Sie sind auf Seite 1von 78

California Law Review, Inc.

The New Forensics: Criminal Justice, False Certainty, and the Second Generation of Scientific Evidence Author(s): Erin Murphy Source: California Law Review, Vol. 95, No. 3 (Jun., 2007), pp. 721-797 Published by: California Law Review, Inc. Stable URL: http://www.jstor.org/stable/20439109 . Accessed: 20/05/2013 12:35
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

California Law Review, Inc. is collaborating with JSTOR to digitize, preserve and extend access to California Law Review.

http://www.jstor.org

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

The New Forensics: Criminal Justice, False Certainty, and the Second Generation of Scientific Evidence
Erin Murphyt
newforensictechnologies such as DNA Accountsof powerful typing, data mining,biometric scanning,and electroniclocation tracking praise thesetechniques for fill thedaily news.Proponents helpingto exoneratethose wrongly accused, and for exposingthe that reliedtooreadily failings of a criminal justicesystem previously evidencelikehandwriting, and hair uponfaulty forensic ballistics, and fiberanalysis. Advocatesapplaud the introduction of a "new paradigm" for forensicevidence,and proclaim that thesenew will revolutionize and how thegovernment techniques investigates tries criminal cases. While the new forensic sciences undoubtedlyoffer an unprecedenteddegree of certainty and reliability, these characteristics alone do notnecessarily render them lesssusceptible In as this Article to misuse. fact, most laudedattributes argues,the of thesenew forms may actuallyexacerbatethe of forensicevidence

Copyright C 2007 CaliforniaLaw Review, Inc.CaliforniaLaw Review, Inc. (CLR) is a Californianonprofit corporation. CLR and theauthorsare solely responsiblefor thecontent of their publications. of Law, University ofCalifornia, t AssistantProfessor Berkeley,School of Law (Boalt Hall). J.D.,HarvardLaw School, 1999. I owe a tremendous debtof gratitude toDavid Sklansky forhis infinite as well as to FrankZimring,Eleanor Swift,Jan wisdom, patience,and insight, Vetter,Jonathan Simon, andChuck Dr. Michael Eisen, and Weisselberg.Dr.Montgomery Slatkin, attorney Bicka Barlow also providedgenerousassistance.Special thankstoAndrea Roth, Todd Edelman, Tim O'Toole, and Eliza Platts-Millsfortheirinvaluablecontributions throughout the process. This work also benefitedfromthe thoughtful commentsof Jennifer Mnookin, Rick Lempert,Sam Gross, Rich Friedman, Brandon Garrett,John Monahan, LaurensWalker, Paul Ed Cheng,Michael Saks, Bill Stuntz,and especiallyCarol Steiker. Giannelli,Bill Rubenstein, Finally, BenWolff,DebbieWon, andLisa Cisnerosprovided wonderfulresearch support.

721

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

722

CALIFORNIA LA WRE VIEW

[Vol. 95:721

conditions that sciences tofall into forensic firstcaused traditional disrepute. Articlechallengesthe This neworthodoxy offorensic science.In so doing,itreframes the debateabout theroleofforensic evidencein thecriminal justicesystem in three respects. First, this Articlesets fortha new taxonomy of forensicevidencethat distinguishes first from second generationforensic sciences. Second, using this framework, this Article illustrates how the particularcharacteristics of the second generation aggravate, rather than relieve, the pathologiesthat the ultimately afflicted first generation. Lastly, this Articlecriticizes current the suggestions for improving use offorensic peculiar characteristics and advocates of the second generation, alternative remedies tailored tothese concerns. specific
INTRODUCTION

evidence in the criminal justice system thatfail to account for the

Forensic science has long capturedthepublic's imagination of criminal From Holmes's trademark to Sherlock magnifying glass the shaky justice.' on theransom note forthe handwriting double Lindbergh baby to theswirling helix of DNA, images of themystical power of forensicscience pervade one of the most-watchedtelevision dramas in the popular culture. Currently, is "CSI: Crime Scene Investigation," and viewersnot satisfied country with just
this offering can also tune in to "CSI: NY," "CSI: Miami," or any number of

otherprogramsshowcasing forensicscience technologies.2 As one scholar the future . . . is largelyto ten consider observed presciently years ago, "[t]o talkabout thecreepingscientization of factualinquiry."3 Today, that"creep"
pours forth in a flood as legal scholars across a variety of disciplines wrestle

with questionsrelatedto science in thejudicial system.4

1.

Black's

esp., evidence Black's Law evidence," application 2. Post,

is Arresting, Wash. the Science Shows, is hip. Science is popular. Science grabs ratings."). Sept. 19, 2004, at N01 ("[S]cience R. Damaska, Evidence Law Adrift 3. Mirjan 143 (1997). In particular, much attention has focused upon 4. the increasing use of empirical to help formulate legal policy. See, e.g., Tracey L. Meares evidence & Bernard E. Harcourt, Transparent 90 J. Crim. and Adjudication L. & Criminology and academic element Social 733, debate Science 735 Research in Constitutional and Criminal of (2000) that treats ("We are calling social scientific for a mode

of scientific or technical knowledge. See, e.g., Jennifer Frey, On Crime-Scene

as "[e]vidence Law Dictionary defines "forensic evidence" used in court; arrived at by scientific or technical means, such as ballistic or medical evidence." Dictionary 597 (8th ed. 2004). This Article uses "forensic evidence," "scientific to to and "forensic science" refer evidence derived from the interchangeably

Foreword: Procedure,

judicial decision-making as a crucial assessment

in constitutional

procedure decisions more transparent."); David to the Law as Science and Policy, Value of Social Science

empirical criminal thereby making decision-making, L. Faigman, To Have and Have Not: Assessing the 38 Emory L.J. 1005 (1989). On efforts

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

723

Of course, traditional forensic evidence, such as handwriting, firearms, and fingerprint has longplayed a role in the bullet, bite, toolmark identification, criminal justice system. But currently on thehorizonare a new generation of and inculpating forensic sciences capable of uncovering criminal offenders at an order of magnitude greater than that affordedby traditional forensic newmethods-such as DNA typing,5 This arrayof exciting techniques. data amarked and biometric mining, locationtracking,7 technologies8represents of old, andwill surelystakea central advanceover therudimentary techniques administration of criminal and indispensable role in thefuture justice.

of expert evidence). to include both nuclear DNA 5. By the phrase "DNA typing" I mean analysis, which as thirteen locations on the genomic strand for repeating sequences of DNA, typically examines two specific regions of themtDNA DNA well as mitochondrial typing, which typically sequences v. South Carolina, 126 S. Ct. 1727 (2006) strand. See, e.g., Holmes (vacating conviction where state held defendant's inadmissible evidence of a third party perpetrator because DNA evidence suggested

Wis. L. Rev. 1113, 1208 (1991) (proposing reforms Evidence, 1991 designed to improve quality

tomollify

the "battle of experts"

so often waged

in civil court, see, e.g., Samuel

R. Gross, Expert

guilt). 6. "Data mining" typically refers to pattern analysis of large quantities of data, and is perhaps better described as a technique or technology rather than a "science." By way of example, the government may check phone records to isolate individuals who frequently call certain foreign those names against flight registry lists. I also use this term to countries, and then cross-check include more 7. Global records. generally the analysis of computer database-generated are currently in use, including satellite-based range of location tracking devices or electronic bracelets), radio Positioning System monitoring (through cell phones A identification location). cell phone

frequency approximate (approving

LEXIS 2500 at *9-10 (Cal. Ct. App. Mar. 18, 2004) (using FasTrak records to document
defendant's Cell Phones 8. travel to site of arson on morning of offense); David A. Lieb, States Seeking to Track Press, Oct. 8, 2005 (detailing pilot programs to for Traffic Conditions, Associated track drivers through their cell phones). "Biometrie between

tags, or cell-site (RFID) triangulation (using cell phone signals to States v. Forest, 355 F.3d 942, 948 (6th Cir. 2004) See, e.g., United site tracking); People v. Ricafort, No. A101531, 2004 Cal. App. Unpub.

matches

use of biometrics, ID Theft, Den v. Post, Jan. 2, 2005 (describing including facial of detecting fraudulent identity card applications); recognition as a means Spencer S. Hsu, D.C. Wash. Feb. Forms Network Post, 17, 2002, at Cl of Surveillance, (describing use of such software during demonstrations, football and other large public games, events); Stephen Targets Thompson,

v. Commonwealth, No. 2005-CA-000815-MR, 1451566 2006 WL identity. See, e.g., Chambers (Ky. Ct. App. May 26, 2006) opinion) (unpublished (noting that the defendant, who gave a false name on arrest, was identified through an iris scan at the jail); Facial ID Technology Makes Gains in Florida, Organized Crime Dig., May 4, 2005 (reporting that use of technology has led to nine months earlier); Arthur Kane, Facial forty-five arrests since implementation Scanning

here refers to techniques that rely upon computer-generated technologies" two samples or between a observed biological either between characteristics, is, in this respect, a venerated "biom?trie sample and a stored image in a database. Fingerprinting forms include facial that, with the advent of databases, has now gone online. Newer technology" recognition or iris pattern analyses that compare digitized images to determine the likelihood of

Facing Security, Tampa Trib., Feb. 9, 2002, at 1 (describing implementation of facial area airport). I would at Tampa the use of software also include under this heading recognition cameras to stored biom?trie profiles?for to record images that are then matched instance, a camera in a government images against recorded building that does not simply record events, but also aims images of suspicious persons that are contained in a database. to compare

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

724

LA WRE VIEW CALIFORNIA

[Vol. 95:721

Many of these new, more reliablemethods have already acquired a as of traditional measure of fameby exposingboth theunreliability techniques tokeep out such failure of thecriminal well aswith theattendant justicesystem illegitimate evidence. Accordingly,it is easy to assume thatthequalities that and desirable will likewiserender make thenewmethods so trustworthy them less susceptible to, if not wholly immunefrom,the problems thatplague traditional sciences. Some criticsof traditional sciences have even touted newmethodologiesas emblematic of a "new scientific certain paradigm" for forensicevidence, one inwhich "untestedassumptionsand semi-informed guesswork [is] replaced by a sound scientificfoundationand justifiable protocols."9 Yet theexperiences forensic sciences in thecriminal of traditional justice new techniques system cautionagainstembracing these withoutanyhesitation. In recentyears, empirical studies and select trial courts have called into of evidentiarystalwarts like handwriting,10 voice question the legitimacy hair and fiber,12 bite and toolmarks,13and even fingerprints.14 exemplars,11 Exoneration studies have demonstrated the shockingdegree to which the from has historically failed topreventthegovernment criminal justice system or fraudulent evidence to aid in the deploying spurioussciences and faulty For example,one studyfoundthat of innocent defendants. defective conviction scientificevidence contributedto over one-half of wrongfullyobtained 15 convictions. It standsto reason that a systemthatfailed to stemtheabuse of untested to safeguard or faulty of forensic the forms evidence mightalso be ill-equipped

J. Saks & Jonathan J. Koehler, 9. Michael The Coming Paradigm Shift in Forensic 309 Science 892, Aug. 5, 2005. Identification Science, J. Saks, Merlin Lessons from the Law's Formative 10. and Solomon: See, e.g., Michael L.J. 1069, 1097 (1998) Encounters with Forensic [hereinafter Identification Science, 49 Hastings formost of this century, a re-evaluation and Solomon] Saks, Merlin ("After standing unquestioned of handwriting Daubertr). 11. identification expertise has resulted from the Supreme Court's decision

in

in Blinded By Science: How Judges Avoid the Science See, e.g., Erica Beecher-Monas, Temp. L. Rev. 97 71 identification also failed the has Evidence, 55, (1998) ("Voice Scientific scientifically valid prong of Daubertr). 12. See, e.g., id. at 86-87 (describing trial court's rejection of hair evidence for failure to meet standards of validation, despite recognition of "long history of admissibility"). Evidence Reference Manual J. ? 1-3.5.1 [2] (Michael Scientific] [hereinafter Annotated (identifying "scores of forensic including bite mark testing that "might not survive empirical test"). techniques," in an Age of DNA Profiling, 61 14. See, e.g., Jennifer L. Mnookin, Fingerprint Evidence L. Rev. of fingerprint evidence and Brook. 13, 39-43 (2001) (detailing the uncritical acceptance 13. Annotated eds., Scientific Saks et al. 2004) to the underlying validity of the technique). chronicling subsequent challenges and the Ethical J. Saks, Scientific Evidence 15. Michael Obligations of Attorneys, 49 Clev. St. L. Rev. 421, 424 (2001) (citing statistics from the Innocence Project indicating that as opposed to fraudulent, forensic science errors play a factor in 63% of wrongful unintentional, conviction forensic cases); see also science errors). id. at 424 (observing that one-fourth of the cases involved fraudulent

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

725

use of more robust,complicatedforms of such evidence,both in terms of assuringits integrity and fostering healthyscientific development.In fact,as this Article argues, thevery traits that make thisnew generation of forensic evidenceso promising servetoraiseconcerns about the use of suchevidence in the future. The seriesof scandals thathave alreadybesieged DNA typing, of thesecondgeneration, arguablythe most sophisticated technique underscore the urgency of this claim.16 Accordingly, this Article sets forth three forensic challengestocustomary evidence in thecriminal to thenotion thatall justice system. First,in contrast forensicsciences share the same essential traitsbut simply range on a more reliable,this continuum fromless to Articledrawsclear categoricallines betweenfirst-generation and second-generation forensic techniques. Article looks to thehistorical of first-generation Second, this experiences forensicsciences in the criminaljustice system to anticipatethe future of evidence. second-generation Contraryto conventional wisdom,which roundly endorses second-generation as superiorto their techniques much-discredited predecessors,this Article argues thattheverycharacteristics thatinstillsuch confidencein the second generation-theirtechnical relianceon complexity, databasing,and breadthof application-in factaggravatetheconditionsthat causedwidespread failuresin thefirst ultimately generation. Thus, thesecond will face thesame concernsabout integrity generation and qualitycontrolthat permeatethefirst generation. Third, this Article examines theproposals typically advanced to improve theuse of forensic evidence in the criminaljustice system, and asserts that these approaches fail to account for theparticular demands of the second generation. This Article contendsthat our current models of criminal justice, even operating at optimaland idylliclevels,cannotadequatelysafeguardthe evidence. widespreaduse of technically sophisticated, highly probative Thus, in thisage of powerfuland pervasivenew forensictechnologies, the criminal justice systemmust reckon anew with how it accommodates scientific evidence.This Article therefore measures specifically proposes responsiveto theconcernsraised by second-generation evidence. Part I definesthetwogenerations of forensic evidence,and illustrates the of second-generation particular characteristics sciences using the most DNA typing. Part II identifies thetwo"fronts" onwhich developed technique: thebattle forqualityassurance iswaged and lost with respectto all forensic evidence: the governmentlaboratory and the courtroom. This Part then demonstrates why the stakes are particularlyhigh and the challenges particularlyacute for the methods of the second generation.Part III conventional acknowledges proposals to improve thecriminal justice system's

16.

See

infra notes

149-156.

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

726

CALIFORNIA LA WRE VIEW

[Vol. 95:721

of forensic processing evidenceand noteshow theyfail to address thedistinct of the second generation. characteristics This Part then sketches solutions to thoseconcerns. tailored

A NEW TAXONOMY OF FORENsIC EVIDENCE

A. Defining the First and SecondGenerations The listof traditional forensicsciences is long and familiar:it includes voice analysisof bite and toolmarks, hair and fiber, ballistics,handwriting, exemplars, and fingerprints. But althoughthesetechniques have longappeared in criminal cases, they have arisenonly inan occasional and sporadicfashion, and usually in a supporting role to other forms of evidence like eyewitness own confession.17 For a myriadof reasons, none testimony or thedefendant's of these first-generation methods ever occupied the full field of criminal adjudication. First, traditionalforensic techniqueshave limited application. They fit For example,handwriting of offenses. typically discretecategories analysis can only aid thosetypes of cases inwhich a writing is at issue,and ballistics evenwithin thedischargeof a firearm. only thosecases involving And then, thosenarrowcategoriesof case type,only a fraction of cases will actually produce forensic evidence.Ballistic evidence requiresthat bullets actuallybe recovered;hair or fiberevidence, even if present, may be easily lost or 18In short, of potential cases amenable to first-generation overlooked. therange both in theory and inpractice,remains evidence, quite limited. Second, first-generation techniquesare experientialand observational, 19 They are neither conceptually rather than technical or experimental. most of those who analyze such nor scientifically complicated rigorous. Indeed, And because first-generation evidencehave no advanceddegreeof anykind.20
17. Terrence F. Kiel y, Forensic Evidence: Science other evidence, and the Criminal Law 56 to

(2006) (reporting that forensic science "along with reconstruct the events surrounding the crime). 18. 19. See Id. at

is used

circumstantially"

and that "a majority of the forensic sciences do not disciplines supported by modern microscopy," rest upon any core scientific or mathematical principles"). to Avert the Ultimate the Forensic 20. Science Community Craig M. Cooley, Reforming L. & Pol'y Rev. 15 Stan. 381, 425-26 & n.271 (2004) Injustice, (noting lack of doctoral or forensic science in the United States, and only two in the world); in criminalistics programs Legal Brendan Koerner, Under the Microscope, Affairs, ("An increasing July/Aug. 2002 in chemistry or molecular and number of forensic scientists hold graduate degrees biology, the University of West Virginia interdisciplinary programs are cropping up at colleges; the science of identifying recently offered the nation's first-ever four-year degree in biometrics, humans by unique physical traits like iris patterns and hand geometry. These students, however, in newer techniques like DNA forensics is still dominated testing. Traditional typically specialize rigorous

id. at 136 (noting difficulty of "[e]fficient and correct fiber recovery"). are observational that "most of the forensic sciences 180 (acknowledging

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

727

most of their techniquesare intuitive, laypeople can readily comprehend results: most people can quickly grasp the notion of matching fingerprint orbulletgrooves. 21 ridges, handwriting slants, techniques are mechanically Third, and relatedly, first-generation unsophisticated. machinery or They do not employ complex interpretive of and rarelyraise questions concerningthe protection instrumentation,22 In fact, proprietary information. most techniques never enduredany rigorous validation testingor study, and so there exists little "science" worth For instance, hair and fiberanalysis needs little more than a protecting.23 and handwriting no microscope and basic chemicals, analysisrequires virtually are alsomore amenable todefense-side Firstgeneration equipment. techniques testing and thecultivation of local expertise, given that criticalanalysisof the technique requireslittle instrumentation and minimal training. Fourth,first-generation techniques are reactive and self-contained in their scope. The analysisof hair,handwriting, investigative fiber, bullets,firearms, a "suspect" for voiceprintsand so on all require that the police identify in theform of an individual personor inanimate object. comparison-whether For instance, ballistics and firearm identification calls for the recovery of a bullet and the identification of a suspected weapon. Likewise, handwriting or hair analysisworks only after isolatinga potentialsuspector match. First forensic a suspect in thefirst sciences lack thecapacityto identify generation thedefendant's to connection theyinstead operate mainly to confirm instance; 24
a crime after other evidence has already identified him as the perpetrator.

by ex-cops 21. L. Rev.

and examiners See, e.g.> David

'educated

DNA Profilingand Toxic Torts,61 Brook. Epidemic ofScientific Began with Sophisticationthat
Ill, characteristics 117 (2001) (noting ballistics that are visible); Mnookin, of fingerprinting science in the Law?Confronting and fingerprinting analysis both result in matching supra note 14, at 32-33 (identifying the "cultural as a contributing factor to its uncritical acceptance); theNew Challenges L. 108 Harv. of Scientific Evidence,

L. Faigman,

at the school of hard knocks.'"). The Tipping Point in the Law's

Use

of Science:

The

plausibility" Developments whereby

Rev. 1481, 1503 (1995) (chiding judges forapplyinga "show-and-tell" principleof admissibility,
a court admits scientific evidence that is intuitively visually comprehensible, such as fingerprint or handwriting, but excludes thatwhich requires interpretation, such as polygraph). 22. Kiely, supra note 17, at 6 (noting that "the great number of the traditionally aided by employed forensic sciences are, in effect, based on and centered in close observation,

modern

id. at 180 (remarking that "a majority of the forensic sciences do not microscopy.,.."); rest upon any core scientific or mathematical principles"). 23. Forensic 34 J. L. Med. & Ethics See, e.g., Paul C. Giannelli, Science, 310, 311

"lack of empiricalsupport" fortraditional (2006) (noting techniques).


24. sciences. sciences such as fMRI

generation techniques generation

In this respect, some technologies, both new and old, bridge the first- and second or neutron activation Old like voiceprints standards and new analysis,

as an respects, itwas available in Fingerprinting has historically shared the characteristics of the first generation: a limited number of cases, had little investigative capacity, and was based on very little scientific validation. increased attention on validating the methodological of However, underpinnings

imaging or digital cameras, embody some of the characteristics of first and some of the second generation. Fingerprinting might be viewed, in some illustrative bridge between the first and second technology generation.

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

728

LA CALIFORNIA WRE VIEW

[Vol. 95:721

Fifth, and finally, because first-generation techniquesare capable of supplying only a narrowslice of information, theytypically do not implicate greater questions of personal privacy. That is, to the extent that these techniquesreveal any information, it tends to be limitedto the facts and circumstances of thecase or suspectat hand.Althoughefforts were oncemade to draw conclusions about race, ethnicity, or mental state fromstudies of fingerprints or handwriting,these endeavors have now been largely Generally speaking, first-generation forensicsciences such as discredited.25 handwriting, fingerprinting, ballistics,bite or tool mark, and fiber analysisdo not reveal information about mental or physical health, biological or demographic characteristics, family relationships, or other intimate information. In contrast, DNA typing, second-generation like datamining, techniques, and biometric location tracking (such as cell site,GPS or RFID tracking), sharecharacteristics thatstarkly scanning(such as irisor facial recognition), differentiate from thefirst them generation. First, second-generation techniquesapply to a wide varietyof offense For example, of cases within those types. typesas well as to a largenumber tocrimesranging DNA typing notonlycan generate from the evidencerelevant most gruesome to the pettiesttheft murder,but italso can do so in a greater percentage of such cases than could its first-generation counterpart, As other second-generationtechnologies like location fingerprinting.26 becomemore sophisticated, thenit tracking, datamining,or biometric scanning is likelythatthey will also becomemore prevalent.It is easy to imaginea

fingerprinting fingerprinting

has pushed it toward the second-generation category. the advent the second generation with formally entered which the record Identification (AFIS), computerized System Fingerprint for greater use of printing as an investigative tool. Simon A. allowed techniques Identification 25. and Early the Criminal Justice

More of

importantly, the Automated and thus

keeping

AFIS). (David Lazer ed., 2004) (describing

System,

in DNA

and

the Criminal

Cole, Fingerprint 74 Justice System

ethnicity, heredity, Identities Suspect that prisoners Some

or the studyof to "graphology," Daubert World, 66 UMKC L. Rev. 251, 259 (1997) (referring
handwriting to reveal personality traits). F. Liotti Thomas & Jamel 26. Lawrence Kobilinsky, Oeser-Sweat, Applications 6 (2005). Compare Forensic & Legal also, e.g., John M. Butler, DNA: Forensic

showed certain print characteristics more frequently than the general population). handwriting analysts also claim the ability to discern facts about individual personality or in the Post emotion by studying handwriting. Andre A. Moenssens, Identification Handwriting

that fingerprints could reveal race, of fingerprinting believed developers and other biographical data, including potential criminality. Simon A. Cole, 103-09 (2002) (reporting, for instance, on claims by one French researcher

DNA Typing 34 (2d ed. 2005) (listingvarious sources of DNA), with Cole, Fingerprint
and the Criminal Justice and the Criminal Justice System, in DNA System, Identification In some supra note 24, at 73 (outlining limitations of fingerprinting as a forensic technique). can render first-generation for methods irrelevant: respects, techniques second-generation there is a fingerprint or hair evidence instance, where to conduct DNA sufficient genetic material typing. or a handwriting sample, there is often

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

729

"EZ Pass" cards, future in which evidenceculled from cell phones,computers, or in and smartidentification which images cardsbecomesmore ubiquitous,27 from a security camera linkedto a database facialrecognition system are used toconvicta hostof offenders acrossa broad spectrum of crimes.28 Second, the second-generationtechniques differ from their first in their generation counterparts scientific robustness. Building and applyingthe methodologiesof DNA typing, biometricscanning,or location monitoring As a result,second requireshighly specialized knowledge and expertise. evenwhile the generation techniques are intuitively inaccessibleto laypersons, of such results methods are typically The underlying viewed as highlyreliable. rigorof second-generation sciences also supportsstronger claims of their probative value thanthat offered by first-generation some second techniques; generation methods purporteven to provideproof to a degree of scientific In short,unlike first-generation certainty. methods that largely rely upon intuitive methods thatlead to findings of general inclusion, second-generation sciences use technicallysophisticated methods that provide individuated related with the findings highestlevelsof confidence. Third, this of second-generation methodologicalsophistication techniques is mirroredby a complementary mechanical sophistication. Whereas first generationsciences reliedon tools no more complicatedthana magnifying glass or microscope, the tools of the second generation are far more costly, For example,DNA typing elaborate,and incapableof ready reproduction. requirescomplex machinery,chemicalkits, and computersoftware;location use software tracking dependsupon satellites and and cell-towers; biometrics image scanners. Conducting an independentanalysis, then, requires a significant capitalexpenditure. Thismechanical sophistication a relatedcharacteristic highlights of the second generation.Namely, because such techniques rest on such a the disclosure or deployment of the technologies complicatedarchitecture, underpinning second-generationtechniques may raise concern about proprietary interests. DNA typing has already weathereda seriesof challenges relatedto thereluctance of privatecompanies to divulge claimedproprietary secrets, such as the chemical sequences used to conduct the analysis.29

Train In Analyzing See, e.g., Andrew Glazer, Police Nationwide Gang Websites, to communicate Press, July 10, 2006 (describing use of websites by gang members information about illegal activities). 28. Of course, counternarratives surrounding such evidence may always be constructed. Jennifer L. Mnookin, The Image of Truth: Photographic 10 and the Power of Analogy, Evidence Yale J. L. & Hum. 1, 4, 17-22 (1998) (noting that initial enthusiasm for photography as a tool of Associated perfect proof truth succumbed that is a "human 29. can be "a potentially misleading to reality that photography form of representation," rather than simply a "direct transcription" of reality). See, e.g., People v. Bokin, No. 168461, slip op. at 15 (Cal. Super. Ct. May 5, 1999) studies against company's claim litigation by defense to obtain data regarding DNA information that should not be disclosed). Google recently made

27.

(addressing that data constituted proprietary

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

730

LA WREVIEW CALIFORNIA

[Vol. 95:721

Similarly, location trackingor biometric devices rely on technologies as intellectual property. developedandprotected most importantly, second-generation sciences-unlike Lastly,andperhaps upon computerized databases to store their first-generation counterparts-rely techniques are therefore not largequantities of information. Second-generation Rather,second just reactive, or limited toconfirming a knownsuspect'sguilt. a suspectab initio-and generation methods are proactive,and can identify even serve to supply theonly evidence of guilt. "Cold hits" in thenational DNA database have both identified suspects in cases with corroborating on thebasis of thegenetic evidence,as well as been used to secureconviction data mining, or biometric information alone.30 Similarly, location tracking, technology may potentiallyeither identify a suspect about whom may be developed additionalevidence, or supportprosecutionon the basis of the scientific evidencealone. raisesa related characteristic that differentiates This database-dependency techniques. Namely, second-generation methods first fromsecond-generation can gravely impacttheprivacy interests of both suspectsand innocent third parties,whereas first-generation techniquestypicallyreveal no information to theevidentiary other thanthat question at hand.A handwriting pertaining more thanthatsomeone reveals little match it.To might sampleor fingerprint theDNA samples collected by the government contain the the contrary, of her existence-and individual'sentiregenetic code-a veritableblueprint even a typedprofile has the power to reveal familialassociations.Other such as the technologieslikewise threaten to compromise privacy interests, collectionand compilation of imagesfora biometric database,or government usersof a certain cell tower of cell sitedata to revealeitherthe theinvestigation undertaken user.And of course, even or thecommunications by a particular such material assuming the government manages to collect or investigate createsa withoutundermining individual process itself privacy,theadversarial with DNA or cell-citeor biometric Once presented secondwave of concern. would be to seek access data, thefirst logical step forany defense investigator whether the to the DNA, cellular, or biometric database to determine conclusion reached is reliable,and whether it containsany otherplausible to individual thethreat privacy. perpetrators, enhancing thereby To illustrate and to convey thesecharacteristics of thesecondgeneration, of criminaljustice, thenext how they will affect the future administration This science:DNA typing. sectionexplores thearchetypal second-generation

the news by contesting a government information from their aimed at data-mining subpoena its trade secrets." popular website, claiming that disclosure of such information would "jeopardize Katie Hafher & Matt Richtel, Google Resists U.S. Subpoena of Search Data, N.Y. Times, Jan. 20, 2006, at Al. 30. See infra note 86.

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

731

Article uses DNA typing as an illustrative example largely because it is the most developedof second-generation sciences,and thus offersthe most fertile groundtoexploretheissues that suchevidence may raise. B. TheArchetypal Second-Generation Science:DNA Typing DNA typing debutedas a forensictool in 1985,when SirAlec Jeffreys recognizeditspotentialtoanswerquestionsof identity centralto theresolution of criminalcases.31After its birth in theUnited Kingdom, the technique quicklyjumped thepond, and by 1988 itappeared in the United States in the firstreported appellate case.32 Since then,thepower of DNA science has even defense dazzled every factionof the criminal justice community, attorneys. which took place in theUnited Consider the followinginvestigation, offan overpasshit a truck Kingdom: a brick thrown passingbelow,killingthe a smallquantity driver.Investigators had no leadsotherthan of blood found on thebrick, A searchof thenationwide which in turn yielded a DNA profile. database containingover twomillion profiles revealed no directmatches. However, a "familial"searchof the same database,which looks forprofiles thatcorrelate highly to theevidentiary profile, yielded a lead. Investigators and thenfoundthesuspect, who of thesuspect, followedthe lead to a relative later confessedandwas convicted.34 Althoughthe perpetrator's profile was not in thedatabase,his relative's whichwould approximate his profileat a profile,

31. in family court Mnookin, supra note 14, at 40. DNA typing has also found application Rivers cases, where paternity is at issue, and even in civil matters. See, e.g., Alabama Tombigbee 2002 WL 227032 Coalition v. Norton, No. CIV.A.CV-01-S-0194-S, (N.D. Ala. 2002) (discussing Species Act that involved DNA challenge to listing of species on Endangered testing of contested

fish).

32. 33.

Andrews Law

v. State, 533 So. 2d 841 (Fla. Dist. Ct. App. enforcement officers approve the ease with

1988). which

DNA

evidence

can

be

its appeal to the public at large collected, processed, stored, and searched. Prosecutors appreciate and the degree to which it is essentially unassailable in court. Judges welcome the scientific rigor with which itwas developed and is usually applied. Even defense lawyers have largely embraced DNA those wrongly technology and sought wider use of DNA testing as a means of exculpating supra note 17, at 6 (reporting "considerable suspected or convicted of offenses. See, e.g., Kiely, . . . such as enthusiasm for the power and potential of twenty-first century scientific advances

DNA director of the FBI lab, has observed wryly that "[o]ne attorney research"). Bruce Budowle, . . . had of DNA [the] position that thousands of innocent people are in jail because typing" and of innocent people are in jail because of no DNA "[t]hat same attorney" thinks that "thousands supra note typing." David Lazer, Introduction: DNA and the Criminal Justice System, in Lazer, observations nicely illustrate both sides of the DNA 24, at 3-4. While Budowle's coin, they fail to the significant differences between exculpatory DNA typing and inculpatory DNA acknowledge

evidence to exclude a suspect has never been in serious dispute?by typing. The power of DNA suspect. analogy, it is easy to determine that a type AB blood sample did not come from an 0+ issues to conclude that a particular 0+ But it raises farmore contestable suspect is the precise, or even a probable, source of the sample. DNA 34. Matthew Falloon, (London), Apr. 20, 2004. Traps Brick Thrower Who Killed Lorry Driver, The

Guardian

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

732

CALIFORNIA LA WRE VIEW

[Vol. 95:721

would the profileof an unrelated individual, much higher frequencythan to theright directed officers person.35 As thisanecdote illustrates, DNA typing has thepotentialto transform bring officersapprehendsuspects,how governments how law enforcement And as one of the most secureconvictions. prosecutions, and how prosecutors sciences,DNA typing provides excellentclues developed second-generation will changecriminal justice,andwhat potential into how thesecondgeneration will raise. concernssuchevidence 1.A High VolumeofCases witha ForensicEvidenceComponent evidence transforms revealshow second-generation StudyofDNA typing Because second-generation thenatureof proofwithin the criminalsystem. are robust and arebroadlyapplicable,caseswith a techniques methodologically forensic evidence component are likely to displace cases without such ina criminal with a large volumeof cases docket evidence, ultimately resulting evidence. involving forensic DNA typingtechniques were both cumbersome For instance, although scientific advancesnow allow rapid when first and expensive conceived,recent at a rate conducive towide-scale use of DNA processing and turnaround Whereas processing used to takeweeks, if not months,with evidence.36 it is expectedthat will soonbe able toprocess analysts robotics and automation In timefor United States,theaverage turnaround up to800 samplesa day.37 the a DNA requesttodayis stilltwenty-three weeks in statelaboratories and thirty

35. familial Denver Dec. Rape

The searches

FBI

has recently changed its own information-sharing policies in the United States' national database. See Mark Hansen, Match Got the FBI to Change Policy and Release Kinship DNA,

to permit such Point: How a 92 A.B.A.J. 48,

Probe

or reagents includes the costly chemicals This estimate Report]. [hereinafter National or equipment. Id. to do the tests, as well as salaries of analysts, but not overhead necessary test in a criminal case as $568.96, while in Another report listed the cost of an in-house DNA of was a Office State Budget & of known offender house $7.58. only sample testing and of Justice, Cost Department Study of DNA Testing North Carolina Management,

Grant 2002-LT-BX-K 003 (2003), at 34, http://www.ncjrs.gov/pdffilesl/nij/grants/203970.pdf

not only reduces it also the time associated with processing DNA; the precise expense of processing DNA it is difficult to calculate the costs. Although a simple, results, some estimates exist. For instance, one report labeled the cost of analyzing Michael Travis C. P. Lovrich, J. Gaffney, typical rape kit as roughly $1,100. Nicholas DNA Forensic Study Report, U.S. Dep't Pratt & Charles L. Johnson, National of Justice reduces

20, 2006. 36. Automation

Analysis 7, Table 4 (2006).

37. Shaila K. Dewan, As Police Extend Use of DNA, A Smudge Could Catch a Thief, N.Y. in state or Times, May 26, 2004, at AI. DNA testing in the United States is largely conducted of law local laboratories. National Report, supra note 36, at 15 (reporting that 80.1% 11.7% use a local agency enforcement agencies use a state laboratory to process DNA evidence, an state laboratories process laboratories). Currently, laboratory, and only 2.9% use private average of 1,284 samples at 28-29. a year, whereas the local laboratories process an average of only 111. Id.

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

733

thenationalcrime laboratory in weeks in local laboratories.38 By comparison, the United Kingdom, theForensic Science Service ("FSS"), tolerates only a timebetween submissionand return twenty-four day turnaround of forensic Offendersamples-those drawn in ideal conditionsfrom samples.39 a known Miniaturization single source-typically take theFSS fivedays.40 processes ofDNA that will soon enableon-sceneanalysis takes only seconds.41 Moreover, advances in collection techniques allow technicians to gather of sourcesthanin thepast. and from a greater samples less intrusively variety thenucleusof cells, is found not only in NuclearDNA, which is obtainedfrom blood but also inhair follicles, skinscrapings, and saliva containing skincells. Buccal swabkits, which demandnomore than a painless scrapeof theinside of a suspect'scheek,are increasingly sensitive and render clear, typeable results. Furthermore, whereas in the past forensic scientists oftenrequired a significant can now generate of biological amount as few material,scientists profilesfrom Modern techniques as six cells, a quantity not even visible to thenaked eye.42 bit of biologicalmaterialand duplicateit to allow analyststo take thesmallest createa testable examine analyststhen multipleplaces, or loci,on a quantity;43 And harnessingsophisticated genetic strandat the same time.44 processing techniques,analysts increasingly are able to "pull-apart"forensicsamples mixturesof more thanone person's DNA, and ascribeparticular containing DNA testing genetic profilesto specificindividuals.45 cannot Finally,ifnuclear be performed because a forensicsample containsdegraded or dead cells,

38. 39. 40. and Wales,

National Id. Christopher U.S. Id.

Report,

supra note 36, at 28-29. of DNA at 15, Technology in England

DNA in England]. FSS processes roughly30,000 offender [hereinafter nij/grants/203971.pdf


samples a month. 41. Alec 1039. In a pilot program conducted in the United Kingdom in 2000-2001, experienced LCN in evidence. DNA responded to all stolen vehicle scenes and swabbed for biological technicians were able to England, supra note 40, at 26. The study showed that experienced recover LCN samples from 51% of the scenes they attended. Id. at 27. At present, however, LCN is not typically considered generally accepted for inclusion purposes, because it raises a number of 42. technicians Liotti sensitivity concerns, although it still has value as a method of exclusion. Kobilinsky, & Oeser-Sweat, supra note 26, at 112-13. 43. The technique of Polymerase Chain Reaction allows scientists to amplify (PCR) a more to amount. material measurable genetic readily produce serious 44. systems now allow DNA Multiplexing analysts to express the genetic information stored at several loci in one simultaneous process, rather than run separate tests for each locus. For example, Y-STR 45. differences between upon the chromosomal typing capitalizes J. Jeffreys, Genetic Fingerprinting, 11 Nature Med. No. 10, Oct. 2005, at

Dep't

H. Asplen, The Application of Justice (2003),

http://www.ncjrs.gov/pdfiilesl/

to amplify only themale fragment of the forensic sample. Such a technique aids and women investigators in rape cases, which often involve mixed samples from both a female victim and a Liotti male perpetrator. See Kobilinsky, & Oeser-Sweat, supra note 26, at 113-17. Forensic experts are also seeking ways to deconvolve mixtures of profiles from persons regardless of sex. See, e.g., Butler, supra note 26, at 525.

men

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

734

CALIFORNIA LA WRE VIEW

[Vol. 95:721

mitochondrial DNA typing can oftenrecover genetic information storedin the 46 cell longafter the nuclear DNA has decomposed. Given theease withwhichDNA evidence is recovered, and theadvances incost-effective, of evidentiary efficient processing of largequantities samples, it is nothard to envisiona future in whichDNA testing plays a centralrole in criminalinvestigation and adjudication. Today, thepublic imagination holds that DNA most commonly applies to theprosecution of seriousoffensessuch as rape andmurder. However, perhapscounterintuitively, DNA evidence may carrytheleastpotential forthesetypes of offenses. Afterall, rapeand homicide cases tendtobe amenable todefenses,including self-defense and consent,that DNA evidence either irrelevant render or less dispositive.47Instead,DNA evidencemay hold the greatestpromise in solving low-level crimes like property andpossessionoffenses.48 For instance, property offenses presently constitute an enormous volume of criminal complaints and costbillionsof dollarsannually, but carry very low ratesof arrestor "clearance."49 According to theDepartmentof Justice's were roughly Unifonm CrimeReportingstatistics, there 10.4million reports of and automobiletheft in 2003.50Of theestimated1.2 burglary, general theft, million motor vehicle thefts, only 13.1% of motor vehicle crimeswere cleared;51of the roughly7 million larceniesor thefts, only 18.1% were were cleared;52and of the2.2 million burglaries, only 13.1% of burglaries are cleared a much higher of violentoffenses cleared.53In contrast, percentage were only 1.38 million reported violentcrimes by arrest.In 2003, there were and aggravatedassault54 -and 46.5% of them murder, rape, robbery, cleared.55 Thus, although propertyoffenses exact a costly penalty on

mtDNA

one exception may be with respect to child sexual abuses cases. Because 47. Although, there are no consent-based defenses to child sexual abuse, the presence of a defendant's genetic in the body cavity of a child can more or less conclusively material prove a case, even without toHelp Solve Thefts, USA Today, Used in property 19, 2006, at Al (describing how in ten states, the "total number of DNA matches the number of matches in violent crime[] [cases]"). crime cases has exceeded In 2003, States billion dollars from nonarson-related 49. the United lost seventeen Federal Bureau of Investigation, Dep't FBI Releases Crime Statistics of Criminal subjecting 48. a child to the trauma of testifying in court. See, e.g., Richard Willing, DNA Database

Liotti & Oeser-Sweat, See, e.g., Kobilinsky, supra note 26, at 120-21 (noting that is useful in examining hair or "items such as teeth and bone, which are often found to of the high copy but "can still produce good results because contain degraded nuclear DNA" number of mitochondrial sequences within"). 46.

Oct.

property crimes. Press Release, 50. Justice 51. 52. 53. 54. 55. Bureau Id. at 57. Id. at 49, 53. Id. at 45, 47. Id. at 11. Id. at 13.

for

2003 (Oct. 25, 2004) (on filewith author).


of Justice Statistics, Statistics 130 (2003).

U.S.

of Justice,

Sourcebook

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

735

or conviction. they are rarely closed by arrest communities, of DNA evidence can However, earlydata suggestthattheavailability thesenumbers.In a typical month, the UnitedKingdom's radicallytransform or "hit"betweenforensic andknownsamplesin their FSS makes an association one rape cases, and a murder cases, thirty databases in roughlyfifteen As one study and drug offenses.56 whopping 770 motor vehicle, property, 1995 to the end of January 2002, themajorityof April observed, "[f]rom ormurder. Instead,the matcheswere notmade in rape,othersexual offenses, largest numbers of matched crimes were commercial and residential out: in 2000 hits bears this Local experience Virginia,of thefirst burglaries."57 murderor rape, whereas 59% helped inno-suspectcases, only 12% were for or robbery.58 Moreover, data indicate crimessuch as burglary solve property offensesand those who some correlation who commit property betweenthose will likely serve to increase commitviolent offenses.59 This correlation toapprehend offenders. property support for allocatingresources DNA evidence can also have a significant offenses, Beyond property whether simple impacton narcotics-or weapons-based possession offenses, a major volume of crimes Such cases constitute possession or with intent. of Justice studyfrom United States.For example,a Department chargedin the feloniesacross largeurban counties 2000 reportsthat40% of state-charged The viabilityof such were weapon or drug possession relatedoffenses.60 on whetherthe suspect is apprehendedin prosecutions, however,often turns thesuspectnot in actual or visible possessionof thecontraband;61 immediate

56. Conviction

Press Release, Forensic Science Service (UK), of Serial Rapist, available (Mar. 4, 2004), see

Forensic at also

Evidence

Proves

Crucial

to

Service (UK), National DNA Database Hits 2 Million Mark (July 15, 2003), available at
http://213.52.171.242/forensic_t/inside/news/list_press_release.php?case=20&y=2003. (describing a typical month as yielding results in 15murders, 31 rapes, and 770 car crimes). in England, DNA 57. supra note 40, at 13. June 11, 2004 58. its 2,000th DNA Cold Hit, Park News, Virginia Lab Records see also Virginia of Criminal Department Justice Services, [hereinafter Virginia Lab]; at http://www.dfs.virginia.gov/statistics/index.cfm DNA Databank available Statistics, (last

inside/news/list_press_release.php?case=22&y=2004;

http://213.52.171.242/forensic_t/ Forensic Science Press Release,

visitedMay

by database hits were in breaking and entering, burglary, grand larceny, or whereas offenses, robbery only 992 were for rape, murder, and rape/murder combined). 59. The state of Virginia reports that 39% of the violent crimes linked through "cold hits" investigations aided linked

VA DNA Statistics] 14, 2007) [hereinafter

(reportingthat 2,284 of 3,614

were

Statistics, Individual

Virginia observation expanding 60. 61. cause

to offenders who had only previous for property offenses. VA DNA convictions in Criminal Justice: supra note 58. But see Amitai Etzioni, DNA Tests and Databases in Lazer, supra note 24, at 206 (commenting on the Rights and the Common Good, and England, while noting one journalist's wry data and similar data from Florida that "[i]f a large percentage of rapists receive speeding tickets, would that justify the DNA database to include those with moving violations?"). in of Justice, Bureau of Justice Statistics, Defendants U.S. Dep't Felony 540 U.S. 366, 372 (2003) (upholding arrest on probable and drugs were found secreted in back armrest, noting

Large Urban Counties iii (2000).

v. Pringle, See, e.g., Maryland of driver of vehicle inwhich money

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

736

LA WRE VIEW CALIFORNIA

[Vol. 95:721

possession of the contraband can elude effective prosecution. DNA typing, however,can conclusivelylinksuspectstocontraband.62 There is even a strong indication that suspects possess no "reasonableexpectation of privacy" in shed
DNA

processing

evidence without worrying about theindividual's Fourth Amendment rights. Consequently, as the costs of deploying DNA decrease, and law enforcementofficers' and prosecutors' awareness of such technology greaterand greater numbers of such cases are likelyto enterthe increases,64 criminal an evidentiary justice system.Indeed, if submitting DNA sample for analysis becomes as easy as it already is to submita sample fornarcotics analysis,thenlaw enforcement officers mightbe expectedto regularly conduct even in cases involving such tests, low-leveloffenses. The stateofNew York has plans toopen a state-of-the-art DNA testing intended laboratory solely for 65
property and other low-level crimes. And already in the United

cells,63 and

thus law enforcement

can easily

gather

such probative

it a reasonable

62.

testing of items including gun found in toilet tank, drugs and clothing found in backpack); v. Squires, Commonwealth 835 N.E.2d 323 (Mass. App. Ct. 2005) (table) (referencing DNA see v. of No. 2005 WL at *5 (Mich. Ct. also Elder, 248287, 562638, testing People drug bag); 10, 2005) App. Mar. (unreported decision) (remarking with regard to ineffectiveness claim that belief was not unreasonable counsel's that evidence against defendant, which included drugs in a jacket with defendant's DNA on it, "was strong and that conviction was likely"). 63. 2002 WL 31518865 See, e.g., People v. Padilla, No. B153331, 13, (Cal. Ct. App. Nov. Fourth Amendment defendant had no 2002) rights because (finding no violation of defendant's semen provided by girlfriend); Molly McDonough, legitimate expectation of privacy in ejaculated

DNA See, e.g.,United States v.Winston, 456 F.3d 861 (8thCir. 2006) (describing

inference that all three occupants

had dominion

and control over contraband).

action, and then testing the saliva on the envelope upon its return); see also Richard Willing, into Giving up DNA, USA Today, Police Dupe Suspects Sept. 11, 2003, at 3A (describing a range ... of trickery to obtain DNA [a] public samples including posing as "a phony dating service ... ... . . . and a a rape counselor a Taco Bell worker health worker Joh, diner"); Elizabeth Reclaiming 857 (2006) 64. "Abandoned" DNA: The Fourth Amendment (reviewing One study revealed that a major technology is simple lack of education issue of abandoned DNA 100 Nw. L. Rev. and Genetic Privacy, and related constitutional issues). factor in the under-utilization of DNA typing and

to Get DNA, A.B.A. J.& Rep., Jan. 27, 2006 (describing court proceedings Cops Played Lawyer in State v. Athan, No. 75312-1, in which trial court upheld ruse by police, posing as lawyers in fictitious firm, to get defendant's DNA by mailing him a false letter inviting him to join a class

and awareness. When databasing surveyed about reasons for failing to submit evidentiary samples, 31.4% of laboratories reported that they did not conduct a suspect had not yet been identified. National Report, supra note 36, at 22. testing because Yet, in the words of the study, "[c]learly scene evidence that need to be submitted In the written of awareness needed. Id. comments of available

these 'no suspect' cases are exactly the types of crime to be effective." Id. at 18. in order for the DNA database a lamentable DNA database lack is

portion of the survey, laboratory remarks demonstrated observations that a national resources, with multiple

at 19. In short, the survey revealed than concerns about funding or that, far more themajor is simple lack of information backlogs, impediment to the investigatory use of databases about their availability. Id. at 22. 65. Justice Public Dewan, Safety, DNA Programs, of Justice, Office of Institute see, e.g., Nat'l supra note 37, at Al; of Justice, DNA in 'Minor' Crimes Yield Major in U.S. Dep't Benefits at http://www.ncjrs.gov/pdfrilesl/nij/207203.pdf available (Nov. 2004) and noting success in retrieving evidence from "the sweatband inside a

that,"[i]n New York, biological evidencefrom 201 burglariesyielded 86 CODIS (reporting


acceptable profiles"

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

737

between2001 Kingdom, roughly 50% of DNA evidence sample submissions and roughly 17% were fordrug and 2002 were for property or theft crimes,66 offenses.67 a more cases submitted DNA testing mean that: Of course, for may simply cases.But given thescarcity of resourcesin thecriminal greater number of total it ismore likelythat DNA-based caseswill displace non-DNA justice system, of cases will dramatically increase. based cases thanit is thattherawnumber must choose onlya fraction Given that of cases topursue prosecutors inevitably from the greater number available, they may develop a bias toward DNA-based
evidence

demands such evidence, the resultof the so-called "CSI effect"evident in crimescene television jurorsexposed tounrealistic shows.69 Thus, prosecutors resources will logically which proofof with limited faced preferthosecases in to scientific certaintyis readilyavailable those that rely only on civilian on overtime witnessesor law enforcement officers pay.70If so, thenthetypical prosecutor'sdocketwill likelycontain a percentageof DNA-based cases to thepercentage of such cases in thepool at large.7' All of disproportionate

in allocating

resources.68 Some have complained

that the community

cap, from the inside of a mask, on a cigarette butt, in chewing gum, on a drinking glass, or from a see also Richard Willing, DNA Database Used toHelp Solve Thefts, USA half-eaten sandwich"); database is being used to Oct. Today, 19, 2006 (reporting that the national DNA "increasingly a USA to in and other crimes" unsolved suspects property Today identify burglaries according review of state crime labs). 66. likewise Virginia "amount of evidence year." National 67. DNA 68. services, the reports that, as more officers use and appreciate DNA submitted by law enforcement for DNA analysis grows by 30 percent every Report, supra note 36, at 22.

in England, supra note 40, at 18. in the past the government might not have charged a Thus, for example, whereas found in a car with a gun in the trunk, because of lack of evidence passenger linking the two, on it. To if the gun had the passenger's DNA today the government would charge that passenger make room for that case, the government might not charge the driver found with a gun under his seat?even concern Today,

Aug. 5, 2004, at 1A. . . Itmay also be that "[o]ne consequence 70. of mathematical may be to shift the proof. focus away from such elements as volition, knowledge, and intent, and toward such elements as . . . ." Laurence H. Tribe, Trial Precision and Ritual in identity and occurrence By Mathematics: L. Rev. the Legal Process, in the case of 84 Harv. 1329, 1366 (1971). Thus, for example,

to though in the past thatwould be the kind of case upon which itwould proceed?due that the lack of DNA evidence renders the case less likely to be successful. "CSI Effect" Has Juries Wanting More Evidence, 69. USA See, e.g., Richard Willing,

evidence, availability of DNA widespread sole question is one of identity?readily

the government might elect to bring cases inwhich the established by the DNA evidence?and dispense with those cases that involve questions of intent. Imagine that a prosecutor can only bring thirty cases due to resource constraints. One hundred cases come in, only forty of which have DNA evidence. cases because of pressing concerns prosecutor may choose to bring a handful of non-DNA raised by the offense or the victim, but the vast majority of the thirty slots are likely to be cases were a minority of the total possible to the DNA-based allocated cases, even though DNA

The

mirror

cases brought. The percentage of cases brought with a DNA element (say, 80%) therefore will not the objective percentage of cases with DNA evidence in the world at large (40%). 71. Of course, the availability of DNA evidence may eventually cause criminals to either take measures to hide their identity or shift to types of crimes in which DNA evidence is less

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

738

CALIFORNIA LA WRE VIEW

[Vol. 95:721

this leads us to the first lesson about second-generation sciences thatcan be ofDNA typing: their gleaned fromthecurrent experience ease of use, breadth of proof renderthemlikelyto appear in a of application, and persuasiveness of criminal cases. disproportionately highvolumeandwide spectrum
2. An Entirely New Kind of Case: The 'Cold Hit'

Advances in second-generation sciences do not just encourage the substitution of caseswith a forensic evidencecomponent forthose withoutsuch a component. They also allow forthe identification of perpetrators even in the absence of any otherevidence.That is, second-generation sciences introduce an entirely new kindof case: one in intothecriminal which the justice system onlyevidence is scientific. In the case of DNA typing,law enforcement increasingly has at its disposal large databases of genetic information. Specifically, as law DNA samples,theyload "profiles," or enforcement officials collectand process intocomputer results of thegenetic testing, databases.These databases contain material collected two typesof files:"forensic"samples thatcontaingenetic fromcrime scenes, and "offender" or "known," single-sourcesamples that contain genetic profiles of offendersor known persons who submitted or pursuant toone ofmany offender-collection materialvoluntarily72 biological statutes.

at least on a broad such a result seems implausible, readily obtained. However, while possible, in a manner little scale. First, many crimes are committed suggesting foresight, and by those for instance, the ease with which a robber whose thought is clouded from intoxication. Consider,

can hide his identity by putting on a mask, yet hardly every robber is masked. Second, unlike trail, even when steps are taken to fingerprints or facial features, it is hard to avoid leaving a DNA do just that. Butler, supra note 26, at 1-2 (describing a rape case in which defendant had the victim shower to eliminate can and an amount technology applies evidence, but inwhich investigators recovered saliva cells from a beer to the naked eye from the bed). Finally, because DNA of semen undetectable across a wide variety of cases, it to displacement caused may be less amenable itwould require abstaining from criminality altogether,

by deterrence, because particular crime. 72. "sweeps"

rather than from a

In fact, recent concerns have arisen over law enforcement's increasing use of DNA or "dragnets" to collect genetic information. In a "dragnet," law enforcement officers on a community and request voluntary submission of DNA investigating an offense descend

samples from the entire eligible population. See, e.g., Pam Belluck, To Try toNet A Killer, Police Ask a Small Town's Men for DNA, N.Y. Times, Jan. 10, 2005, at Al. Voluntary contributors to continued retention of the genetic sample after such efforts have later balked at the government's the case is closed. Tim Potter & Mar. 9, 2005 (describing motion Eagle, in a "dragnet" related to search for BTK Serial Killer Dec. Search: Some Stan Finger, Motion to return DNA Asks: to DNA?, Wichita What Happens man filed who submitted DNA sample by Privacy Issues New

From

Police DNA Dragnets, USA Today, Sept. 16, 1998;Keith O'Brien,Men Seek ReturnofDNA
Claim Police Bullied Them for Swabs, Orleans Times Picayune, 73. 28, 2003. For a comprehensive Justice N. Meyer, listing of such statutes, see David Lazer & Michelle in Lazer, and Debate, supra note 24, at 372 System: Consensus the constitutional and statutory limitations on the collection of genetic material,

killer); Richard Willing,

is the Catch for

DNA

and

the Criminal

73. Unfortunately,

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

739

In theUnited States, theDNA database is nicknamedCODIS, or the "CombinedDNA Index System,"and it exists at threelevels: local (LDIS), state(SDIS), and national (NDIS).74While statutes and regulations governing NDIS circumscribe the information thatmay be uploaded and require meet certain laboratories thatload profilesto the local proficiency standards,75 and state counterparts often include material obtained under less stringent As ofMarch 2007, thenationaldatabase, standards.76 which ismaintainedby the FederalBureau of Investigation (FBI) and to which everystatecontributes, over 4.5million profiles.77 Of these,179,763were forensic contained samples, The states are not far and 4.3 million were known or offender profiles.78 most advanced behind:Virginia,which iswidely recognizedas one of the

have

state statute authorizing 484 (Minn. Ct. App. 2006) (holding unconstitutional 722 N.W.2d C.T.L., v. Vermont of Dist. Ct. App. 24. 2006) blanket DNA Watkins, (Vt. charged defendants); sampling on state constitutional grounds the "suspicionless collection and (No. 6805-2-04) (invalidating samples from all convicted nonviolent felons); see also D.H. Kaye & Michael banking" of DNA E. Smith, DNA Identification Databases: Legality, Legitimacy, and the Case for Population-Wide L. Rev. 413, 415 (advocating for a population-wide 2003 Wise. database as the most Coverage, of preserving privacy and social justice interests). Suffice it to say that many and interesting questions, ranging from privacy concerns to the scope of the Fourth Amendment an arise from the and search of individual's information. collection, storage, genetic beyond, 74. See 42 U.S.C. ?? 14131 et seq. (2006). "CODIS" actually refers to the software used effective means

material to DNA 2004) (en banc) (upholding statuterequiringconvicted felons to submit Welfare of Matter of the database);Nicholas v. Goord, 430 F.3d 652 (2dCir. 2005) (same); In the

and the proper use of such material, exceed struggled with this very question. See,

the scope of this article, although scholars and courts 379 F.3d 813 (9th Cir. e.g., United States v. Kincade,

to search for information, but ithas emerged as a nickname for the database. For instance, rules govern what constitutes an appropriate "crime scene" or "forensic 75. unknown" any profile of fewer than thirteen loci sample, and laboratories cannot load into NDIS System DNA Index samples and ten loci for forensic unknowns, National for Acceptance of DNA Data, at 7, 9 (Jan. 1, 2000) Standards must Laboratories (outlining protocol for DNA typing and setting forth criteria of acceptance). standards issued by a technical working group affiliated with also comply with quality assurance for convicted offender NDIS (NDIS), the FBI. Butler, supra note 26, at 441 & App. individual IV (reproducing DNA Advisory Board Quality Standards). For example,

Assurance 76.

not qualified to submit material nationally, at http://www.lrc.ky.gov/kar/502/032/010.htm of samples from laboratories to qualify

or to store profiles to state standards); pursuant insufficiently for in the national database. inclusion Code Regs. 2.04 See, e.g., 515 Mass. complete state database, six of loci into and searches (1987) profiles (permitting loading on four loci); N.Y. Regs. at Codes R. & tit. 9, ? 6192.3, based Comp. available (last profiles visited 14, May to be loaded with

states may elect to include profiles extracted in laboratories available see, e.g., 502 Ky. Admin. Regs. 32:010(6), 14, 2007) (permitting submission (last visited May

- NDIS 77. CODIS Statistics Clickable Map Statistics, FBI, available at (last visited May 14, 2007). In the United http://www.fbi.gov/hq/lab/codis/clickmap.htm loaded theprofilesof the entire"criminally active Kingdom, theFSS claims ithas effectively
population." Programme and Pathology Science Office Unit, DNA Expansion (UK), Forensic Achievement 3 (2005-2006) in 2004 2000-05: Reporting (reporting achievement of goal of obtaining DNA of "criminally active population," estimated at roughly 2.5 million). 78. Id. Home

http://www.criminaljustice.state.ny.us/legalservices/section6192.htm#3 2007) (setting laboratory testing standards at national level, but allowing only six loci).

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

740

LA WRE VIEW CALIFORNIA

[Vol. 95:721

jurisdictions indealingwithDNA issues,79 has loaded over 253,986 offender 80 f20,Clfmahdlae As of 2004, Californiahad loaded samples and 7,044 forensicsamples.81 A 82 274,000 knownprofiles and 9,300 forensic samples. The accessibility and expansionofDNA databaseshave given rise to the "cold hit" case inwhich the major or only evidence is biologicalmaterial has no linkingthedefendantto theoffense.In thesecases, thegovernment investigatory leads,but develops a geneticprofilebased upon somematerial leftat thecrime scene. The government thenruns thatforensic profile in a database and uncoversa "match"-a storedsample associatedwith a known person or offender. As of December 2006, federalinvestigators had used the nationaldatabase to make roughly 47,000 "cold hits."83 And as thedatabases have grown, thematch capacityhas skyrocketed: whereas it tookVirginia nearlyeightyears, from1993 to2001, to reach its first 1,000 "cold hits," the statereacheditssecond 1,000 inamatter of eighteen months.84 Since 2001, the has averagedat leastone "cold hit" a day, and as of July laboratory 2002, that figure had doubled to twoand one halfhitsa day.85 From a cold hit, thegovernment either develops further factsto implicate or else brings thecase on thebasis of thisevidence alone. To be thesuspect, will endeavor to collect sure, in themajority of cases, the government additionalevidencebeyond the forensic proof.For instance,in one case, the a suspect,foundtwo a "cold hit" and,afteridentifying established government

79. Ethics

See

WhatShould (andWill) the Under the FourthAmendment? Med. & SupremeCourtDo?, 33 J.L.
nation 102, 104 (2005) (detailing inDNA database expansion"). Department Virginia 80. available CODIS history of Virginia collection statutes, which have "led of Criminal Justice Services, at http://www.dfs.virginia.gov/statistics/index.cfm Clickable DNA Databank

Tracey Maclin,

Is Obtaining

an Arrestee

'sDNA

a Valid

Special

Needs

Search the

Statistics,

(last visited May

14,

at available FBI, Statistics, Map Virginia these numbers do 14, 2007). (last visited May Notably, http://www.fbi.gov/hq/lab/codis/va.htm not include databases because they need kept at the state level, which are often more expansive not comply with federal laws in processing or reporting information. 81. Statistics 82. 83.

2007).

an unknown, but identical, sample found at another crime scene. 84. Virginia Lab, supra note 58; Karin Brulliard, Va. Gets U.S. Funds for DNA Backlog, Wash. Post, Sept. 22, 2004, at B01 (reporting that as of July 31, law enforcement inVirginia had there was no evidence?including found suspects in over 2,000 cases in which 1,200 burglaries and robberies?through 85. Brulliard, Criminal likewise available their DNA database). Tests and Databases in Etzioni, DNA supra note 84; see also Amitai and the Criminal Justice System, supra note 24, at 200. California Justice, in DNA of Attorney Gen'l, reports one cold hit a day. Bureau of Forensic Serv., California Dep't at http://caag.state.ca.us/bfs/ (last visited May 14,2007).

at http://www.fbi.gov/hq/lab/codis/ available Success, FBI, Measuring "Cold hits" can be either an offender-to-scene match, 14, 2007). (last visited May a known offender fits an unknown profile recovered at a crime scene, or a scene-to-scene meaning that the profile derived from an unknown sample from one crime scene matches match, meaning CODIS success.htm

Cal-DNA Database (Oct.27, 2004) (on filewith author). Through

Press Release,

Attorney

General

Lockyer

Announces

More

than 1000 Hits Obtained

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

741

theday witnesseswho claimed to recall thesuspecthavinga cuton his finger by thevictim.86 But, in to a wound inflicted of the murder thatcorresponded may proceedon thesole basis of geneticevidence some cases, thegovernment to evidence,suchas thesuspect'sproximity ormarginally probative additional or thirty was available-sometimes as far back as twenty genetic typing
years.88 the scene of the offense.87 In some cases, the offense occurred long before

of "cold hit" cases have even respondedto the influx Some jurisdictions by authorizing"JohnDoe" warrants intended to circumventstatuteof seeking an arrest warrant Typically, investigators limitations restrictions.89 warrantauthorizeslaw must specifically by name theperson thatthe identify to arrest. However,where a name is not available,but a genetic enforcement theissuance of an arrest warrant profile has been developed,some states permit geneticprofile.In suchcases, fora "John Doe" identified onlyby a particular if Doe's geneticsample is later should"Doe" everbe identified (for instance, warrant may be executedeven though enteredintoa database) thenthearrest of limitations would have otherwisepassed. InWisconsin, the the statute of limitations with the statute where the state legislature dispensed entirely warrantina sexualassaultcase onDNA evidence.90 bases an arrest

at http://crimmaljustice.state.ny.us/forensic/dnacasehighlights.htm in Monroe County). Bryan R. Hawkins 87. Courts have not reached consensus on the question whether available more, suffices

86. New York State Division of Criminal JusticeServices, DNA Case Highlights,
(describing case of

genetic evidence, without v. State, 16 S.W.3d to support a conviction. See, e.g., Roberson 156, 170 (Tex. Crim. App. 2000) that "the perils of eyewitness identification testimony far exceed (observing those presented by DNA expert testimony" and affirming that verdict can be based on DNA

defendant at trial, and rejecting evidence, even given that complainant misidentified only on DNA no is not "infallible" and thus cannot stand alone because argument that DNA "[virtually evidence conviction In the United Kingdom, is absolutely conclusive"). of a man found guilty by a jury solely on the basis based the Court of genetic

based alone); People v. Rush, 672 N.Y.S.2d 362, 363 (App.Div. 1998) (upholdingconviction
of Appeal the quashed evidence indicating that sample was about one in in

the random match four million; the United Law, discussed

probability of his genetic profile to the evidentiary on those statistics, the court concluded that he was one in seven to tenmales Redmayne,

Kingdom with such a profile. Mike 2003 Mich. St. L. Rev. 849, 879-80 in Mike

Redmayne previously

toMd. Rapes, Wash. 88. See, e.g., David Post, Apr. 27, Snyder, DNA Links Ga. Man 2005, at B5 (describing how a profile entered into the national database by a New York lab turned as well as two rapes in the New to a string of rapes in the late 1980s inMaryland, up "matches" York area in the earlier 1970s). J. Imwinkelreid, The Relative Priority that Should Be Assigned to 89. See, e.g., Edward and the Criminal in DNA Trial Stage DNA Justice System, supra note 24, at 94-95 Issues, John Doe (listing states where legislators have proposed legislation, along with those in which prosecutors 90. have Butler, sought such warrants even without express legislative authorization). DNA Warrants: supra note 26, at 446; see also Veronica Valdivieso, A

Smithv. United States,348U.S. 147, 152 (1954).

(2003) to Reason, L. Rev. 65 Mod. 19 (2002)). Professor Redmayne, Appeals noted that the accused also had no connection to the area. Id. The Supreme Court has held that an uncorroborated confession, without more, cannot support a conviction.

and Evidence Rationality, Naturalism, v. an R Lashley, (citing unreported case

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

742

CALIFORNIA LA WRE VIEW

[Vol. 95:721

For several reasons, includingthe lack of centralrecordkeeping, it is difficultto determinethe frequency with which the government presently bringscases inwhich theonly evidence is genetic First,although material.91 laboratories increasingly recordtheir "cold hit" matches, most fail to follow up on thenumber of cases thatactuallyproceed to prosecution and disposition. Second, of those jurisdictionsthathave trackedprosecution rates,92 none appears to track whetheror not additionalevidence was subsequently adduced in thecase. Third, the"cold hit" is stilla relatively recent phenomenon, and thusthecasesmay nothave yet wended their way through thecourts.93 Finally, itseems likelythatina greatnumber ofDNA cases, theexistence of damaging geneticevidence resultsin a guilty plea, as is thecase forthevastmajorityof criminal cases overall, which precludes appellate challenges and thereby decreases the likelihoodof a readilyvisible judicial trial. In Virginia, for on thebasis of only a "cold hit" pleaded guilty example,an inmateidentified 94 and accepted the deathpenalty. "Cold hit" cases are,however,clearlygoing forward.95 The first apparent case, which occurred in the United Kingdom,was a rape case inwhich the no evidenceotherthanthegenetic information prosecutionintroduced and the factthatthedefendant InLouisiana, a had access to thearea of theoffense.96 DNA dragnetresultedin thearrest of Derrick Lee Todd in May 2003.97 In August and October 2004, he was convictedof two separate murders and

Panacea

90 Geo. L.J. 1009 (2002) for Old, Cold Rape Cases, 91. N. Meyer, DNA See David Lazer & Michelle

(discussing John Doe warrants). and the Criminal Justice System:

Consensus database

that "[tracking and Debate, LAZER, supra note 24, at 379. Others have observed a high priority," as there exists must become hits and prioritizing case management to judge the overall effectiveness of DNA data on which data banking programs." "inadequate Frederick R. Bieber, into Earned Runs: the Effectiveness Turning Base Hits Improving of 34 J. L. Med. & Ethics Databank 222, 222 (2006). Programs, Statistics, See, e.g., VA DNA supra note 58. In Virginia, 93. in 2003 of the outcome of the first 1000 for example, a survey conducted cold hits revealed that 100 resulted in convictions through plea or trial, 7 yielded not guilty 92. and 53 were never prosecuted; 752 were pending at the time of the survey. VA DNA supra note 58. is Executed, DNA Comparison Frank Green, Patterson in Slaying, Led to Conviction at Times Dispatch Bl Mar. 15, 2002, (VA), (reporting on execution of James Earl See, e.g., State v. Scarborough, 201 S.W.3d 607, 624-25 (Tenn. 2006); Hartsfield DNA

Forensic

verdicts, Statistics, 94. Richmond Patterson). 95.

v.

Criminal 2001

State,200 S.W.2d 813 (Tx. Ct. App. 2006); State v. Hunter,2006WL 2790248 (Ohio Ct. App. Sept. 29, 2006); People v. Harrison,2005WL 2429974 (N.Y. App. Div. Oct. 4, 2005), appeal denied, 843 N.E. 2d 1162 (2005); see also David Lazer & Michelle N. Meyer, DNA and the
Justice System: Consensus and Debate, LAZER, supra note 24, at 379 (reporting that a first 102 cold hits found that four had resulted in convictions and that study of New York's charges were pending in fourteen others, but there was no data about the remaining cases). 96. 97. Regina Melinda v. Adams, [1997] 1 Crim. App. Testimony 377 DeSlatte, Jury Hears (So. Miss.), Oct. 5, 2004. of DNA (A.C.). Evidence inDerrick Lee Todd Case,

Sun Herald

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

743

sentencedto death.98InVirginia, a murder thatoccurred in 1992 remained unsolved foryears,and investigators had no leads.Four years later,the state requiredoffender Mack Reaves to submit a DNA sample, but backlogs preventedit from being processeduntil2001. Once analyzed,thesample was matched to a samplecollected in the1992 case, andReaves pleaded guiltyin 2001 to avoid thedeath penalty.99 Clearly, the "cold hit" case has stakeda will likely justicesystem, and it onlyexpandas courts and place in thecriminal prosecutorsgrow increasingly reliantupon, and comfortable with, DNA databases.l00 a lessonabout thefuture of second-generation Thus,DNA typing offers the thisevidence sciences:unlike itspredecessors of first-generation, may in many cases be the sole proof of guilt that exists.Yet the present legal framework forhandlingforensic evidencehews tonotionsbettersuitedto the notkeptpace with The lawhas simply first, rather thanthesecond,generation. advances in forensic science. The Supreme Court last addressed the in 1985, in constitutional forexpert assistanceto indigents Ake v. requirements in which the the to Court recognized barestentitlement expert Oklahoma, only of the importance of preserving physical advice.'01And the last articulation evidence in a criminalcase came almost twenty years ago inArizona v. destruction of Youngblood,102in which the Court held that government was not Due ProcessClause so longas it physicalevidencedid notviolate the 103 done inbad faith. In short,contemporary perspectiveson scientificevidence reflecta of theroleof forensic on science in criminal conception adjudicationfounded The current the characteristics of the first generation. view is thatforensic But theseantiquated evidence is auxiliary, occasional, and nondeterminative. ideas of forensic evidence ignoretheemerging realitythatsecond-generation in forensicevidence is increasingly central,pervasive, and determinative

98. 99. Police

Id. Tom

Led Jackman, Va. Man Receives Life Sentence for '92 Slaying: Va. DNA Database to Suspect Eight Years After Shopkeeper's Post, at B02. In another Virginia Death, Wash. case, a man was arrested and charged for a rape that occurred twenty-two years earlier. David toRape Charge, Manassas Journal Messenger, Jan. 11, 2005. Stegon, DNA Cold Hit Leads to 100. there are presently over 500,000 backlogged Notably, evidentiary samples believed to testing, in homicide, v. Oklahoma, rape, and property crime offenses. National 68 (1985); see also Caldwell v. Mississippi, Report, 472 U.S. supra 320

be amenable

note 36, at 14. 101. Ake

470 U.S.

103. Id. at 57-58. The Supreme Court recently granted a stay of execution, only to then of Starr, the former Solicitor General deny certiorari, in a capital case petition filed by Kenneth the United evidence. Lovitt v. True, 545 U.S. 1152 States, regarding the destruction of DNA sentence was an Execution later commuted to life without Times, Nov. parole by the governor. David at A19. Stout, Clemency Stops

(1985). 102. 488 U.S. 51 (1988).

(2005) (granting stay),cert,denied Lovitt v. True, 126 S. Ct. 400 (2005) (Mem.). The death
in Virginia, N.Y. 30, 2005,

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

744

CALIFORNIA LA WJRE VIEW

[Vol. 95:721

criminal adjudications. Given therole thatsecond-generation are technologies of criminal apt toplay in theadjudication cases, a close examination of how the criminal justicesystem handlesforensic evidence is inorder. II
THE SINS OF THE FATHER: Two FRONTS, Two FAILURES

Can conventional models of criminalprocess ensure the integrity of second-generation forensic evidence,especiallygiven thatsuch evidence may be the onwhethercurrent onlyproof insomecases? The answerrests models of criminal process in factvouchsafe theproduction, and subsequent adversarial testing, of second-generation forensic evidence. This Part examines thetwoprimary sites forevaluating and safeguarding each of theseaspects of evidentiary the scientific integrity: process and the judicialprocess or, in shorthand, the laboratory and thecourtroom. To ensure theproduction of reliableforensic must guaranteethatthe evidence,each site used to interpret theevidence is generallyreliableas a method,and technique was executedreliablyin a particular thatthe technique This Part first case.104 accurateappraisalof bothof these diagnoses thestructural problemsimpeding as they arosewith regardto thefirst-generation aspectsof reliability sciences, and then considers whetherthecharacteristics peculiar to thesecondgeneration will aggravate or alleviatethoseconcerns.

16 Calif. Precedent, Daubert subscribes

are not entirely distinct from one another"). that "conclusions and methodology observing Professor Michael of the distinction between Saks, examining the criminal law, finds maintenance and methodology conclusions beneficial, and subdivides the inquiry one degree further: At the highest level of abstraction are scientific theories, the basic concepts and explaining a field's empirical knowledge. One step down are to the real world of the theory, that is, broad applications general applications

as Legal John Monahan, Social Facts: Scientific Methodology Social Facts]. 877, 885-87 (1998) [hereinafter Walker & Monahan, to the same basic to and methodology" structure, finding "conclusions 579, 595 (1993) implicate distinct interests. Daubert v. Merrell Dow Pharms., 509 U.S. (placing not on the conclusions that they "the focus, of course . . . solely on principles and methodology, But see GE v. Joiner, 522 U.S. 136, 146 (1997) generate"). (breaking down distinction and Laurens Walker & L. Rev.

104.

underlying

or measures of procedures, that follow from the techniques, relationships, are At most concrete the level of the field's theory. specific applications to the case at bar. tools, and procedures knowledge, Michael Elise J. Saks, The Aftermath of Daubert: An Evolving Jurisprudence 40 of Expert Evidence, J. 229, 233-347 L. Faigman, [hereinafter Saks, Aftermath]; see also David (2000) Porter & Michael J. Saks, Check Your Crystal Ball at the Courthouse Please: Door, the Past, Understanding the Present, and Worrying About the Future of Scientific

Jurimetrics

with regardtoDNA Evidence, 15 Cardozo L. Rev. 1799, 1822, 1827 (1994). By illustration,
typing is that the cells of human beings contain typing: the "scientific theory" underlying DNA to and capable of evaluation. The "general material that is each individual unique genetic or capillary chain reaction of the theory is that, for example, polymerase application" the of the results express genetic typing. Finally, electrophoreses effectively "specific application" refers to the effectiveness of executing the technique in a specific case.

Exploring

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS A. TheLaboratory

745

The established method fordistinguishing good from bad science is to consideritsresilience when challenged. The scientific method, thecornerstone Good of reliability,105 asks whether a method is testableand falsifiable.'06 science thrives, 107 Open debate spursthe and evolves, inan open environment. the and thwarts of the bad. 108 of soundnew principles development propagation Competition inspires scientists to challenge orthodoxy and engage in to rigorous experimentation. Diversityfurther subjectstheories peer reviewand undervarious ensures thattheysurviveclose scrutiny which in turn testing, of conditions. But while all of this may be true science generally,it has neverdescribedthefield of forensic science. unfortunately 1.A Diagnosis of the FirstGeneration and laboratories The listof first-generation forensic analysts caught up in or anotherisboth The worst scandalsof one variety well-documented and long. storiesare of methodologies seemingly concocted fromthinair, such as the who purported "Cinderella"expert tomatch footand shoe impressions based On theotherend of the on a method developedby and knownonly toher.109 which have long been spectrumare techniques such as fingerprinting, embraced in the absence of any scientificvalidation even though such 110 But even setting aside thevalidity validationseemsat leastpossible toattain. of a particularly methodology,the ignominious past of the first generation of evidence,falsification of includestalesof fabrication and improper handling and reports, ormisleading results rogueor incompetent analyses,and corrupt 111 testimony. across thecountry? forensic scienceoccur in laboratories Why does faulty forensic Themost likely answeris that sciencehas neverbeen ordinary science. The techniquesof forensicscience rarely find analogues in academic or

105. community substantive

See

Knowledge 37 (5thed. 1989)).


106. 107. 108. Daubert, Id. Id. at 596. 509 U.S.

defined good science as thatwhich is subject to "falsifiability, and Refutations: Conjectures 593 (quoting K. Popper, at 593.

at 593 ("[Submission to the scrutiny of the scientific 509 U.S. Daubert, in part because it increases the likelihood is a component of 'good science,' that will be detected."). Court specifically flaws in methodology Indeed, the Daubert or refutability, or testability." Id. at of Scientific The Growth

while

4 Va. J. Soc. Pol'y & L. 439,458-62 (1997). Independent Crime Laboratories, 110. supra note 24, at 268-74. Cole, 111. and See, e.g., Giannelli, supra note 109, at 442-468 (analyzing a range of "sciences" associated scandals); Cole, supra note 24, at 281 (noting that "the first external proficiency tests on American police fingerprint laboratories" resulted in only 44% of examiners scoring perfectly, 22% reported false positives; in later tests, the false positive rate ranged from 3% to 15%).

109. Paul C. Giannelli,TheAbuse ofScientific Evidence inCriminalCases: TheNeed for

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

746

LA CALIFORNIA WRE VIEW

[Vol. 95:721

As commentators commercialsettings. have observed,"[t]here is virtually no 12 and there other 'market'foridentification "are no industrial uses of tests,"1 not only what forensicidentification scientists Thus, thegovernment do."'113 creates forensic science, but also almost exclusively executes forensic Unlike scientific techniquesthatemerge from collaborative or procedures.114 competitive environments spanning bothpublic and privaterealms,almostall forensic science,and almostall forensic scientists, claim commonancestryin the government."Peer review" in forensic science approximates self who "validate" a particulartheoryor and the scientists congratulation,115 who oftenstandtobenefit methodology are those from itsapproval.116 thanfinding ina robust Thus, rather motivationand regulation community of peers, the forensic scientistis beholden to the internal demandsof police So long as these clients remain and government investigators attorneys.117 or self need not engage in any new development satisfied,the laboratories Rather, crime laboratories primarily engage in applied science, criticism.118 to themechanical processing of government limitingtheirresponsibilities who hold no more than an undergraduate evidence.119Indeed, technicians degree staff many police crime laboratories,120 and thesepersonnelare often

112. and Solomon, supra note Beecher-Monas, supra note 11, at 73; see also Saks, Merlin or industry) where 10, at 1132 (commenting on "the lack of other institutions (such as academia or as well as create incentives for critical evaluation improved knowledge competition might improved 113. 114. laboratories technique"). Saks, Merlin See Giannelli, and Solomon, supra note 10, at 1092. n.182 supra note 109, at 470 &

of crime (reporting that 79% are governed by law enforcement). 115. Annotated that the conceptions Scientific, supra note 13, ? 1-3.5.1 [2] (observing as centerpieces of scientific validity, as expressed inDaubert, of "peer review" and "publication" are not very rigorous in the forensic disciplines); Social & Laurens Walker, cf. John Monahan Authority: 477, 500 suspect"). 116. Obtaining, (1986) Evaluating, (characterizing Social and Establishing scientific findings made Science in-house in Law, 134 U. Pa. L. Rev. and unpublished as "highly

Court's

Porter & Saks, supra note 104, at 1829; Paul C. Giannelli, The Supreme Faigman, Hall L. Rev. Daubert Cases, 33 Seton "Criminal" 1071, 1101 & nn.173-74 (2003) it lacked true peer review in that "law (referring to case rejecting toolmark testimony because enforcement technicians" wrote the relevant validating studies). the "cultural difference between supra note 9, at 893 (observing Koehler, and forensic science" and cautioning that "[w]hen individuals who are not steeped in the culture of science work in an adversarial, crime-fighting culture, there is a substantial risk 117. Saks & science prevail"); succinctly see also Saks, Merlin and Solomon, supra note 10, at as: 1093 &n. 118. 109. One commentator identifies the eight problems of forensic laboratories

normal

that different set of norms will

to Improve Forensic Science, 20 Eur. J. L. & Econs. Roger Koppl, How 119. See Saks, Merlin and Solomon, supra note 10, at 1092. Of also contributes to this phenomenon. 120. Kenneth G. Furton, Ya-Li Hsu, and Michael D. Cole,

poor quality control, inappropriate government monopoly, government budgetary dependence, information sharing between the government and technicians, insufficient distinction between the number of defense experts, lack of a analytical and interpretive function, lack of an adequate of laboratories. See those custom that and exist, among experts public ownership competitive 255, 257 course, (2005). lack of resources Background

What Educational

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

747

by or analysis,even ifencouraged research ill-trained to conduct independent "grewup in forensic science that As a result, or incentives. adequate resources a case of "arrested development." from 121 law" suffers thecriminal methodological peer reviewnot only stuntsthe The lackof meaningful to evade the science also enables forensic but growthof forensicscience, Many onmost scientific endeavors. standards imposed stringent qualitycontrol they do standards: fail toadhere to evenbasicmonitoring forensic laboratories and or proficiency routine testing,122 not engage invalidationstudies undertake than publish them resultsin secrecyrather do tendto shroudtheir those that words of one 123 In theoft-quoted disciplines. publiclyas in other scientific must meet higher standardsto be "clinical laboratories renownedscientist, mustmeet toput a defendant labs thanforensic allowed todiagnose strepthroat own errors, and catch their 124 Thus, forensic laboratories rarely on death row." or as monitoring accreditation such rigorous they face fewexternalincentives, aimed at study more exacting to adopt practices.Indeed,in a recent standards, inforensic testing of implementing thefeasibility blindproficiency ascertaining revelation were compromised by "clandestine researchers' efforts laboratories, personnel"-in other law enforcement of thetestto the labby thecooperating were made to conductblind testing,the police words, even when efforts was a revealingto thelab thatthesample thetest by deliberately compromised test. 125 of robust At the same time,structural barriersimpede thedevelopment forensicresearchand practices. Although defense testing "defense-oriented" market to drive the does and can occur, there is generallyno centralized "defense-side" forensic testingor research developmentof institutional must relyeither on the defenseattorneys Without such institutions, facilities.126

Do

44 J. Foren. Sei. 128, 129-31 (1999) Crime Laboratory Directors Require From Applicants!, (reporting findings of survey of crime lab directors that, for all positions combined, 63% required a B.S. and 27% a B.A., and for firearms, document, or fingerprint examiners, 17% of directors & degree of any kind); see also Joseph L. Peterson, Steven Mihajlovic required no college-level Uses,

The Capabilities, Joanne L. Bedrosian, Sci. 30 J. Forensic 10, 18 Laboratories, degree, and 17% personnel held a bachelor's See Saks, Merlin 121. and Solomon,

14, at 40-43. For example, many to both validity and proficiency fingerprinting readily lend themselves to studies in either respect. full-scale has failed conduct wholly discipline 122. Saks & Koehler, supra note 9, at 894 (noting that "blind see also Beecher-Monas, supra note 11, at 84. nonexistent"); 123. Saks, Merlin and Solomon, supra note 10, at 1093-94.

Criminalistics and Effects of the Nation's (1985) (reporting that in 1982, 48% of laboratory had not finished college). supra note 10, at 1091-92; see Mnookin, supra note commentators such as have observed that forensic sciences testing, and yet the

tests are practically

125. of External Blind DNA Proficiency Testing. Joseph L. Peterson, et al., The Feasibility II. Experience with Actual Blind Tests, 48 J. Forensic Sci. 32, 38 (Jan. 2003) (reporting that, in one of five labs tested, "police contact person revealed the plans for the blind test to laboratory management"). 126. For example, a "defense" testing and research center might do everything from

on Trial, 339Nature 501, 505 (1989). DNA Fingerprinting 124. Eric Lander,

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

748

CALIFORNIA LA WRE VIEW

[Vol. 95:721

127or find independent benevolence of government laboratoryanalysts, analysts, who are oftensimply retired government technicians. Moreover, to the extent thatdefense attorneys endeavor to obtain an independent theirinquiries or requestsforrawdata are often examination, met with thehostility and reluctance of an adversaryratherthan thecandor and 128 neutrality of a scientist. In the words of one commentator, "[w]here science advancesby open discussionand debate, forensic sciencehas been infected by the litigator's preferencefor secrecy."129 Forensic scientists often feel the pressuretoproduceresultsthat will please their central and even sole client,the and to shield their government, processes fromthedefenseor even thepublic domain.'30 Thus, defenseresearchis almostnonexistent, and defensetesting is piecemealand sporadic. 2. ThePathologiesof the SecondGeneration At first blush, itmight seem that the second generationof forensic sciences would avoid, rather thansuffer thepathologiesoutlinedabove. from, After all, many second-generation techniques derive from technologies to the world outside thepolice precinct, pertinent and scientists with expertise in theseareas populatenot just crime laboratories but also researchinstitutions and private industry. Advances inDNA researchfill thenews everyday, and datamining,and locationtracking biometrics, upon technologies rely generated which presumably are equally responsiveto by and used inprivateindustries, bidder. any legitimate But closer examination reveals that the characteristics of second in fact generationtechniques aggravatetheproblemsalready extant in first generationforensicsciences. In fact,an attorney confronted with a second generationscience report-whetherclaiming that the crime scene sample matched the of theclientina database,or that profile a biometric scan matched theclient to the imageon thesecurity cell phone triangulation camera,or that

to conducting and verifying government its own analysis of analysis, studies aimed at challenging government orthodoxies. few government labs will accept testing requests from defendants. One However, . . .would that "fifty-seven percent study of 300 crime laboratories concluded only examine evidence submitted by law enforcement officials." Giannelli, supra note 148, at 1331 (quoting evidence, 127. independently checking to undertaking Peterson, Effects of theNation's 128. Giannelli, 129. Joseph L. Steven Mihajlovic & Joanne L. Bedrosian, Criminalistics 30 J. Forensic Laboratories, Sci. The Capabilities, 10, 13 (1985)). Uses, and

see also Giannelli, supra note 109, at 470,473; supra note 162, at 117-18. and Solomon, supra note 10, at 1092-93. Saks, Merlin 130. Saks & Koehler, supra note 9, at 893 ("'All [forensic science] experts are tempted, many times in their careers, to report positive results when their inquiries come up inconclusive, or indeed to report a negative report as positive.'") 84 J. Crim. L & (quoting A.A. Moenssens,

Crimol.

Rosenthal, Problems

The Daubert/Kumho Implications of Observer Effects in Forensic and Suggestion, L. Rev. 90 Calif. 1, 27-41 of Expectation susceptibility of forensic analysts to bias from government influence).

1, 17 (1993)); Michael Risinger,Michael J. Saks, William Thompson & Robert


Science: Hidden (2002) (describing

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

749

placed the client's cell phone at the crime scene-will be even more ill confronted equipped to assess theaccuracyof such evidence thanan attorney orballisticsreport. First,the Why? Three reasons. handwriting with an ordinary or research sciences lackcommercial of second-generation application forensic of the technology generally;second, market robustness analogs, despite the or they rely on databases and research in the controlof the government andheightens or adversarial inquiry independent which both frustrates industry, information; privacyand proprietary concernsabout safeguarding legitimate expertise,financialinvestment, and third, theydemand a degreeof technical and of informal thedevelopment thatinhibits andmechanical sophistication these three each of Parts will examine advisors.The following independent problemsin turn. Research Forensic andNonforensic a. TheGap between Despite the aura of commercialapplicationthatlooms around second use of such techniquescan be readily generationtechniques,the forensic of application counterpart. First,theforensic fromitsnonforensic differentiated use. For instance, variesmarkedly fromitscommercial a general technology many research scientists,pharmaceutical with respect to DNA typing, in genomics-based work, but companies,and othergroups takegreat interest from that of the differs significantly overallobjective typically thegeneticist's generallylooks forareas of thegenetic Whereas a geneticist scientist. forensic theforensic diseases, or characteristics, human attributes, strandthatregulate materialhas no most commonly studiesthoseplaces atwhich genetic scientist establishedloci)."3' the thirteen or purpose (typically, function demonstrable in genomicsvalidatesDNA broader interest To suggestthatthegeneticist's market for widespread thatthe purposes is like suggesting forforensic typing of an electric chair. electricitysomehow ensures the proper functioning to identify may be used by private industry Similarly,a biometrictechnique mean it isvalidated does not workplace,but that knownemployeesina secured instance. for use identifying unknown personsin thefirst developed a particular Second, themere fact that private industry degree thanthegovernment alone,does not ensurea greater rather technology, of openness or methodological soundness.Any company thatdevelops a allies closelywith its primary technologyfor forensicpurposes inevitably The reason is clear: once a companydevelops thegovernment.132 customer,

131. LiOTTi, & Oeser-Sweat, KoBiLiNSKY, supra note 26, at 104. But see Barry Justice System, Steinhardt, Privacy and Forensic DNA Data Banks, in DNA and the Criminal itmay turn out that these loci in because supra note 24, at 173 (rejecting the term "junk DNA" fact code for some useful purpose).

Orchid Cellmark, a leadingprivateDNA lab, employ the law firmof SmithAiling Lane to
promote its interests in government. Chris Asplen, a vice president at the firm, in turn has played a

132.

For

instance, Applied

BioSystems,

which

develops

technology

for DNA

typing, and

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

750

CALIFORNIA LA WRE VIEW

[Vol. 95:721

and markets a revenue-generating forensic product,it strivesto protect the product and ensure that it is universallyembraced and adopted. Thus an adversary of the government-forexample, a defense attorney-is also an of thecompany. adversary Moreover, even apartfrom government allegiance, private companies may have proprietaryinterestsin protectingnew technologies, which further discouragepermitting open access. For example, forensicscientists typically DNA typing conduct using "kits"andmachines developed and sold by private companies.133 However, thesecompanies vigorouslyguard the methods and validationstudiesunderlying their as intellectual technologies and property,134 have successfully resisted disclosing thescientific theories that underpintheir techniques.'35 Similarly,private cell phone companies, email providers,and search engines mightbe reluctant to revealhow they collectand store data forfear of 136Think of the recent grantingcompetitorsaccess to such information. controversy surrounding Google's refusalto disclose the search terms users enteredinto its searchengine: thecompany'sprimary claimwas not privacy, but rather theneed toprotectitsproprietary information.'37 AlthoughGoogle
role in advancing the prevalence of DNA to the company's website, Mr. typing. According "worked closely with both Attorney Generals to develop DNA [sic] Reno and Ashcroft of Justice," has "testified before numerous state and city legislative policy for the Department and "testified before Congress to help appropriate over $160 million for forensic DNA bodies,"

major

Asplen

Thomas, Staff, http://www.sal-gov.eom/Staff.html#4 testing." See Gordon, Honeywell, (last visited May 14, 2007). With regard to another company, one independent scientist described how, although he purchased DNA-typing analysis software from a private company, the company twice refused to allow him to enroll in their software training course, because he was not government affiliated. Dr. Simon (2002). 133. See, e.g., Butler, supra note 26, at 97 (describing "[t]wo primary vendors for STR . . . and as "Promega kits used by the forensic DNA community" Corporation Applied at id. 359-63 various instruments to used Biosystems"); (describing perform capillary and theirmanufacturers, electrophoreses along with software used to interpret data). Service for the District Ford, Lexigen of Columbia Science and Law Consultants, Lecture at the Public Defender

134. See Butler, supra note 26, at 100-01 (comparing Promega corporation, which its primer sequences, with Applied which "has repeatedly refused to published Biosystems, ... release the primer that this information is proprietary"). sequences claiming Applied claimed that "they would lose revenue if generic brand products were produced by Biosystems other entities Manufacturing DNA the using Convictions: revealed primer information." are Entitled Id. Jennifer N. Underlying Why Defendants to theData Notes, Mellon, Forensic DNA

Kits, 51 Duke L.J. 1097, 1099 (2001) (reviewingtheresistancetodiscoveryexhibited by private


kit manufacturing companies and arguing for greater disclosure). 135. See, e.g., State v. Traylor, 656 N.W.2d 885, 900 (Minn. 2003) ("[W]e hold that disclosure of the primer sequences and unlimited access to Perkin-Elmer's validation studies are not necessary for the scientific community to validate the Profiler Plus and Cofiler kits and,

due process right to a fair trial has not been violated."). therefore, that [the defendant's] 136. Still a Fantasy, Toronto See, e.g., Lynda Hurst, Bio-security Star, Jan. 24, 2004, at Al in development of biotechnology, and reporting that the "proprietary right on (noting missteps the algorithm used in iris scanning is held exclusively" that is by a New Jersey company considering a request to open up the technology). 137. Katie Hairier & Matt Richtel, Google Refuses to Hand Over Search Data to US, Int'l

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEW FORENSICS

751

eventually managed to resisted the government'srequest, the government obtain the same data fromthreeotherprivate search engineswithout any opposition. However, one can imagine that, if even the government thendefensecounsel obtainingsuch information, occasionallyhas difficulty would be hard-pressed to convince a court to honor a subpoena forsimilar case. access ina criminal b. Access to Databases sciences further it of second-generation renders The databasedependency ifever, unlikelythata completeappraisalof theevidencewill be frequently, or private industry undertaken. That is, even assuming thatthegovernment a second-generation underpinning open access to the technologies permitted thegovernment is apt to or DNA typing, likebiometricscanning technology the imagesor genetic material retain a tight hold on thedatabases containing used for comparison. storage, and search of DNA typing requiresthecompilation, For example, These databases are critical to large quantities of genetic information. thelikelihood of a profileappearingin thepopulationat largeand determining scientists to making "matches" between samples. But nongovernmental ifever,can access thisdata.138 and rulesof Statutory protections infrequently, source materials and raw data in specific discovery protectthegovernment's
cases,
139

or outright This inevitablyinhibits preventsdefense the instant dispute.140

and judges rarely require disclosure beyond

thematerials

relevant to

Herald

138. using Professor used

supra note 11, at 78; requests can fall on deaf ears. See Beecher-Monas, L. Rev. 793, 816 (1991) (discussing and DNA, 44 Vand. Criminal Discovery, Scientific Evidence, evidence and concluding that need for greater discovery of "predicate materials" underlying DNA Pat Smith, Hearings in DNA Discovery "the rules do not require adequate discovery"); Begin Spat, The Recorder Feb. 2, 2005 (describing hearing inwhich public defenders (San Francisco), order allowing broader than case-only discovery). Indeed, the defendant sought jurisdiction-wide in one case received "greater discovery under the FOIA [Freedom of Information Act] after his 16 prior to trial." Giannelli, trial than he could have received under Rule supra, at 816 (referring to United States v. Stifel, 594 F. Supp. 1525,1528, 1531-38 (N.D. Ohio 1984)). 140. of theNew C. Thompson & Simon Ford, DNA Typing: Acceptance and Weight See, e.g., William v. Wesley, Genetic Identification Tests, 75 Va. L. Rev. 45, 105 (1989) (citing People

in the particular case. For example, Federal Criminal Procedure Rule to provide only a description of "the witness's the bases government opinions, Fed. R. Crim. P. 16(a)(1)(G). those opinions, and the witness's qualifications."

very well have great interest in research Interview with Dr. Montgomery Slatkin, of Integrative Biology, University of California, Berkeley (Mar. 3, 2006). to thatwhich is 139. The rules of discovery often limit the scope of mandatory disclosure 16 requires and reasons the for

Trib., Jan. 20,2006. For instance, population geneticists might access. such data, but at present are foreclosed

Thus, counsel's Paul C. Giannelli,

and undisclosed introduction of previously unpublished studies)); see also Saks, by government's Merlin and Solomon, supra note 10, at 1092-93 (noting that the defendant has little access to those few studies generated by government scientists).

defensechallengecountered 140 Misc.2d 306, 239-30 (Albany CountyCt. 1988), and describing

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

752

CALIFORNIA LAWREVIEW

[Vol. 95:721

attorneys and independent researchersfromchallengingthe validityof the government's conclusions. The same holds true for other formsof second-generation science, in although some cases a privateparty,rather thanthegovernment, may hold therelevant information. For instance, facialrecognition or irisscan techniques databasesof recorded dependon government-compiled biometricinformation, and radio frequency of cell site information or vehiclemovements tracking relieson data collectedand stored by privatecompaniesfrom particular towers access to such data to examine it forunusual patterns,inaccuraterecord or errorsindataprocessing. keeping, Of course, the courts and government have sound privacy reasons to tightly regulatesuchmaterials. While a defendant confronted with location tracking data might request access to a database to determine what other personswere in the same area at the same time, the disclosure of such information obviously implicatestheprivacy interests of such persons.DNA databases can likewise reveal "familial" connections, therebyexposing information about personsnot even included within the immediate scope of
authorized intrusion. 142 Raw DNA or stations.14 But here too, independent researchers are unlikely to gain broad

essence of personhood: a person's phenotypiccharacteristics, gender, age, health,and genealogy.143 Thus, even apart fromany statutory laws limiting who comes along. Ifmost people shudderto thinkthattheir any researcher social security number would be known to the tojustify world, imaginetrying
access,144 the government is understandably reluctant to open up databanks to

samples have the power to divulge

the very

141. 2006, passport and who

at A4

to Get 'Biometrie' Scan, Toronto e.g., Jim Bronskill, Passports Star, July 24, for (reporting complaints with regard to Canada's adoption of biom?trie technologies security, including that "We don't really know much about how these databases get made is programming them"). See,

142. P. Riordan, Nanibaa' A. Garrison, & Joanna L. See, e.g., Henry T. Greely, Daniel to Catch Offenders' Kin, 34 J. L. Mountain, Family Ties: The Use of DNA Offender Databases Med. & Ethics 248 (2006); Frederick H. Bieber, Charles H. Brenner & David Lazer, Finding

DNA ofTheirRelatives, 312 Science 1315-16(June CriminalsThrough 2006).


143. 34 J. L. Med. surname). 144. Pilar N. Ossorio, About Face: & Ethics 277, 278, 283 Forensic Genetic (2006) (adding

Testing for Race and Visible Traits, that itmay also predict an individual's

seem to leave the door states have inadequately defined privacy laws, which Many of use or study by third parties, or for non-law-enforcement open for some measure purposes. See the breadth of most of these statutes allows law Steinhardt, supra note 131, at 175-80. Yet for non-law-enforcement officials, to use the database purposes. To it tends to be limited to "humanitarian non-law-enforcement usages, or missing persons identification. Most states lack an organized regime through which purposes" defense-oriented research entities (non-law-enforcement and non-public officials) can gain access enforcement, or other public the extent that they authorize

to government databases. The wide range of vague and confusing statutory requirements leaves unclear the parameters for a private researcher. See Seth Axelrad, Survey of DNA at Database American of & available Statutes, Law, Ethics, Society Society 14, 2007). (last visited May http://www.aslme.org/dna_04/grid/statute_grid.html

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

753

movements. and cell phone the broadcast of their thirteen-loci genetic profile c. TechnicalComplexity, Mechanical Sophistication, and the Dearth of Independent Analysts of Finally, the technical complexity and mechanical sophistication means that broad-basedindependent research along second-generation sciences of govemment conclusionsare unlikelyto occur with case-based verification material, defense widely. Even assumingopen access to all theunderlying in finding lawyers would encounter an expertqualified to conduct difficulty or review. research Whereas the fingerprint or ballistics analystat the local and starttakingdefense-sideconsulting sheriffsoffice might retire jobs at home, the local second-generation analystcannot readilydo the same. For necessarytoexaminethedata generated by a example,justbuyingthesoftware testsof the raw biological DNA lab,without conductingany independent sample,requiresan experttomake a substantial capital investment. Actually conductingindependent researchprojects or experiments requiresaccess to data and funding far in excess of that typicallyavailable to indigent of defendants.145 Similarly, it is difficultto imaginea robustcommunity or in the of location data biometric accuracy expertsspecializing checking It is fareasier to imaginethat once thegovernment puts theevidence scanning. without it will be accepted forward, questionas true. most developed second-generation of the science, Furtherexamination DNA typing, illustrates thisdynamic.Independent methodologicalresearchis of nongovemment all but nonexistent, and thereis only a small community Yet numerous and significant with questionsremaintobe answered experts.146 regard to DNA analysis-concerning, for example, how to disentangle

William

Given that roughly 80% of defendants in the criminal justice system are indigent, see J. Stuntz, The Uneasy Relationship Between Criminal Procedure and Criminal Justice, L.J. 1, 28 (1997), 107 Yale there simply does not exist a robust, paying consumer base for consistent and widescale defense work. For instance, one expert reported that he was able to conduct useful, albeit informal, studies regarding DNA transfer only because a wealthy defendant 145.

forwhom the study might prove beneficial subsidized his work. See William C. Thompson, Simon L. Raymer & Dan E. Krane, Evaluating Forensic DNA Evidence: Ford, Travis E. Doom, Michael Essential Elements of a Competent Defense, The Champion, Apr. 2003, at 26. to entertain 146. There do exist a handful of individual academics and scientists willing defense-side

consulting work and review government reports with an objective eye, but of course they are still restricted to the data disclosed by the government. Perhaps themost successful such at a in 2002, which consists of an automated analysis service, available entity is one established that provides an independent review of a CD-ROM of the reasonable price to defense advocates, raw data. See Forensic provides service 14, Bioinformatics, (last visited May http://bioforensics.com counsel with a thorough report of all of the genetic defense testing, rather than just the government's gloss on the "relevant"

government's This 2007). information

recorded during this service reviews cases from a information, and highlights possible problem areas. Because wide variety of labs, and a broad array of cases within a lab, it also has produced a data set from which research conclusions may be analyzed, and has a limited potential, if used consistently, to spot recurrent or systemic errors, at least as regards the raw data.

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

754

LA CALIFORNIA WRE VIEW

[Vol. 95:721

or mixturesof genetic samples from more thanone person,how commonly easilygenetic material is transferred, or to what extent populationsubstructure Yet fewnongovemment have thetime, affects match probabilities. researchers resources, interest,or capacity to conduct such inquiries. Similarly, independent testing of evidence in individual cases is not terribly common,'47 primarily and when such tests are performed,it is often by laboratories 148 beholdentogovernment contracts andhostile todefenseinterests. This lack of testingdoes not reflecta justifiedconfidence in DNA evidence. Afterall, scandalshave revealedsystemic problems in a number of "flagship" DNA laboratoriesand horrific tales of false-positive DNA as an analyst accidentally matches. 149Errors as small and unintentional or forgetting to changegloves after squeezinga pipetteintothe wrong tube,'50

147. In fact, inmany jurisdictions, a defendant had no right to obtain physical evidence in control of the state for purposes of independent testing. In specific response to the availability of DNA testing, many legislatures have granted to defendants a legal entitlement to testing both pre Senate trial and post-conviction. See, e.g., D.C. Code ? 22-4133 (2006); Helen Dewar & Dan Morgan, ' Tackle Busy Agendas, Wash. Post, Oct. Approves Bill on Victims Rights: Both Chambers

some courts grant the defendant's request to have a defense expert present right. Nevertheless, during testing that is likely to consume the entire sample. 148. 168461, slip op. at 15 (Cal. Super. Ct. May 5, 1999) See, e.g., People v. Bokin, No. went in favor of prosecution because bias inadmissible laboratory analyst's (holding DNA to indicate outright hostility to defense "beyond advocacy" in a Po^i-Daubert, The Right to Expert Assistance Oklahoma: Rev. in the context of DNA 1305, 1396 (2004) (describing, ambiguous

of federal law appropriating funds for defense DNA 10, 2004, at A05 (reporting on passage this right to test is second to the government's right: if the government elects to testing). However, the sample, the defendant typically cannot claim violation of that test, and such testing consumes

function); Paul C. Giannelli, Ake v. Post-DNA L. World, 89 Cornell analysis, that "[w]hen faced with an

situation, where the call could go either way, crime lab analysts frequently slant their v. Ake in ways that support prosecution theories") [hereinafter Giannelli, interpretations see also Janet C. Hoeffel, Note, The Dark Side of DNA Unreliable Profiling: Oklahoma]', 42 Stan. L. Rev. the Criminal Defendant, 465, 499-500 (1990) Scientific Evidence Meets (describing extremely U.S."). 149. Problems Mnookin, charged released how "[a]nalyzing lucrative business" William C. biological in which evidence the chief aim an and testifying about it in court has become to crime laboratories in the is to "sell ...

loci conducted after itwas revealed that illness prevented based on testing of additional him from having committed the crime); see also Mary arm Spoto, Murder, Rape Charges Dropped to Botched DNA Evidence, Star-Ledger Due (Newark), Feb. 7, 2006, at 28 (reporting that cold the match had examined evidence from hit case must be dropped because the analyst who made the old case, possibility

on the 'Gold Standard': Tarnish Recent Understanding Thompson, DNA Testing, The Champion, Jan./Feb. 2006, at 10-12 (listing scandals); case of Raymond Easton, who was arrested and supra note 14, at 49-50 (describing at six loci, but his DNA after officers linked him through a "cold hit" that matched in Forensic

along with a new case involving the defendant, on the same day, raising the Annie Sweeney & Frank Main, Botched DNA Report Falsely of cross-contamination); It Reports Lab Findings, Chi. Sun State to Change How Case Compels Implicates Woman; Times, Nov. 8, 2004, at 18 (noting that a laboratory forensic profile "matched" woman, based on after what ultimately turned out to be only partial match, was revealed erroneous when woman, arrest on warrant, was shown to be incarcerated at time of offense). Must Crime Lab Botches Murder 150. Paula McMahon, Drop Inquiry: Prosecutors is Contaminated, Sun-Sentinel (Ft. Lauderdale, FL), June 24, Charges after DNA Evidence

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

NEWFORENSICS THE

755

an extraction, can compromisecritical evidence.151In Texas, a scandal Houston crimelaboratory. News accountsrevealedthe ragesover the currently laboratory's deplorablephysicalcondition and shoddy practices, which inpart resulted in the misplacement of 280 boxes of evidencecovering approximately 8,000 criminalcases.152In other laboratories, improper handlingof evidence has turned up "matches"that appear to resultfrom contamination, rather than Similar problems have emerged in laboratories across the actual guilt.153 theeliteFederalBureau of Investigation DNA lab,155 including and country,154

at 1A (announcing dropping of murder and robbery charges due to "someone the squeezing or into the wrong vial" and noting disagreement eye-dropper regarding whether government in Police DNA Lab, Las defense attorney caught error); Keith Paul, Audit Calls for Changes 2003, Vegas defense Sun, May, expert caught 23, 2002, at 1 (reporting results of audit conducted after independently hired forensic lab in mistakenly labeling DNA typing results with name of

a case that recently arose inMichigan. of a grown man There, the DNA turned up in the testing of evidence related to a thirty-six-year-old murder case. On its face, the and reliable. But because theman, who was four at the time of the evidence appeared reasonable murder, lived one hundred miles away from the scene and would have somehow had to drop blood on the deceased broader review victim for the profile to appear, the evidence raised suspicions. And, in fact, a of the laboratory records revealed that the man's DNA was being tested by the in the old case. laboratory around the same time as the evidence was processed

innocent man). 151. Consider

very same the analysts insisted that no contamination had occurred, and the age of the man at the Although time of the offense precluded any argument that he murdered the woman, it is easy to imagine a different outcome had the evidence been from a contemporaneous crime. See, e.g., Teresa Mask, How See DNA Evidence May Decide Unsolved Killing: 1969 Slaying Trial Continues Detroit Free Press, 19,2005. Today, July In Texas, Oversight for Crime Labs 152. is Urged, N.Y. Times, Jan. 5, Ralph Blumenthal, 2005, at Al8. Jurors 2005 Test Results Still a Mystery, Jackson Citizen Jan. 19, Patriot, case in which DNA thirty-six-year-old murder testing revealed profile of apparent culprit, as well as an utterly unrelated, then four-year-old boy); see also infra note 156 case in Australia, a profile on a in which genetic testing matched (describing the famous Leskie 153. Steven Hepker, DNA (describing

murdered

child's bib to a clearly unrelated rape victim whose sample had been tested by the same analyst in the preceding weeks). 154. Vic Ryckaert, Judge Asked toHalt DNA Retests: Crime Lab Less Than Candid About Cases Under Review, Attorney Says, The Indianapolis Star, Aug. 13, 2003, at IB (describing fall-out from publication of prosecutor's in sixty-four evidence request that crime lab retest DNA

Apr.

DNA Mix-up Prompts Audit at Lab, Las Vegas Puit, Police Forensics: Review-J., audit at Las Vegas 19, 2002, at IB (discussing laboratory after switched names on DNA at the State Patrol imprisonment of "suspect"); DNA Testing Mistakes profiles led to year-long a series of errors ranging Crime Labs, Seattle Post-Intelligencer, July 22, 2004 (cataloguing him); Glenn from cross-contaminations samples across and within cases, including a vaginal sample with semen of positive control, along with other errors). Not even the private laboratories have proven exempt from such corruption. See, e.g., Rick Orlov, Lab Used by LAPD Falsified DNA Data, L.A.

cases believed compromised by analyst); Keith Matheny, Supervisor Accused of Passing Off DNA City Record-Eagle, Dec. Test, Traverse 19,2004 (detailing internal investigation of supervisor inMichigan State Police Crime Lab DNA unit that had a subordinate take a proficiency test for

DNA

Daily News, Nov. 19,2004, atNl (describing dismissalof Sarah Blair fromOrchid Cellmark, DNA data); Jeff after Coen & Carlos Sadovi,CrimeLab Botched allegationsthatshe manipulated
Tests, State Says, Chi. Trib., Aug. 19, 2005, at Cl (noting that Illinois state police found numerous errors in results reported from Bode Technology, an independent lab based inVirginia). 155. Richard Willing, Mueller Crime Lab After Questionable DNA Tests, USA Defends

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

756

LA WRE VIEW CALIFORNIA

[Vol. 95:721

As this catalog of scandal and malfeasance reveals, around theworld.156 of first sciences have not been spared the ignominies second-generation mechanical generation sciences, and given their technical complexity, andprivacyandproprietary concerns,it is database-dependency, sophistication, will change. unlikelythat B. TheCourtroom Even assuming,however, thatforensicevidence lives primarilyin the need scrutiny of government science, this lackof scientific gated community Yet as this Part explains,severaldistinctive not imply a lackof legal scrutiny. of forensic characteristics of thecriminaljustice systemcause legal scrutiny evidence to falter. the In the landmark Merrill Dow Pharmaceuticals,157 case ofDaubert v. Supreme Court announced its regime for assessing scientificevidence, of thejuryand of theadversary expressingitsconfidencein "the capabilities of presentation system generally," and in "[v]igorous cross-examination, on theburdenof proof," to protect contrary evidence,and carefulinstruction or fraudulent Daubert of faulty scientific evidence.158 against the introduction test to determineadmissibility: whether a scientific outlined a four-factor whetherithas been subjectedtopeer withstandstesting; technique successfully whether ithas a known errorrate and standardsto review and publication; control its operation; and whether it is generally accepted in a scientific 159 community. of Yet while Daubert addressed the legal standardsfor admissibility how those standards shouldoperate in scientific evidence, itdid not specify

did not catch 1, 2003, at 3A (noting that purported quality control guidelines Today, May coworker controls in 100 DNA failure to run negative technician's cases, caught only when state crime lab, have also arisen about work done by the Virginia revealed the problem). Questions one of the leading laboratories in the country in the DNA field. See, e.g., Maurice Possley, Steve to Touches Even Elite Labs: Flawed Scandal & Flynn McRoberts, Mills Work, Resistance Scrutiny Seen Across U.S., Cm. Trib., Oct. 21, 2004, at Cl. at The Age, Nov. in the Dock, 156. See Leskie Bib Puts Science 22, 2004, available l/21/1069027328463.html?from=storyrhs. http://www.theage.com.aU/articles/2003/l Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 596 (1993). 157. 158. 509 U.S. at 596. Daubert Daubert, replaced the longstanding standard of admissibility 509 in Frye v. United States, 293 F. 1013 (D.C. Cir. 1923). Daubert, in federal courts enunciated at 585-89. have

In the wake of Daubert, which applied only to federal courts, many states, some of on the federal rules of evidence, adopted its standards. See evidentiary rules modeled J. David E. Bernstein & Jeffrey D. Jackson, The Daubert Trilogy in the States, 44 Jurimetrics 351, 355-56 (2004) (observing that by mid-2003, roughly twenty-seven states had adopted a test consistent with Daubert)', cf. Minor v. State, 914 So.2d 372, 400 (Ala. Crim. App. 2004) (noting U.S. which state's continued adherence requires 159. five to be evaluated state law evidence, which doctrine, except as regards DNA to Daubert). pursuant at 593-94. Some scholars note thatDaubert 509 U.S. actually enumerated to Frye the question of error rate is analytically distinct from that of elucidated

Daubert, criteria, because standards or protocols.

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

757

practice. How shouldcourtstreat multiplerequeststoadmitscientific evidence, within a single case or across cases, especially ifurged by the same party? Must each courtroom entertain challenges to theadmissibility of a technique each timeit is used, ormay a judge properlytakejudicial noticeof findings or nations?'60 made inothercases, courtrooms, When should counties,states, courtsconsider an admittedtechnique"scientificlaw" and thus theproper when might new developments object of judicial notice,161 and, conversely, Who should justifysubjecting existing"scientificlaw" to renewedscrutiny? bear theburdenof putting forth evidence that calls into questionthecontinued reliability of an established methodology? What shouldbe therelevance of a laboratory's error rate, as opposed toamethodology'serror rate,indetermining a technique's admissibility? This Part identifies the ways inwhich thedistinctive characteristics of criminal processundermine the properfunctioning of this model in thecriminal that surroundthe justice system,firstby describing the legal structures then of forensic and admission first-generation evidence, by asking how evidence will fare within thosestructures. second-generation Specifically,the Partexaminestheexperience first of first-generation and concludes techniques thatstructural features impedethejudicial system's with monitoringfunction regardto thefirst of partiesto a generation. Namely, thestructural asymmetry criminal of resources, case, alongwith thescarcity weak discovery practices, and high rateof plea bargaining, adversarial renders processes an inadequate of the integrity of forensic the safeguard science.Building on thisdescription, next Part thenexplains how these shortcomings are especially acutewhen consideredin light of thecharacteristics peculiarto thesecondgeneration. 1.A Diagnosis of the FirstGeneration Under the current evidentiary regimegoverningcriminalcases, judges approachmethodological questions as questions of law and case-specific applications of these methods as questionsof fact.'62 Accordingly, when faced

160. Court to duck the question, the Daubert seemed Indeed, expressly noting that the Frye decision itself focused exclusively on 'novel' scientific techniques, we do not "[although ... to apply specially or exclusively read the [Federal Rules to of Evidence] requirements unconventional evidence." Id. at 592 n.ll. The Court went on to assume that "well-established propositions are less likely to be challenged than those that are novel, and they are more handily defended," and observed that "theories that are so firmly established as to have attained the status of scientific law ... properly are subject to judicial notice." Id. 161. 162. Id. See, e.g., Saks, Aftermath, supra note 104, at 232 (noting the "long practice, especially among state supreme courts, which have had considerable experience with expert evidence over the past century, of treating decisions about the admissibility of scientific evidence as a matter of this is the established Scientific, supra note 13, ? 1-3.8 (same). While law"); Annotated

rules that balance fluidity(the individual treatment of a case) with stability Seeking evidentiary
(consistency and efficiency), Professors JohnMonahan and Laurens Walker put forth a functional

system

in the criminal

law context,

it has only recently been

extended

to the civil

law context.

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

758

CALIFORNIA LA WRE VIEW

[Vol. 95:721

with forensic methodologies,trialcourts"rely inpartupon legalmemoranda, scientific documents,and precedent-ratherthan factualhearings with live witnesses-to determinetheir Indeed, trialcourtsroutinely admissibility."163 findscientific methodologiesreliablesolely on thebasis of judicial notice,164 and appellate courtshave endorsedparticular methodologies and techniques or appraisal of relevant based solely upon approval in other jurisdictions 165 literature in thefield.

are more to argue that methodological and textual analysis whereas "law-like," questions are more "fact-like." See Laurens Walker & involving the application of methodologies questions 86 Va. L. Rev. John Monahan, Scientific Authority: The Breast Implant Litigation and Beyond, Social Facts, supra note 104, at 877, 888-90; Laurens Walker & JohnMonahan, Monahan, A New in Law, Frameworks: Use Science 73 Va. L. Rev. 559, 559-60 of Social & Walker, & Monahan, Social Frameworks]', Monahan supra note [hereinafter Walker 479. Professors Monahan evidence establish, Monahan, extension Litigation, for example, Social Facts, and Walker in the civil arena, Social

Walker & Monahan, Breast Implant Walker & 801, 802, 819-21 (2000) [hereinafter Litigation]',
(1987)

to the use of social initially limited their proposals to the use of empirical evidence although they briefly nodded in criminal obscenity trials. Walker community standards of decency

115, at scientific to &

to hard science questions limited to social science "which was research," authority model, resolved by science panels). 163. See, e.g., Hayes v. State, 660 So.2d 257, 262-64 (Fla. 1995) (vacating death sentence after taking "judicial notice" of the National Research founded on unreliable DNA evidence

the supra note 104, at 880-81. Recently, however, they have advocated of their argument to the hard sciences, as well. Walker & Monahan, Breast Implant to "extend our earlier work" by applying their scientific supra, at 803 (proposing

States v. Porter, 618 A.2d of DNA 629, 635 (D.C. 1992) acceptance testing"); United review of admission of DNA evidence and noting that "[i]n doing so, we (conducting appellate may consider not only expert evidence of record, but also judicial opinions in other jurisdictions, as well as pertinent legal and scientific commentaries"). that Professor Saks likewise observes general to admissibility de novo review treatment of forensic evidence includes law-like applying decisions, judicial approval of opinions based upon on extrinsic sources, and categorical deference to binding precedent finding a particular methodology admissible. Saks, Aftermath, supra note 104, at 232. 164. because See, e.g., United States v. Havvard, 117 F. Supp. 2d 848, 854 (S.D. Ind. 2000), affd

v. Commonwealth 1992 forensic science report and citing cases in other jurisdictions); 640 A.2d Crews, 395, 400 (Pa. 1994) (rejecting defendant's complaint regarding trial court's on judicial decision from other jurisdictions to establish the scientific community's "reliance Council's

proceedings"); People v. they "have been tested for roughly 100 years" by "adversarial 145 Cal. Rptr. 466, 472 (Cal. Ct. App. Palmer, 1978) (approving gunshot residue evidence based upon a scan of literature in field). 165. See, e.g., Porter, 618 A.2d at 635 (D.C. 1992); United States v. Beasley, 102 F.3d

260 F.3d 597 (7thCir. 2001) (acceptingfingerprint evidence despite lack of scientifictesting

to permit the courts of this circuit to take judicial notice of it in is sufficiently well established future cases"); People v. Chandler, 536 N.W. 2d 799, 803 (Mich. Ct. App. 1995) ("Courts of this state may continue to take judicial notice of the admissibility of the RFLP method of DNA testing, see also People v. Richie, No. B158254, 2005 WL 1340382, *8 including the statistical analysis"); request to take judicial notice on appeal of four more (Cal. Ct. App. 2005) (granting appellant's recent outside States scientific v. the record in support of position, because "we can consider scientific literature to determine whether a scientific technique is generally accepted"); cfi United for 171 F.3d 587, 591, 593 (8th Cir. Iron Cloud, 1999) (reversing and remanding studies of scientific methodology, into question). based upon appellate judicial

of thePCR method ofDNA analysis Cir. 1996) ("[W]e believe thatthereliability 1440, 1448 (8th

evidentiary hearing on admissibility notice of cases calling themethodology

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

759

At the same time, and in contrast to this "law-like" status of methodologies,trial courtstypically refuseto lookat evidenceof a laboratory's reliability,or lack thereof, when resolving case-specific questions of ormethodologicalsoundness.166 admissibility Rather,courtstreat such attacks in a fact-like a case-specific manner; the fact-finder considersthemthrough lens as relevant only to the "weight"of the evidence. Indeed, some courts refuse even toallow counselanyaccess to,or argument about,a laboratory's or to thespecific analyst'serrorsinothercases, finding such evidence irrelevant reliability questionathand.167 a trial Upon initialanalysis, thisrubriccarriesgreat appeal. Permitting courttoadoptpreviousfindings can save on costlyand repetitious hearings and promoteuniformity among different courts and judges.Moreover, because much robust debate in the scientific in appears inwrittenformat community orpapers, live a witnessesarenotnecessarily essentialtocommunicate journals range of perspectives to a court.And allowing a court to determinean admissibility questionby lookingoutsideof a factualrecordadduced by the parties,as a judgemight lookoutside therecordto sourcesof legal authority, a technique that as illegitimate diminishes thelikelihood criticized will roundly to somehow penetratea courtroom due (even strategic)lack of vigorous opposition. After all, ithardlybehooves thejustice systemif,forexample,a judge rulesastrology reliableand admissiblesimply because shewas boundby

166. States v. Morrow, 374 F. Supp. 2d 51, 67 (D.D.C. See, e.g., United 2005) ("A of its past proficiency and is of little value in determining laboratory's error rate is a measure whether a test has methodological the defendant has sought to do here is flaws. . . .What challenge the proficiency of the tester rather than the reliability of the test. Such the weight of the evidence, not its admissibility.") (internal quotation omitted); challenges go to State v. Adams,

at all because itmight be propensity evidence, and if admissible, would might not be admissible be relevant only to the weight of the evidence at hand. The last type of error, the "lab was sloppy in this case" error, would be admissible only as toweight, unless the sloppiness in the case was so grave that itundermined 167. (unpublished the exclusion See, e.g., People the reliability of themethodology altogether. Id. at 68. v. Funston, No. C032472, 2002 WL 313198, *4 (Cal. Ct. App. 2002) (finding no error in trial court's exclusion on prejudicial/probative grounds

the first type of error?what 374 F. Supp. 2d at 66-67. Regarding specific case at hand." Morrow, the "generally sloppy lab" argument?the Court noted that the past error rate might be considered

relates to the validity of the underlying scientific principles, not the correctness of the expert's The Court inMorrow identified three possible conclusions"). types of error: "(1) a laboratory's past error rate; (2) the error rate that results if an analyst follows the ... protocol and uses properly calibrated instruments in the specific case at hand; and (3) the possibilities of human error in the

817 N.E.2d 29, 48 (Ohio 2004), cert,denied, 544 U.S. 1040 (2005) (noting"reliability inquiry

Probability, 34 J.Legal
constitute

that lab had twice in 277 case reviews made reports that turned out to be "false positives," where trial court reasoned that "the question is one of relevance" and "the fact errors the in not made the is of the issue whether it erred in the instant [that past lab] probative" An Empirical case); cfi Dale A. Nance & Scott B. Morris, Juror Understanding ofDNA Evidence: Assessment Formats for Trace Evidence With a Relatively Small Random-Match of Presentation character evidence, without impermissible taking a position, or Neglected Imwinkelried & D. H. Kaye, DNA Typing: Emerging Issues,

opinion) of evidence

Studies 395, 435 n.61 (2005) (notingquestionwhether errorrates


and citing Edward J. L. Rev. 413, 16 Wash.

461-63 (2001)).

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

760

LA WRE VIEW CALIFORNIA

[Vol. 95:721

therecord andno party presented evidence to thecontrary. thatcourtsshould treat Conversely, it seems fitting questionsabout the properapplicationof an establishedtechniquein a fact-like manner-left to adversarial challengeand determination Afterall, erroris an by thefact-finder. inevitable partof scientific testing, and a particular error need not undermine thelegitimacy of the methodas a whole. The execution of a particular scientific testis arguably well determined by lookingonlywithin a record, and only to the evidence judged relevantto the question at hand.Moreover, assigning weight to theevidence in a particular case-taking intoaccountall its flaws, contradictions, orweaknesses-is thefact-finder's verypurpose. But despite the initialappeal of thisbifurcatedregime,thehistoryof forensic science suggeststhatit falters when placed in action in thecriminal the introduction justice context. Rather thanstreamline of forensic evidence, thesystem effectively railroadsit. In thisrespect,it is perhapssignificant that theSupreme Court expresseditsconfidenceinjudicialprocess ina civil,rather The problem and nature of than criminal, case. 168 may restin the verystructure criminal process. The prosecutorialfunctionin every jurisdictionis consolidated into a For the federalcriminalcourts,theofficeof the centralfigurehead. Attorney In thenation.169 General coordinatestheactionsof all prosecutorsthroughout on a local or statewidelevel. The cities and states,officesare coordinated in turn is a repeatinstitutional prosecutor player in thesystem, handlinga wide varietyof cases inwhich an issuemay arise.'70Across thenation, then,a ina particular a single is essentially forensic technique's proponent jurisdiction theprosecutor.171 science is a government litigant: Indeed,given thatforensic
dominated field, even the government's chief proponents of the technology, the

168. raised Criminology

the same

a bill circulated all but ignored criminal prosecutions."). Perhaps wary of this prediction, that exempted criminal evidence from the proposed codification of the in Congress unsuccessfully 1st Sess. (1995). Nevertheless, test. H.R. 988, 104th Cong., Daubert empirical evidence suggests in the Daubert that "whereas civil defendants prevail challenges, most of the time criminal ... lose." Peter J.Neufeld, The to Criminal Justice: And defendants (Near) Irrelevance o/Daubert Some Suggestions for Reform, 95 Am. J. Pub. Health, S107, S109 (2005). has on the Limits of Cf. Marc Galanter, Why the 'Haves' Come Out Ahead: Speculations Law Soc'y Rev. 95 9 & (1974). Change, v. United States, 483 U.S. at 592 n.10 (citing Bourjaily 509 U.S. 171. See Daubert, 171, 175-76 (1987)). The rules do not specifically place the burden upon the proponent, but that iswho it falls to naturally. 170.

evidence was admitted, United States v. Jakobetz, 955 F.2d 786 (2d criminal case inwhich DNA In the remand o? Daubert, 834 (1992)). ruefully Cir.), cert, denied, 506 U.S. Judge Kozinski for that the Court's announced criteria would unconsidered observed pose newly problems the highly 43 F.3d at 1317 n.5; Giannelli, forensic evidence. Daubert, supra, at 109 ("Despite visible efforts to reform the rules governing experts in the civil arena, the 'junk science' debate

In fact, the Court had previously denied certiorari in a criminal case that would have "Junk Science": The Criminal Cases, 84 J. Crim. L. & issue. Paul C. Giannelli, that the Supreme Court denied certiorari in a 105, 110 & n.33 (1993) (observing

169. 28 U.S.C. ?519(2006).

Social

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

761

172 figures. witnesses,can become institutional scientific iswell-positioned role,the government institutional From thiscentralized, prosecutorcan The new accept methodologies.173 to ensure that the courts case possible foradmission.She consolidateresourcestopresentthestrongest with scientists work collaboratively can assign specializedormultiplecounsel, admission, or technique's of the in support theories and to develop arguments can are optimal. The prosecutor decide not to seek admissionuntilconditions by choosing thosecases and actively"forumshop" a new forensictechnique those courtrooms-indeed,thosejudges-most likely to be receptiveto the might accepted eager to see a technology A prosecutor proposed technology. towhich defense lawyersseem least even choose testcases with reference 174 adversary. likelytopose a formidable well equip the dynamics that At the same time, the very structural undermine evidence in turn of novel scientific to argue in support prosecutor Unlike theprosecutorial meaningfully against it. thedefense's abilityto fight function,the defense function is typicallydiffused among paid private defense or localized central offices. In many jurisdictions, practitioners, but are not even repeatplayerswithin thecriminaljustice system, attorneys 175This rather take criminal cases only when required by the courts. and comprehensive of thedefensefunction impedesconcerted decentralization when they moment at thecritical techniques efforts to respondtonew forensic gain momentum.Coordination of the initialdefense response to a new

New York & Erie R..R., 62 U.S. (21How.) 88,101 (1858)).


173. Prosecutors retain wide to discretion in almost See to resource allocation tactical decisions. Prosecutorial "unreviewed 174. such choices; 86 J. Crim. Discretion, discretion is the norm" for prosecutors). Indeed, one need not subscribe the prosecutor who believes

there is an abundance 172. By contrast, in civil cases the perceived problem is the opposite: of experts able and willing to testify to "any" opinion. See, e.g., Gross, supra note 4, at 1129-30 are too readily available"); has shown that opposite id. at 1130 ("Experience ("[E]xpert witnesses v. (quoting Winans opinions of persons professing to be experts may be obtained to any amount.") every aspect of their work, from charging L. Misner, Robert Recasting generally L. & Criminology 717, 736-37 (1996) (demonstrating that to a dark view of prosecutors to think theymight make in the integrity of the scientific technique, as such a

or atmosphere if for no proponent should, would logically choose a less formidable adversary and effort. of time other reason than to prevent unnecessary expenditure at 2, Bureau of Justice See Steven K. Smith & Carol J.DeFrances, 175. Indigent Defense, J. DeFrances, 1996 (describing "ad hoc" appointment Statistics, Feb. system); see also Carol State-Funded

of Justice Statistics (Sept. 2001) Services, 1999, at 2-3, Bureau Indigent Defense that "[t]he decentralized and diverse ways of delivering indigent defense services (commenting information nationwide make collecting difficult," and identifying the three primary mechanisms In one study of the as public defender systems, assigned counsel, and contract appointments). to relying twenty-one states that funded 90% or more of their public defense services (as opposed upon federal or local funding), only sixteen states had a state-centralized public defense program; control to local branches. Id. Even within states with the remaining three states had devolved centralized assigned not handle level, the central public defender office may programs, at either the local or national all cases. Id. at 3 (reporting that nineteen of the twenty-one states also used ad hoc counsel programs, and eleven of the twenty-one states also used contract programs).

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

762

CALIFORNIA LA WREVIEW

[Vol. 95:721

government technique therefore 176 occurs,ifat all,withmuch less frequency. Thismay be particularly truein theearlystagesof a technique, when the government has its tightest gripand theonly literature about thedevelopment or validation 177 of the method is that generated by thegovernment. As a result, thedefenseattorney may acquire a distorted perspective of a methodology's legitimacy, and even theskeptical defenseattorney may encounter a dearth of extrinsiccriticalanalyses.178 Unarmed with legitimate contrary voices, and often confronted with judicial misperception that the staff of forensic laboratories are neutral"scientists" than rather partisanadvocates, thedefense 179 is ill-positioned to mount an effective challenge. In addition, evenwhere coordination amongdefenseattorneys is possible or desirable, the natureof the defense rolemay preclude it. The defense unlike theprosecutor, attorney, meets forensic evidence reactively: she cannot pick or choose theperfectcase or theperfectforuminwhich tomount an opposition.Furthermore, pragmaticand ethical limitations thwart effective poolingof data.For instance, an attorney would be hard-pressed to advocate a third party'sretention and storage of the ballisticevidence ina client'scase for thepurposeof conductingsystematic studies. Nor could an attorney use the findings made in one case to eithersupport or attackthe findings in another, a breachof clientconfidentiality withoutrisking or a conflict of interest. Moreover, ethical rules bind defense attorneys to the zealous representation of each individual which further thedefense client,180 constrains

Michael

176. Indeed, at least preliminary data bears this out. In his study of federal and state court to expert evidence, Professor Risinger that "[t]he most observed challenges striking contrast between the state and federal numbers is the prosecution's higher loss rate in state courts." D. Risinger, Navigating Expert Reliability: Are Criminal Standards of Certainty Being Left on theDock?, 64 Alb. L. Rev. 99, 111-12 (2000) (citing the expert win rates of the government in criminal cases as 90% in the federal system, and 75% in state courts). Professor Risinger attributes this difference in part to the difference in the types of cases brought in state versus federal court. the greater resources and geographic dispersion of federal prosecutors versus state can perhaps also explain the more resources and options at their disparity. With

Id. However,

prosecutors disposal when rate. To

putting forth scientific evidence, federal prosecutors naturally succeed at a higher the contrary, at a local or state level, prosecutors have fewer resources and options. Similarly, the defense response is perhaps strongest and best coordinated at the local level. 177. See supra Part ILA. 178.

See Giannelli, Ake v. Oklahoma, supra note 148, at 1386 ("There is a special need for there is often a lack of experts when novel scientific evidence is introduced. Paradoxically, defense experts in these cases precisely because the procedure is new."). outside 179. sound

supra note 10, at 1135 ("In many of the cases we have reviewed, the courts were presented with of a given kind of asserted scientific evidence. only one-sided questions regarding the adequacy Prosecutors typically offered the novel forensic science and defendants typically offered no reply of substance. The courts in these cases often said they were Canon impressed 7 (1969). at the 'uncontradicted' expert testimony."). 180. Model Code

is Judges further view the lack of controversy in the field as proof that the principle and well-accepted, rather than as possible of "absence evidence of vigorous inquiry, an research tradition, lack of resources, or stagnation." and Solomon, Saks, Merlin impoverished

of Prof'l

Responsibility

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEW FORENSICS

763

attorney'schoice of whether to challenge admissibility. Consider a defense presented with a novel scientific attorney techniquein a homicide case. That attorney, knowingthat ultimately thetrial will turn on self-defense rather than identity, might choose tomount a lackluster challengeor no challengeat all theevidence. In sucha case, thedefenseattorney when thegovernment tenders with limitedresources would be remiss,both practicallyand ethically,in wasting precious time and effort carefullyopposing the admission of the her failureto do so will make it scientific evidence,even if she knows that harderin a future case to convince thesame judge thatthevery samekind of It isnothard evidenceadmitted earliershouldnow be considered unreliable.181 cases inwhich the to conjecturethata defenseattorney might accede to fifty defendantagrees under oath during a plea colloquy that certain forensic tohimbeforeattempting evidencecorresponded toargue ina singlecase going to trialthat thesame forensic method is entirely andunreliable. untrustworthy In addition,theefforts thegovernment expends at theearly stagesof a technique's acceptancereapprolongedrewards, because thedecisions in these initial hearings oftenserveas thefoundation for widespread acceptanceof the technique. Once a techniquetakes root,both practical and legal obstacles preclude itseasy extirpation. For preciselythereasons Monahan andWalker but insteadsimplyadopt the findings of earliercourts admissibility hearings, and rule the technique admissible.Moreover, a court confronting an decided understandably admissibility questionpreviously feels less compelled to requiretheprosecutor, perhaps thesame prosecutor who previously held a courtroom or even that same courtroom, to re-enact complexhearinginanother theearlier This practicearguably even pays heed to theprinciples of hearings. consistency and equal treatment underthelaw:when a court deems a technique admissible in one court likewiseadmissible in another,it treatslike litigants 183 alike and avoids theawkwardness of disparateresults.
cite,182 trial courts typically choose not to undertake lengthy or complicated

181.

The defense

cannot

subvert the zealous

greater good of all defendants generally. 182. See Walker & Monahan, Social factual

pursuit of a single client's

defense

even to the

Frameworks, supra note 162, at 583-84 (noting that treatment of social science evidence, requiring "[t]he same testimony about the same in case after case" research studies ... is "an inefficient use of court time"); see also Saks,

the Supreme Court's in Joiner in part decision Aftermath, supra note 104, at 233 (criticizing because "it is inefficient to allow parties to relitigate the same general question over and over"); a higher court determines, on the Scientific, supra note 13, ? 1-3.8 ("Once cf. Annotated scientific merits, that a . . . forensic identification technique can do what it purports to do (unless sense in allowing there is a change in the state of scientific knowledge), there is not much the same question to be revisited by the trial courts in case after case."). In contrast, the Supreme Court in General Elec. Co. v. Joiner, 522 U.S. 183. 136 (1997), announced that the abuse of discretion standard governed review of trial courts' appellate

admissibility decisions. courts the deference recognizing

Id. at 141-43. In applying this standard, the Court bestowed upon lower even while reserved for partly factual determinations, traditionally the potential for them to reach different decisions with regard to the same evidentiary

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

764

LA CALIFORNIA WRE VIEW

[Vol. 95:721

of scientific Thus, the law-like treatment methodologies, in effect, set forth reversestheburdenof evidentiary inDaubert: rather admissibility than askwhethera proponent of scientific evidencehas proven thetechnique's reliability by a preponderance of the evidence, criminalcourtspresume a techniqueadmissible unless a party demonstrates by some unascertainable standard thatother courts erred in admittingit, or that the science has undergone a significantchange thatwarrants revisitinga prior court's Given this shiftin thedynamicsof admissibility, and combined findings.184 with thecustomof determining a technique notice, admissibility by judicial levelof approvalbefore the law's impulsetoward need only gain a threshold and consistencytakes hold and a science admissible in enough efficiency 185 jurisdictions becomespresumptively admissibleinall others. drive courts to While considerations of consistency and equal treatment current little reason to the also have support regime, prosecutors challengethis inotherproceedings, prevailing wisdom.Able to relyon thefindings and act inpreserving has an interest only responsively upon challenge,thegovernment thestatus as evidenceof quo. Afterall, since the law tendstoview uncertainty falsehood,new theoriesonly call into question the legitimacy of those previouslyaccepted and proven.Questioning an establishedtheory becomes it servesonly to provide opposing counsel,or thecourts, counterproductive;

issues. Id. at 142.; see, e.g., Beecher-Monas, supra note 11, at 78 & n.153; Saks, note at In Saks has observed that the abuse of 233. this Professor 104, supra Aftermath, regard, in Joiner technically permits one court to uphold the discretion standard of review announced another court finds the minority-endorsed legitimacy of a majority-endorsed technique, while admissibility technique legitimate, thereby leaving the public baffled. Saks, Aftermath, supra note 104, at 234; see also Janet C. Hoeffel, The Sixth Amendment's Lost Clause: Unearthing Compulsory Process, courts are deciding the same issues differently. A particular in one court and denied in another."). 184. See, Scientific, e.g., Annotated determinations makes expertise or scientific method may

2002Wis. L. Rev. 1275, 1324 (2002) ("The second effect of the Daubert trilogyis thatlower
be admitted supra note 13, ? 1-3.8 (agreeing that a higher court no sense "unless there is a change in the state of v. Leon, 468 U.S. 897, 927 (1984) J., (Blackmun,

States Cf. United introduced in support of view of social science evidence (agreeing with majority's as "a provisional the court's decision one" subject to reconsideration outcome, but describing should experience call into question the empirical assumptions upon which the decision rested). concurring) Of evidence is excluded, incentives remain high to improve upon the science and & Monahan, Breast See, e.g., Walker supra note 162, at 823 Implant Litigation, in which court discouraged blind (relating trial court's rejection of plaintiff s scientific evidence, new event "in the that and the effect" but rather revisiting question "precedential encouraged course, where try again. conclusive questions studies emerge"). Finally, it should be noted that some forensic science admissibility are decided by statute, thus obviating this concern altogether. See Paul C. Giannelli,

revisiting admissibility scientific knowledge").

Admissibility ofForensic Science Evidence, 28 Okla. City U. L. Rev. 1, 5 (2003) (describing


various and polygraph including hypnosis, battered-wife syndrome, DNA, techniques, that received legislative validation). 185. Unfortunately, "reducing the variability and dynamism across cases severely limits the and experts across cases and for adversarial testing of diverse scientific evidence opportunities evidence over time." David Expertise S. Caudill & Richard E. Redding, Courts, and Interdisciplinarity inFederal Junk Philosophy 57 Wash. & Lee The Paradox of Science? L. Rev. 685, 750 (2000). of forensic

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

765

of the necessary to defeat thecontinuedadmissibility with the ammunition knows, "if it ain't broke,don't fix it." As everygrandmother technique.186 methodologiesactuallydiscourages of forensic Thus, the law-like treatment of and development research from scientists engaging in further government that forensic technique,and subverts the innovationand experimentation 187 development. characterize scientific typically discourage lawyerseven on the defense side from Strong incentives evidence,bothwith respectto a technique's raisingchallenges to scientific ina particular The very and to itsreliability case.188 methodologicallegitimacy that defense evidencebestowsan airof reliability of forensic "scientific" nature Counsel may simplybe unwillingto may be loathe to confront.189 attorneys and the thattheforensic technique, arguments spend timeadducingsufficient it,are infactillegitimate. cases endorsing precedential And, just as ithas been argued thatelaborate legal regimesencourage of legal arguments,190 factualinquiriesin favor to disregard defenseattorneys

186. reveal

had gone before,andwhat the future risk raisingjudicial doubts about all that discoveries may
about forensic the present. No advances means raising . . . that have been identification sciences fewer doubts. We largely frozen in court."). appearances of have seen examples in time, with little if any Similarly, "the law's desire

See

Saks, Merlin

and

Solomon,

supra

note

10, at

1131-32

("Fundamental

new

fundamental progress since their foundational for finality not only impedes the disclosure of available science, but also militates against the open Sheila that lead to the production of new scientific knowledge." communication and exchange Jasanoff, Transparency Probs.21,40(2006). 187. Of course, in Public Science: Purposes, Reasons, Limits, 69 J. L. & Contemp.

that the vast majority of counsel do just that. Giannelli, supra note 168, and observing that one testimony on future dangerousness (describing "junk science" Saks, supra note 15, at expert testified up to 127 times without meeting meaningful opposition); in which 431 that "out of 90 state court opinions handwriting (citing study indicating to the admissibility of the identification evidence was proffered there was not a single challenge see also Saks, Merlin and Solomon, forensic handwriting examiners"); supra note 10, at 1132-33 low-tech variety, suggests at 114-15 & forensic scientists usually n.347 testify, even to baseless (describing freedom with which or commercial there is no academic community to hold them accountable propositions, because and lawyers fail to attack). Indeed, studies have shown that the defense calls an expert in only a over a span small percentage of cases. In a survey of appellate court cases decided after Daubert to the prosecution's scientific of six years, one scholar found only 213 state court challenges in federal court. Risinger, supra note 176, at 125; see also evidence, and eighty such challenges Stuntz, supra note 145, at 42 (citing study of appointed counsel inNew York City, which revealed that the defense used experts in only 17% of homicide cases, and in only 2% of other felony cases). 189. Annotated Scientific, supra note 13, ? 1-3.5.1 [2] ("It appears assumed that historically

there are some situations in which the government may desire progress or a persuasive change. For instance, the government might prod scientific inquiry in response to attack made by defense lawyers on the basis of a technique's shortcoming, or a court's adverse a technique's forensic capacity. ruling on admissibility, or the prospect of increasing the history of scientific evidence in the criminal justice system, even of the 188. Again,

. . . [T]he routinely relied on by prosecutors. what they claimed they did."). 190. Stuntz, supra note 145, at 15, 21.

[defense]lawyersbroughtfew challenges to thebasic validityof a wide range of techniques


lawyers mainly that these experts could do

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

766

CALIFORNIA LA WREVIEW

[Vol. 95:721

so too might it be observed that the entrenchment of law-like scientific methodologies steerdefenseattorneys away fromscrutinizing the fact-based resultsof forensictesting carefully. Rather thanchallenge the evidence, an "overworked, underpaid,"court-appointed counsel-who may also lack the time, knowledge,or energyeven to screen the case for the reliability of its scientific -may simplytryto incorporate the findings conclusions91 intothe a prospective more meritorious defenseattackon theevidence'smethodology or application more likelyit is thatthegovernment may seem, the will obviate theattack by offering a plea that cannot be refused. In thisrespect, althoughtheadversary model conceives thesystem as "a disputebetween two sides in a positionof theoretical equalitybeforea court which must decide on the outcome of the contest,"'94the reality flatly contradicts this ideal. The adversaryin the criminaljustice systemtends to perform simplya screening function, winnowingout thosefewcases that will actually make it before a fact-finder for resolution.195 And even if defense counselmightbe able to mount a fruitful attack,resourceand role constraints inhibit defense counsel from undertaking it.This administrative, ratherthan adversarial,characterrenderstheDaubert Court's primarysafeguard-the advocateand theadversarial process 96-trulyvigilantinonly a small fraction of cases.
theory of the case or,192 more likely, negotiate a plea bargain.193 In fact, the

work

One scholar remarks that under-litigation in the criminal field "could be the result of a of factors," including that "criminal defense lawyers . . .have seen little profit" from such due to the lack of judicial L. Faigman, The Law's challenges, receptivity. David Scientific on the Law's Revolution: and Ruminations Use of Experts in Year Seven of the Reflections 57 Wash. & Lee L. Rev. "most criminal defense Revolution, 661, 661 n.2 (2000). Moreover, 191. couple is conducted by over-worked, and under-resourced underpaid, public defenders," whereas forensic science expert testimony is a time-intensive and expensive proposition." "[challenging Id. Thus, "[p]ublic defenders simply might not have the time and money to do it effectively." Id. 192. For instance, rather than challenge in a rape case, counsel might the DNA recovered

the DNA tampering, rather than exclusively relying upon an argument questioning typing results. See Richard Lempert, After theDNA Wars: Skirmishing with NRCII, 37 Jurimetrics J. 439, 444

point to the location of recovery (for instance, a stain on the bed versus in the living room) as evidence that the sex was consensual. In the highly publicized O.J. Simpson trial, the defense stain on the socks of the defendant argued that certain aspects of a blood suggested police

46 (1997).
193.

L. Rev.

2117,

of plea bargaining, but that "there is no real dispute that... the vast majority of cases of without a formal trial"). to Conviction 194. Mirjan Damaska, Barriers and Two Models Evidentiary A Comparative Procedure: Study, 121 U. Pa. L. Rev. 506, 563 (1973). 195. 196. Lynch, Daubert, supra note 193. 509 U.S. at 596.

See, e.g., Gerald Lynch, Our Administrative System of Criminal Justice, 66 Fordham 2121 (1998) (noting that "[m]eaningful statistics are elusive" with regard to the rate are disposed of Criminal

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

767

2. ThePathologiesof the SecondGeneration alleviate scientific techniques Itmight be argued thatsecond-generation many of the concerns raised in thepreceding section.Advocates of DNA have argued thatitprovidesa new standard bywhich to typing, for instance, ofDNA science will judge all forensic They suggestthattherigors science.197 forensic and that sciences, from the embarrassments traditional spare it plagued even advocate the"DNA paradigm"as a tool forreassessingfirst-generation 198 techniques. most advancedsecond-generation But evenwithin theshortlifetime of the science, DNA typing, examplesof bothquestionable methodologicalassertions in theearly1990s, For and erroneous technical application abound.199 instance, one expert testified that"in the experienceof theentireforensiclaboratory he did not know of a single instance'where different individuals community, DNA profilesforthree areunrelated or that have been showntohavematching would be highlydubious: in 2004, four Today, such a statement probes."'200 DNA testing, Sir Alec Jeffreys, declared the founder and pioneerof forensic a ten-loci and recommended thatfifteen that probewas "no longerfoolproof," or sixteen Both markers be used to safeguardagainst false inclusions.201 for yearson thebasis of falsely Virginia andTexas wronglyjailed individuals 202 DNA evidence. For nearly every laboratory mistake or inculpating were lawyers and judges who failedtocatch it.203 malfeasantact, there In short,itmay seem that the characteristics thatdefine the second presence complexity, scientific certainty, recurrent generation-theirtechnical

197. 198. 199. 200. Oklahoma,

Saks & Koehler, Id.

supra note 9, at 893.

who claimed, without contradiction, that there was "no disagreement Spencer v. Commonwealth, in the scientific community about the reliability of DNA print testing" even though two National of Science reports indicated several large areas of disagreement). Academy 201. Upgrade, methods greater Science, Alok The Jha, DNA Guardian Fingerprinting

notes 149-156. See supra text accompanying v. Crews, 640 A.2d 395, 402 (Pa. 1994); see also Giannelli, Ake v. Commonwealth execution case, supra note 148 (describing testimony of analyst in the first DNA

'No Longer Foolproof: Pioneer Calls for of Process (London), Sept. 9, 2004, at 5. Even though different DNA-typing versus STR such as VNTR power, typing have different degrees of discriminatory on DNA to determine uniqueness. Committee Forensic than three loci are necessary National Research Council, The Evaluation Of Forensic Dna Evidence 161,

Threaten

DNA Testing Woes Across State See, e.g., Steve McVicker, More DPS Labs Flawed: Mar. Thousands Houston audits Chron., 27, 2004, at Al of Cases, (describing failures at forensic laboratories across Texas, initiated after DNA revealing widespread retesting that an analyst at the Houston of biological evidence revealed laboratory falsely inculpated a 202. Adam Liptak, You Think DNA Evidence is Foolproof? Try Again, N.Y. Times, 16, 2003, ? 4, at 5 (discussing exoneration of Josiah Sutton, whom an analyst at the Houston crime lab wrongfully supra note 149; Butler, supra note 26, at 390 inculpated); Thompson, scandal). (discussing Sutton case and Houston convicted man); 203. Thompson, supra note 149.

NRC II]. 34 (1996) [hereinafter

Mar.

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

768

WRE VIEW LA CALIFORNIA

[Vol. 95:721

in a wide range of cases, and database-based comparisons-would justify But, closer confidencein their wide-scale use in thecriminaljustice system. forensic evidence examination of thehistoricalexperienceof first-generation reveals such optimismtobe misplaced.As is alreadyapparentfromtheshort many of the characteristics thatmake second historyof DNA typing, at equal, ifnotgreater, generation sciencesso appealing in fact risk places them errorin thecurrent for regime. thetechnical complexity First, with regardtoadmissibility determinations, make close and continuous of of second-generation techniques judicial scrutiny their methodological soundness less likely.Judgesconfronting sophisticated and material resourcesto scientific evidencemust investgreaterintellectual Even of second-generation conducta comprehensive examination techniques. or well-meaningjudgesmay struggleto comprehend complicatedscientific mathematical principles,204 and the heightened likelihood of errormay discourage a court from delving too deeply into such complicatedscientific knowledge. Judicialreluctance, however,only rendersthe initial hearingson a new technique more decisive, since few laterjudges will retreadthe treacherous at a resultsuspiciously to contrary path-especially if it is only to riskarriving that reached earlier.205Yet at the initial stages, second-generation

as well

that expressed by the Crews majority, which likened "match" testimony to testimony run over this dog." Crews, 640 A.2d at 402-03. According to the saying "I saw a blue Chevrolet the defendant's blue car, it Crews court, even though the testimony cannot establish that itwas Id. The difference, of course, is that jurors identification evidence." remains "useful, admissible about the trial can rely upon their intuitive or experiential knowledge sitting in the dog homicide often mirrors frequency of blue Chevrolets Detroit might assign different the contrary, a juror who first evidence from a DNA analyst to the evidence; in the area to assign weight for instance, jurors in To than jurors in San Francisco. probative value to such evidence who

401-02 (Pa. 1994); Statev. Bible, 858 P.2d 1152, 1193 (Ariz. 1993).The reasoningin such cases

cases allowed testimony regarding the testing and sampling, For example, early DNA as testimony that the two samples "matched," but refused to admit statistical evidence due v. Crews, 640 A.2d 395, to lack of general consensus in the community. See, e.g., Commonwealth 204.

is unique, and then hears testimony that every person's DNA in the suspect's the reports that the DNA sample "matched" forensic sample, can infer only that in fact the forensic sample came from the defendant. In this than the evidence warrants. Absent personal sense, such testimony may in fact be more damaging the juror simply concerning the frequency of certain genetic profiles in the population, knowledge hears has no other independent or experiential to rely in determining the upon which knowledge significance of the "match" statement. for instance, the enormous backlash that attended the trial court's decision 205. Consider, in United States v. Plaza, which held aspects of fingerprinting evidence insufficiently reliable after

an opinion of an expert witness that a of the suspect, but precluding "testimony expressing particular latent print matches, or does not match, the rolled print of a particular person and hence as obtuse and unsophisticated, and Ostracized is, or is not, the fingerprint of that person"). lambasted for breaking with one hundred years of precedent, the trial court eventually determined to save face and reverse course despite the wealth of scholarship supporting the court's initial

test. 188 F. Supp. 2d 549 (E.D. Pa. 2002), withdrawn from bound of the Daubert application at *19 (noting that the government may at 2002 WL introduce volume but available 27305, of fingerprints and to similarities between the latent print and that evidence attesting to uniqueness

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

769

and of mechanical sophistication methodologies,with their requirements knowledge,are even less likelyto have withstood the specialized technical of auditors.Indeed,thegeneralrigor of of an independent community scrutiny an air of "mysticinfallibility" sciences may also lend them second-generation that discourages critical inspection,and the existence of "real world" thatthe nonforensic applications in factdeploy analogues, notwithstanding A judgewho meaningfully different methodologies, may bolsterthissense.206 or irisscans or DNA testsgenerally thinks thatcell phones orGPS satellites when put to work in theworld may be less inclinedto questionwhether, remain sound. forensic purposes,the methodological underpinnings cases only exacerbates this The high volume of second-generation impulse. Resource constraints may ultimately persuade the "amateur of thebench,particularly those inclinedto intellectual timidity scientists"207 scientifictechniques,to lean heavily upon the with regard to sophisticated "law-like" statusof other courts' rulingsratherthan spend precious time a seeminglylegitimate Of course, themore that methodology. deciphering

conclusion. opinion). As

United

places upon courts. In its motion in turn imposes upon that prosecutorial watchfulness for pressures reconsideration in Plaza, the government argued not only with regard to its "prosecutorial that "other cases in which effectiveness" in the case at bar, but also pleaded fingerprint identification could be expected to play a significant role ... would be seriously compromised by Id. at 552-53. in effect, the the preclusion of [the requested] Thus, testimony." opinion government premised fear of the widespread its argument for admission repercussions of an opinion

States v. Plaza, 188 F. Supp. 2d 549, 576 (E.D. Pa. 2002) (vacating earlier illustrates nicely the import the government the second Plaza opinion them law-like precedence, and the admissibility rulings in a regime that accords an aside,

the court highlighted the particular importance of live testimony as one of the decisive Moreover, to reconsider. Id. at 575 (observing factors in its decision at the that one of the witnesses reconsideration hearing, FBI print unit chief "Stephen Meagher, heretofore a name in a transcript, became

In its reconsideration preferred form of proof-taking in admissibility decisions. opinion, the trial court claimed benefit from materials outside the record; however, the judge felt constrained to earmark them as such. Id. at 554 (interposing a "Historical Note (not drawn from testimony)").

thePlaza opinion also illustrates well thedifficulty infixing with specificity the Incidentally,

not only on the facts of the case at bar, but on its precluding the introduction of such testimony.

a real person, and through his live testimony I was able to get a substantially more rounded picture of the procedure"). 206. United States v. Addison, 498 F.2d 741, 744 (D.C. Cir. 1974). In the words of one scholar, "disputing the technology is like disputing the law of gravity." Hoeffel, supra note 148, at 466. As stated by a defense lawyer confronting DNA typing evidence, "[w]hen an expert comes in and says there's a one in 700 million chance that your man is not the one . . . it just kills you." Id. at 466 n.10. Thus, in the aftermath of the trial of the first man inNew York state to be convicted in part based upon DNA-typing evidence, one juror observed, "The DNA was kind of a sealer on the thing. You can't really argue with science." Id. at 515 & n.297. at 601 (Rehnquist, 207. 509 U.S. Justice Rehnquist, of course, J., concurring). Daubert, coined this phrase as an expression of skepticism at the propriety of having federal court judges

resolve complicated scientific disputes, many of which remain unresolved by experts in the field. That debate, regarding the capacity of judges to render decisions with regard to highly technical evidence, continues to rage. See, e.g., Faigman, supra note 191, at 684 (looking with optimism into "the next twenty years or so," in which and judges will become "lawyers increasingly sophisticated consumers of science").

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

770

CALIFORNIALA WRE VIEW

[Vol. 95:721

which thedefendant admitsguiltand forensic evidence is approved incases in more itaffirms thebelief thatthiskindof theevidencegoes unexamined,the and reliable. trustworthy evidence is typically But even an intellectually entrepreneurial judgewilling to fullyexercise her gatekeeper role underDaubert might not findmuch help from the of second-generation sciences render adversarialparties.The characteristics scrutinyof either general methodological legitimacyor specific case vast of cases. application unlikelyin the majority like of second-generation defenseattorneys, Given the rigor techniques, the simplyto trust judges, may findthemselves susceptibleto the temptation or "open integrity of theevidence,thus making thecase seem insurmountable to studysuch evidence in thehopes of a reward, an effort, and reaps too little more technically And the uncoveringa flawed methodological approach.209 more likely itbecomes thateven a well form,the complex theevidentiary thescienceregardless of of comprehending meaning attorney may be incapable can be expected to After all, not every attorney the effort she expends.210
and-shut."208 Many lawyers will reasonably conclude that it requires too great

208.

to pursue defense even of guilty client, while doubting ethical criminal defense counsel zealously 91 defense" Luban, Are Criminal Defenders Different?, tactics); David propriety of "aggressive Mich. L. Rev. 1729, 1729 (1993) require criminal, but (arguing that institutional considerations not civil, defense tactics). But see Darryl K. Brown, Rationing lawyers to pursue aggressive Criminal

theethicalobligationof Defense, 91Mich. L. Rev. 1703, 1703 (1993) (recognizing ofCriminal

Of course,

thatwould

violate

the ethical rules. See,

e.g., William

H.

Simon, The Ethics

An Argument From Institutional Design, 104 Colum. L. Rev. Entitlements: Defense 801, 801 (2004) (arguing that scarce defense resources should be allocated upon express, rather than covert, bases such as "factual innocence"). The methodologies this is happening. that just 209. evidence Anecdotal suggests in only a handful of cases; in techniques have been robustly challenged underlying DNA-testing the admissibility of DNA the first appellate criminal case challenging evidence, the defense called v. State, 533 So. 2d 841, 847 (Fla. Dist. Ct. App. no experts. Andrews 1988); Hoeffel, supra note in 1987, "there were evidence 148, at 499 & n.193 (describing how, after the introduction of DNA no

176, at 125 (noting that only two of 213 federal court cases studied posed direct, rather than to the DNA evidence, and in only 44% of the state court cases and 18% of derivative, challenges to DNA evidence at all). Data there any challenges the federal court cases were regarding the would of the testing?which by defense attorneys to the execution challenges mounted harder to come by. rather than by an admissibility hearing?are reflected in cross-examination In the words of one court: 210. DNA Without Without be

cases in "many" for the defense" involving DNA typing, up until the expert witnesses to the v. Castro, No. 1508/87 (N.Y. Sup. Ct. landmark hearing in People 1989)). Challenges of these technologies may receive even less scrutiny. See also Risinger, supra note application

special training, the defense would be at the mercy of the prosecutor's in the procedures used or in the interpretation of results. discern weaknesses to State-Funded DNA Experts: A. Jay Zollinger, Comment, Defense Access

a trained expert fully understands. printing is a highly complex process which only cannot properly prepare for trial, or understand defense counsel this understanding, to question results or cross-examine experts testifying for the prosecution. appropriate avenues expert, unable

to

Due

Considerations of v. Edwards, 868 S.W.2d L. Rev. 85 Calif. 1803, 1812 (1997) Process, (quoting Tennessee 682, 697-98 (Tenn. Crim. App. 1993)) (internal quotation marks omitted)). It is worth noting that of for by the availability assistance is often not readily compensated the lack of outside

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

771

master the methodological details facial recognitionsoftwareor DNA amplification and testing. of the technique Similarly,the seeming(or actual) impenetrability may discouragetheattorney from conducting a thorough inspection forerrorsin its application.Faced with the choice between spending time searching for possible errors explicable andmeaningfulenough to sway a juryand simply a defensecompatible acceptinga plea bargainor crafting with the scientific course. may quite reasonably choose thelatter evidence,theattorney Of course, an attorney may always requestexpert assistance to help interpret evidence. But, due to thehighvolumeof second-generation scientific of both low-andhigh-level evidence,and itslikely appearanceina range cases, suchassistanceis likelytobe less,not more, availingthan with respectto first A defenseattorney one hundred sciences. generation handling cases, a majority cannotfeasibly ofwhich aremisdemeanors, petitionforassistance in theforty cases that Courts jealouslyguard containsecond-generation scientific evidence. and that limited most jurisdictions require counsel to demonstrate budgets,21 their and thatthe issue is "likely tobe a significant requestis "reasonable"212 factorin [the]defense."213 Absent experteyes, thesophisticated technologies even articulating of thesecondgeneration counsel from theneed may prevent 214 forassistance even if she tried: a judge in one case denied an untrained request as "no more than a plea thatDNA evidence is simply too 'complicated."'215 a need in everycase, the But even if counselwere able to demonstrate courtsfrom sheervolumeof second-generation prevents techniques appointing

government experts, because many government See supra note 140. for assistance. 211. "[jJudges Giannelli, routinely supra deny note 148, at lawyers' requests

scientists refuse to accommodate

defense

requests

1312 (citing numerous for expert/investigative

fees"

studies demonstrating that even in capital cases R. (2d

(internal quotation omitted)). v. Mississippi, 212. 472 U.S. See, e.g., Caldwell 320, 323 n.l (1985); see also Wayne Jerold H. Israel, & Nancy J.King, Criminal LaFave, Procedure, ? 11.2 & nn. 180-84 ed. 1999).

213. 470 U.S. Ake v. Oklahoma, states, along with the federal 68, 82-83 (1985). Many government, statutorily. For example, the implement the constitutional right to expert assistance to indigent federal defendants when "necessary Criminal Justice Act provides for expert assistance for an adequate 18 U.S.C. As many See, e.g., ? 3006A(e)(l) representation." (2006). commentators e.g., have noted, John F. Decker, is frequently difficult for an indigent to obtain. See, expert assistance in Services the the Constitutional and Expert Defense of Criminal Cases:

supra note 148, at 1365. Statutory Rights of Indigents, 51 U. Cin. L. Rev. 574 (1982); Giannelli, 214. 835 F.2d See, e.g., Little v. Armontrout, 1240, 1244 (8th Cir. 1987); see also as "unusually & Ford, supra note 140, at 52 (describing DNA evidence Thompson complex, series of procedures, drawn from molecular requiring a complicated biology" which may require ... to consult experts in a variety of fields, including population genetics, chemistry, and "lawyers microbiology"). 215. Zollinger, Texas, 939 S.W.2d supra note 210, at 1812-13 (citing Cade v. Florida, 658 So.2d 550, 555

(Fla. Dist. Ct. App. 1995);NorthCarolina v.Mills, 420 S.E.2d 114, 118 (N.C. 1992); Taylor v.
148, 151 (Tex. Crim. App. 1996)).

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

772

CALIFORNIA LA WRE VIEW

[Vol. 95:721

independent experts in all, or even most, cases. Not only would such appointments be inefficient, but they would also be extraordinarily costly.The same forces thatgeneratedemand for technicalexpertise in turn work to decrease supply. The mechanical sophisticationand technical expertise associated with second-generation techniques like DNA typingor facial recognition technologies all but preclude the development of plentiful independent expertshops. Simply readingraw data in a DNA case requires and the softwarethatcan cost thousandsto tensof thousands of dollars,216 of thetechnology rapidevolution can render a largecapital investment obsolete ingranting withina short number of years. Judicial parsimony requestsfurther shrinkstheavailable expertpool.217 Those expertsactuallyassigned are also likely to be more expensive for second-generation sciences: the time the data and conveying commitment is greaterbecause both interpreting technical results tocounsel takeslonger.In short, second-generation experts are likelytobemore scarce thantheir first-generation counterparts, and evenwhen are likelytobe stretched thin and demandcostlyfees. available, they Lastly, even assuminga judge receptiveto such a challenge,a defense attorney capable and well-resourcedenough to pursue one, and an expert it remains available forappointment, unlikelythattheexpertcould undertake theexamination necessaryto truly theintegrity safeguard of theevidence.That is because themechanical sophistication and technicalcomplexityof this The infrastructure evidence all but foreclosesindependent research. necessary for true methodological testingis simply lacking.At the same time, the of second-generation database-dependency sciences, and the privacy and secrets concerns they raise, effectively proprietary prohibitaccess to the
material necessary for independent research. Manufacturers of DNA typing

or biometricscanningsoftware kits,cell phone or searchengine technologies, may bristleat disclosingbroadly the technology particular underlyingtheir of even under a court"gag" order.Similarly,the relinquishment techniques, in databanksforexploratory data stored indiscriminately purposes-whether DNA profiles, or cell records-understandably irispatterns, raises legitimate concerns aboutpersonalprivacy. The database dependency of second-generation also means technologies thatscrutiny of these techniquesforcase-specificerrorsin application itself access to largevolumesof data,whichmay not be feasibly requires disclosed,
or feasibly reviewed,

thelocation of a phone at a particular timerequires report accuratelyidentified

in every case.218 For instance, verifying that a cell-site

216. 217. 218.

contamination

to the laboratory's logs and corrective action files, laboratory protocols, maintenance logs, proficiency supra note 139, at 815-16 & n. 152 testing results, caseworker files, and so on. See, e.g., Giannelli, cases). (explaining need formore extensive discovery inDNA

supra note 3 6. and Solomon, supra note 10, at 1092 & n. 112. Saks, Merlin a skilled reviewer must have broad access At a minimum, See

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

773

theaccuracyof the towerlocation, verifying all theprecursor data, including clarity of signal,lackof interference with signalreception, and correspondence to actual physical terrain; this is obviouslydifficult to scrutinize closely in everycase.219 Or consider,forexample, theFBI DNA-lab scandal concerning analyst of "negative controls" Jacqueline reports Blake, who pled guiltyto falsifying the data used to demonstrate thatno contamination has takenplace during a problem with the fileson Blake's computer.221 Similarly,an analystfired froma private laboratory for substituting clean controlfiles in problematic was discoveredonlywhen a reviewer noticed that her negativeblank samples an auditof a fileswere strangely identicalin every case.222 More recently, Massachusetts crime lab revealed "instances in which laboratory officials entered the same genetic profile under two different ID numbers in the database," and in which an analyst reported"DNA results in four cases matched the geneticmaterial fromold rape kits when theyhad not."223 reviewof thedocuments Independent relatedtoa singlecase simply could not 224 have capturedthese errors. Review of the analyst's entirebody of work might have caught the suspiciousdata, but of course no court would have on thechance that mandated suchbroaddisclosureex ante,simply theanalyst's workwas notup to snuff.
testing.220Her actions only came to lightwhen a coworker working late noticed

219. Associated

See,

Press,

e.g., David A. Lieb, States Seeking to Track Cell Phones for Traffic Conditions, to track drivers through their cell Oct. 8, 2005 (detailing pilot programs tracking technology). Release,

phones and explaining 220. See Press Filing False DNA

however, Ms. Blake apparently substituted a completed file in 103 cases, and that copied file as a blank run in the case. See Maurice Possley, Steve Mills & Scandal Touches Even Elite Labs, Chi. Trib., Oct. 21, 2004; Richard Willing, Flynn McRoberts, DNA Mueller Crime Lab after Questionable Tests, USA Today, May 1, 2003, at 3A. Defends Notably, ordinary sample contamination occurs so frequently thatmost labs require their analysts blank injections, misrepresented to keep their own DNA profiles on file, so that they can be compared against findings. See, e.g. this practice). Kobilinsky, Liotti, & Oeser-Sweat, supra note 26, at 99 (advocating of Justice, Office of the 221. Dep't The U.S. FBI DNA Inspector General, and Practice of Protocol Vulnerabilities Laboratory: Review ii (May 2004), available at http://www.usdoj .gov/oig/special/0405/final.pdf. 222. See Aff. of Dr. Robin W. Cotton, Ph.D, Maryland 223. v. Kenneth Ernest Abend, Nos. K

to of Justice, Former FBI Biologist Pleads Guilty Department (May 18, 2004). Specifically, Laboratory Reports negative controls are blank to safeguard against and expose any contamination thatmight have occurred injections designed in the testing process. If a blank comes back with stray material, then the analyst knows that the results of a test, especially of an "unknown," may be the result of contamination. Rather than run

02-506 andK-0401903 (Nov. 2,2004), at 3& attachment B (onfilewith author).

Feb.

tubes were

case ofwoman implicated in sister'sdeath when sample (Greenville), Aug. 28, 2005 (describing
erroneously Post, Aug. mislabeled); 21,2005, Tom at Cl (cataloging Jackman, Paternity Suit Raises Doubts a list of faulty DNA tests). About DNA

at Crime Lab, Boston Jonathan Saltzman, US Audit Found More Problems Globe, 1,2007, at Al. inMurder 224. Journal Reporter See, e.g., Phoebe Zerwick, DNA Mislabeled Case,

Tests, Wash.

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

774

LA CALIFORNIA WRE VIEW

[Vol. 95:721

materialsnecessaryto findsuch flaws ineverycase Yet disclosure of the technologies are used is all but impracticable. Not inwhich second-generation only would requiringsuch sweeping document disclosure in every case a jurisdiction,225 but it would also demanddisclosure of an effectively bankrupt largequantity of paperwork.226 unsustainably, and perhaps even impossibly, And no expertor attorney could, as a matterof practice,undertakesuch a reviewineverycase. thecourts'demonstrated reluctance to approve the From this perspective, means necessarytoeffectively evidence isnot in the inspect second-generation end pathological; itmay in some respectsbe quite reasonable.But this make second-generation reasonablenessinvites danger: theveryqualities that will neverencounter technologies so desirable make itall the more likelythey on is troubling adversarial scrutiny of anykind. And while thislackof scrutiny when consideredin light of thevery more troubling itsface, itbecomes all the in power of these technologies, real possibilitythat, given the investigative 227 onlyactualevidenceof the defendant's guilt. many cases they may be the III
WHERE Do WE Go FROM HERE?

When itcomes to second-generation evidence, somemay argue thatthe As notedabove, in effective lackof scrutiny is,statistically speaking,tolerable.

225.

from $1,000 to $10,000); Giannelli, Ake v. Oklahoma, supra note 148, at 1398 see also id. at 1363 ("If the standard [for appointing an (reporting expert costs as high as $28,000); the right is gutted. If the standard is too lax, the costs skyrocket."). expert] is too demanding, of Sciences recommended the first National Academy Nevertheless, just that: they suggested that expert as ranging with because few attorneys can deal in all cases involving DNA, experts be appointed in on DNA Technology Committee Research this type of science. National Council in Forensic NRC Science 147-49 (1992) Forensic DNA Technology Science, I]. [hereinafter The subsequent report recommended appointment of experts, either to the court or to the parties, defense DNA and noted

costof a DNA U. Cm. Legal F. 395, 396 n.l 11 (1998) (citingthetypical ExpertAssistance, 1998

John Devlin,

Comment,

Genetics

and Justice:

An Indigent Defendant's

Right

toDNA

of multiple evidence might require the appointment that the complexity of DNA II, supra note 201, at 169-70. experts. NRC cases reveal courts' struggles to the sufficiency of discovery inDNA 226. Early challenges to to strike the right balance between the defense's interest in obtaining comprehensive material interest in controlling assertions, and the government's challenge the validity of the government's the burden of amassing documents. See, e.g., United States v. Yee, 129 F.R.D. 629, 630 (N.D.

with Ohio 1990),affdsub nom United Statesv. Bonds, 12F.3d 540 (6thCir. 1993) (commenting
... the to accept request that "the defendants appeared government's regard to broad discovery within Fed. R. Crim. P. that they are seeking are not encompassed contention that the materials the request, mainly because 16"). In Yee, the magistrate judge ultimately granted the defendants' case posed one of the initial challenges to the admissibility he of DNA evidence. Moreover,

and scientific assessment cited as support the lack of "extensive independent specifically as well as that have been developed by the F.B.I.," replication of the reliability of the procedures in bona fide questions about each of the categories the "fact that the defendants have developed which they are seeking discovery." Id. at 631. 227. See text accompanying supra note 95.

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

775

no errortookplace, and second thevastmajorityof cases, it is likelythat more crediblesourcesof evidence thanthe generation sciencesare at base far 228 has neverbeen traditional forensic sciences. But thecriminal justice system edict goes: with being a randomgame of chance; as the familiar satisfied one innocent suffer."229 It is "[b]etterthattenguiltypersonsescape thanthat as theuse of scientific so too will evidence increases, reasonabletoexpect that, two increase thenumberof errorsattributed to its use. More importantly, all the evidence make meaningfulscrutiny more aspectsof second-generation indispensable. First,thescale of errorthat can occur among second-generation thanthat which occurred among the techniques is an orderofmagnitude larger may wrongly inculpate first-generation. hair comparison Whereas a faulty someone in one case, a wrongly calibrated machine can chum out large volumesof erroneous information and tarnish multiplecases.Or considersome DNA typing: of theerrors that amanufacturer a may contaminate may occur in an analyst or a technician may fail to runpositiveor negativecontrols, kit,230
may erroneously

or unrelated multiplerelated cases aswell. not justa singlecase, but evidence is apt to be faultyin fewer Second, even if second-generation will be at their That is not overall cases,when itdoes fail thestakes highest. appear so irrefutably probative, just because second-generation technologies more than but also because theyallow criminalcases to be built on little chargesare routinely brought based upon only a forensic proof: for instance, "cold hit" DNA match.232 Moreover, in some cases the crimemight have more difficult to defense all the occurredyears before, making an effective
233

input data into a database.

231

Such mistakes

can compromise

muster.

At

the very

least, in cases

involving no evidence

but forensic

Some might even argue that the error rate of DNA typing ismore favorable, and thus see to v. But of that Ake Giannelli, supra note 148, at 1396 Oklahoma, preferable, eyewitnesses. ("A British study (albeit small) found that '38 per cent of defence [sic] lawyers who had obtained 228. an independent analysis' of DNA evidence received reports that 'differed from those of the prosecutions' expert.'"). 229. William Blackstone,

4 Commentaries

(1970) (Harlan, J., concurring) ("[I]t is far worse man go free."). at the manufacturing 230. Contamination another incident

*358; In re Winship, to convict an innocent man

397 U.S.

358, 372 than to let a guilty

level has occurred in the United Kingdom, and States. See, e.g., Becky Pallack & Kim Smith, recently arose in the United Contaminated DNA Strikes Three Cases, Az Daily Star, Dec. 13, 2005 (describing how same unknown sample turned up in testing at Tucson crime lab and then in two Florida crime labs, causing officials to conclude that the tubes used at a factory). for testing were contaminated the court to preclude the defense from even that it unduly prejudiced the jury with regard

case moved in the Tucson Interestingly, prosecutors to the the contamination jurors, mentioning arguing

to the reliability of testing in that case; their request was denied. Id. 231. For example, a Las Vegas lab inadvertently switched two DNA profiles as it entered as a result, an innocent man spent a year in jail awaiting prosecution them into the database; for sexual assault. Glenn Puit, Police Forensics: DNA Mix-up Prompts Audit at Lab, Las Vegas Apr. 19, 2002, at IB. See supra Part I.B.2. What innocent person

Review-J., 232. 233.

could

recall why

they frequented

a certain

location

or made

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

776

CALIFORNIA LA WRE VIEW

[Vol. 95:721

evidence,justice dictates implementation of the most exactingsafeguards. If theonly evidence ina thousand cases across thecountry is thetestimony theft of a forensicanalyst that the defendant matched the evidence when the likelihood of a random matchwas one in 240 billion, thensurelyeverything shouldbe done toensurethat such testimony is in fact accurate. However, existing recommendations forimproving thequalityof forensic within theconventional science incourttendto stay framework, askingonly to moremoney forexperts,234 shore itup by granting bettertraining for providing requiring more elaboratehearingsand discovery,236 selecting more lawyers,235 has its own merits, and if implementedcould these recommendations evidence in thecriminaljustice dramatically improvethequalityof scientific 239 do notaddress, much less rectify, the of system. Yet they particular economy the criminal justice system, which perpetuates the introduction of faulty forensic evidence.Instead,theconventional fixesrely view of upon an outdated nature the of forensic evidence, where case-specific review plausiblysuffices to ensure thequalityof evidence.They assume: thatan attorney iswilling and able (or even obligated) to engage in extensivepretrial investigation and from winnow contestable uncontestable maneuveringto cases; that judgeswill conductan adversarial a motionshearingor of somekind (whether proceeding is much less possible or trial) in thosecontestable cases; and thatit efficient,
competent juries,
23723

and allowing

for greater independent testing.238 Each

of

certain purchase or undertook other such activities on a random day many years earlier, or even to verify their assertions? locate thewitnesses in the Paul S. Milich, Science 234. Controversial Giannelli, supra note 109, at 475-76; 43 Emory L.J. 913, 925 (1994). Courtroom: Daubert and the Law's Hubris, Giannelli, supra note 109, at 475-76. C. Thompson, Evaluating William the Admissibility of New Genetic Tests: Lessons 22, 99-100 (1993); Christopher G. Shank, from the "DNA War", 84 J. Crim. L. & Crimonology to Protect in Criminal Trials: Modifying the Law's Approach the Accused Note, DNA Evidence 34 Ariz. L. Rev. 829, 870 (1992). Genetic Evidence, from Prejudicial 235. 236. L. Rev. 237. Graham C. Lilly, The Decline 53, 85 of the American Jury, 72 U. Colo. (2001). Of course, debates rage concerning whether scientific evidence exceeds fair expectations scientific disputes often of jurors' abilities, or whether juries are up to the task of resolving unresolved among the experts in the field. Compare, e.g., Joseph Sanders, Scientifically Complex L. Rev. 355, 360, 363 48 DePaul Cases, Trial By Jury, and the Erosion Process, of Adversarial . . .poses (1998) and Lilly, supra, at 61 (arguing that "long-term trends in the nature of litigation to their traditional serious questions about the potential of American juries adequately perform roles"), with David W.

Scientific Evidence: 238. 239.

The Standard the 37 JurimetricsJ.23 (1999); Edward J. Imwinkelried, Jurybox, for Admitting
A Critique from the Perspective of Juror Psychology, 28 Vill. L. Rev. 554,

Shuman,

et al., Assessing

the Believability

of Expert Witnesses:

Science

in

570-71 (1982-83).

are specifically designed for a civil justice cannot into the criminal be and system justice system. For instance, some reasonably transplanted juries or qualifying specialized impinge upon other rights of criminal defendants: suggestions could the defendant's from determination certain impermissibly prejudice jury removing questions Sixth Amendment right to a jury of peers.

Giannelli, supra note 139, at 816-17. Of course, some of these recommendations

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

777

toreviewthe of all scientific desirable,toassignexperts outcomes testing. Moreover, such proposals fail to acknowledge that,even if resource constraints entirely disappeared,the monitoring of second-generation sciences in a requires a scope of inquiry broader thanthat accorded to each defendant case, and a recalibration of thebalance of power between the singlecriminal centralized and decentralized government defense. Against thisbackdrop, thisPart attemptsto set forth a nonexhaustive concerns. catalogof recommendations keyed to these particular Althoughnone of theserecommendations alone offersa completesafeguard, if implemented together, they have thepotentialto improve theuse of forensic dramatically evidence in thecriminal justicesystem. A. Loosening the Government's Grip on the Technology As argued above, forensicsciences generally,and second-generation technologiesin particular,require reviewsof greaterdepth and breadth to uncover flaws in either the underlying methodological techniqueor the execution of thattechnique ina particular case. Comparinga recovered writing to the suspect'swritingexemplar may be the sole basis of a first-generation thereliability of conclusionsdrawn inDNA typing finding. By contrast, may or match from material derived databasesof genetic dependupon probabilities examination of the work an analysthas done across cases. upon comparative Thus, effective monitoringof second-generation evidence demands close in not justat the individual case level court, but also across theentire scrutiny But given theappearanceof thesetechnologies in a high rangeof operations. volumeof cases and the privacyand proprietary broaddisclosure concernsthat howmight suchreviewstake may raise, place? Scholars and advocates have urged perhaps the singlemost important of the forensiclaboratory to ensurebetter change:wide-scale reform system, of a neutralscientific qualitycontroland recast thecultureto that lab rather thanan armof thegovernment.240 forensic laboratories are Truly independent

have recommended laboratories with higher quality creating independent and Giannelli, 109, at 469; supra note accreditation, strengthening protocol, supra note 11, at 100-01; and proficiency review of labs, id. at 474-75 & n.202; Beecher-Monas, One studies, Saks, Aftermath, supra note 104, at 239 & n.41. ongoing validation encouraging recent innovative approach to this argument is to introduce a system of "competitive self Scholars technicians,

240.

or corroborative supra note 11, at 90 n.250. The testing of scientific evidence. Beecher-Monas, cannot alone ensure the integrity of all right to independent or duplicative testing, however, forensic evidence. First, inmany cases, the DNA sample is exhausted by government testing, and no evidence remains for independent submission. Second, a costly and time-consuming procedure such as duplicative of verifying the integrity of the testing cannot serve as the ordinary means government's results. Third, there are many strategic reasons why defense counsel might elect not

to create "rivalrous in labs, such that technicians would know that regulation" redundancy" sent tomultiple is periodically evidence laboratories for testing and quality control. Koppl, supra note 118, at 256, 267. Some also recommend the institution of a legal entitlement to independent

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

778

CALIFORNIA LA WRE VIEW

[Vol. 95:721

essential in part because theyformthe firstlineof defense against shoddy forensic science.241Such labs might readily find homes in large public universities, or innot-for-profit organizations. system, even assumingthat But thecreation of an independent laboratory be accomplished,isnot alone enoughfortworeasons. such a featcould in fact First, to fosterthe maturationand criticalexamination of complex second no single institution generationtechniques, ought to be allowed to operateas the sole custodianof the toolsnecessary to develop and challenge scientific orthodoxies.Second, supervisionover the proper implementation of those technologies requires constant and ongoing scrutiny at the wholesale, not just retail,level.The criminal justicesystemtherefore needs institutions capable of each kindof oversight. and empoweredtoundertake 1.Centralized Oversight Agencies Methodological development monitoring requiresthat and qualitycontrol a neutraland bipartisan entity have thepower topursue,and encourageothers or Board, should to pursue, researchand auditingfunctions. Such an entity, the have members drawn fromall relevantcommunities:the government, defensebar, and private industry, Arned academia, or forensiclaboratories. of with a research Board would oversee theequitabledissemination budget,the researchfundsforstudies.242 The Board would also have access to all private data related to a particular technique,and could award or proprietary circumscribed access to researchers consonant with the needs for New techniques or theprotection be of tradesecrets. would first confidentiality to the submitted disseminate Board,which could then proposed methodological The Board could and peer review. approachesforthepurposeof close scrutiny that also issue periodic "state-of-the-technology" reports clarifytheongoing in a technology's use or development, areas of uncertainty and outline the

a to conduct routine independent testing. For instance, to the extent that a jurisdiction bestows on the it also includes a proscription that right loses meaning unless right to retest evidence, comment on the exercise, or failure to exercise, such a right. Until courts resolve the government's questions of confidentiality and evidentiary use surrounding independent testing, defense lawyers will be reluctant to submit all evidentiary items?particularly those already shown to "match" the confirmatory testing. 241. from civil liability incurred from substandard or They also help insulate municipalities fraudulent work. See, e.g., Brandon L. Garrett, Innocence, Harmless Error, and the Federal L. Rev. 35, 98-99 (2005) (reporting on a Cleveland civil Law, 2005 Wise. Wrongful Conviction suit lodged after an analyst falsified hair and blood evidence, and inwhich the settlement included provision 242. of a "permanent, independent scientific monitor"). Institute of Justice, a branch of the U.S. The National of Justice, currently Department the of such research, but this entity is clearly inadequate. First, because the work, the government also defines what projects are interesting or worthy defendant?to

solicits a limited amount government solicits of being undertaken, determinations.

rather than allowing a vibrant and diverse research community tomake such the government selects the recipients of such grants, it is able to Second, because skew the awards toward researchers sympathetic to its interests.

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

779

deemed acceptableor unacceptable.It could also variousapproachescurrently expound standardsand protocolsof model practices for the executionof forensic techniques. particular proposed by Professors mirrors thatfirst This model inmany respects model to thehard social authority Monahan andWalker, who extendedtheir could resolvequestions sciencesand outlineda "NationalScience Panel" that In Walker's and pertaining to causation in breast implant litigation.243 of such panels law-like Monahan's model, courts could give the findings but would be capable of being "overturned," deference: the findings In thecivil arena,these kindsof panels aim toproducea correct. presumptively occur throughout thenation, questions that coherent responseto thescientific But in the more efficiently and consistently. so that litigation is conducted criminalarena, such panels have the potential to producemore than just to the government as a counterweight they might also function efficiency: of crimelabs) whetherde jure (as theoperator of forensic science, domination consumer of forensic services). or de facto(as the primary of amodel forsuch have something already techniques Second-generation of experts who a panel.The debutofDNA analysiscatalyzedtwoconvocations became the"how use ofDNA evidence that manuals fortheforensic produced scientist Eric Lander exposed After distinguished to" guides for courts.244 in an evidence inPeople v. Castro, resulting numerousflaws in the forensic and defense experts statement from the government joint unprecedented concluding that the evidencewas unreliableand eventuallyresultingin its theNational Academy of Sciences respondedto his call fora exclusion,245 DNA typing.246 to investigate forensic committee of government Accordingly,in 1992, theFBI, alongwith a consortium a committee charged a that it had commissioned from issued report agencies, of of scientific knowledgein thefield and analyzingthestate with summarizing over thefirst conclusions, Aftercontroversy report's DNA evidence.247 erupted the meritsof issued in 1996.248 Scholarlydiscussion regarding anotherreport What is clear, abound.249 thepanels, and of theconclusionsthat each reached, of thepanels served to inform and educatejudges however, is thatthereports areas of dispute in thefield, and provideda and litigators about the legitimate

national

Walker & Monahan, Social Facts, supra note 104, at 823-24 (describing operation of science panels). 244. The American Bar Association also recently endorsed national standards for the use of forensic DNA evidence. 243. 245. ed., 2006). 246. 247. 248. 249. Jennifer Mnookin, Id. at 227. NRC NRC See, I, supra note 225. II, supra note 201. e.g., Lempert, supra note 192, at 465-68. People v. Castro, in Evidence Stories 226-27 (Richard Lempert

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

780

CALIFORNIA LA WRE VIEW


250

[Vol. 95:721

useful summary and reference forbest practices. Even thedebate between thedisparate conclusionsreached by thefirst and secondpanels has contributed a richnessto theconversation criminal that justicehas otherwisesorelylacked with regardto forensicscience. In this regard,thebeneficialeducative role played by neutral expertpanels, particularly with respect to resolving the complicateddisputes likelyto characterize the second-generation sciences, is illustrative and instructive.251 But an isolated or sporadicconvocation of expertsisnot enough.Second generationtechnologiesrequireconstant New monitoringand development. technologies may arriveon the scene, and old technologies can stillpresent novel or innovative For instance, while thefoundations questions.252 ofDNA

250.

United Statesv. Adams,Nos. 03-2108, 03-2152, 2006WL 1888737 (3d.CirN.J. 2006) (referring to standards setbyNRC); People v.Watson, 789N.E. 2d 375 (111. App. Ct. 2003) (same); affd by 2d. 271 (111. 2005); United Statesv. Trala, 162F. Supp. 2d 336, 348 (D. People v.Watson, 214 111. Del. 2001) (same),affd byUnited Statesv. Trala, 386 F. 3d. 536 (3d.Cir. 2004); vacated byTrala
v. United convened by a neutral party, and so while the committee may have strived to complete a fair and balanced report, and may have achieved it cannot be said that their origins were neutral. Of course, that goal, the of such panels would garner additional conclusions legitimacy if they were deliberately composed 251. of each for example, the "two-step process" used by the by a neutral and disinterested body. Compare, and Monahan. & Monahan, See Walker judge in the civil case studied by Professors Walker a In that case, the judge first designated supra note 162, at 808-809. Implant Litigation, to provide "names of neutral, impartial persons who have indicated expertise" "Selection Panel" and would be able to communicate well and serve, and then chose the four-person panel from that list of names. Id. be appointed legislative branch. 252. There were and actuality of fairness, such panels Ideally, to preserve both the appearance and monitored by a neutral party, such as a member of the judicial or even remains States, 126 S. Ct. 1078 (2006). Neither of the prior two committees was

See,

e.g., United

States v. Ewell,

252 F. Supp.

2d

104, 109 n.8

(D. N.J. 2003),

affd by

members

Breast

would

some question to draw whether the population used samples The randomized. insufficiently large and not demonstrably government researcher who published the study upon which the frequency tables are based looked at sample I, supra note 225, at 91 (1992); P.J. Bickel, group sizes in the low hundreds. See, e.g., NRC conclusions Discussion Nat'l Acad. Sci. 5497 (May 1997) the that assume (observing [with] assumptions" modeling that the data is drawn from a "random sample[]" of the relevant population and that no linkage is present). For instance, many charge that such a small sample size is insufficiently random, and at large are inappropriate: the frequency thus inferences about the composition of the population table for genetic characteristics in the African American population from 203 for Caucasians, and from 209 for Hispanics. on the Thirteen CODIS Bahamians, Hispanics, was See developed from only 210 et al., Bruce Budowle, Loci in African Americans, of The Evaluation of Forensic Evidence, that "many scientists would not agree 94 Proc.

profiles;

Data Population U.S. Caucasians,

Core Short Tandem Repeat 44 J. Forensic and Trinidadians, Sci. 1277, Jamaicans, 1278 (2001). A subsequent study attempted to address these concerns, and drew upon data from 1749 African-Americans, 1511 U.S. and 1421 Hispanics; that study Caucasians, roughly concluded that these populations were in Hardy-Weinberg B. equilibrium. See Bruce Budowle,

& R. Chakraborty, CODIS STR Loci Data from 41 Sample Populations, 46 J. Shea, S. Niezgoda Forensic Sci. 453, 453-89 later criticized. See Dan E. Krane, Travis E. (2001). That study was L. Raymer, William M. Shields, & William Laurence Mueller, Michael C. Thompson, Doom, 49 J. Forensic Sci. 453 (2004). At present, laboratories use a "theta correction" to Commentary, account for the possibility Liotti of substructure among certain populations. & Kobilinsky, it is worth noting that the same questions have Oeser-Sweat, supra note 26, at 156. Finally,

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

781

typing are now fairly firmly entrenched,new controversies emerge 253 constantly. By means of illustration, consider the following concrete example from a controversy currently unfolding in the DNA research most commonly used in theUnited community. The DNA-typing technique States todayexaminesgenetic information at thirteen different places, or loci, 254 on thegenetic strand. Regardless of thenumberof loci developed,DNA of a "match" analyststypically calculate thesignificance between theforensic and known samples using a method known as the "product rule."255 The reliesupon dataFBI scientists the productrule,in turn, developed todetermine of geneticinformation, frequency of particular alleles,ornumerical expressions in the population.The validityof thismethod depends upon two critical that thefrequency tablesderivefrom a sufficiently assumptions: first, largeand random sample toallow for generalconclusions,256 and second,that thereisno of DNA typingintoevidence and across thenationhave accepted theresults ruled theproductrule,and the frequency tablesunderlying it,an acceptable thesignificance of amatch. way to represent Yet, recent evidencecalls into question theaccuracyof using theproduct match probabilities.258 How thatevidence rule to convey was uncovered,and
link or correlation between each piece of information.257 At present, courts

DNA evidence. Criticshave chargedthatthe mtDNA database dogged theuse ofmitochondrial


similarly Frederika Limitations 253. contains A. too few samples to adequately See capture the true population frequencies. L. Roth & Edward J. Ungvarsky, Database Kaestle, Ricky A. Kittles, Andrea on the Evidentiary DNA Evidence, Value of Forensic Mitochondrial 43 Am. Crim. L. in miniaturization, YSTR

Rev. 53 (2006).
familial

See supra Part LB (referencing developments and so on). searching, mixture deconvolution, 254. Butler, supra note 26, at 111. 255.

typing,

DNA

Butler, supra note 26, at 501; Frederick Bieber, Science and Technology of Forensic and The in DNA Criminal Justice System, supra note 24, at 35; see Profiling, Liotti & Oeser-Sweat, Kobilinsky, supra note 26, at 167-69. For an excellent primer on see Kobilinsky, Liotti & Oeser-Sweat, supra note principles and techniques of forensic DNA, 26, at 1-196. the analogy if the frequency used above, tables were conducted Following by a two at shoe hundred then store, sampling people large-sized naturally the purported frequency of large-footedness would fail to reflect the actual frequency of large-sized feet in the population. Or 256.

in the general population with crossed eyes if, for some reason, it should turn out that all people also have that each variable was prove large feet, then the assumption independent would incorrect. 257. Kobilinsky, See, e.g., Butler, Liotti & Osser supra note Sweat, supra

26, at 501 (describing product rule); see also note 26, at 135, 341 (defining "linkage" and For example, when analysts look at only themale fraction assumption of "linkage equilibrium"). of DNA, the product rule cannot be used, because the various pieces of genetic information are Liotti & Osser Sweat, supra note 26, at 116. Emerging independent research also indicates that the second assumption?that at the various not hold true for all populations. After three years independence loci?may 258. its frequency to be linked. Kobilinsky, of of

known

the data upon which it based battling government refusal to disclose experts obtained a fraction of the data and conducted independent experts uncovered that in certain Native American populations,

analysis. As it appeared that correlations were

tables, defense a result, these

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

782

CALIFORNIALA WRE VIEW

[Vol. 95:721

what has happened since thediscoveryillustrate thespecific problemssecond generationsciences raise.Notably, questionswere first raisedwhen an alert analysthappened to observe,and thento pursue out of laudable intellectual curiosity, the fact that theDNA samples of two unrelatedindividuals(one Caucasian, oneBlack) matchedat nine loci. Under thestatistical models thenin place, a personpicked at random of would match that nine lociprofileat a rate 1 in754million inCaucasians, 1 in561 billion in African Americans,and 1 in analystchecked therestof the60,000 persondatabase forsuchmatches, and uncoveredninety pairs of individuals who matched at nine loci, and several But althoughsuchmatches prompted pairs at ten and even eleven loci.260 serious questions about the accuracy of the populations statistics used in criminalcases, theanalystcould takeno further action,because she had "no or thefunding to look intoitanymore."261 time Hence, theresearch stalled. a defenseattorney, a man of thesefindings, Upon learning representing with a "cold hit" crimeon thebasis of a nine locimatch,262 charged attempted to gain access to the data to conduct further investigation. Predictably, however, thegovernment vehemently opposed the requestand cited to the arguments elaboratedabove: privacy concerns,263 the burden that itwould a single place on thegovernment "to requiretheState todo a searchto satisfy was outside the scope of the and the fact thatsuch research Defendant,"264 In another analyst's duties.265 hearing,thegovernment responserevealed the intimacyand degree of state and federal coordination: themanager of California's databanktestified thatthefederal had warned thatif a authorities courtordered of thedatabase forresearch disclosure purposes,thestate"could lose our authorityto use the software,"since "if the FBI pulled our
113 trillion in Southwest Hispanics.259 Simply because she was curious, that

evident between several of the loci that had been glossed over by government researchers. Dan E. L. Raymer, William M. Shields, & William Krane, Travis E. Doom, Laurence Mueller, Michael 49 J. Forensic of data C. Thompson, Sci. 453 (2004) Commentary, (noting that "examination . . . shows ... for two Native American reported populations significant departures from HWE at three loci for each population" and arguing that "these loci Equilibrium] [Hardy-Weinberg should not be used when genotypes 259. author); to compute the frequency the product rule is employed in these populations"). at 21-22, In theMatter of theApplication See Transcript of Proceedings also Kathryn of multi-locus of the State of DPS Crime Locus

An Order, (Oct. 17,2005) (No.MISC-001) California for


see Poster Presentation Laboratory, Match Between Two Apparently 260. 261. 262. Plus" Transcript, /?.at 25.

Arizona and Brian Koeneman, Troyer, Theresa Gilboy, at Promega A Nine 12th International Symposium: Unrelated Individuals, Phoenix, AZ (2001). supra note 259, at 23, 26-30.

[hereinafter Transcript](on filewith


STR

than an hour, and the other only a couple of months. 265. Transcript, supra note 259, at 58, 70.

some states routinely do only the "Profiler As a means of saving time and money, test,which looks to nine loci, rather than also do the "Cofiler" test to reach the full thirteen. 263. Transcript, supra note 259, at 70. 264. Id. at 6. Ironically, the analyst testified that one requested search would take a less Id. at 57-58, 71.

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

783

it is just like authorization, Microsoftwho saidwe can't useWord anymore. Of course,thefederal flies We are shutdown."266 government's interpretation in the face of the federalstatuteenabling theCODIS, which specifically provides that the databases be made available "if personally identifiable information is removed,fora populationstatistics database, foridentification research and protocol development purposes, or for quality control courtruledin favor and purposes."267 Nevertheless,thetrial of thegovernment, againstfurther inquiry.268 Litigationon thesame issuehas occurredinvarious jurisdictions around 269But underthecurrent it is left the nation,tovarying success. to legal regime, statetrial and to individual individual defenseattorneys, judges, topress foran answer to the question,"how many multilocimatches are in the state and case litigation is the nationaldatabases, and why?" But althoughindividual only place to seek theanswer to that question, it is hardly thebest place. A togrant courtisunderstandably reluctant motion topursue a defense singletrial such broad researchsimplyfor thebenefitof a singledefendantin a single case.Moreover, judgesmightunderstandably be wary about issuing ordersto a comprehensive examination of the open thedatabaseswithoutundertaking privacyinterests involved, thecaliberof theresearch proposal,or theestimated a busy trial courtis typically scopeof the project:all questionsthat ill-equipped At thesame time,thegovernment, to consider. which controlsthedata, lacks most critically, and perhaps the interest, the money, time, expertise to conduct As a consequence,critically is essential inquiries. scientific important inquiry thwarted. effectively sucha questioncould be asked of a neutralentity, Imagine, however,that which then had the power to grant appropriateaccess to independent This body could efficiently researchers capableof answeringit.270 negotiatethe

266.

in DNA Database, Laura Ernde, Defender J. (S.F.), Jan. 11, Cites Mistakes Daily to the Office, appealed attorney, Bicka Barlow of the San Francisco Public Defender's Court of Appeals, which subsequently denied a writ of appeal. Davis v. Super. Ct. San Francisco, No. Al 16603 (Cal. Ct. App. Mar. 8, 2007). 268. 2007. The to disclose the government the court ordered only the barest of Ultimately, a summary of the report of the analyst's findings. See Minute Entry, In theMatter of as DPS for an Order Requiring of Records the Application of the State of California Custodian 269. information:

ofKenneth Konzak). 190226) (testimony 267. 42 U. S. C. ? 14132(b)(3)(D) (2006).

Transcript

of Proceedings

at 31-32,

People

v. Davis,

(Jan.

18, 2006)

(No.

SCN

Product Documents/DNA Oct.

associations individual offices in a particular jurisdiction, professional organizations, or even non-profits erected National of Criminal Defense Association Lawyers, conduct defense office

v. Lopez Database Unit, Arizona (No. CR-20051252) (Super. Ct. Az. lab to conduct a specific search of state database, and report 26, 2005) (ordering Arizona results to defense attorney) (on file with the author). 270. The request could logically come from a range of places, including public defender

such as the

these types of reviews. Data generated by a researcher commissioned or professional be protected by attorney-client necessarily organization would in such a study or report might not be such that access to the information contained privilege,

to precisely by either a specific

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

784

CALIFORNIA LA WRE VIEW

[Vol. 95:721

and confidentiality concernsraisedby such privacy,proprietary information, research. The research could be made available on a broadbasis, to thebenefit burden of of defendants as a class.Rather than have a singlejudge shoulderthe a large,long-term research benefit of ordering what is essentially projectforthe has enabled one defendant, thejudge could instead askwhetherthegovernment and studythrough scientific development broader-scaleefforts mediated by a neutralscientific panel.More generally,such a panel could also periodically surveythefieldand identify desirableareas forcontinuedstudy.271 Ifproperly thepanel could even award grantsforresearch,the results of which funded, criminal could then justicecommunity. bemade available to theentire Anotherkey component of sucha panelwould be the periodicpublication toassist judges, lawyers, with of guidancedocumentsintended and technicians Where varying critical and emergingquestions related to a technology. legitimate theoriesexist, thesedocuments would highlightthe conflictand in favorand againsteach position. In thisrespect, a summarizethearguments thangloss over, conflictsin the neutralnationalpanelwould embrace,rather of variousmethods or scientific community with regard to the desirability monitoringinstitution would be a natural techniques. Similarly,a centralized of forensic forthecollectionand analysisof data about theefficacy repository methodologies. Woefully absent from the public debate about the use of are anydata on the in thecriminal various technologies justicesystem degree to and towhat effect, which forensic evidence is actuallyused in investigations, A centralizedoversight result in conviction. whether those investigations evidence is used properly;itcould agencycould not only ensure thatforensic data necessarytoensurethatit isused intelligently. also compile the The panel could also serve as a promoterof best practices by promulgating protocols and standards. It could oversee the auditing and sound techniques proficiency testing necessaryto ensurethat methodologically
are implemented properly on a lab-by-lab and case-by-case basis.
272

Even

disclosed 271. fields"); databases

broadly. whether and Solomon, Saks, Merlin 10, at 1132-33 supra note (questioning of scientific decisions of courts are instruments too blunt to guide the development in reliance upon & Ford, supra note 140, at 100-07 (identifying problems Thompson Cf.

"admissibility

and study). and suggesting areas of necessary development The Office of the Inspector General, which scandal, noted investigated the FBI DNA of for quality control in DNA the general lax enforcement mechanisms typing labs. U.S. Dep't of Protocol and Practice Vulnerabilities A Review Justice, The FBI DNA Laboratory: 272. standards required for participation audits often did not follow

in the national database, and that even those who id. at 21 (noting the recommendations presented); and that stricter enforcement measures in . . . procedures did not prevent the "weaknesses that led to the FBI lab scandal, even though the lab had received clean reports from protocols" assurance relied on external both internal and external requires

could "self-certify" theircompliancewith quality 17-21 (May 2004) (notingthat labs initially

auditors, and was accredited at the time). Effective forensic analysis is valid that the methodology of numerous kinds, including: quality assurances the for executing that the laboratory's protocols tailored to a particular purpose); (including

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

785

though most statesrequirelicenses foreverything from nail salons to fishing, no nationalor statewidelicensingstandard or board exists forforensic crime
laboratories.
273

laboratories, and forblind proficiencytesting and baseline qualification of ratherthan simply relyingupon voluntary or ad hoc national analysts,274 accreditation of thevery processesrun by professional organizations comprised
technicians under review.
275

A national panel could set out standards for regular auditing of

for such oversightagencies,with militates in favorof a tiered structure of statewide to thenationalpanel in theform structures complements oversight within each laboratory. aimed at ensuring And even if the quality control failed to act nationally,theexistenceof an assortment of federal government inboth standard-setting stateor regionalpanelsmight stir"competition" and research. As scandalshave erupted across thenation,severalstates have reacted by For instance, in responseto scandalsat itspremier convening just suchentities. laboratory, Virginia quickly enacted legislationcreating a Departmentof ForensicScience,headedby a director by thegovernor, and separate appointed from theDepartmentof Criminal Justiceservices.276 The legislationalso wake of scandalsat providedfora stateForensicScience Board (FSB) in the The FSB is responsible for adopting Virginia's premiere laboratory.277 of various forensic regulationsfor the administration disciplines, including DNA, drug,and breathalyzer testing, and forsetting forth goals and standards At thesame time, forthedepartment. Virginia conveneda Scientific Advisory

In this respect, comprehensive monitoring

likely

that the laboratory's training, oversight, and error prevention); is generally reliable (including blind testing, quality assurance in methods, and regular review of corrective action files); and that the execution of a methodology a particular case is reliable. recent study of forensic laboratories that conduct DNA testing revealed that only and 87% of those had been accredited the professional by ASCLD, association. of 2001, at 2-3, Bureau Steadman, Survey of DNA Crime Laboratories, Justice Statistics, Jan. 2002. A comprehensive list of current state quality assurance regulations 273. A 63% were accredited, Greg W. for DNA typing can be found DNA Laboratories, at Seth Axelrad, Am. Soc'y State L., Regulations Med. & on Quality Ethics, Assurance available for at

are valid (including methodology actual execution of that protocol

Forensic

of external blind proficiency tests of forensic several obstacles, the 125, at 32. Despite encountering researchers blind proficiency laboratories is testing in forensic DNA even though they did not recommend it. Id. at 39. possible," 275. the American association, Presently, the professional Society of Crime Laboratory Directors or ASCLD, accredits laboratories through its Laboratory Accreditation Board. Butler, laboratories. Peterson, supra note that "external concluded supra note 26, at 395. Their accreditation process does not require any regular blind proficiency error logs to centralized testing, nor does it appear even to require that laboratories maintain record (and presumably analyze and correct) errors made in and across cases.

http://www.aslme.org/dna_04/reports/axelrad2.pdf. 274. A recent study explored the feasibility

276.

for providing forensic laboratory services to government agencies responsible 1101. The defense may petition the court for laboratory services. Id. ? 9.1-1104. et seq. (2005). 277. See Va. Code Ann. ? 9.1-1109

Va Code Ann. ? 9.1-1100 (2005). The DFS is not entirely in thatit is independent,
only. Id. ? 9.1

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

786

CALIFORNIA LA WRE VIEW

[Vol. 95:721

who serveforfour members Committee(SAC), composedof thirteen years.278 The SAC reviewslaboratory operations of the Department of ForensicScience, on protocols,testing, andmakes recommendations qualifications, and quality
control.279

Virginia's efforts are a step in theright directiontowardcomprehensive A centralized state body can execute regular auditing procedures, oversight.280 as well as commission spontaneous open-file, big-picture reviews of laboratories' materials,including comparisons across cases. Inmany instances, will be theonlymeans of uncoveringred flags thatpinpoint such reviews 28 Such an entitycan also certain labs, or analysts, for closer scrutiny. concernstopreventinappropriate skillfully navigateconfidentiality disclosure. At thesame time,theentity couldmake publiclyavailable theresults of quality 282 measures. control and assurance come intobeing?283 As a practical matter, howmight suchan entity The Justas Congress would be through easiest response legislativeenactment.284

278. 279. 280.

Id. ?9.1-1111 Id. ?9.1-1113. The

et seq.

of the Justice for All Act of 2004 rendered states eligible for grants if passage that they used "generally accepted established laboratory practices and procedures, or appropriate certifying bodies." 42 U.S.C. ? 3797k et seq. And by accrediting organizations indeed, several states have undertaken gestures of reform, both before and after the 2004 federal they certified legislation. committee); See, MASS. Comp. Stat. 3981/5 e.g., 20 III. (2005) (creating laboratory advisory 6 ? 184A (creating a board composed GEN. L. ANN. of government to collect data and report on operation of forensic services in state); N.Y. EXEC. ? (creating ? 38.01 on forensic science); Tex. Crim. Commission composed et seq. (creating Rev. Code ? 43.103.010 (same); Wash. (2005) state forensic investigations council). in Other efforts were unsuccessful Sess. reforms. See, e.g., S.F. 3273, 84th Leg., 2nd Reg. (Minn. 2005) diversely

representatives 995-a & -b (2006) Proc. Ann. Code

government-constituted enacting the proposed (proposing (Mo. 2006) committee); a forensic

S.B. 768, 93rd Gen. Ass., 2nd Reg. Sess. laboratory oversight commission); (setting forth system of crime lab oversight, including a diversely composed oversight H.B. 1380, 2d year of 159th Sess. (N.H. 2005) (establishing forensic science oversight Teichroeb, Mar. are Flawed, one analyst, Problems at that revealed

Crime Lab Tests They Sit in Prison?But internal audit of 13, 2004, at Al (describing which revealed flaws in 30 of 100 cases); Jonathan Saltzman, US Audit Found More Crime Lab, Boston Feb. 1, 2007, at Al Globe, (reporting audit of state crime lab systemic problems visible only across cases). 282. As one commentator abnormalities anomalously examination. 283. in a

commission). 281. See, e.g., Ruth Seattle Post-Intelligencer,

the potential sources of authority for discuss as as treatment for such well the of the results by both district and panel's convening panels, Breast Implant Litigation, supra note 162, at 825-830. appellate courts. Walker & Monahan, under the the constitution of a DNA Advisory Board 284. (DAB), Congress mandated DNA forensic Act of 1994, which standards for convened for five years and addressed to At five the DAB's duties transferred the close ofthat year period, effectively testing. an FBI-led the Scientific Working committee (SWGDAM), Group on DNA Analysis Methods that could not fairly be characterized as independent. Butler, supra note 26, at 394-95. Identification

"high Koppl, supra note 118, at 270-71. and Monahan Professors Walker also

has observed, a "[sjtastistical review" is essential to determine a case. Labs with cases, not just within laboratory by looking across or low numberfs]" on relevant criteria can then be targeted for closer

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

787

create createdCODIS, and so followed the states,so too could legislatures Alternatively,theNational Research Council, the CODIS's chaperone.285 armof the NationalAcademy of Sciences, couldpick upwhere it left operating pressurethe convene an expertpanel,whichmight in turn off and regularly resources. branchesfor access to therequired and legislative executive forcethecreation of suchpanels by refusing Courts could also effectively without proof that the to find admissible any forensicevidence proffered has been subject datanecessaryto testits methodologicalsoundness underlying After all, Daubert itself specifically scrutiny. to nongovernment-related of the scientific is a community that"submissionto the scrutiny instructed that the likelihood in because it increases of science,' part 'good component It further will be detected."286 acknowledged flawsin methodology substantive consideration that the"scientific projectisadvancedby broad andwide-ranging Precisely to enable such wide-ranging of a multitude of hypotheses."287 must require thegovernment open access todatabases-perhaps consideration, but certainlyto a qualified and limited not to every individual litigant, the multilocus matches The litigation of researchers. surrounding community just as the decision in theCastro case provides just such an opportunity: NAS panel, so toocould an adversedecision in several trial catalyzedthefirst courtcases-either orderingthedisclosureof thedatabases or precludingthe executive of theevidencedrawnfromtheir matches-prompt either admission Such issue forcing might also prompt by courts or legislativeresponse.288 information tocooperate more about proprietary privatecompaniesconcerned although the analysts.For instance, with defense-sideor independent fully DNA private companies thatdeveloped the primer sequences for forensic deemed information, to release what proprietary refused they initially typing some companies realized a competitive advantage in publishing such would elect their over those technologies information, because lawenforcement notopen forinspection.289

285. laboratories. Laboratory

authorization for access to the federal database However, (Hammond). requires federal, not just of Sciences state, action. Indeed, just before this article went to press, the National Academy one such general committee as a result of congressional action. See Identifying the convened at The of Science National available Needs the Forensic Academies, Community, http://www8.nationalacademies.org/cp/projectview.aspx?key=48741 509 U.S. at 593. 286. Daubert, 287. 288. Id. at 597. Of (last visited May 14, 2007).

In fact, various states have considered legislation that tightens scrutiny over forensic See, e.g., 2005 Vt. S.B. 249 (introduced by Senator Illuzzi) (creating a Forensic 1380 111.H.B. 5241 (Durkin); Mo. H.B. 1330; N.H. H.B. Oversight Commission);

course, to be effective, a decision likely would have to be rendered in numerous cases. The government can always simply choose not to use the forensic evidence, or to forego the case entirely, if itwishes to avoid enforcement of such an order. 289. Butler, supra note 26, at 101 (describing how, after courts in California, Colorado of primer sequences, excluded DNA evidence absent disclosure the Promega and Vermont . . . and have done so to "made the decision their STR kit publish Corporation primer sequences

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

788

LA CALIFORNIA WRE VIEW

[Vol. 95:721

2. Granting Access to Basic Information At the local level, each forensic agency should be held responsiblefor providing easy access tocertain universaland critical documents.In the words of one scholar, discussingthe increasing presenceof science inpublic debate, toensurethat "[i]t is important good scientific information notonly is available in theabstract, but also ismade available to theright people, at theright times, in theproduction, and inways that promoteaccountability transmission, and use of knowledge."290 With theadventof theInternet, little justification exists on thegrounds thatroutine for withholding generaldocuments production of such information would unduly burdenthelaboratory. Forensicagencies should storekey operating materials such as protocols,analysts' resumes,and the results testsin an electronicformat of validationstudiesand proficiency that could be downloaded at any time. Agencies could likewise make the reports issued by independent entitiesavailable in an electronic formatthateven indigent defendants could access. Agencies could evenmake proficiency test and audit results and corrective action logs available online. At present,such is nearly impossiblefordefense investigators to obtain.291 information To the extentthatsuch information might includesensitive material, thegovernment can be providedto counsel inan could easilyprotectit through passwords that case by courtorder. a laboratory individual Alternatively, could simplyhold hours"during which such items "visiting are available forinspection. B. Loosening the Law Courtroom's Grip on the and availability of forensic Changes, like those above, to theoversight will unquestionably theoverallqualityof evidencepresented evidence improve incriminal thelaboratory cases. But simplytightening structure isnot oversight enterthe will likely enough.In thecomingyears,second-generation techniques court in a highnumberandwide rangeof cases. And as scientific techniques will changeas well. scientific about legitimate evolve,understandings practice much less address,eitherof Yet thecurrent fails to embrace, legal structure will stillsuffer thestructural from theserealities. The defensefunction features of scientific and the thatdiscourage robust testing evidence, prosecutorial function will stillbenefit fromitsdominance of the market in forensic services. But severalsmall shiftsin legalobligations, aimed at eveningtheplayingfield between thegovernment and thedefense, could shoreup theadversary process

since"). 290. 281,

291. revealing

Jasanoff, supra note 186, at 133-34. Crime Lab Tests are Flawed, Ruth Teichroeb, They Sit in Prison?But at Al series of audits that uncovered flaws in numerous cases, (recounting that defense counsel and the defendants

supra while

note also

materials

But see 111. Sup. Ct. R. 417(b)(iii-vi) (2004) (expressly providingfordefensediscoveryof such
inDNA cases).

in those cases had not been notified of findings).

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

789

as a safeguardnot only against faultyforensicevidence, but also against accurateforensic evidenceused ina faulty fashion. 1.GreaterCentralization ofDefense Functions thedesirability of consolidating Questions have long lingered concerning indigent defenseservices withina particular jurisdiction.292 Advocates of such consolidation,typicallyin the form of public defender officesor agencies, argue that effectiverepresentation requires the pooling of resources and centralized experience available only in formal, Otherproponents agencies.293 point to the institutional power realized by the centralization of defense when suchofficescontroltheir services, own internal allocationof particularly resources.294 The riseof second-generation forensic evidence lendsfurther credenceto officialcentralization. arguments favoring However, even stoppingshortof formal of the of such services, reorganization provision some informal changes inhow defenseattorneys approach theirresponsibilities can be implemented more immediately. forensic evidence requires Specifically,second-generation that defenseattorneys coordinate their not justacrosscases, but efforts broadly, acrosscounty and state borders. Fortunately, thevery technologiesthatdefine the second generation of forensic evidencealso enable a second-generation aids in response:technology stateand nationalcohesion among an otherwise diffuse practiceof criminal defense. For instance, inresponseto thecomplexity and pervasiveness ofDNA twopracticing typing, joinedwith a renowned academic expert in attorneys

292. services

Robert

variousways inwhich indigent defense States, 58 L. & Contemp. Probs. 31 (1995) (outlining
recommendation for greater deliberate collaboration holds true across tend to confront fewer of the structural defendants bar, although well-resourced impediments tomounting effective challenges of evidence. 293. Individual Actor v. Institutional Player: Alternating See, e.g., Kim Taylor-Thomas, Visions of the Public Defender, 84 Geo. L.J. 2419, 2432 (1996) (relating successful collaboration the defense D.C. and Cook The Death funding). supra note 293, at 2449-57 (describing how centralized are performed). My

L. Spangenberg

& Marea

L. Beeman,

Indigent Defense

Systems

in the United

Counsel

Yale

between Washington, for thePoor:

L.J. 1835, 1849-57 (1994) (detailingproblems raised by lack of centralizeddefense


and adequate See, e.g., Taylor-Thomas,

Illinois public defender offices); Stephen Bright, County, Sentence Not for theWorst Crime But for the Worst Lawyer, 103

programs 294.

wield politicalpower particularly officescan better with regardto funding issues).Opposition to


centralizing the defense function tends to center around the perception that to do so would be both defense lawyers would costly and contrary to the ultimate goal of justice, because litigate?and even win?more cases. Note, Gideon's Promise Unfulfilled: The Need for Litigated Reform of

mount

statewide defense also argue that consolidation the services). One might heightens possibility of conflicts of interest among defense lawyers, both in ethical terms as well as in their own apportionment of time and effort to particular cases. See, e.g., Darryl K. Brown, Defense to Ration Services and Shortchange Some Clients, 42 Brandeis L.J. 207 Attorney Discretion

Indigent Defense, 113Harv. L. Rev. 2062,2067 (2000) (recounting politicaloppositiontobills to

(Winter 2003/2004).

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

790

CALIFORNIALA WRE VIEW

[Vol. 95:721

DNA in sounding a call todefenseattorneys tocollaborate nationally on issues of DNA evidence.295 They identified the "most fundamental need" as an "ongoing, comprehensive, national repository of defense-oriented forensic science information."296 Accordingly, in conjunctionwith the National Association of CriminalDefense Lawyers and theNational Legal Aid and DefenderService, theattorneys formed an onlineForensicLibrarytoprovidea forum for defenseattorneys to share materialsrelatedto forensic evidence. Such collaboration servesas a model forthekindof poolingof resources necessary to ensure thatthedefensebar respondseffectively to complicated, challenging, and changingforms of second-generation forensic evidence.But defensecounselneed greaterresourcesto foster and encouragecollaboration. Legislaturescould aid in suchefforts by specifically funding activities aimed at nationalcollaboration, and nonprofit organizations could designateindividuals responsible for coordinatingsuch efforts on a broad scale. Finally, some adjustments in the understandings of confidentiality and work-product privilegesare also essential,so thatcommunications and information-sharing among defense attorneys might be appropriately protectedthe sameway in which communications between lawyersindifferent officesof the Department of Justice areprotected. 2.DefenseEntitlements to Access evidence depends so largely Secondly,because second-generation upon data outside thescope of the individual case, access to thisdata is essentialto safeguard theintegrity of theevidence.If the government may use a database to make conclusionsabout thedefendant, shouldhave access thenthedefendant to that within reason,toconfront database, thoseconclusions. Thus,within the bounds of an individualcase or investigation, should be defense attorneys topetitionthecourtfor permitted equal access to therelevant databases.Courts should not accept the excuse that third partieshold thesedatabases: if the governmentobtained information, then the defense should have equal todo thesameor else theevidenceshouldbe deemed inadmissible. opportunity Similarly,and especially in the absence of neutral entities like theBoard described above, courts should reject claims that disclosure of database materials is undulycumbersome or invasive, and should insteadsimplycabin thescopeof disclosureappropriately. Again, the experienceof DNA evidence proves illustrative. Presently, Illinois is one of thefew statesin thenation toprovidea statutory framework

295. Defending Aug. 2005, 296.

See, With at 39. Id.

e.g., Richard (And Against)

S. Schmechel, William C. A Call Forensic Evidence:

& Edward J. Ungvarsky, Thompson to Share Resources, The Champion,

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEW FORENSICS

791

for defense access to genetic databases in individualcases.297Thus the defendant, upon a particularshowing,can exercise a parallel capacity to investigate an offensethrough a searchof thedatabase.Such a search may aim an evidentiary a geneticprofileother to show, forinstance, that itemsuggests or that thanthedefendant's, a mixtureof profiles points to a different possible todue processand theassistance perpetrator. rights of counsel The defendant's clearly contemplatesuch searches,and theyare essential to the equitable 298 administration of justice. For example,one lawyerin Missouri reported that a man with no priorviolentconvictions who was "matched" she represented a "cold hit" in a database to a case in through which thegovernment intended the databasewith regard to an intimate sample taken fromthevictim, the over theobjections of theprosecutor, to run attorney persuadedthelaboratory, thesecondprofilethrough thedatabase.300 When thesecondprofileturned up a match to a convictedsex offender, thegovernment dismissed thecase against In the absence of explicit statutory for granting her client.301 frameworks defenseaccess, however,attorneys shouldnot have todependon thekindness of govemment technicians. 3.GreaterGovernment Duties Given thedegree towhich thegovernment dominanceof the market for in ternsof knowledge forensic science inevitably advantagesthegovernment thegovernment and resourcesabout such techniques, should assume several with the dutiesconsistent prosecutorial dutyto see that "justicebe done" rather
to seek the death penalty.299 Noticing that the government had failed to check

Illinois statute permits the defendant in any case where DNA may be relevant to the court for an order requiring the state police to investigation or at trial to move or searches within the database. conduct certain genetic tests, or tomake certain comparisons 725 III. Comp. Stat. ? 24-4-63 ? 5/116-5 (2005); see also Ga. Code Ann. (2005) (providing similar 297. The the defense rights to Georgia defendants investigation, preparation, American Bar Association for DNA evidence. ABA to the to the DNA data bank ismaterial upon a showing that "access or presentation of a defense at trial or in a motion for a new trial"). The in its articulation of general standards recently endorsed this approach on DNA Standards Evidence Criminal Justice 8.3, (Standard

of Delegates, approved by ABA House Aug. 2006). Some states provide oblique entitlements in that they specifically curtail database access generally, but then provide an exception for defense counsel pursuant to court order. See, e.g., Haw. Rev. Stat. ? 844D-82 (2006); N.C. Gen. Stat. ? scope of such access, (2006). The permissible statute specifically states that only the material is far from clear, especially where the however, "related to the case" may be disclosed. See, e.g., instruct a witness to "sequester" in St. Louis,

15A-266.8 (2005); Cal. Penal Code ? 299.5(g)-(h) (West2005); N.J. Stat. Ann. ? 53:1-20.21

Iowa Code ?81.8 (2005).


298. defense,

By comparison, just as the government cannot so too should the government not be allowed Public Defender

to refuse to talk to the biological "witnesses,"

particularly potentially exculpatory ones. 299. Email from Cynthia Dryden, Assistant 300. 301. Professor Id. Id.

March 1,2006 (onfilewith author).

of Law,

University

of California,

Berkeley?School

to Erin Murphy, MO, of Law (Boalt Hall),

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

792

CALIFORNIA LA WRE VIEW

[Vol. 95:721

thanto simply win everycase.302 an affirmative First,the law should imposeupon thegovernment duty to disclose any departures from protocol thatgovernment analystsundertakein reachingtheresults at issue in thecase.303Such an affirmative duty,like the to disclose exculpatory dutyof thegovernment evidence to the defense,304 places the obligation to observe and reportany deviations from standard practiceupon the party bestpositionedtobringthese deviationsto theattention of the court. In many cases, the government will be able to justifysuch on thebasis of sound scientific departures practice or evolving standards; nevertheless,the government should identify and reportthem,ratherthan leaving to the defense-who is often leastwell-positioned to notice-the responsibility of uncoveringthem. Once disclosed, the defensehas several of thetechnique in light options: it may elect tochallengethelegitimacy of the modification; argue to the jury that themodificationwas ill-advisedor or forego use of theinformation invalidated theresults; altogether. This obligation of disclosureis also inkeeping with thelaw-like deference evidentiary rules accord to the findings of validity of a particular thecourtinitially Presumably, admitted results of a particular methodology.305 was executed in a techniqueon the premise that the approved technique with general standards.If, however, some particularcase in conformance ormodification was required,thenthegovernment has a duty to "tweaking" disclose thedeviation-much as an advocateof one positionhas an obligation to disclose bindingcontrary or as thegovernment has an obligation authority, to disclose information in itspossession thatcontradicts the statements of its thanthedefense's carefulreviewor witnesses. In short, thegovernment, rather good luck, should call to attention any deviations from the protocols that method in thefirst acceptanceof the place. garnered to scientific Second, encourage progress,courts should place upon the with the obligationsof good affirmative government obligationsconsistent science. in finality, and consistency tends Although thelaw's interest certainty, to value precedent of over innovation, theseprinciplesill-servetheenterprise Rather than science,which thrivesinsteadon noveltyand experimentation. entrench forexperimentation, the methodologiesand penalize thegovernment to engage in researchand law should create incentivesfor the government

Berger v. United States, 295 U.S. 78, 88 (1935). in a DNA For example, case, the protocol may require that the analyst disregard as spurious any peaks lower than a certain cut-off level, or in particular position to another peak, or at a particular height-ratio to another peak. Yet, in a particular case, the government may attempt 302. 303. peak's disclose

some other information justifies, in the government's to incorporate those peaks because eyes, the to In the present is under no obligation inclusion. legal framework, the government to the defense to override standard practices; in the proposed its decision regime, the decision to light.

government would be required to bring that discretionary 304. 373 U.S. 83 (1963). Brady v. Maryland, 305. See supra Part II.B. 1.

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

793

new evidence insupport of itstechniques. andbringforward development, shouldcarry a burden of placingbeforethe Accordingly,thegovernment court continuedevidence of a technique's legitimacy. Rather than render once answered,is rarely admission of a methodology a one-time questionthat, to provide asked again, the law should affirmatively requirethegovernment evidenceverifying the technique'scontinued This is not to say that viability. thegovernment shouldbe expected to reinvent the wheel by conducting full scale admissibility thanstartfroman hearings in every case. Instead,rather would regularly ask assumptionthat"no news is good news," thisapproach me lately?" "whathave you done for While thedisclosureof new validation admissionof the studies mightnot be essentialto thecontinued methodology, to supplya court thefailure with evidenceof continued withinthe development fieldwould, aftera substantial amountof time,cause courts to view such evidence with increasingskepticism.Likewise, the absence of evidence themethodology's continuedvalidity could alone constitute demonstrating evidenceof itsobsolescence, and justify whereas courts exclusion. In short, venerate an ancientlegalprinciplefor of time, should havingstood thetest they greeta similarly antiquated scientific technique skeptically absentevidenceof ongoingviability. forthetheory thansimplyselecting and advocating that suits Third,rather it best, the government evidence and should bear a burden of presenting Law imposes derivedfrom all legitimate, theories. disclosingresults competing and thenimparts one result over upon sciencean unrealistic degreeof certainty, another withoutdue regardfor legitimate conflict.306 rules settle Evidentiary for the"generalacceptance"of one methodwhen, in fact,authentic conflict and more than one method may have attained a threshold of exists,307 In science, it is not unusual thattwoopposingpositions reliability. may find equal support in legitimate argumentand proof. In such cases, law-like deferenceto one positionat theexpenseof theotherthwarts and distortsthe actual state of thescience. Such conflicting, but equally legitimate, methodologicalapproaches merit with equal play beforea jury. However, thedefense is ill-poised, particularly to identify or seekout respectto second-generation techniques, areasof conflict and retainexperts in supportof its position. Placing the burden on the

306. time. Walker

over time just as law changes over time"); id. at 822 n.119 (quoting Heidi Li inMass and Uncertainty 74 Tex. L. Rev. Feldman, Science 1, 16 (1995) Exposure Litigation, ("As scientists acquire new data and change their collective judgments about which background to hold constant, they revise and replace even well-established scientific theories. assumptions Scientific 307. theory does not achieve absolute finality.")) Frye v. United States, 293 F. 1013 (1923); (footnote omitted). Daubert v. Merrell Dow Pharms., 509 U.S.

contingent science changes

Law, like science, must remain receptive to new information, allowing it to adapt over & Monahan, Breast Implant Litigation, supra note 162, at 822 (arguing for the treatment of certain scientific results, while "law-like" noting that "[i]nevitably,

579, 593-94 (1993).

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

794

CALIFORNIA LA WRE VIEW

[Vol. 95:721

government properlyacknowledges both the government'sdominationof of bestowingexpertassistance upon forensicscience, and the impossibility ineverycase. Failure toproduce suchevidence,likefailureto everydefendant in itspossession,could constitute disclose exculpatory information groundsfor precluding theevidencealtogether.308 By means of illustration, takea question regarding theproper means of calculating the randommatch probability in a "trawl" case, where the government matched thedefendant after making a "cold hit" in a database.At present, thereexists reasonable debate regardingthe preferred method for in such cases. As the DNA AdvisoryBoard calculatingthe match probability has explained: There are alternate methods forassessing theprobative value ofDNA evidence.Rarely is there onlyone statistical and approachto interpret of theuser, the explain theevidence.The philosophyand experience legal system,thepracticality of theapproach, thequestion[s]posed, availabledata,and/or assumptions all affect thechoiceof approach.309 For example,some scholarsargue fora likelihood ratiothattakes intoaccount of a "trawl"aremore thesize of thedatabase.310 Some suggestthattheresults reliable thanin a simpleconfirmation case, because theanalysthas compared thegeneticprofile to a database and excluded a largenumberof persons.311 a simple"counting method" ismost appropriate,312 Some think that and some as theanalystlooks in of a "falsepositive" increases contendthatthelikelihood a database for a match, and thus the statisticalprobability should be accordingly discounted by thisrisk.313

308.

particular approach, some commentators disputes the equal

can argue in favor of a the government theories exist, however, it finds most persuasive. Of course, leaving the fact-finder to decide which to resolve methodological find it appalling that juries should be allowed if presented with that even expert scientists cannot resolve. See supra note 237. However, Where competing

to then jurors are in many respects in the best position legitimacy of both positions, of the case. For instance, jurors might be approach best fits the circumstances adopt whichever more conservative in a "cold hit" case than in a case with corroborating evidence, and for good cause. Thus, drawing on Professors Monahan's like treatment of scientific methodologies might for instance, whereas law and Walker's model, entitle a judge to instruct a jury that one particular is sound, a revised instruction might instead inform jurors that several separate and scientific approaches or techniques are sound. Butler, supra note 26, at 614. II, supra note 201, at 40. See, e.g., NRC See, e.g., Butler, supra note 26, at 618 (citing Evett, Foreman & Weir, Biometrics

technique competing 309. 310. 311. the Legal

97 Mich. L. Rev. 931, 933 (1999). Consumption of Scientific Evidence, NRC I, supra note 225 ;Butler, supra note 26, at 515. Franco the C. Thompson, Taroni & Colin G. Aitken, How 313. See, e.g., William a J. Forensic Sci. 52-53 Positive the Value 48 False DNA 47, Evidence, of Affects Probability of one scholar (Jan. 2003). With regard to the use of likelihood ratios to express match probabilities, 312. has that such ratios should incorporate error rate data. See Jonathan similarly argued persuasively Ratios Should Account for Error J. Koehler, Likelihood (Even When a National Why DNA

DNA Database Searches and 56(4), at 247-76 (2000)); PeterDonnelly& RichardD. Friedman,

J.425, 429 (1997) (arguing that Research Council ReportSays TheyShouldNot), 37 JURIMETRICS

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

795

may pick lawyers government proceduralenvironment, In the current and effectively demonstrate its reliability, they prefer, methods which of these raisedby an opposingparty.In a voices unless specifically ignore anycontrary orwell-resourced mightbe knowledgeable select fewcases, opposingcounsel to the but in thevast favorable defense, to the method offer enough itself toargueor perhapsevenknow to will notundertake of cases, counsel majority to pick itspreferred view. Yet allowing thegovernment argue any contrary and leaving to thedefense competitors, among legitimate methodologyfrom with the least theories, saddles theparty theobligationto uncoveralternative an equally valid resourcesand least access with theburdenof introducing with regardtootherforms operate While sucha burden might rightly approach. science,the world of forensic in thelopsided history suggeststhat, of evidence, a legal well. Rather,bestowingon thegovernment bear it defensecan rarely basis that have legitimate calculations any to all statistical obligation present todefensecounsel-diminishes the but also directly not just to thefact-finder nature process of criminal or theadministrative inequities riskthatinstitutional evidence. ofmisleadingscientific will resultin the presentation 4.Error andAdmissibility Finally, in every case, courts should considerwhether the laboratory as as a legaland then first levelof competence operatesat a sufficient generally of either At present, courtstypically view theerrorrates a factual question.314 as factualquestions of themethodology itselfor the executing laboratory 315 by a jury, ratherthan as legal questions of "weight" to be determined

into probability determinations). At present, this question is often treated either as irrelevant when the defense cannot make a showing of malfeasance expected to affect the results in the case at bar, or else as a matter 374 F. Supp. the evidence. See, e.g., United States v. Morrow, for the jury to decide in weighing 314. 2d 51, 67-68 United States (D. D.C. v. Ewell,

error rates should be calculated

. . . not relevant for "weight exclusion"); 2005) (finding such evidence 252 F. Supp. 2d 104, 113-14 (D. N.J. 2003) ("What the defendant has the proficiency of the tester rather than the reliability of the test. sought to do here is challenge 109 Such challenges go to the weight of the evidence, not to its admissibility."); People v. Reeves, Cal. Rptr. 2d 728, 750-53 (Ct. App. 2001). The second National Research Council report refused to even recommend consideration statistical calculations

as a qualifier on the of laboratory error at trial, whether or as independent evidence regarding reliability. NRC II, supra note 201, at some courts have found that there exists some level at which the general 185. Alternatively, cease a to so be below the standards that fall far of "reliable," they acceptable laboratory practices and thus should be excluded. (reversing testing procedures meant used in the case); of the testing procedures" acceptance should (N.Y. Sup. Ct. 1989) (holding that admissibility properly executed). 315. See Dale A. Nance & death conviction

v. Florida, 838 So.2d 1073, 1081 (Fla. 2002) See, e.g., Murray errors and contamination in evidence because based in part on DNA in demonstrating the general its burden "the State did not meet 985, 986 People v. Castro, 973 N.Y.S.2d turn in part on whether the technique was

Assessment of Presentation Empirical 34 J. Legal Random-Match Probability,

An Scott B. Morris, Juror Understanding of DNA Evidence: With a Relatively Small Formats for Trace Evidence Stud. 395, 398 (2005) (citing cases).

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

796

LA CALIFORNIA WRE VIEW

[Vol. 95:721

As a result,laboratories admissibility. have littleincentive to documenttheir rateof error and courtshave still less incentive to requiretheir provisionand
disclosure.316 But

processed at laboratories thatfallbelow a reasonablestandard of operational A restaurant efficiency. thatservedup roaches in its spaghetti on fiveearlier occasions is not,after all, a place youwould go to eat,nomatterhowmuch it assuresyou thatitprepared your meal in a sanitary manner. In thesamevein, when a person's freedom, and not just good digestion,is on the line,a similar standard shouldapply. A facility with a demonstrated history of improper storage or handling of evidence,or an inexcusablerateof failure on proficiency tests, simplycannot generateresults reliableenough todiscounttherisk of error, regardless of how meticulously its personnelhave performedthe tests in an individualcase. Moreover, staking admissibility of evidenceon a laboratory's generalreliability not only creates incentives on the part of the laboratory to complywith published standards of operation, but it also gives theprosecutiona vested interestin competently managing the laboratory, therebyencouragingan oversightrole in place of unquestionedallegiance.Finally, if reliability were as a threshold treated questionof admissibility, onwhich theproponent of the evidence carried the burden, then the government would have to submit evidenceof the laboratory's errorrate.This requirement would thereby foster the implementation of testing and oversight procedures necessary to quantify sucha rate. And, if thequestionof a laboratory's threshold competencelevel is a quintessentiallylegal one-does this laboratory generally operate at a thresholdlevel of reliability?-thencourtswould have to set appropriate standards of operationallegitimacy. Thiswould have thesalutary effect of both such standards, articulating an incentive and giving laboratories tohew to them. Moreover, while earlier decisions findinga particular lab competent may warranta degree of deference, counsel should always retainthe capacity to prove an approvedlabunreliable. a tainted Similarly, laboratory might redeem itself by demonstrating that it has instigated changes to remedya systemic problem. Regardless, once a laboratory meets theappropriate threshold standard, thecourtsshouldnonetheless admitevidenceof erroras factualevidence for
See United Statesv.Morrow, 374 F. Supp. 2d 42, 46 (D.D.C. 2005) (notingthat "the
indicates that possible contamination issues go towards the legal precedent than the admissibility?of DNA U.S. v. Lowe, 954 F. Supp. 401 (D. evidence"). (noting that the "FBI does not compute a systemic laboratory rate" yet nonetheless of

as a threshold matter, judges

should not admit evidence

316.
great weight weight?rather Mass. 1996)

evidence on finding evidence admissible, observing that "no federal cases ... have excluded DNA account of a theoretical error rate alone"). Simon Cole compiled existing data on the rate of false for fingerprinting, and determined that, assuming the resultant .8% false positive rate, positives U.S. laboratories "reported in 2002 alone." Simon Cole, More Than Zero: 1,905 false positives Accounting for Error in Latent Fingerprint Identification, 95 J. Crim. L. & Crimin. 985, 1034

(2005).

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

2007]

THE NEWFORENSICS

797

This principlecarriesparticular the jury toweight as it deems appropriate. hit" weightwith regardto "cold cases, given thattheexistenceof extrinsic mistakeor error on thepartof the evidence is less likelytocompensatefor any testingauthorities. Of course, the government,in turn,could introduce or errors in thelaboratory are routinely mitigating evidence-that anyproblems and efficient cited dealtwith in a professional manner,or thatthe laboratory's errors did notaffect the particular case.
CONCLUSION

Although thisArticle uses DNA typing to illustratethe problems and recommend technologies possiblemeans presented by second-generation italso intends tobegin a more generalconversation about ofmitigatingthem, will accommodateevidence fromthenext how the criminaljustice system a cell of forensic science-whethertheevidence inquestion is that generation datamining uncovers phonewas used at a particulartimeand place or that thecrime.The distinct characteristics evidence thatthedefendant perpetrated of second-generationforensic sciences-including theirmethodological breadth of application, scientificcertainty, complexityand sophistication, and relianceon databases for interests, implications privacyand proprietary must consideras and scholars elicit a host of concerns thatcourts,lawyers, such evidence continuesto infiltrate criminalcases.Moreover, the fact that evidenceof an individual's thesetechnologies have the power toprovidestrong makes thetaskofmonitoring guilt,even in theabsenceof any otherevidence, The recommendations in theaccuracyof suchevidenceall the more important. this to strike an efficient balance among thevariouscompeting Article attempt and legal rules to betteraccommodatethe concerns.In adjustingevidentiary of forensic Article aims to fashiona justice second generation evidence, this of evidencethat enterintoit. of theinnovative forms system worthy

This content downloaded from 111.68.111.198 on Mon, 20 May 2013 12:35:44 PM All use subject to JSTOR Terms and Conditions

Das könnte Ihnen auch gefallen