DNA and distrust.
Over the past three decades, government regulation and funding of DNA testing has reshaped the use of genetic evidence across various fields, including criminal law, family law, and employment law. Courts have struggled with questions of when and whether to treat genetic evidence as implicating individual rights, policy trade-offs, or federalism problems. We identify two modes of genetic testing: identification testing, used to establish a person's identity, and predictive testing, which seeks to predict outcomes for a person. Judges and lawmakers have often drawn a bright line at predictive testing, while allowing uninhibited identification testing. The U.S. Supreme Court in Maryland v. King, for example, held that entering arrestee DNA in databanks does not implicate substantial Fourth Amendment concerns, since police do not test for genetic predispositions "not relevant to identity. " We argue that policy implications of genetic testing laws cannot be so neatly demarcated. For example, federal welfare laws require states to use DNA to establish paternity to collect child support from "deadbeat dads, " which may be relevant to identity, but also creates potentially destabilizing effects on families. We explore how genetic testing has been regulated across a variety of fields. We identify two dominant modes of regulatory action dealing with genetics: data-driven and ethics-based. Data-driven legislation is ostensibly focused on short-term benefits of gathering a population's genetic information. Ethics-based legislation, in contrast, is concerned with long-term consequences, such as effects on privacy. We particularly critique data-driven legislation, and we argue that judges, legislators, and scholars should focus squarely on the individual and government interests at stake. We set out a list of five factors that legal actors should consider when considering genetics regulation: (1) equality, (2) accuracy, (3) privacy, (4) finality, and (5) federalism. In particular, equality concerns permeate the short history of DNA regulation. In each of the areas explored, comparatively disadvantaged groups such as arrestees, convicts, juveniles, noncitizens, and welfare recipients have received the most intrusive regulation and collection of their genetic evidence, while comparatively privileged persons benefit from enhanced genetic privacy. We conclude that the regulation of genetic evidence deserves far more careful legal scrutiny, since the ways that genetic evidence is deployed can profoundly affect constitutional rights and the structure of legal and social institutions.
DNA testing, widely available for over twenty-five years, has revolutionized the way local, state, and federal governments understand identity by making it inexpensive to obtain a person's genetic profile and link people to biological evidence and to each other. With the benefit of different types of DNA testing, the state can now say with greater certainty whether a particular suspect was the culprit of a crime or whether a particular person is the biological parent of a child. DNA testing has been embraced with enthusiasm by courts, legislatures, and agencies, state and federal, across areas of law ranging from criminal law, employment law, family law, and health law because it is easy to obtain and offers apparent certainty. This Article critically assesses these developments, focusing on the seemingly unobtrusive collection of genetic data, and argues that heightened legal scrutiny of genetic regulation is needed.
As with any new technology, genetic testing has captured the imagination of scholars. Early on, some scholars predicted that genetic tests would supplant traditional legal tests. Many legal standards, particularly constitutional tests, are broad and vague, while DNA tests have the appealing ability to seemingly make evidentiary determinations certain. In 1992, Rochelle Cooper Dreyfuss and Dorothy Nelkin called the trend to reduce questions to genetics "genetics essentialism." (1) More recently, scholars have asked whether there is a "genetics exceptionalism" in which policymakers over-privilege the importance or uses of genetic evidence. (2) Still others have predicted that
DNA testing will reinvigorate the eugenics movement, as imagined in films such as Gattaca. (3)
"Genetic testing," however, is not a legal subject. As many scholars have observed about other new areas of law, such as "cyberlaw," the law of genetic testing is a version of "the law of the horse." (4) It is not a unified field but the analysis of a particular technology in relation to a cluster of existing legal fields. DNA testing has not led to universal genetics essentialism, genetics exceptionalism, or a new eugenics; instead, its uses have been inconsistent and variable depending on the social and legal context in which DNA testing is adopted. Rather than a story of genetics overtaking the law and rendering complex legal questions deceptively simple, DNA testing has itself been caught up in preexisting regulatory relationships between federal and state and local governments, privacy advocates and scientific researchers, law enforcement and social services, each with competing goals. And, as Gaia Bernstein has astutely observed, "technological innovation can both create and bring to the forefront legal values that for years lurked in the shadows of legal discourse." (5)
These preexisting regulatory contexts often result in legal structures that reflect the concerns and perspectives of elites at the expense of other members of the political community. The amount of attention given to genetic privacy, we argue, depends in large part on the context in which a genetic regulation is initially conceived. Equality concerns permeate the short history of DNA regulation. In each of the areas we explore in this Article, comparatively disadvantaged groups such as arrestees, convicts, juveniles, noncitizens, and welfare recipients, have received the most intrusive regulation and collection of their genetic evidence. In contrast, more privileged persons are not subjected to government DNA collection and may instead benefit from legislation protecting their genetic privacy.
As John Hart Ely famously argued in his classic book, Democracy and Distrust: A Theory of Judicial Review, the legislative process, through its enactment of the majority will, can lead to discrimination against the politically powerless and block the channels of political change. (6) According to Ely, when courts attempt to intervene, they often mistake specific problems "as isolated pockets of concern," slotting them into "familiar pigeonholes." (7) These seemingly separate issues, however, may form a "unity," or a larger problem in which "uncontrollable discretion" risks harm to undervalued individual and minority rights. (8)
Such, we argue, is the story of genetic regulation. Just as Professor Ely observed of the death penalty that "the system is constructed so that 'people like us' run no realistic risk of such punishment," (9) in genetic regulation, legislatures and judges have been far more careful and forward-thinking in instances where they, or people of their social class, are likely to be subject to regulation. This has led to some "pigeonholes" of genetic regulation being treated very differently from others.
Consider, for example, a case decided using the "pigeonhole" of Fourth Amendment privacy rights in criminal law. In Maryland v. King, the U.S. Supreme Court was faced with the question of whether taking and analyzing a cheek swab of an arrestee's DNA is "like fingerprinting and photographing, a legitimate police booking procedure." (10) In his majority opinion, Justice Kennedy concluded that it was constitutionally reasonable for the state to take DNA evidence because of the "negligible" physical intrusion and the limited use of determining whether the individual was associated with the crime scene or victim. (11) A DNA test, according to the Court, was simply not very different from a fingerprint or a photograph. The Court understood the issue to be simple expansion of Fourth Amendment jurisprudence, not a new foray into issues of genetic privacy.
Even the majority opinion, however, revealed, perhaps unintentionally, that cabining off DNA evidence into the criminal law pigeonhole might not be as simple as it sounds. Although the opinion emphasized how current DNA testing of arrestees examines "noncoding parts of the DNA that do not reveal the genetic traits of the arrestee," it also acknowledged that "science can always progress further, and those progressions may have Fourth Amendment consequences," particularly if they seek to determine "an arrestee's predisposition for a particular disease or other hereditary factors not relevant to identity." (12) This application of DNA testing may not be so different from predicting future behavior or outcomes, when DNA databases are used both in the present, but also to potentially link a person to future crimes. (13) The line between identification testing and predictive testing is not always so clear.
Federal lawmakers clearly believe such bright lines can be drawn. For example, Congress passed the Genetic Information Nondiscrimination Act of 2008 (GINA), which prohibits changing the terms of health insurance coverage based on genetic information and forbids employment discrimination based on genetic information. (14) As one federal court has put it, "[t]he basic intent of GINA is to prohibit employers from making a 'predictive assessment concerning an individual's propensity to get an inheritable genetic disease or disorder based on the occurrence of an inheritable disease or disorder in [a] family member.'" (15) In contrast, few such privacy protections exist in the welfare context, in which Congress encourages states to conduct potentially intrusive paternity testing to promote child support collection. (16)
We argue that drawing a bright line between uses of genetics to identify and to predict is, at best, useful only as a starting place forjudges, legislators, and policymakers. A court should not find a use of genetics "not relevant to identity," and stop there, as the Supreme Court largely did in King. In this Article, we examine whether a clearer and more consistent framework for answering such questions is possible or desirable. Our framework makes the political process problems in current DNA regulation transparent and provides guidance to inform judicial review to intervene where important political constituencies are unrepresented.
This Article takes a broad look at how DNA has altered the legal landscape by canvasing several disparate areas to identify common themes and common questions. (17) We analyze how legislatures have approached genetic testing, and use our analysis of these approaches to introduce new typologies that will help categorize and illuminate the regulatory choices genetic testing invites. In particular, we identify two modes of genetic testing: identification testing, which focuses on establishing a person's identity, and predictive testing, which focuses on predicting outcomes for a person based on his or her genetic code. These two forms of testing have tended to result in different modes of regulation.
Next, we identify the two dominant modes of regulatory action in the genetics arena: data-driven and ethics-based. By "regulation," we broadly refer to legal uses of genetic information, not only regulations promulgated by administrative agencies, but also statutes enacted by legislatures, as well as judicial decisions and the exercise of executive discretion.
In general, we argue, data-driven regulation tends to concentrate on the short-term benefits of developing comprehensive information about a population's genetic information; ethics-based regulation, in contrast, is centrally concerned with the long-term consequences of using DNA, such as its effects on privacy or individual rights. In addition, data-driven regulation is often enacted or approved of with little or no discussion of its effect on genetic privacy; instead, the focus is often on some other, more immediate legislative goal, such as collecting child support payments or solving a crime. Ethics-based regulation, in contrast, tends to be made with great deliberation and with an understanding that individuals have an interest in their own genetic privacy.
We argue that where a particular approach to DNA testing falls within this schema depends in large part on the area of law at issue, the political power of the constituents likely to be affected by the law, and the legal and social context in which the regulation arises. Data-driven approaches, we find, are more likely than ethics-based approaches to reflect a flaw in the political process that makes the population most affected by the regulation the least heard. In addition, we find a strong correlation between identification testing and data-driven regulation, and a similar connection between predictive testing and ethics-based regulation. This connection, at least so far, is likely the result of whether a crisis exists at the time a law is passed. Data-driven laws are often passed in response to a perceived emergency--the threat of millions of "deadbeat dads" failing to pay child support or a particular rape or murder that could have been solved had the perpetrator's DNA been available in a databank, for example. But, we argue, as time goes on, there is a danger that the rules applied to identification testing may be expanded to cover predictive testing, and these rules, depending on the legal "pigeonhole" in which they originated, may offer little or no protection to individual genetic privacy.
The Article proceeds as follows. In Part I, we introduce the typology described above, first by describing the different types of DNA testing that can be conducted currently, both to identify personal genetic data and to predict outcomes, and then introducing how identity is litigated in different legal settings resulting in policy choices that may or may not directly regulate genetic evidence directly or intentionally. We suggest that one must ask (1) whether genetic regulation is predictive or for purposes of identity and (2) whether the regulation is data-driven or ethics-based regulation.
In Part II, we contrast genetic regulation in three fields: (1) employment and health law, where with the passage of GINA, Congress focused on direct regulation of genetics with a focus on ethics concerns; (2) criminal law, in which the Supreme Court has ratified the data-driven and direct regulation of genetics, through a federal project of assembling a vast national DNA databank, but in which legislators, law enforcement, and the Court have adopted a hands-off approach towards questions of ethics, neglecting privacy questions and reluctantly permitting DNA testing that might free the innocent; and (3) family law, in which the federal government has required states to focus on establishing paternity, incidentally regulating genetic evidence as part of accomplishing that data-driven concern, and to the exclusion of other important ethical concerns of real importance to state actors.
Part III considers the commonalities of these areas. When legal actors focus on collection of genetics data, they may do so directly and assess whether there are ethical concerns with doing so; we view that as preferable to indirectly regulating genetics while treating ethical concerns as not implicated. We set out a list of five factors that legal actors should consider when examining genetics regulation: (1) equality, (2) accuracy, (3) privacy, (4) finality, and (5) federalism. We conclude by considering what the inconsistencies in regulating genetics suggest about our commitments to using information about genetic identity to regulate. We argue that far more legal and policy scrutiny is required to justify broad government collection and testing of genetic evidence.
I. DNA TESTING, IDENTITY, AND REGULATION
DNA testing, like blood-typing, fingerprinting, or any other means of biometric identification, is a scientific method for identifying personal characteristics of a living being. Unlike some of these earlier forms of biological testing, however, DNA testing provides the opportunity to both identify a particular person with near certainty and to predict characteristics as well as propensities that may be associated with that person's genetic makeup. The nuclear DNA of a person contains twenty-three pairs of chromosomes, and that DNA sequence is found in almost all of our cells. That nuclear DNA contains genes that provide important instructions for the functioning of each cell, but it acts along with other proteins, enzymes and ribonucleic acid (RNA), in complex ways that are still being studied. This genetic code can be used for many purposes. It can, for example, constitute proof that a particular person's genetic material was deposited, through blood, semen, or hair, in a particular place.
Genetic information is also shared; one's genetics are inherited from one's biological parents. (18) Therefore, genetic tests can show that a particular person is related to another and even provide proof of ancestry, as the descendants of Thomas Jefferson and Sally Hemings have discovered. (19) DNA can also be used to show propensities--a propensity to have a particular kind of cancer or perhaps even a propensity to commit a crime. Test results may be relevant not just to an individual, but also important to other blood relatives, if the tests indicate that a person inherited a genetic condition.
Standing alone, a DNA test may mean little. Scientific data require interpretation to be understood, and not just by the laboratory analyst that conducts the testing and reports the results, but by medical, legal, or government actors that seek to use that information. Perhaps even more importantly, scientific data, like any form of evidence, require that the actors using that data have a theory of why the data are probative. But in order to understand the limitations of and interpretative choices inherent in the use of DNA, we must first understand, in its basic outlines, the current science of DNA testing.
These results have occurred because DNA tests are meaningless without interpretation. The interpreters of DNA's meaning--forensic analysts, judges, legislatures, prosecutors, police, and parents--are embedded within existing legal institutions with various regulatory goals and purposes. As Dorothy Roberts has put it well: "There is nothing either precious or sinister about the genetic tie by itself. The genetic tie's precise social import depends on the type of relationship to which it becomes relevant and the prevalent social conditions that influence that relationship." (20) The availability of DNA has shaped these legal institutions, which simultaneously shape how DNA evidence is used and is itself regulated.
This feedback loop means that the interaction between DNA and legal change has differed markedly across various contexts. For example, in contrast to the narrow focus in Fourth Amendment challenges to DNA testing in criminal cases, in a family law case in which a putative father also brought a Fourth Amendment challenge to a DNA test, a federal judge emphasized broader policy concerns in finding: "DNA testing for paternity is not unreasonably invasive in light of the compelling state interest to protect the welfare of children." (21)
A. Types of DNA Testing
There is not just one type of DNA test. There are many forms of genetic testing that examine different portions of the genetic sequence, making different comparisons, and for different reasons. Two general types of DNA tests, broadly speaking, are designed to identify specific individuals, as opposed to those designed to test for other genetic information, such as hereditary diseases or other health risks.
1. Identification Testing
In criminal cases, DNA tests are (currently) of the first type, used to try to identify individuals. The purpose of such DNA tests is to identify a culprit by comparing a person's genetic material to trace evidence left at a crime scene. DNA material can remain intact for years or even decades in good conditions or if samples are stored properly, but humidity and exposure to bacteria, heat, or chemicals can cause DNA to degrade quickly. Most criminal cases do not have biological material left at a crime scene by a perpetrator; such testing is particularly useful in sexual assault cases or in cases where the culprit left a piece of clothing or was known to have touched an object. Before DNA testing became common, ABO blood typing could sometimes be done in criminal cases, and it could readily exclude a suspect; however, since large portions of the population share each of the four blood types, it was not very probative evidence that a person had the same type as evidence tested from a crime scene. (22)
DNA testing can provide highly probative evidence when there is relevant material that can be tested. While nuclear DNA contains many genes that do important work providing instructions for how to encode proteins, there are "noncoding" regions of the genetic sequence; in fact the vast majority of the sequences is non-coding, sometimes called "junk DNA," and at minimum, its purpose is not known. (23) In the 1980s, geneticists identified particular noncoding regions of the DNA sequence that vary a great deal from person to person; the more variable the region of the DNA is the more useful it is to differentiate between people. In certain portions, there are quite variable repetitive sequences in which the same pattern of alleles (sequences of DNA at a particular position or locus) is repeated over and over. (24) Beginning in the 1980s, scientists developed techniques to isolate those repetitive alleles or regions of the DNA strand. Early Restriction Fragment Length Polymorphism (RFLP) testing required a large quantity of nondegraded genetic material, and interpretation of the results was potentially subjective. (25)
By the mid-1990s, however, Short Tandem Repeat (STR) testing could be conducted, and unlike RFLP testing, it can be performed on very small samples. During the 1990s, polymerase chain reaction (PCR) methods permitted scientists to amplify small samples for STR analysis by making billions of copies of the sample, and then new capillary electrophoresis technology permitted rapid and largely computerized analysis of genetic material extracted from samples. (26) Those tests are now standardized, and in the United States, analysts now compare thirteen loci, or portions of a person's genome, and examine the number of repetitions of particular alleles at each loci (soon to be expanded to include twenty loci). (27) A person's DNA profile, as used in a criminal case, consists of thirteen pairs of numbers (one for each chromosome), with each number representing the number of "repeats" at each loci.
The results of that DNA testing do not provide a "match," even if all thirteen pairs consist in the same numbers, but rather a probability that genetic material could be expected to randomly match. Using population statistics, scientists can determine whether one person in many millions, or even billions or trillions could be expected to randomly match a particular DNA profile. One area of regulation, or lack thereof, lies simply with the calculation of those population statistics. In criminal cases at least, analysts rely on allelic tables prepared by the FBI. The FBI has been unwilling to share the underlying population data with researchers, who have criticized this unwillingness, and in 2015, it emerged that there were a series of errors in those tables, and those errors have resulted in thousands of notifications being sent about criminal cases in which the calculations may have been erroneous. (28)
Other DNA tests sometimes used in criminal cases are less probative; DNA testing can be done on the Y chromosome (Y-STR testing), but all members of a paternal lineage would share the same results, and mitochondrial DNA tests, useful when hairs are found at a crime scene, also provide weaker findings since mitochondrial DNA is matrilineally inherited. (29)
In family law, DNA tests are also of the first broad type and are used to identify individuals, but specifically to examine whether they have a familial relationship with each other. Genetic testing is typically performed using different types of DNA tests that involve a far wider array of genetic markers than the thirteen loci used in criminal cases, comparing the DNA profile of a child with that of the possible parent, with results that are 99.99% accurate. (30) A paternity test, for example, can cost as little as a few hundred dollars and as much as two thousand dollars and can be conducted prenatally (with lower cost "curiosity testing" and various companies offering tests for under $100). (31)
In both family law and criminal law, the goal is to identify a person using a genetic test. DNA testing was not such a fundamentally new technology that the law had to radically shift to take account of it. Less precise blood typing could be used in the past, and in criminal cases, there may be a wide range of evidence connecting a person to a crime; for example, many criminals are caught in the act. In criminal law, identity matters in the sense that in some criminal cases (though not crimes involving non-strangers), there is a question who committed the crime. In family law, though, the identity of a biological parent may be relevant information, but sometimes the biological tie may not be so relevant after all, depending on, for example, who has formed a relationship with the child, who has taken on child-care obligations, whether there has been an adoption, and a range of other factors. The biological link standing alone is not at all dispositive as to child custody, and therefore, the advent of more precise modern DNA testing would not be expected to transform family law.
2. Predictive Testing
A second broad category of genetic testing examines a person's DNA, not to identify that person, but in order make predictions about the person's likely health or medical predispositions or risks, drug sensitivity, or perhaps even the person's future behavior. (32) The term phenotypingrefers to the identification of characteristics of a person that can be observed as having derived from their genetic traits, in interaction with the environment. Our focus in this Article is on genetic identity used to identify the relevant person and not in the sense of identifying a person's phenotype or genetic characteristics. Some of those genetic characteristics may be readily identifiable. For example, all fifty states require genetic testing for newborn infants, using DNA, protein, or metabolite analyses to screen for genetic disorders such as hypothyroidism, phenylketonuria (PKU), and sickle-cell anemia. (33) These tests are conducted on blood samples in order to identify and treat conditions early on. Legal challenges regarding the collection, storage, and retention of these blood samples have been rejected, although the Minnesota Supreme Court found consent must be obtained to use those samples for research. (34)
There are currently genetic tests available for over 1,000 different diseases, but as National Institute of Health Director Francis Collins puts it well, "the field is moving so quickly that any genetic risk predictions based on today's understanding will need to be revised in the context of new discoveries tomorrow." (35) Of course, family health history has long been used to assess risk factors for a range of medical conditions. Given the increasingly low cost of genetic testing, more sophisticated efforts to engage in "personalized medicine," both to predict and treat conditions, are increasingly possible. (36) However, for a wide range of common medical conditions and behavioral diagnoses, genetics may play a part, but so may environmental factors, and the precise role of genetics is the subject of intensive research. (37)
B. Regulation of Genetics
Identification and predictive testing have been treated quite differently by the law. In addition to the distinction between identification and predictive uses of DNA, we will also explore an important distinction in how legislatures operate when they are regulating genetic testing. Sometimes, a legislature is intensely concerned with the morality and long-term consequences of using genetic information. We call this "ethics-based regulation." By ethics, we broadly refer to a family of normative, moral, and individual-rights concerns, including privacy, equality, family and child welfare, dignity, and procedural due process concerns. Such ethical concerns have prevailed, we argue, where the long-term consequences are likely to be visited on wide swaths of population across socio-economic groups. Predictive uses of genetics may also make ethical issues salient, although we do not at all view it as necessarily the case that identity-focused uses raise fewer ethical concerns. In contrast, legislatures sometimes act in what we call a "data-driven" fashion. In data-driven regulation, the primary goal is the collection of genetic information to identify and track individuals. This form of regulation tends to be more concerned about short-term gains (often political ones) without attending to the long-term ethical problems that might arise. Data-driven regulation also tends to focus more on genetic identity, a narrower band of information about a person, than it does other types of genetic information. In the long run, data-driven efforts can lead unintentionally to breaches of privacy when information intended for one purpose is mined for another. Because data-driven regulation often bypasses ethical issues, it often disproportionately affects those who do not have political capital or an adequate voice in the conversation at the outset.
In fact, data-driven regulation can often be incidental, in that lawmakers or policymakers may not even be aware that they are regulating genetics. Instead, they think they are regulating something else--welfare law, for example, or immigration law. (38) In these instances, genetic information is one piece of a larger legislative goal lawmakers assume that a genetic tie stands in for another kind of relationship, for example, that a genetic father should automatically be a legal father, or a genetic parent should automatically confer citizenship on her foreign-born child. Here, the problem is less the long-term consequences of data-gathering (although it can be) but rather that the initial, unintentional genetics legislation skips the important step of thoughtfully considering what it means to be engaged in genetics regulation. As with data-driven regulation generally, incidental regulation can also have the drawback of further disadvantaging the already disadvantaged.
II. COMPARING APPROACHES TOWARDS GENETIC IDENTITY
We will now take a close look at the path of genetic regulation in three areas in which the law has responded to genetic testing: (1) health and employment law, (2) criminal law, and (3) family law. In each area, as noted, DNA tests have been used in multiple ways to predict health issues or establish identity, and DNA tests have also been sidelined or discouraged by regulation in important ways. Each area provides a set of important lessons about the interaction between genetic technology, evidence law, institutional interests, policy, and regulation.
A. Predictive Testing: Regulating Genetics in Health and Employment Law
Predictive genetic testing is useful for assisting individuals and their health providers in preventive care. It is also widely used to assist future parents in deciding whether to conceive and whether to carry a pregnancy to term. Predictive genetic testing has other potential uses--particularly, in the criminal law area, for predicting propensities to commit crimes--but so far, the areas of law in which it has been heavily regulated have been primarily health- and employment-related.
The primary concerns of these laws are privacy and discrimination. If the government, private insurer, or employer knows that an individual carries a gene that makes them more likely to suffer from a particular disability or disease, that entity might not want to provide health coverage or employ that particular individual. In 1996, Congress enacted the Health Insurance Portability and Accountability Act (HIPAA). (39) Although the better-known provisions of HIPAA concern protections for patient privacy and the ability of employees to take their insurance with them when they switch jobs, HIPAA also prohibits employer-sponsored health insurance from establishing eligibility rules based on a person's "genetic information." (40) More recently, Congress passed the Genetic Information Nondiscrimination Act of 2008 (GINA), which expanded the provisions of HIPAA to prohibit changing the terms of health insurance coverage based on genetic information. (41) And in 2011, Congress passed the Patient Protection and Affordable Care Act ("PPACA," "ACA," or, more colloquially, "Obamacare"). This law prohibited health insurers from denying insurance based on "pre-existing conditions." (42) Under GINA, an insurer could not deny coverage based on a genetic test but could deny it based on an already expressed genetic disease. The ACA closed that loophole. (43)
In passing GINA in 2008, Congress not only expanded anti-discrimination measures in health insurance coverage but also extended these protections to cover employment. GINA amended the federal employment discrimination statutes by adding "genetic information" to the list of forbidden types of discrimination in employment. (44) Employment decisions ranging from hiring, firing, layoffs, promotion, benefits, or any term or condition of employment may not be made based on genetic information, (45) harassment based on genetic information is prohibited, (46) and an employer may not retaliate against an employee who complains of such discrimination. (47) Nor can an employer request or obtain genetic information about employees, (48) with some exceptions (including DNA testing for law enforcement purposes). (49) Further, EEOC regulations make clear that employers cannot seek genetic information from family members to, for example, find out about their medical histories and propensity for particular diseases or disorders; "family members" is defined broadly to include "persons related from the first to the fourth degree of an individual." (50)
An important goal of GINA was to prevent discrimination in employment decisions, as well as discrimination by healthcare providers and insurers at the federal level, and to promote additional genetics research without individuals fearing that results of genetic tests might cause them to lose their jobs. (51) However, one central purpose of the law had nothing to do with workplace privacy, but rather was to encourage people to get DNA testing to identify medical conditions--and support research in genetics so that the future role that genetics plays in certain medical conditions can be better understood. The legislative findings noted that the law had the purpose to "protect the public from discrimination" but also to "allay [the public's] concerns about the potential for discrimination, thereby allowing individuals to take advantage of genetic testing, technologies, research, and new therapies." (52) Then-House Speaker Nancy Pelosi commented that "[b]ecause of this legislation, Americans will be free to undergo genetic testing for diseases such as cancer, heart disease, diabetes, and Alzheimer's, without fearing for their job or health insurance." (53)
Federal intervention in health- and employment-related genetic testing is an excellent example of ethics-based regulation of predictive genetic testing; for that reason, scholars have cited to it as an example of "genetics exceptionalism," in which uses of genetics are treated as "special" despite similar or even greater concerns with regulation of other types of medical information. (54)
We categorize this regulation as ethics-based because of the strong concerns about ethical issues exhibited in the test and legislative history of the law, including both privacy concerns and concerns about workplace discrimination. GINA in particular was the result of thirteen years of extensive advocacy by members of Congress who imagined a future world in which genetic testing could be used for nefarious ends. (55) The law was not passed, as so many are, in response to a particular tragic event or out of fear of an immediate threat. (56) In fact, a common argument against GINA was that it was premature, since genetic discrimination was infrequent. (57) That said, more litigation under GINA may be coming. In 2015, in a case the district judge dubbed the "mystery of the devious defecator," a federal jury awarded over two million dollars in a genetic discrimination case as compensation for the claimed invasiveness of submitting to the cheek swabs demanded by an employer seeking to conduct DNA tests on fecal matter left at a warehouse to try to identify the person responsible. (58) The case, however, addressed an attempt to identify individuals, though, not to predict future medical outcomes.
Why were lawmakers so focused on the unknown future when passing GINA and other genetic-privacy legislation? The predictive nature of the testing, bringing with it fears of eugenics, coupled with the intentional nature of the discussion, appears to have led to an ethics-based analysis. Both the predictive and intentional aspects of the inquiry were crucial. Predictive testing--testing for propensities, not actual identity--is far more speculative than identity testing. (59) A person could carry a gene associated with developing a certain type of cancer, for example, but that gene might never express itself. In contrast, a person whose DNA test demonstrates that his DNA matches the biological material found at a crime scene or establishes paternity of a particular child is actually that person, not simply likely to be. Certainly, the legislators were more concerned with protecting the rights of the person being tested than they are with identity testing. With predictive testing, then, the chances of being "wrong" seem much higher. (60)
The intentional nature of the discussion mattered, too. The legislators debating GINA knew that they were legislating about genetics. Granted, the legislation concerned health law and employment law, but legislators were specifically focused on the effect genetic testing could have on individual people and how the law could foster or deter scientific progress and how it might lead to invasions of privacy and discrimination.
Taken together, these two factors leant themselves to an ethics-based approach. Although many bioethics scholars might lament the degree to which their work is undervalued, the attention to ethical concerns during the passage of GINA is stunning when compared to what has happened in other legislative debates discussed in this Article. (61) Legislators and activists brought up eugenics, especially Nazi-era racial eugenics, concerns about human cloning, etc. Granted, some of this involved a lack of scientific or ethical understanding--many members of Congress reportedly did not understand the difference between genetic testing and human cloning--but the concern with the long-term implications of legislation was there. (62)
In addition to the concern with eugenics, legislators were also concerned with the rights of the disabled. This concern has also been the focus in debates over prenatal genetic testing. (63) If anything, the leading criticism of GINA is that it goes too far in using an ethics-based approach.
B. Identification Testing: DNA and Criminal Law
The story of GINA and the ethical debates surrounding predictive genetic testing are fairly well known. In contrast, Congress's regulation of genetic testing in criminal law and family law has been less studied and critiqued. We argue that there are several reasons for this. First, the genetic testing at issue in those fields is identification testing, not predictive testing. Identifying a person, rather than predicting who they might become, appears at first glance to be less ethically fraught. Second, the approach has been data-driven, not ethics-based. The passage of laws allowing for the creation of DNA databanks and mandatory paternity testing occurred in response to perceived crises, not in response to concerns about the influence of new technology on individual rights. Taken together, these factors have led to unintended consequences, errors, and potential ethical quandaries.
The use of DNA testing in criminal law is an apparent success story, in which genetic evidence has made it possible to far more conclusively deter mine guilt or innocence in serious criminal cases. The regulations are data focused, with the goal to identify past and future culprits. The legislation, regulations, and policies of law enforcement, together with judicial review of them, have all been direct and intentionally regulating genetic evidence. In so doing, such regulation has largely neglected ethical and individual rights concerns.
Moreover, the legal and regulatory story has been far more mixed. Relatively few criminal cases are susceptible to DNA testing. The databanks have done much good: they have helped to solve crimes and to free the innocent. But there are significantly diminished returns from expanding DNA databanks to include broad categories of felons, misdemeanor convicts, and arrestees, at the cost of overburdening crime labs, many of which face long backlogs in conducting testing on evidence from crime scenes, and with great costs to privacy. (64) Use of DNA is asymmetrical, with widespread use to assemble giant repositories of genetic information that raise real privacy concerns, but far more limited use of DNA tests to solve crimes and to potentially free innocent convicts, raising due process concerns as well as accuracy concerns, given unwillingness of law enforcement to permit study of those databanks. To return to our typology, not only have ethical concerns been sidelined, but data-driven regulation has not worked as well as had been hoped. Federal spending can be a blunt, unpredictable, and ill-managed instrument. Judicial review could have addressed at least some ethical concerns regarding privacy and due process, but so far, highly deferential review has largely elided those concerns.
1. DNA Arrives on the Scene
Beginning in the late 1980s, law enforcement in the United States began to use DNA testing to link evidence from crime scenes to suspects. The federal government played a key role in the development and standardization of techniques used. The British geneticist Sir Alec Jeffries pioneered the genetics research that made DNA testing possible in criminal cases by focusing on short repetitive sequences within the genome that are highly variable between individuals and can be isolated for testing. Jeffries later licensed technology to conduct tests isolating those sequences to Imperial Chemical Industries (ICI), which in turn opened its U.S. operation, called Cellmark, to initially focus on paternity testing. (65) Cellmark quickly began to market the new technology to law enforcement, promoting "DNA FINGERPRINTING" as a way to get "CONCLUSIVE RESULTS IN ONLY ONE TEST!" (66) Soon, competing private labs entered the market, the FBI began to perform DNA tests, as did state crime labs. (67)
The FBI took a leading role to determine the form DNA tests would take in criminal cases. (68) The FBI chose an incompatible DNA testing system using different enzymes than those used by Cellmark and the other leading lab at the time, Lifecodes. (69) The FBI trained forensic scientists on the new protocols, created a national DNA database, and built a new DNA Analysis Unit. (70) Having taken the lead "largely by fiat" in developing standards for DNA testing, the FBI left out not only private labs, but also academic scientists, defense attorneys, and others. (71) Because of its head start and leadership role, crime labs around the country sent evidence for DNA testing to the FBI lab well into the 1990s. (72)
DNA technology improved dramatically by the mid-1990s, and as a result, the use of DNA tests in criminal cases accelerated. (73) Courts quickly ruled DNA evidence was admissible evidence in criminal cases--admissible, typically, to show the guilt of the defendant. Meanwhile, the federal government and some states enacted statutes permitting statutes of limitations to be relaxed to allow prosecutors to bring charges years later based on DNA evidence, or by extending or abolishing statutes of limitations in rape cases. (74)
Concerns regarding unclear and inconsistent standards used to reach conclusions about forensic DNA, in part because of the lack of transparency and validation as the FBI adopted its DNA testing protocols, led to high-profile court challenges and academic disputes, nicknamed the "DNA Wars" of the 1990s. (75) The result was a different type of federal intervention: two National Research Council, National Academies of Sciences committees convened leading scientists to study the problem and arrive at standards for forensic use of DNA testing. (76) In addition, in 1993, Congress intervened and enacted legislation authorizing the FBI to assemble DNA samples in a national databank, the Combined DNA Index System (CODIS). (77)
2. DNA and Claims of Innocence
Legal efforts focused on collecting genetic data from potential criminal culprits were largely one-sided. While legal actors moved quickly to allow DNA to provide powerful evidence of guilt, the use of DNA testing to free the innocent moved quite slowly. Although the first few convictions relying on DNA tests were quickly followed, in 1989, with the first exonerations relying on post-conviction DNA testing, (78) in the 1990s only two states had statutes providing a right to access post-conviction DNA testing. (79) Many of the people freed by DNA tests in the first decade and a half of its use waited many years to obtain those tests. Today, all fifty states have enacted statutes providing access to DNA and post-conviction relief. (80) However, many of those statutes contain sharp restrictions on access to DNA testing, including bars on testing if the defense lawyer should have requested testing at trial, limits on access to persons convicted of certain felonies, and bars to testing of persons who pleaded guilty or did not litigate the issue of identity at their trials. (81) Further, courts have interpreted the statutes in strained ways that have made it still more difficult for prisoners to obtain testing. (82)
In contrast to these developments in the states, beginning with its 1993 decision in Herrera v. Collins, the Court has remained steadfastly unwilling to recognize, but has assumed the hypothetical existence of, any constitutional right to challenge a conviction based on "truly persuasive" evidence of "actual innocence." (83) In the 2006 case of House v. Bell, when the Court had the opportunity to directly confront the question whether to recognize an actual innocence claim, it chose not to do so, again simply assuming its availability. (84)
In District Attorney's Office v. Osborne, the Court squarely confronted whether there is a constitutional right to secure DNA testing. (85) William Osborne, a man convicted of rape in Alaska, sought post-conviction DNA testing that the State conceded could conclusively establish his innocence. Alaska was one of three states at the time that had no post-conviction DNA testing statute and refused to grant the test. The Court ruled that Osborne had no freestanding constitutional right to DNA testing. However, the Court did say he had a procedural due process right to DNA testing based on "a liberty interest in demonstrating his innocence with new evidence under state law." (86) The Court largely relied on federalism reasons in ruling that ability to obtain a post-conviction DNA test would be dependent on state law remedies. A state has no obligation to make post-conviction DNA testing available in the first place, but if a state does so (as all states, the federal government, and the District of Columbia have now done), a litigant like Osborne would have to show that state statutory mechanisms were arbitrary and violated procedural due process. Some litigants have successfully challenged state DNA testing laws on the basis that they arbitrarily exclude categories of individuals from access to testing, but other courts have ruled that the Osborne decision leaves wide latitude to states. (87) In effect, the Court, while recognizing the remarkable power of DNA testing, chose to leave the regulation of post-conviction DNA testing to the states, even where the state in question, Alaska, could offer no reason, much less a compelling reason, as Justice Stevens pointed out in dissent, for denying Osborne the DNA test he requested. (88) Once again, judicial review of rights in genetic evidence was highly deferential, and not based on any careful balancing of government and individual interests or analysis of individual rights.
3. Federal Support for DNA Databanks
In contrast to the slow and uneven response to the use of DNA to potentially free the innocent, a raft of federal legislation funded and expanded the use of DNA to identify the guilty. This legislation was intentionally and directly focused on collecting genetic data. The national DNA databank, CODIS, is a federal creation, but it is also a joint state and federal effort. (89) The effort began in 1990 informally, and CODIS was established in 1994, following federal legislation, the 1993 the DNA Identification Act, that allowed the FBI to assemble DNA samples in a national databank. (90) Each state collects DNA profiles, which can then be accessed and searched by the FBI software, with any resulting "matches" then provided to the law enforcement agency. The 1996 Anti-Terrorism and Effective Death Penalty Act (AEDPA) provided the FBI with authority to add to CODIS the DNA profiles of federal offenders, but the DOJ was uncertain as to its authority. (91) In 2000, the DNA Act provided authority to collect profiles from federal offenders. (92) CODIS now contains over eleven million profiles, and it continues to expand. (93) The FBI reports that it "has produced over 293,808 hits assisting in more than 279,741 investigations." (94) These DNA profiles currently contain information from thirteen loci that are very useful for identification of individuals, but not for predicting future medical outcomes or characteristics of individuals.
All fifty states and the federal government have now enacted laws permitting collection of DNA from those convicted of serious felonies. Federal legislation has continued to expand the list of qualifying crimes, with the 2001 USA PATRIOT Act expanding qualifying federal crimes (95) and the 2004 Justice for All Act expanding collection to all federal felons. (96) In twenty-nine states arrestees and detainees--many of whom are never charged or convicted--can have profiles entered into the database; for example, over a dozen states require that all felony arrestees have DNA entered, half of the states require that murder and sex crime arrestees have DNA entered in the CODIS, while several states extend DNA collection to misdemeanor arrests. (97) These requirements were prompted by 2005 federal legislation permitting arrestee DNA to be entered into CODIS. (98) However, few of those states allow defense lawyers access to the databases for searches that might show clients' innocence; only nine states do so. (99) Very few states have conducted even ad hoc reviews of closed cases to discover potential wrongful convictions. One of the few such audits, conducted in Virginia, resulted in several exonerations. (100)
DNA collection efforts have been challenged in lawsuits raising a series of privacy and constitutional challenges, all of which have proved unsuccessful. (101) This litigation culminated in the case of Maryland v. King, in which the Supreme Court found it constitutional under the Fourth Amendment to take DNA from persons arrested for "serious offense[s]." (102)
Federal largesse has facilitated DNA collection. In 2000, when CODIS was expanded yet again, Congress enacted the DNA Analysis Backlog Elimination Act to supply grants to states for DNA lab work. (103) The findings included the statement that although "States have received millions of dollars in DNA-related grants," nevertheless, not only had there been inadequate use of DNA testing post-conviction to potentially free the innocent, but "more funding is needed to improve State forensic facilities and to reduce the nationwide backlog of DNA samples from convicted offenders and crime scenes that need to be tested or retested using upgraded methods." (104) The federal funding allocated was substantial. For example, between 1999 and 2007, the federal government allocated $560 million to the states to eliminate backlogs as DNA database demands began to tax the capacities of crime labs. (105) In 2003, President Bush announced a one-billion-dollar "DNA Initiative." (106) However, not all allocated money was spent. In 2010, the National Institute for Justice (NIJ) awarded less than $65 million in DNA Backlog Elimination Program grants--less than half of the $151 million that Congress authorized. (107)
Underfunding explains in part why backlogs continue to tax state and federal law enforcement, with nationwide backlogs of over 500,000 cases. (108) These DNA collection efforts have been the victim of their own success; the result has been even greater backlogs in processing all of these DNA tests. (109) The voluminous DNA processing of convicted offenders and arrestees can be more straightforward and less expensive, in a given case at least, than DNA testing samples from crime scenes, which may involve mixtures of several person's profiles, deteriorated samples, and more need for analysis and interpretation. The backlog problem is quite urgent at some labs. Consider, for example, the case of Cody Davis, an innocent man misidentified by the eyewitnesses to a robbery in Florida. The ski mask worn by the robber was not DNA tested before trial, due to backlogs at the crime lab; only four months after the trial was the mask DNA tested, and the results cleared him and led to his post-conviction DNA exoneration. (110) The most recent statistics available on crime lab activities show the same problem: DNA tests are only a third of the work of public crime labs, but account for seventy-five percent of backlogs. In turn, although DNA analysis in actual criminal casework is less than twenty-five percent of the DNA testing crime labs performed, over one third of the 1.1 million backlogged requests were for DNA samples from casework. (111)
Analyzing DNA from actual crime scenes can be far more expensive and complex than simply loading profiles into a database, as noted. (112) Congress has intervened to encourage more use of federal funding to do DNA testing in live criminal cases. However, these efforts have not been very successful. Money under the Act often remains unspent, which can result in denial of grant requests for the next year. Reasons may include cumbersome reporting requirements, poor reporting by local jurisdictions, the burden of growing backlogs, reluctance by local law enforcement to conduct DNA testing in criminal investigations, and perhaps most important of all, poor local law enforcement training and resources to collect DNA from crime scenes. (113) In 2004, the DNA Backlog Elimination Act was renamed the Debbie Smith Act, after a rape victim, (114) and in name and in its intent, the Act changed the focus to criminal investigations. (115) The text of the Act requires that forty percent of grant money be used on "samples from crime scenes." (116) Yet that was a reduction: it had been the case that fifty percent of grants had to be used for testing of crime scene evidence. (117) Federal grants are designed in a way that makes backlogs worse. Although federal grants have been chiefly allocated to crime laboratories, the demands of entering profiles into the CODIS extend far beyond the costs of the testing itself. Prison officials must track who must be tested and who has already been entered in the database, samples must be taken from prisoners, and then samples must be brought to the crime lab. The costs for enacting DNA collection from arrestees may be even more substantial. Some states have decided not to go that far specifically for cost reasons. (118) On the other hand, police departments may create their own DNA databases that go farther. (119) States can also defray costs through criminal fines; many states now charge DNA databank fees to convicts. (120) States vary in their participation in CODIS; some upload profiles more aggressively than others. (121) A 201 Senate Report noted that "a significant backlog of DNA samples and rape kits remains in public crime laboratories" and expressed concerns that large sums of grants approved by the National Institute of Justice have been awarded to "entities of dubious merit," including policy groups that did not themselves do DNA testing. (122)
That is not to say CODIS does not serve important goals. In many thousands of cases, CODIS identifies culprits or clears innocent people during criminal investigations. Forty-five percent of the first 250 DNA exonerations also involved the post-conviction identification of the culprit, often due to a "cold hit" in CODIS. (123) Sometimes the exoneration was motivated in part because DNA showed not only innocence but the guilt of an identifiable person. CODIS is a powerful tool. However, its size and the resources allocated to it have come at a cost to not only the ethics-based concerns but also the very goal that legislators were focused on: the goal of using data to solve crimes.
Delays or failure to conduct DNA testing may undergird decisions not to pursue cases, and they may discourage victims from cooperating and lead to failures to prosecute. (124) A 2009 National Institute of Justice survey found that in "18 percent of open, unsolved rape cases, forensic evidence had not been submitted to a crime lab." (125) One of the few surveys conducted of use of forensics in criminal cases, led by Joseph Peterson, sampled over 4,000 crimes in five jurisdictions, and found that "most evidence goes unexamined" and no evidence is collected and no forensics tests are conducted, even for quite serious offenses. (126) After all, crime scene investigators have "varying levels of training and experience." (127)
Individual rights are sidelined; for example, defense rights to access CODIS data to potentially prove innocence. CODIS does permit disclosure of data for "criminal defense purposes, to a defendant." (128) However, only nine states provide defendants with access to the DNA databases, and absent such statutory authority, defense access may depend on whether prosecutors agree to request such a test. (129) DNA testing for the defense does not receive the same federal largesse. The "Justice For All Act" allocated five million dollars a year for conducting post-conviction DNA testing, under the Kirk Bloodsworth DNA testing program (named after the first death row inmate exonerated by DNA). (130) That funding is small compared with over $100 mil lion allocated each year for backlog elimination (which itself is apparently often not fully spent). The Bush Administration had "resisted implementing the program for several years," and the funding remained entirely unspent for four years. (131)
Indeed, not only defense lawyers, but prosecutors may be left in the dark should evidence of innocence surface. CODIS is not well connected to local prosecutors. CODIS hits in cold cases result in notification to local police, who likely closed the case, but not to prosecutors, who would know if there had been a conviction (possibly of the wrong person). (132) Many jurisdictions have no policy in place to respond to a DNA hit in a closed case. (133) As a result, CODIS hits may receive no follow-up. (134)
The philosophy of "more data is good" has meant that labs have not focused on which data are the most useful or important. The labs themselves may make mistakes as they invest in processing entries and not in quality control and independent auditing. (135) Separate money was allocated for grants to improve the quality of forensics work. (136) That grant money went unspent entirely for years after the statute was passed. (137) An audit by the Office of the Inspector General found NIJ was not requiring recipient labs to comply with requirements that they put in place independent auditing mechanisms. (138) To this day, the Department of justice still has not insisted laboratories comply with those rules. Also troubling may be the spillover effect on the vast majority of the work that crime labs do that is not DNA testing. (139) A landmark National Academy of Sciences report in 2009 explored the urgent need for national scientific standards and stronger oversight and quality control of our entire system of forensics, calling for the creation of a National Institute of Forensic Science. (140) None of those recommendations has yet been enacted in federal legislation. (141)
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Abstract through II. Comparing Approaches Towards Genetic Identity B. Identification Testing: DNA and Criminal Law 3. Federal Support for DNA Databanks, p. 757-785|
|Author:||Abrams, Kerry; Garrett, Brandon L.|
|Publication:||Notre Dame Law Review|
|Date:||Dec 1, 2015|
|Previous Article:||Why the right to elective abortion fails Casey's own interest-balancing methodology - and why it matters.|
|Next Article:||DNA and distrust.|