JUDGES, FBI, DNA, & LIES
ARE COURTROOMS AND LABORATORIES
SLOPPY ACROSS AMERICA
Volume 1 of 2
BY
JOHN W. FIDLER
Copyright 2011 John W. Fidler
Smashwords Edition
Published by JWF Publishing, Inc., BOSTON, MA
The information in this book is true and complete to the best of my knowledge. It is offered without guarantee on the part of the author, publisher or printer. The author, publisher, and printer disclaim all liability in connection with this book.
All rights reserved. No part of this book may be reproduced or transmitted in any form whatsoever without prior written permission from either the author or publisher except in the case of brief quotations embodied in critical articles and reviews.
2009 TXU001-656-065 Library of Congress Registration
ISBN # 978-0-615-41901-5
ACKNOWLEDGEMENTS
There are a few people who I wish to thank individually for the trust they had in me, for allowing me to act as either an investigator on a case or as a legal assistant. In addition I thank them all for encouraging me to write this book.
First and foremost, is Attorney Albert F. Cullen Jr. who trusted me to work with him on my son’s case and not to allow any conflict in the case because he was my son. Attorney Cullen was meticulous in every aspect of his representation. He taught me to read every little detail concerning documents and material. He taught me to be patient and also meticulous when interpreting documents. Thanks Al.
I wish to thank Attorney Angela G. Lehman who trusted me to work with her concerning the subpoena served on my son. John was served with a subpoena compelling him to submit to combing and plucking his hairs at the FBI Headquarters in Boston. In addition, Attorney Lehman, as a young attorney, protected John’s sisters, who had to appear before a Grand Jury. Thanks Ang.
I wish to thank my friend Attorney Timothy P. O’Connell who also trusted me to investigate many of his cases and act as a legal assistant in others. Attorney O’Connell had the same qualities as Attorney Cullen and Lehman. Tim is meticulous, honest and well educated not only as a street lawyer but a lawyer you could count on in the courtroom. Thanks Tim.
I wish to thank my two friends for their editing skills, Angela Gunn and Doctor Mary O’Neill. If it wasn’t for them this book would have never got off the ground. Together they have made what was a bunch of words into a fine book for students to learn from. Thanks Angela, Mary and Charlotte.
Last and not forgotten my good friend Arthur Page who was the first person to read JUDGES, FBI, DNA, & LIES and complimented and criticized some of the writing, which we corrected together. Thanks Snooze.
I also wish to thank Steve, Paul, and Candace for all their help in the design of the cover and printing.
PREFACE
Many of us first became aware of DNA through the widespread television coverage of the O.J. Simpson trial and the seemingly ever ending CSI programs on television today. But, what is seen on television and what transpires in Forensic DNA laboratories across America, when analysts compare two biological stains of DNA, is a far cry from the imaginary circumstances presented by television viewing.
In 2005 Congress asked The National Academy of Sciences to assess the state of the forensic techniques used in court proceedings. On February 4, 2009 the New York Times carried a story by Solomon Moore concerning forensic laboratories across America. In the article Professor Paul Giannelli stated with regards to DNA, “Enforcement officials…‘choose to say they never make mistakes but they have little scientific support, and this report could blow them out of the water.’” Professor Giannelli is a Weatherhead Professor of Law at Case Western Reserve University who teaches on evidence and scientific evidence. He also authored nine books which are cited in court case.
The report was published as a prepublication, entitled Strengthening Forensic Science In The United States: A Path Forward, which was made available at the end of February. The final version has just been published at the end of 2009.
The report did not have much to say about DNA mixtures. In chapter three of the report it stated, "However, this does not mean that DNA is always unassailable in the courtroom. There may be problems in a particular case with how the DNA was collected, examined in the laboratory, or interpreted, such as when there are mixed samples, limited amounts of DNA, or bias due to the statistical interpretation of data from partial profiles." Strengthening Forensic Science In The United States: A Path Forward, at 3-12.
The report casts doubt on the reliability of other scientific evidence, [e.g. tool mark identification, fingerprints identification, and etc.] introduced into courtrooms and recommended further scrutiny by the scientific community and the legal system, just as was DNA evidence is, before putting it before a jury. In addition, it appears that the committee sought to have one universal legal standard for each scientific field of evidence. For instance, the standard in accepting scientific evidence, such as fingerprints, in federal courts should also be the same standard as set in Massachusetts and other States.
Interestingly on the same day that the Times carried the above story the Boston Globe carried a story about the Massachusetts State Police Crime Laboratory, written by Jonathan Saltzman. Some of his comments were astonishing to me. He reported, “Mary Kate McGilvray, a veteran of the lab who was appointed acting director after Carl Selavka resigned under pressure in March 2007, has reduced the backlog of DNA evidence awaiting analysis. But she is often relying on inexperienced analysts, and that alarms employees who fear it could lead to mistakes.” Mr. Saltzman stated the Inspector General questioned, “…whether the lab should continue reporting to the State Police because scientists might feel pressured to tailor findings to suit criminal investigators.”
Several days later the Massachusetts State Police Crime Laboratory was accredited as being a reliable laboratory by the Laboratory Accreditation Board of the American Society of Crime Laboratories Directors (ASCLD-LAB).
DNA testing and the matching of biological stains to one another is a complicated issue, which involves several steps to get a match or partial match, for this book those issues and steps are hopefully simplified in order that the reader can have a clearer picture of how analysts come to their conclusions of what makes a DNA match and partial match. Judges, FBI, DNA & Lies do not bog down the reader with a bunch of theories and math formulas. However, I strongly recommend to attorneys who are defending clients, judges making decisions about DNA evidence, criminal defendants, and persons in paternity matters, should consult the scientific literature, the actual case law used in this article, and cases pertaining to their particular issues as well as recent court decisions.
This book is based on actual case law decisions, several scientific articles and recommendations quoted from scientific groups. Some of the comments and decisions are from trial judges while others are from appellate judges. Hopefully, after reading this material, individuals will question all aspects of a chemist/analyst/ population geneticist's methods, calculations and conclusions of connecting two biological DNA stains.
As you read this material you will see that laboratories have different protocols and methods when interpreting DNA mixtures. Doctor Peter Gill a member of The International Society for Forensic Genetics (ISFG) an organization for forensic investigations in Europe and other parts of the world stated, "If you show 10 colleagues a mixture, you will probably end up with 10 different answers." See Human Identification E-Symposium, April 14, 2005.
The Restricted Fragment Length Polymorphism (RFLP) method of testing DNA was accepted by courts in the 1990s. However the calculations of a statistical probability of a match between two biological stains were rejected. The rejection was largely based on the non-reliability of the databases used in authenticating the statistical calculations of the DNA match, which was determined by allele frequencies from certain population groups. Today I believe this still to be true.
That rejection did not last long. Between early to mid 1990s The Department of Justice, the Federal Bureau of Investigation, the FBI’s Laboratory’s Scientists and their colleagues, discovered ways to appease the gatekeepers [trial judges] into allowing testimony that the statistical probabilities of a DNA match were scientifically reliable. This happened even though during this same time period, scientists as well as population geneticists did not agree about the reliability of the databases. In fact, in 2005 these same persons are still debating how best to interpret a mixed DNA profile with that of a person’s DNA profile. See MIX05.
When Congress passed the DNA Identification Act in 1994, it gave the Federal Bureau of Investigation through the Department of Justice, full authority to implement methods of DNA typing and permission to start databanks to store DNA profiles of individuals and profiles of collected samples of evidence.
The RFLP typing method yielded to Short Tandem Repeat (STR) typing in the late 1990s. Laboratories now mostly use the STR typing method when comparing biological stains, which is explained throughout this book and more detailed in Chapter VII. STR-PCR is highly susceptible to contamination.
Part way through this Article there will be a review of the first several Chapters.
CONTENTS
Chapter I. The Early Stages of DNA and its Assumptions
Chapter II. Early DNA Cases
Chapter III. Setting Standards for DNA Laboratories and Testing
Chapter IV.Federal Courts Accepting DNA Testing and U.S. v. Lowe
Chapter V. Massachusetts Cases Setting the Standard
Chapter VI Review
Chapter VII. Short Tandem Repeat (STR)
Chapter VIII. My Son and the FBI
(Two Attempted Armored Car Robberies in One Day)
Chapter IX. Potential Major and Minor Contributors of a DNA Profile
Chapter X. The Wrap
(Two more Armored Car Robberies and a DNA Mixture)
Chapter XI Epilogue
Appendix to Chapter VIII: Motion to Dismiss with supporting FBI Documents
CHAPTER I
The Early Stages of DNA and its Assumptions
Early in the twentieth century Godfrey Hardy and Wilhelm Weinberg stated evolution is simply a change in frequencies of alleles in the gene pool of a population. They concluded that gene pool frequencies are inherently stable, but that evolution should be expected in all populations and at all times. In other words even though evolution continues in the population, the gene pool frequencies will not change as long as the below conditions are accountable for when determining allele frequencies:
1. natural selection and mutation is not occurring;
2. the population is infinitely large;
3. all members of the population breed and mating is totally random;
4. everyone produces the same amount of offspring;
5. there is no migration in or out of the population.
Hardy and Weinberg devised an equation of simple binomial (for two alleles) or multinomial (multiple alleles/product rule) distribution of the genotype frequencies as p2+2pq+q2a1a=1. Whereas, p allele is the dominant frequency and q allele is the recessive frequency. This principle is known as the Hardy-Weinberg Equation, "HWE".
However, with random mating the population approaches a state in which the frequency of a multilocus genotype is the product of the genotype frequencies at separate loci. When the population has arrived at such a state it is said to be in linkage equilibrium, LE, which is different from HWE. One of the differences in these proportions is that HWE are attained in a single generation of random mating, whereas LE is attained gradually. The long and short of the HWE and LE are that they both assume certain conditions occur in a population. For a more comprehensive reading of the HWE and LE, see The Evaluation of Forensic DNA Evidence, Chapter 4, National Research Council 1996, referred to as NRC II. Also see, State v. Cauthron, 120 Wn.2d at 902.
There were other concerns of how the population deviates from the above equilibriums as an indication of the intensity of the above external factors. These deviations are solvable by other mathematical calculations. One is the theta, with regard to correcting relatedness within the population, another is the chi-square, and then there are the confidence intervals with regards to an unknown portion of the population.
Although the above deviations are important for calculating the correct allele frequencies, for this paper the theta, chi square and confidence intervals explanation and mathematical calculations need not be expressed. However, one should positively make sure the above factors used to correct and calculate allele frequencies in a population database are used correctly by laboratory personnel and the computer program used for those factors.
Because of the inconsistencies within forensic laboratories when testing, typing, and matching biological stains The National Research Council (NRC) created a committee in 1989, The Committee on DNA Technology in Forensic Science. The Committee was commissioned to establish certain procedures for laboratories, scientists and analysts to follow with regards to testing, typing and matching biological [DNA] stains. In addition, the Committee also made recommendations of how one should calculate the statistical probabilities of a match with biological stains.
The Committee also had to demonstrate to the legal system the justification when a scientist/analyst/geneticist concluded, that to reasonable scientific certainty the DNA from (x) and from (y) came from the same individual using statistical probabilities. I.e. "Reasonable scientific certainty means that you are (x%) certain that you would not see this profile in a sample of (y) unrelated individuals." John M. Butler, (2005), Forensic DNA Typing, c. 21 at 514.
The Committee in 1992 published its findings and made certain recommendation for laboratories to follow. One of the committee’s recommendations was for laboratories to use two principles when it wanted to know the frequency of an allele within a certain population group, i.e. Caucasian, Black, Hispanic, etc. The two principles the Committee recommended were: 1) the Ceiling Principle; 2) the Interim Ceiling Principle. Laboratories were to establish databases from sub-population groups and from different races within the population. The publishing is referred to as NRC I or NRC 1992.
The Ceiling Principle:
“The ceiling principle (NRC 1992, p 82-85) places a lower limit on the size of the profile frequency by giving thresholds for the allele-frequency values used in the calculation. To determine the thresholds, the report recommended that 100 persons be sampled from each of 15-20 genetically homogenous populations spanning the racial and ethnic diversity of groups represented in the Untied States. For each allele the highest value among the groups sampled, or 5% which ever was larger, would be used. Then the product rule would be applied to those values to determine the profile frequency…The necessary ground work for applying the ceiling principle has not been done, and there have been few attempts to apply it. We share the view of many experts who have criticized it on practical and statistical grounds and who see no scientific justification for its use.” See also NRC II, at 156-157.
However, NRC I recommended that until the ceiling principle's data could be collected and put into effect, the interim ceiling principle should be used.
The Interim Ceiling Principle:
“In contrast to the ceiling principle, the interim ceiling principle has been widely used and sometimes misused. The rule (NRC 1992, p14-15, 91-93) is: 'In applying the multiplication rule, the 95% upper confidence limit of the frequency of each allele should be calculated for separate US 'racial groups and the highest of these values of 10% (which ever is larger) should be used. Data on at least three major 'races' (e.g., whites, black, Hispanics, east Asians, and American Indians) should be analyzed.' The report also stated that the multiplication (that is, product) rule should be applied only when there is no significant departure from HW and LE even though the ceiling principle was introduced specifically to accommodate deviations from HW and LE.'" See also NRC II, at 157.
Using the Restricted Fragment Length Polymorphism "RFLP" method with the above principles an allele frequency was established with the “Floating Bin" or "Fixed Bin” methods. Most laboratories used the "Fix Bin" method to establish an allele frequency. Those systems need not be explained for this article. However, if one wishes to see how binning works, I suggest reading, The Use of DNA Typing in Forensic Science, Laurence D. Mueller, Department of Ecology and Evolutionary Biology, University of California, Irvine, California. In addition See NRC I and II.
Today using the Short Tandem Repeat "STR" method an allele frequency is determined by how many times an allele is observed in a population database. The database of 216 persons multiplied by 2 (because persons in the database are assumed to have inherited one allele from each parent) is 432.
As shown below the allele repeat 10 and 11 were observed 109 and 134 times, in the used database out of the 432 alleles. Dividing the observed repeat by the size of the database repeat (109/432) determines the allele frequency for that particular repeat. Shown below are the allele frequencies for allele repeats 10 and 11 at the locus CSF1PO:
Once the allele frequency is determined for each allele, then the Multiplication/Product Rule is used to determine the locus frequency by multiplying each allele's frequencies. The locus frequency for the above CSF1PO locus is 0.2523 x 0.3101=0.16.
To determine the Genotype Profile frequency for two loci, simply multiply each of the loci’s frequency to obtain the Profile frequency, i.e. 0.16 x 0.28 = .045. See Below.
The analyst then will compute the above profile (.045) to see how many persons would randomly (by chance) match that Genotype Profile in a certain race. I.e. 1000/0.045, computes to 1 in 22,222 Caucasians would have the same profile at the above two loci. As more loci are used in the profile the larger the 1 in ?? computes.
The analyst will also establish the minimum frequency for rare alleles. To determine the minimum an analyst will implore 5/2N method. Five is the inflation number and 2N is the number of alleles in the database. NRC II states this will be a conservative approach in determining a less frequent allele. I.e. if the allele 22.2 at locus FGA was only observed twice out of the 432 alleles in the database using the regular method of determining the frequency you would simply divide the 2/432 producing .005, and then using the 5/2N method 10/432 producing .023
However getting back to the Ceiling and Interim Ceiling Principles, these Principles were hotly debated among scientists and population geneticists who were called to testify in courts of law.
In the first two cases judges did not allow DNA evidence because of the disagreement within the scientific community, including a number of population geneticists, who had concerns about the validity of Cellmark Laboratories and the FBI Laboratory’s databases used to determine the statistical probabilities of a DNA match. The Massachusetts Supreme Court agreed with the trial courts in those cases. (See Comm. V. Curnin, 409 Mass. 218, January 1991). See Also Commonwealth v. Lanigan, 413 Mass. 154, 596 N.E.2d 311, July 1992.
At this point in time, allowing scientific evidence into a court room was subject to the Frye Rule, established in 1923, Frye v. United States, 293 F. 1013, (D.C. Cir. 1923). The party introducing the scientific evidence had to meet stricter standards then today. Under Frye there were two prongs to meet that judges had to ponder whether to allow scientific evidence in at trials:
a) The underlining theory is generally accepted as reliable in that relevant scientific community; and,
b) The method or technique used to apply the theory is generally accepted in the scientific community.
The latter prong is what was hotly debated throughout the scientific community and population geneticists; that is whether the ceiling and interim principles of obtaining an allele frequency was correct or incorrect.
Although some courts still use the Frye test because of its stringent standard , in 1993 under the United States Supreme Court’s decision of Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, the Frye Rule was relaxed. See also Lanigan II, 419 Mass. 15, at 24-26. See Chapter IV for details.
CHAPTER II
Early DNA Cases
1.Commonwealth v. Curnin, 409 Mass. 218, January (1991)
This is the first case in which the Supreme Judicial Court of Massachusetts (SJC) considered whether to allow in at trial the DNA evidence and its statistical probability, regarding a match with the defendant’s DNA and the DNA evidence collected at a crime scene.
The case involved the rape of a minor child. At the time the standard for allowing scientific evidence in at trial, at this time, was subject to the Frye Test, Frye v. United States 293 F. 1013 (D.C. Cir. 1923). Prior to trial Judge Dortch held an evidentiary hearing to determine whether to allow the DNA evidence and its statistics probabilities in at trial. Cellmark Laboratories did the DNA testing. Cellmark determined that the defendant’s DNA and the DNA found on the victim's nightgown produced the probability of one in 59,000,000 that random individuals in the Caucasian population would have the same DNA profile.
Cellmarks' expert testified about the reliability of DNA testing, but also stated she was not qualified to give an opinion in the field of population genetics concerning the allele frequency from the database used by Cellmark. In addition she testified that there was a question as to, whether the alleles used behaved in a standard way in the population as well as the fact that there was no scientific study of the database used. She stated the database had never been published for its acceptance in the scientific community, violating the first prong under the Frye Test. The database consisted of 200 Caucasians' blood samples, collected from a New York City blood bank.
Dr. Laurence Mueller, a population geneticist, testified for the defendant. Dr. Mueller testified that Cellmark’s database was not adequate for the purpose of producing estimates of allele frequencies in finding a particular genotype of the human population. Dr. Mueller’s main concern was that the database used for the allele frequency of the Caucasian population had sub-populations, or in his words substructures.
The lower court ultimately found that the method of gene process to DNA fingerprinting was generally accepted within the scientific community. However, the court did not comment on the statistical probabilities of a match between the defendant and the nightgown. See Curnin at 220, note 4. Curnin was convicted and appealed. The appeal was heard under the de nova [new] standard.
The Supreme Court (Wilkins, J) with Liacos C.J., Abrams, J. Nolan, J., and O’Connor, J.J. concluded that it did not have to resolve the propriety of the forensic DNA testing conducted because Cellmark had not demonstrated any general acceptance in the database used. Nor had Cellmark demonstrated an inherent rational that the process Cellmark used [product rule] when it arrived at one Caucasian in 59,000,000 would have the DNA components (alleles) disclosed by the test that showed an identity between the defendant’s DNA and the DNA found on the nightgown. Curnin at 221-222. In other words, using the Product Rule to calculate the statistical probabilities was not an approved scientific theory, and Cellmark had no expert to testify otherwise. The SJC remanded the case back to the trial court.
2.Commonwealth v. Lanigan and four companion case; Breadmore and thirty-three companion case, 413 Mass. 154, 596 N.E.2d 311, July 1992. These cases are referred as Lanigan I.
The issue in Curnin as well as these cases was with the databases and how they were misleading when used by Cellmark and the Federal Bureau of Investigation when calculating the statistical probabilities of a match. The defendants all filed motions to exclude the State from introducing any evidence regarding the DNA tests, because of the misleading statistical probabilities with regards to the databases used. Most of the defendants argued that the DNA tests were speculative and prejudicial. They also argued that the Commonwealth failed to establish; (1) "that the scientific community generally accepts the theory that DNA tests produce reliable results; (2) [that] techniques currently exist in DNA testing that are generally accepted by the scientific community and that are capable of producing reliable results; and (3) whether the laboratory that performed the test used these techniques in analyzing the samples in the instant case." Lanigan at 158.
Judge Elizabeth Brown consolidated the defendants' Motion in Limine. The facts in Lanigan were:
Allegedly Lanigan’s semen was found on a minor child’s clothing. The FBI did the testing on Lanigan and the evidence. The peculiar aspect of this case is that after one test was completed the FBI ran Lanigan’s DNA profile within two different databanks [database]. The first databank was made up of 200 Caucasians, the second 700 Caucasians. The 200 databank produced a one in 4.0 million match, while the 700 databank produced a one in 2.4 million match. Lanigan I at 312-313. By using 500 more persons in a database it lowered the chance of a random match by 1.6 million persons. So much for allele frequencies, in that, the frequency doesn’t vary with the amount of individuals in the databases!!
There were four experts for the Commonwealth, one of which was a population geneticist, Dr. Neil Risch. In opposition, the defendants were allowed to submit the transcript of Dr. Laurence Mueller, another population geneticist, who testified in the Curnin case about a similar issue with Cellmark’s databank.
It should be noted, that the Commission's Report “NRC I” was published two months prior to this decision in April 1992. NRC I recommended using the ceiling principle as a way to determine allele frequencies. However, Cellmark and the FBI did not use the ceiling principles instead they were still using some sort of multiplication rule [product rule] from databases they had collected. Dr. Mueller testified in Curnin that because these databases had sub-populations, and that, "There are disagreements about using the product rule in the circumstances." See Curnin at 225. At the same time other scientists debated the authenticity of the ceiling frequency (ceiling principle) for determining the allele frequencies. See National Research Council of the National Academy of Sciences, entitled “DNA Technology in Forensic Science”. See NRC II at 35.