobanks.
BIBLIOGRAPHY
Annas, G. (1993). “Privacy rules for DNA databanks”. Journal of the American Medical Association, 270(17), 2346-2350. (Reprinted in Penelope Barker (Ed.), Genetics and society (pp. 100-112). New York: H.W. Wilson Company, 1995.)
Bernhardt BA, Geller G, Strauss M, Helzlsouer KJ, Stefanek M, Wilcox PM, et al. (1997). “Toward a model informed consent process for BRCA1 testing: a qualitative assessment of women's attitudes”. J Genet Couns. 6: 207-22.
Chadwick, Ruth (2013). “Ethical issues in personalized medicine”. Drug Discovery Today: Therapeutic Strategies 10.4: e171-e174.
Cranor, Carl (ed.) (1994). Are genes us? The social consequences of the new genetics. New Brunswick, NJ: Rutgers University Press.
Crick, Frederik (1994). The Astonishing Hypothesis: The Scientific Search for the Soul. New York, NY: Touchstone.
Everett, Margaret (2004). “Can you keep a (genetic) secret? The genetic privacy movement”. Journal of Genetic Counseling 13.4: 273-291.
Farkas, Daniel H., Hans S. Goerl, and Randall N. Hyer (1997). “Genetic privacy legislation: Two views”. Molecular Diagnosis 2.1: 83-87.
Garrison et al. Forthcoming. “Genomic Contextualism: Shifting the Rhetoric of Genetic Exceptionalism”, American Journal of Bioethics.
Goodman, Bryce (2016). “What’s Wrong with the Right to Genetic Privacy: Beyond Exceptionalism, Parochialism and Adventitious Ethics” in Brent Daniel Mittelstadt and Luciano Floridi (eds.) The ethics of biomedical big data. Dordrecht: Springer, 139-167.
Gostin, Lawrence O. (2017). “Genetic privacy”. Genetics and Gene Therapy. Routledge, 241-251.
Green, Michael J., and Jeffrey R. Botkin (2003). “Genetic exceptionalism in medicine: clarifying the differences between genetic and nongenetic tests”. Annals of Internal Medicine 138.7: 571-575.
Gutmann, Amy and James W. Wagner (2013). “Found Your DNA on the Web: Reconciling Privacy and Progress”, Hastings Center Report 43: 15-18.
Hamburg, Margaret A., and Francis S. Collins (2010). “The path to personalized medicine”. New England Journal of Medicine 363.4: 301-304.
Hellman, Deborah (2003). “What Makes Genetic Discrimination Exceptional?”, American Journal of Law and Medicine 29: 77-116.
Hewitt, Robert E. (2011). “Biobanking: the foundation of personalized medicine”. Current opinion in oncology 23.1: 112-119.
Knoppers B (2005). “Consent revisited: Points to consider”. Health Law Rev 13: 33-38.
Manson, Neil C., and Onora O'Neill (2007). Rethinking informed consent in bioethics. Cambridge University Press.
Miller, H. (1998). “DNA blueprints, personhood, and genetic privacy”. Health Matrix: J Law Med, 8(2), 179-221.
Murray, T.H. (1997). “Genetic Exceptionalism and ‘Future Diaries’: Is Genetic Information Different from Other Medical Information?” in Genetic Secrets: Protecting Privacy and Confidentiality in the Genetic Era, M.A. Rothstein (ed.), New Haven, Conn.: Yale University Press.
Nelkin, Dorothy & M. Susan Lindee (1995). The DNA Mystique: The Gene as a Cultural Icon. Ann Arbor, MI: University of Michigan Press.
Rothstein, Mark A. (2005). “Genetic exceptionalism & legislative pragmatism”. Hastings Center Report 35.4: 27-33.
Ursin, Lars Øystein (2008). “Biobank research and the right to privacy”. Theoretical medicine and bioethics 29.4: 267-285.
Weiner, C. (1994). Anticipating the consequences of genetic engineering: Past, present, and future. In Carl Cranor (ed.), Are genes us? The social consequences of the new genetics (pp. 31-51). New Brunswick, NJ: Rutgers University Press.
[1] This paper will offer theoretical background and rationale for another, experimental philosophy paper.