Transcript An Update on Daubert Hearings on Fingerprints

An Update on
Daubert Hearings
on Fingerprints
IAI
87th International
Educational Conference
Las Vegas, NV
August 4-10, 2002
Stephen Meagher
Unit Chief
Latent Print Unit III
FBI Laboratory
The DAUBERT Decision
• 1) Whether the theory or technique can be
and has been tested.
• 2) Whether the theory or technique has been
subjected to peer review and publication,
– although the fact of publication, or lack thereof,
in a peer-reviewed journal is not a dispositive
consideration.
The DAUBERT Decision
• 3) The known or potential rate of error of a
particular scientific technique.
• 4) The degree of which the technique or
theory has been generally accepted in the
scientific community.
• 5) The existence and maintenance of
standards controlling its operation.
FEDERAL RULES OF EVIDENCE
RULE 702. Testimony by Experts
Post-Daubert and Kumho
• If scientific, technical or other specialized knowledge
will assist the trier of fact to understand the evidence or
to determine a fact in issue, a witness qualified as an
expert by knowledge, skill, experience, training or
education, may testify thereto in the form of an opinion
or otherwise, provided that (1) the testimony is based
upon sufficient facts or data, (2) the testimony is the
product of reliable principles and methods, and (3) the
witness has applied the principles and methods reliably
to the facts of the case.
Daubert Hearings on Fingerprints
-a brief look back• 1999; U.S. v Mitchell
– Daubert Hearing in June, 1999
• 5 days, 11 witnesses (7 government, 4 defense)
–
–
–
–
Trial in February, 2000
Hearing on Brady and new evidence
Sentencing July, 2002
Notice of appeal
Daubert Hearings on Fingerprints
-a brief look back• 2000 U.S. v Alteme
– 2 days, 4 witnesses (3 government, 1 defense)
• 2000 U.S. v Havvard
– 1 day, 1 witness
• 2000 - 2002
– 37 additional challenges in many forms
– probably more than reported
U.S. v. Llera-Plaza
• Philadelphia, PA
– Private attorneys as lead
• Filed standard motion to exclude fingerprint
evidence in 2001.
– Federal Public Defenders Office included for hearing
• stimulus for all challenges
• Both sides agreed to argue the issue in writing by
filing summaries of U.S. v. Mitchell.
– No expert witnesses would testify.
U.S. v. Llera-Plaza
• Judge Pollak ruled (January, 2002) to limit
fingerprint testimony.
– Evidence could be entered, but…
– Fingerprint experts could only testify to similarities
and/or dissimilarities, but…
– could not testify that the prints are a “match.”
• Is this a Daubert victory?
– Yes, but, fingerprint community considers this to be
significantly less than a Daubert Hearing victory.
U.S. v. Llera-Plaza
• Government requests Judge to reconsider.
– Case went to DOJ, Washington, D.C., for
consultation.
• Judge Pollak agrees to reconsider.
• New hearing in March, 2002
U.S. v. Llera-Plaza
• Judge Pollak reverses earlier decision,
– with interesting comments:
• judge must review latent prints prior to trial, and
• [perceived] limitation to only FBI fingerprint
experts.
• Trial in progress.
U.S. v. Llera-Plaza
• Use of U.S. v Mitchell did not address some
of the issues raised by Judge Pollak
• Second hearing had limited testimony
– methodology
– practitioner error rate
• Ken Smith
• Defense witnesses
– Bayle, Haber, Arvizu
When can it occur?
• Pre-Trial
– Written defense motion
setting forth issues.
– May or may not have
testimony by experts.
– If testimony, provide
detailed information
with supporting
documentation.
• During Trial
– Complicates your
testimony to know
when to give brief and
concise answers versus
detailed information.
– Consideration for Jury
yet need to convey
adequate and
compelling detail.
Tactical Change by Defense
• Unless a case is one of first impression, the
defense will not file for a Daubert Hearing,
but wait until trial to aggressively challenge
the latent print expert.
– First, on your knowledge of the scientific basis
during your qualifying phase (This is focused
towards the Judge primarily, the jury is
secondary at this point.)
– Next, after you testified to the identification.
(This is focused on side-tracking the jury.)
Pre-Trial Meeting
with the Prosecutor
• Discuss the “key words” with the
prosecutor prior to trial so both of you
recognize when the Daubert issues are
being raised.
– “Tested technique” or “methodology;” “peer
review;” “error rate,” “general acceptance” and
“standards.”
What are the defense tactics?
• The claim of absolute certainty
– no possible or probable conclusions permitted
• Lack of standards
– for the ACE-V methodology
• individualization, exclusion, inconclusive
– maintenance (SWGFAST)
– no standard for claiming a latent print “of value”
Latent Prints “of value”
–
A + C = E
– A(50%) + C(50%) = E(100%)
– A[25% + 25%] + C(50%) = E(100%)
– A(25% + 0%) + C(0%) = E(25%)
therefore, establishing a predetermined number of points to claim a
latent print of value during the analysis phase is non-relevant in a
scientific sense.
What are the defense tactics?
• NIJ Solicitation
– June 20, 2000 NIJ letter
• Proficiency testing
– 1995 CTS test results
• Lack of testing
– statistical probability models
– prior studies and 50K study
• how different are fingerprints
• how much information do you need to “match”
fingerprints
Statistics
• Historically, there have been numerous
attempts to determine the statistical
probability of two individuals (two fingers)
having the same friction ridge arrangement.
• Most of the studies were limited to level 2
detail, and most did not even use all of the
level two details.
Statistical Probability Studies
“On the Individuality of Fingerprints” by Pankanti, et al
STUDY
Galton (1892)
Henry (1900)
Balthazard (1911)
Bose (1917)
Wentworth/Wilder (1918)
Pearson (1930)
Roxburgh (1933)
Cummins/Midlo (1943)
Gupta (1968)
Trauring (1963)
Osterburg, et al (1980)
Stoney (1985)
# of Minutiae=36
1.45 x 10-11
1.322 x 10-23
2.12 x 10-22
2.12 x 10-22
6.87 x 10-62
1.09 x 10-41
3.75 x 10-47
2.22 x 10–63
1.00 x 10–38
2.47 x 10-26
1.33 x 10-27
1.2 x 10-80
# of minutiae=12
9.54 x 10-7
3.72 x 10-9
5.96 x 10-8
5.96 x 10-8
4.10 x 10–21
8.65 x 10-17
3.35 x 10-18
1.32 x 10-22
1.00 x 10-14
2.91 x 10-9
3.05 x 10-15
3.5 x 10-26
Statistical Probability Models
“On the Individuality of Fingerprints” by Pankanti, et al
• Assumptions and limitations
• Conclusions
Statistics
• Review of Statistical Modeling to Date
– No previous model adequately captured
complete comparative process
– AFIS is a statistical model, but very limited for
latent prints
– Need to take existing AFIS data and perform
statistical analysis
Statistics
• Lockheed Martin Corp., under the
direction of the FBI, conducted a 50K vs.
50K fingerprint comparison test.
– 50,000 left sloped loops, white males only
– used IAFIS matcher algorithms to run each
fingerprint against the 50K file.
– Level 2 detail only
Statistics
• Test 1
– fully recorded fingerprint (known)
– 2.5 billion comparisons
• Test 2
– 21.7% of fully recorded fingerprint
(simulated latent)
– 2.5 billion comparisons
Statistics
• Conclusions - Test 1
– The probability of a non-mate rolled
fingerprint being identical to any particular
fingerprint is less than 1/10 (1 followed
by 97 zeros)
– world’s population = 5.9 billion
• x 10 fingers = 59 billion fingers
– The probability of any two fingerprints being
the same is 59/10, or 1 chance in 10.
Statistics
• Conclusions - Test 2
– The probability of a minutiae subset of a non-mate
fingerprint being identical to a minutiae subset of any
particular fingerprint is less than 1/10 for a small
number of minutiae (4 level 2 detail), decreasing to less
than 1/10 for larger numbers of minutiae (18 level 2
detail).
– world’s population = 5.9 billion
• x 10 fingers = 59 billion fingers
– The probability of any two fingerprints being the
same is 59/10, or 1 chance in 10.
Statistics
• BOTTOM LINE:
– Not even taking into consideration all of the
level 2 detail and none of level 3 detail the
statistical probability of two individuals
having the same ridge arrangement remains
beyond reasonableness of the entire worlds
population times 10 fingers.
What are the defense tactics?
• Methodology v. Practitioner error rate
– Lack of testing to determine practitioner error
rate
• One Dissimilarity Doctrine
– just make up an explanation to dismiss a
dissimilarity
What are the defense tactics?
• Amount of area needed
– Fully rolled print vs. partial print
• 50K study
• Control of the discipline
– law enforcement v. scientific community
• Dichotomy between “Point Counters” and
“Ridgeologists”
What are the defense tactics?
• Literature
– published, peer reviewed
– what do you rely on, citations
• 1999 FBI survey
– General Acceptance
– Uniqueness
What are the defense tactics?
• On the horizon
– Is the known exemplar a true and accurate
image of the finger?
• Inked prints
• live-scan prints
– Lights Out
What’s next?
• Daubert Hearings
– Las Cruces, NM; Monday, August 12, 2002
– Ft. Worth, TX; Thursday, August 15, 2002
– Pending; Baltimore, MD