By Michael T. Risher

Of the hundreds of thousands of arrests every year in California on suspicion of a felony, nearly 320,000 in 2006, approximately 30% never lead to any conviction.[1] A disproportionate number of these innocent arrestees are people of color. In the U.S. justice system, people who are arrested but never convicted are presumed innocent; yet, as of January 1, 2009, all arrestees have been forced to let the State of California take a DNA sample, analyze it and include the resulting profile in a criminal database, to be compared evermore with crime-scene evidence. There are procedures for some of these people to try to get the samples and profiles expunged; however, these procedures often require arrestees to wait three or more years before even requesting expungement and necessitate the help of a lawyer.  As a result, the overwhelming majority of people arrested but not convicted of any crime are unlikely even to try to get their samples destroyed. Tens of thousands of profiles taken from innocent people will thus remain in these criminal databases. The consequence will be a magnification of the current racial disparities in our criminal justice system as more and more people of color's DNA profiles are included in databases that make them potential suspects whenever DNA is recovered from a crime scene. 

The effects of this disproportionate inclusion of people of color in the databanks are made clear by the other papers in this series on genetics and race.  What is perhaps less clear is how our criminal justice system, which promises equal justice under the law, can tolerate this injustice.  This paper, after outlining the legal growth and transformation of DNA databanks, examines how various steps in our criminal justice system create and magnify racial disparities, and how the law makes it nearly impossible to effectively address the problem.  It also looks at how taking DNA samples at various stages in this process may affect these disparities and the factors that cause them.   I use as my primary example California's system because it is one of the world's largest criminal justice systems in one of the nation's most diverse states.  It is also the system in which I have practiced law for the last decade, and is representative of where DNA databanks throughout the country will likely be in the next few years as more and more states and the federal government collect DNA from arrestees.[2]

DNA databanks comprise two distinct components: the actual biological samples and the computerized database of the profiles generated by analyzing these samples.  In criminal-justice databanks, the biological samples are collected from crime scenes (forensic samples) and from known individuals (known samples).  Until recently, known samples were usually obtained by drawing blood, although now most states and the federal government primarily obtain samples by swabbing the inside of the person's cheek to collect skin cells.[3] 

The government analyzes both forensic samples and known samples to create DNA profiles, which are essentially a digitized description of 26 parts of the DNA molecule.  The profiles are then uploaded to the Combined DNA Index System ("CODIS"), a centralized, searchable law enforcement database accessible to state, federal, and international law enforcement agencies. CODIS was created by the FBI in 1994 after Congress authorized it to establish a national DNA database to link existing state and local databanks.  The biological samples themselves are retained by the local police or crime lab for later testing. 

Once an arrestee's profile is uploaded into CODIS, it is immediately compared to the thousands of crime-scene samples in the CODIS forensic database.  As long as the arrestee's profile remains in CODIS, any new crime-scene samples will be searched against it. When an arrestee profile exactly matches a crime-scene profile, CODIS automatically notifies agencies that provided the sample. Then that agency will usually provide the identity of the arrestee to the agency with jurisdiction over the crime so that it can follow up.

DNA databanks have grown exponentially in the last decade as new laws have expanded the range of people subject to having their DNA forcibly seized, analyzed, and the resulting profile databanked.  California's databank is a good example of this. It was originally conceived as a way to connect people convicted of serious violent crimes with other such crimes in which DNA evidence is most useful.  The original 1989 DNA-collection law established a databank and required people convicted of murder or a felony sex offense to provide DNA samples before they were released from custody.[4] The state department of justice had the authority to analyze these samples and include the resulting analysis in the new statewide databank.[5]  From today's perspective, this program seems quite limited: the only people subject to having their DNA databanked had been convicted of very serious crimes, either by pleading guilty or after the charges had been proved beyond all reasonable doubt to a jury.  

But the law soon began to expand to include more people.  The first steps were modest:  in the late 1990s, new crimes were added to the list of qualifying offenses, and the law was amended to require that samples be taken immediately after conviction, rather than just before release. The latter change was enacted as the focus shifted from preventing new crimes to solving old crimes. In 2004, California voters enacted Proposition 69, drastically expanding the database.  The two biggest changes were that, as of November 2004, every person convicted of any felony - which can include simple drug possession, shoplifting, or even intentionally writing a check without sufficient funds to cover it - has had to provide DNA samples.  And, as of January 1, 2009, every person arrested for a felony in California must give a DNA sample.  Prop. 69 thus radically changed the database from one comprising profiles of individuals convicted of violent felonies to one that includes profiles from suspected shoplifters. 

This huge increase is not distributed equitably among all people. African-Americans comprise 6.7% of California's population, but 21.5% of those arrested for felonies in the state.[6] Although, as discussed below, the possibility of race-based decision-making at all levels of the criminal justice system makes it impossible to know whether changing from a database of people convicted of felonies to one including everybody arrested for felonies will result in an increase in the proportion of people of color in the database it will clearly result in a significant increase in the absolute number of minorities included. 

Racial disparities fluctuate depending on the stage of the criminal proceeding from the high-level initial decision to make certain acts criminal, to a police officer's decision to contact or arrest an individual, to the decisions made by prosecutors, judges, jurors, and defense lawyers.   Thus, the stage at which DNA samples are taken will affect the racial disparities in the databank, albeit in unpredictable ways. 

The first and broadest stage at which racial disparities are introduced into the criminal justice system is at the legislative level, where crimes are defined and classified. The basic question is, of course what conduct is considered criminal: why is a person who possesses drugs subject to criminal sanctions while a business that puts its workers or consumers at risk with dangerous or unsanitary facilities subject only to civil sanctions?   Even beyond that basic issue, our criminal justice system treats very similar conduct differently in ways that create racial disparities.  The most notorious example of this is the crack versus powder forms of cocaine  disparity in the federal system, which for years punished people convicted of crack cocaine offenses (well over 80% of whom are African- American) much more severely than powder cocaine offenders (72% of whom are white or Hispanic).[7]

The legislative establishment of "drug-free zones," often around schools, parks, or public-housing projects, can also have racially disparate effects.[8] These laws mean that people who live and commit drug crimes in dense urban areas, where few locations are not close to a school or park, will be punished more harshly for the same conduct than are their suburban or rural counterparts.  Because urban areas usually have higher proportions of people of color, these harsher punishments will reinforce racial disparities.

Laws like these interact with seemingly race-neutral DNA collection laws to produce great disparities in the databank.  A databank that includes all persons convicted of felonies will include every person - primarily people of color - convicted of possessing cocaine or heroin, no matter how small the amount; but it will not contain samples from people - primarily white - convicted of minor methamphetamine offenses that were prosecuted as misdemeanors. Conversely, a databank that includes only violent crimes or sex crimes - as many originally did - should result in fewer disparities than an all-felony database for the reasons just described.  Excluding non-violent crimes is reasonable since DNA evidence is almost never involved in non-violent offenses. 

DNA databanks themselves create a feedback loop that further magnifies these disparities.  Well over half of all serious crimes go completely unsolved, with the police never even identifying a suspect. If DNA databanks work as they are intended, they will identify suspects for at least some - perhaps many - of these crimes.  But a racially skewed databank will produce racially skewed results;  because racial disparities in the criminal-justice system have led to the inclusion of a disproportionate number of profiles of African-Americans in CODIS, the databank will return a disproportionate number of matches to African-American suspects.  In contrast, crimes committed by members of groups that are underrepresented in CODIS will escape detection, particularly as the police spend an increasing amount of their limited time and resources focusing on cases where they have found a DNA match. 

The U.S. Constitution, as interpreted by the courts, does not prohibit this shift, regardless of the racial disparities it introduces.  The courts have held that the Fourteenth Amendment's promise of equal protection of the law prohibits only intentional discrimination, which means that challenges to criminal laws that result in racially disparate impacts are extremely difficult.  In the words of the U.S. Supreme Court, discriminatory intent means "more than intent as volition or intent as awareness of consequences.  It implies that the decision-maker, in this case a state legislature, selected or reaffirmed a particular course of action at least in part because of, not merely in spite of, its adverse effects upon an identifiable group." If the governmental body would have acted the same way even without the discriminatory intent, the law stands.

The United States Court of Appeals has applied these same principles to reject an argument that the racial disparities in the federal DNA database made it unconstitutional.[9] No matter how disparate the impact of the database, without indications that Congress enacted it in order to adversely affect African-Americans, the challenge failed. 

A second policy-level set of decisions also creates racial disparities:  the allocation of law-enforcement resources.  The clearest big-picture example of this is the so-called "war on drugs," which is largely responsible for filling our prisons with men and women of color over the last 30 years.[10] A war on securities fraud or tax evasion would result in the arrest and prosecution of a very different demographic.  Yet resources for combating white-collar crimes have been cut, despite evidence that violations are common and devastating to our society, as evidenced by the current global impacts of finance fraud.  On a smaller scale, police decisions to conduct buy-bust operations in specific neighborhoods - where undercover officers attempt to buy drugs from people on the street and then arrest anybody who sells them the drugs - mean that the police choose who will be targeted based on what neighborhood is chosen for the operation. These operations usually occur in poor, urban neighborhoods with large minority populations. 

Racial disparities also enter through racial profiling by individual officers.  Studies have shown that some mixture of unconscious racism, conscious racism, and the middle-ground use of criminal profiles often leads law enforcement to focus its attention and authority on people of color. This can include everything from discriminatory enforcement of traffic laws to detainment and arrests of people of color without sufficient individualized suspicion.

As with challenges to legislative actions, challenges to racial profiling under the Constitution are extremely difficult because of the need to show discriminatory intent.  The difficulty is magnified because the law gives police officers substantial discretion as to who they approach, stop, question, or search. The Supreme Court has held that the police may lawfully make pretextual stops - for example, singling out one speeding driver among many because the officer has a hunch that they may be carrying drugs. This means that, although the police may not stop a person based solely on race, there are many explanations an officer can give if called upon to explain a stop: the driver or passenger's nervous glance, reduced speed upon seeing the officer, a pedestrian wearing a heavy coat on a warm day, and so on. None of this behavior alone would justify the stop of a car, but such seemingly innocent actions are enough to justify the officer's decision to stop this particular car for driving a few miles per hour over the speed limit while ignoring all the others that did the same, or to stop a particular individual for jaywalking while ignoring similar violations.  Even if a court determines that an officer did make a stop based on nothing more than the driver's race, the only remedy is the possibility of a civil suit against the officer. Unless serious harm was done, this is highly unlikely to occur, so it is not a significant deterrent to such police abuse of power. 

Arrestee sampling adds another incentive for police officers to make questionable or outright illegal arrests.  Whether or not the arrest leads directly to charges being filed, the arrestee's DNA profile will automatically be included in the database and run against all crime-scene evidence, now and in the future.  Because of the barriers to having DNA samples removed, few arrestees will be able to have their samples and profiles expunged, thus allowing a single law enforcement officer the power to place people under lifetime genetic surveillance. 

The low level of proof required to make an arrest, combined with the difficulties of preventing arrests that are illegal for lack of proof or for discriminatory enforcement of laws, means that allowing DNA collection immediately after arrest will lead to large databases full of innocent people.  Furthermore, given the ubiquity of racial profiling, people of color will largely populate the databases.  The bottom line is that police end up with enormous discretion to determine who is in a database, with absolutely no review of many of their arrests. The consequence of the arrest of a plainly and indisputably innocent person will be not only a short stint in jail, but a lifetime of genetic surveillance.                                             


Michael Risher, JD, is a staff attorney at the American Civil Liberties Union of Northern California where he handles a wide range of cases involving freedom of expression, criminal justice, and other civil liberties issues.  Risher was a Deputy Public Defender in Alameda County from 1998 to 2005.



1. California Department of Justice, Division of California Justice Information Services, Bureau of Criminal Information and Analysis, Crime in California 2006 Data Tables, Table 37, available at (hereinafter "Crime in California"). 2006 is the most recent year for which complete data are currently available. 

2. In the last 7 years, the number of states that require every person convicted of a felony to provide a DNA sample has gone from 22 to 50. Compare State Laws on DNA Data Banks Qualifying Offenses, Others Who Must Provide Sample (2009) with Fighting Crime with DNA (2002), both available from the National Conference of State Legislatures at 

3. See Cal. Penal Code § 295(e), 28 C.F.R. § 28.12(f)(1); 73 Federal Register 74935 (December 10, 2008) ("the states that collect DNA samples from arrestees typically do so by swabbing the inside of the person's mouth (''buccal swab''), and many states use the same method to collect DNA samples from convicts"). 

4. Former Cal. Penal Code § 290.2, enacted by Stats.1989, ch. 1304, § 1.5, pp. 5176-5178. For a discussion of the enactment, amendment, and eventual replacement of this provision see People v. King, 82 Cal.App.4th 1363, 1369-70 (2000). 

5. Former Cal. Penal Code § 290.2; see People v. King, supra, 82 Cal.App.4th at 1369-70. 

6. See Crime in California, supra, Table 42. Because these data do not distinguish between felony and misdemeanor convictions, they overstate the number of people who are convicted of felonies. Some of this overcount can be corrected by subtracting the number of people sentenced only to jail, which indicates that the person was convicted of a misdemeanor. See Cal. Penal Code § 17. 

7. See Crime in California Table 37. This ratio comports with national statistics, nationally, less than 20% are violent crimes or residential burglaries. United States Department of Justice, Bureau of Justice Statistics, State Court Sentencing of Convicted Felons, 2004 - Statistical Tables, Felony Sentences in State Court, available at 

8. There are serious questions about whether compulsory seizure of DNA from people who have merely been arrested for crimes violates the Fourth Amendment to the U.S. Constitution, which prohibits unreasonable searches and seizures. See In re Welfare of C.T.L., 722 N.W.2d 484 (Minn. App. 2006) and United States v. Purdy, 2005 WL 3465721 (D.Neb.,2005) (both holding that arrestee testing violates the Fourth Amendment). 

9. Crime in California, supra, Table 31; U.S. Census quick facts, supra.

10. Cal. Penal Code § 299. 


Search: GeneWatch
Created in 1999 by the Council for Responsible Genetics, the Safe Seed Pledge helps to connect non-GM seed sellers,distributors and traders to the growing market of concerned gardeners and agricultural consumers. The Pledge allows businesses and individuals to declare that they "do not knowingly buy, sell or trade genetically engineered seeds," thus assuring consumers of their commitment.
View Project
Created in 1999 by the Council for Responsible Genetics, the Safe Seed Pledge helps to connect non-GM seed sellers,distributors and traders to the growing market of concerned gardeners and agricultural consumers. The Pledge allows businesses and individuals to declare that they "do not knowingly buy, sell or trade genetically engineered seeds," thus assuring consumers of their commitment.
View Project