Comments on Proposed Amendments to Codify a Familial DNA Search Policy in New York State

The Criminal Courts Committee, the Criminal Justice Operations Committee and the Corrections and Community Reentry Committee (the Committees) disapprove of the proposed amendment to Part 6192 of Title 9 NYCRR to codify a familial search policy, because such a policy will unconstitutionally target specific racial and ethnic groups. Even if it did not, the Committees believe that the technique has not proven to be a reliable source of information for criminal investigations, and impermissibly invades important privacy rights.   The Committees also disapprove of the proposed amendment because it exceeds the rulemaking authority of the New York State Division of Criminal Justice Services (DCJS).

Should the proposed amendment find support of an appropriate rulemaking body, the Committees recommend specific changes to certain sections that: (1) are overbroad in the designation of offenses that qualify for familial searching; (2) are overly vague in the description of what law enforcement steps must be taken before conducting a familial search; and (3) fail to incorporate judicial oversight for applications for familial searching.

Under current law, when DNA is recovered from a crime scene or a specific piece of evidence, the profile is compared to all profiles in the “convicted offender” database, otherwise known as the New York State DNA Index (SDIS).[1]  If the recovered DNA “matches”[2] that of an individual currently in the database, that individual’s name is transmitted to law enforcement for further investigation. However, if the recovered DNA does not match anyone in the database, this information is not transmitted.  In other words, if the recovered DNA does not match that of an individual in the database, then the individuals in the DNA database are not considered to be potential perpetrators of the crime, and law enforcement must use other investigative tools.   Law enforcement cannot compare recovered DNA in the state population at large because, among other reasons, the New York State Legislature never intended to create a DNA database that would have included residents who were not convicted of any crime.

The proposed rule would allow SDIS, upon request of law enforcement, to release names of individuals in the DNA database whose DNA does not match DNA recovered from a crime scene or evidence sample, but shares a small portion of genetic code with the recovered DNA.  The purpose of releasing that person’s name is not to investigate the person in the database as a potential suspect, but instead to search for that person’s male family members using public records, such as those maintained by the DMV.[3]   Once these “family members” are located, law enforcement will take steps to obtain their DNA samples either surreptitiously – by gathering “abandoned DNA” [4] – or through a warrant.[5]

It is important to note that these suspected “family members” become law enforcement targets merely because they are presumed to share a portion of their genetic code with someone who is in the DNA database, not because they have any established connection to the crime.[6]   Indeed, these “family members” need not be close relatives; in fact, they may not be “relatives” at all, since law enforcement’s tools for searching for presumed family members may be inaccurate or imprecise.[7]

GENERAL OBJECTIONS

The Committees raise four general objections to the proposed amendment.  First, the agencies that have promulgated it, namely DCJS and the Commission on Forensic Sciences, lack the authority to do so, since the amendment would effect a change to the DNA collection statute and would therefore constitute legislation.  Second, the proposed expanded search threatens New Yorkers’ rights to privacy in their genetic and family information and relationships.  Third, the proposed amendment will disparately impact communities of color, implicating the Equal Protection Clause of the constitution.  Finally, law enforcement has established no need for this rule change that would justify this significant, discriminatory privacy intrusion.

1. The Proposed amendment to 9 NYCRR § 6192 cannot be enacted without participation of the state legislature.

New York State Executive Law § 995-c authorized DCJS to “promulgate a plan for the establishment of a computerized state DNA identification index.”[8]  This plan was implemented in 2005 with the approval of “the DNA subcommittee … and the speaker of the assembly.”[9]  Codified in 9 NYCRR § 6192, et seq., the plan defines the individuals whose DNA is eligible for indexing, and sets limits on the purposes for which law enforcement can use that DNA.[10]

Through Executive Law § 995-c, the Legislature delegated authority to DCJS to “establish” the DNA index, including expanding it to add categories of individuals subject to DNA indexing.[11]   The Legislature did not however, grant the same latitude concerning how the DNA data is to be used.  The Court of Appeals held more than 40 years ago that “[a]dministrative agencies can only promulgate rules to further the implementation of law as it exists; they have no authority to create a rule out of harmony with the statute.”[12]  Such agencies cannot “add [] a requirement not found in the existing State statute.” [13]

The proposed amendment grants to law enforcement powers that are far broader than those contemplated in the Executive Law.  Permitting expansion of the means by which law enforcement uses DNA samples, as opposed to collecting and indexing it, steps outside the Legislature’s delegation of authority to DCJS.  Such a large-scale change to law enforcement’s use of DNA evidence should be publicly debated and voted on by the legislature, which is accountable to the electorate, and not affected by the appointed members of a state agency.

2. This Amendment threatens the privacy rights of all New Yorkers.

Familial DNA searching, as set forth in the proposed amendment, should not be allowed in New York State because it infringes on deeply-engrained privacy rights.  It would pave the way for people who have never been convicted of a crime to become possible suspects in an investigation solely because a family member – even a very remote one – has been convicted of a crime in New York State.  Currently, an individual convicted of a crime is required, as a consequence of that conviction, to give up a certain amount of privacy by having his or her DNA put in the database.  Under the proposed amendment, the individual’s family member, who has not agreed to forfeit rights to genetic privacy, will relinquish these rights with no recourse.  This raises Constitutional challenges under the Fourth Amendment where case law is still developing to confront these privacy issues.[14]

A person identified as a purported “family member” by law enforcement will be placed in the stressful and anxiety-provoking position of excluding himself as a suspect of a crime that he likely had nothing to do with.  (See infra, discussing “false positives” in familial searching.) The   requirement that an individual engage in this process is fundamentally unfair. Even a dystopian scenario featuring a universal database of everyone’s DNA would infringe on privacy less than the scenario created by the proposed amendment.[15] The National Research Council of the National Academies of Sciences, Engineering and Medicine has recommended the prohibition of familial DNA searching for these very reasons.[16]

3. This Amendment will have the impact of unreasonably and unfairly targeting communities of color in New York City and beyond.  

Minorities, particularly Blacks and Latinos, are disproportionately represented in New York’s criminal justice system.  This disproportionality leads to a racial disparity in the State’s DNA database.  Since familial DNA searching connects a person with a conviction to multiple family members, the racial disparity found in the criminal justice system is exacerbated by a familial searching process that naturally leads to a greater number of minorities under law enforcement surveillance.[17]  Stanford University Professor Henry T. Greely predicts that if familial searching were to be adopted nationwide, approximately one in every three Black persons would be under surveillance.[18]  This amplification of pre-existing racial disparities has led some to label familial searching as “genetic stop and frisk.”[19] Fundamental fairness and respect for citizens’ privacy and for maintaining a fair and just society should prevent New York State from implementing this questionable practice.[20]

Proponents of familial searching cite to a 2004 Department of Justice study that found that “almost half of all prison inmates surveyed report having a close relative who had also been in prison.”[21]  In citing this study, they argue that familial searching will be effective because, essentially, crime runs in families.  But the 2004 study does not take into account and effectively ignores the voluminous and well-documented research showing that over-policing and over-criminalization of underserved communities and communities of color causes the imprisonment of people in the same family.[22]  It is not someone’s genetic similarity to a person in prison that makes them a likely crime suspect, the research shows; instead, it is increased and misplaced police scrutiny.  Therefore, genetic similarity would not discover whether someone has committed a crime, but instead would only compound the existing problems facing communities of color.

4. No law enforcement need for familial searching is significant enough to outweigh countervailing rights to privacy and fundamental fairness. 

Considering the serious concerns about its drafters’ lack of rulemaking authority, together with the equal protection violations and privacy invasion that will come with familial searching as set forth in the proposed amendment, the Committees believe that law enforcement must demonstrate a need for it.  To date, they have not.  The Committees are not aware of any meaningful studies showing that familial searching has provided sufficient value to law enforcement to overcome these Constitutionally-protected concerns.

To the contrary, studies have shown that although termed “familial” searching, the method proposed by this amendment cannot precisely predict shared familial genetic code, as opposed to genetic code that two unrelated people share by mere coincidence.  As one researcher observes: “[a]s it turns out, even an individual’s close relatives, such as parents or full siblings, can display a remarkable amount of genetic difference, and strangers can share remarkable similarities. With only limited biological information it can be very hard to discern whether an offender profile is likely to indicate a true lead or not.”[23]

Even in jurisdictions where familial searching is permitted, it has shown only limited success.   The California Department of Justice has been using familial searching for almost ten years, but, as of April of this year, yielded only eight “hits” out of more 134 searches.[24]  Even such a small “hit” rate can be further diminished by false positives.[25]  For example, in New Orleans, police obtained a warrant for a man’s DNA based on a familial association between evidence from a rape and murder victim and the man’s father.  This man remained under police suspicion of a gruesome crime for months until he was ultimately excluded after additional DNA testing.[26]  In Colorado, the Denver District Attorney’s Office maintains a list of what it terms “successful” familial DNA investigations.  Yet this list, which goes back more than a decade and spans the entire United States, United Kingdom and New Zealand, contains only 58 cases.  Of those 58 cases, many do not actually involve “familial” searches, but instead “partial match” searches similar to what New York already has in place.[27]

Although New York State has not yet used familial searching, it has had a similar “partial match” policy in effect since 2010.[28]  This partial match policy (which does not include the familial search policy’s provisions on additional testing and data release), has resulted in only three cases that reached prosecution.  Of those three cases, one of them involved an innocent person whose partial DNA profile was found on a cigarette in his neighbor’s yard.[29]  The second was a property crime case. Only the third case – a “cold hit” from 1990 – was for a serious crime, and resulted in a conviction.  A rate of one serious conviction in almost 10 years of use does not, in our Committees’ opinion, support there being a sufficient law enforcement purpose for this drastic invasion of privacy.  Notably, the case that promoted the push for familial searching in New York, the murder of Karina Vetrano in Queens, was solved through traditional law enforcement methods alone.[30]  Familial searching was not used.

The purported law enforcement purpose of familial searching also must be weighed against the serious and frightening cases where an innocent person is implicated in a major crime through a false or misleading DNA match.  Such a false match can happen through lab errors, which have been well-documented throughout the country, and in New York. [31] But it also can occur simply because a DNA match is not conclusive proof of guilt.

Finally, the proposed rule, and familial searching itself, embraces the misguided notion that a DNA match necessarily is evidence of guilt. Recent cases have shown the error in this preconception.  Lukis Anderson was arrested and charged with murder in California in 2013.[32]  The sole evidence against him was his DNA profile under the fingernails of the victim.  Mr. Anderson, however, had an airtight alibi: he was in the hospital at the time of the murder.  Further investigation revealed that the same paramedic who treated Mr. Anderson earlier in that day also treated the murder victim and inadvertently transferred the victim’s DNA on their medical equipment.  Mr. Anderson was lucky to have had such an airtight alibi, otherwise the exact match of his DNA on the victim likely would have led to his wrongful conviction for murder.  In another case in New York City, the Queens District Attorney’s office prosecuted a man for robbery based solely on his DNA match, even though surveillance video showed someone else committing the crime.  This man was held at Rikers Island for one and a half years before he ultimately was acquitted by a jury.[33]

While familial searching may be a tool that law enforcement can use, it is not a sufficiently useful tool, when weighed against the constitutional interests of New Yorkers.  The hope that this new tool may yield a “hit,” that such a “hit” may yield a “match,” and that such a “match” would be to an actual person who committed the crime is far outweighed by the invasion of important privacy and constitutional rights that familial searching entails.

SPECIFIC OBJECTIONS

The Committees also take the position that, even if legislation is proposed to permit expanded searching along the lines of the proposed amendment, specific changes must be made to the current draft to help ensure its fair, impartial, and predictable application.

First, section (h)(1)(iv) provides that any “crime presenting a significant public safety threat,” can trigger familial searching.  The term “significant public safety threat” is not defined, making this section too broad and ripe for abuse.  While sections (h)(1)(i) through (h)(1)(iii) are limited to serious felonies such as homicides, violent felony sex charges, and class A felonies, section (h)(1)(iv) would allow for misdemeanors and other far less serious crimes to trigger familial searching.  Section (h)(1)(iv) would essentially act as a catch-all that could apply to most crimes and make the narrowly tailored application of the previous listed groups of crimes effectively meaningless.  Because its key term is subject to interpretation, and because the drafters of the proposed rule have included no provision for judicial oversight of applications made to the division (see below), this section is ripe for abuse.  

Second, section (h)(2)(i) permits familial searching after only “reasonable investigative efforts have been taken.”   However, the term “reasonable investigative efforts” is left undefined.  This section should be changed to require that all other reasonable investigative efforts have been exhausted before a familial search can be performed.  Adding “all” requires law enforcement to follow every reasonable lead.  “Reasonable” allows for law enforcement to use their experience and judgment.  “Exhausted” adds a requirement that these steps not just be initiated, but completed before resorting to familial searching.  These modifications would give law enforcement more guidance, and is completely consistent with the requirement that familial searching be a last resort for a serious, unsolved case.

Third, section (i) provides that an application for a familial search be made to the Division of Forensic Science, instead of a judge or magistrate.  Determining whether an application for a familial search is lawful, though, involves analysis and application of law, and must therefore be handled by a judge or magistrate.  Delegating this function to the executive branch is improper and impractical. Without judicial oversight, the application for familial searches would be ripe for abuse.  Requiring a judge or magistrate to review applications for familial searches would prevent this problem.

*  *  *

In conclusion, the Committees have substantial concerns about the proposed amendment to 9 NYCRR § 6192 both as a whole and in specific parts.  The Committees therefore disapprove it

Corrections and Community Reentry Committee
W. Alex Lesman, Chair

Criminal Courts Committee
Kerry Ward, Chair

Criminal Justice Operations Committee
Sarah J. Berger, Chair

October 2017