we review an article written from the perspective of a defense attorney looking to attack and discredit the forensic scientist and his/her discipline.  Although it deals with more than just fingerprints, it lists some specific questions and addresses the concepts surrounding those issues.  The understanding of the defense position provided by this article makes its length seem more bearable.  If you are interested in what defense attorney's are saying about the below questions, this article will be an interesting read:

Exactly How and Why Was Peer Review Performed?

Did the examiner’s case file have all the preliminary crime scene reports and investigative reports?  If so, what examination-irrelevant data was contained in these reports? 

Did the examiner document any of her supporting data?  If documented, in what manner was the data documented?  Can the laboratory’s activities, observations, and results be reconstructed exclusively on the basis of the available records?

Is the examiner required to identify a certain number of corresponding points of similarity before an identification can be rendered?  If a minimum number of matching points is not required, how does the examiner go about rendering his conclusion? 

Has it been empirically established that the evidence can be individualized?

What was the condition of the evidence?

_________________________________________
Psychological Influences & the State Employed Forensic Examiner: How to Elicit Evidence Concerning Observer Effect Errors through Cross-Examination and Discovery
Illinois Association of Criminal Defense Lawyers Newsletter, Summer 2003
http://www.law-forensic.com/iacdl_newsletter_summer_2003.htm
by Craig M. Cooley
Investigator, Illinois Appellate Defenders Office, Death Penalty Trial Assistance Unit.  J.D. candidate (2004), Northwestern University School of Law; M.S. (Forensic Science), University of New Haven; B.S. (Psychology), University of Pittsburgh.  Mr. Cooley can be contacted via his website: www.law-forensic.com

If one were to question state employed forensic scientists about their objectivity and affiliation with law enforcement or the prosecution many would contend, as Dr. Irving Stone does, that,“[T]he role of the modern forensic scientist has moved from that of a prosecution or defense advocate to an objective reference source for the judicial system.”[2] While this statement symbolizes the definitive objective of any forensic investigation [i.e., objective fact finding], it fails to represent the profession’s true and current state of affairs. All too often, given the forensic community’s current organizational configuration, conscious and unconscious biases saturate forensic examiners’ alleged unprejudiced interpretations. These overt or covert influences profoundly impact their ultimate conclusions. This Article’s objective will be to discuss the covert forms of bias and error that are ubiquitous to the forensic science community—observer effect errors [or expectation/examiner bias]. Once discussed, attention will be directed at how criminal defense attorneys can elicit evidence of these seemingly imperceptible biases and errors through discovery and cross-examination.

A. Forensic Science & the Principles of Psychology:

Observer Effect & Expectation Bias

Forensic science pioneer Paul L. Kirk once commented that, “Physical evidence cannot be wrong; it cannot be perjured; it cannot be wholly absent. Only in its interpretation can there be error.”[3] As the defense bar is well aware, forensic interpretation, regrettably, represents a significant, if not the most important, issue in many criminal cases. The unfortunate aspect of any endeavor that necessitates interpretation is that such tasks require a “scientific observer.” As cognitive psychologists have repeatedly documented, however, “the scientific observer [is] an imperfectly calibrated instrument.”[4] The scientific observer’s imperfections stem from the fact that subtle forms of bias, whether conscious or unconscious, can easily contaminate the observer’s seemingly objective perspective. Identifying and curtailing such biases are significant when one considers the forensic community’s affiliation with law enforcement and the prosecution. This relationship has fashioned an atmosphere where scores of forensic professionals have become biased for the prosecution.[5] So strong are these biases that forensic examiners have deliberately fabricated evidence or testified falsely just so the prosecution can prove its case.[6]

Outside premeditated falsification, however, forensic practitioners remain susceptible to far more invasive but normally overlooked inaccuracies. These errors are by-products of context effects (a.k.a. observer effects) where an examiner’s perceptions and inferences are affected by domain-irrelevant data (e.g., an investigator’s desired outcome, the conclusions of other experts on the same case).[7] In certain respects, these errors are more bothersome than deliberate fraud and misconduct because they are often undetectable if certain procedural safeguards are not incorporated into forensic examinations. Given the indiscernible nature of observer effects, accomplished and well-intentioned forensic examiners can offer genuine conclusions that are imprecise and erroneous even when they are employing well validated forensic techniques. Again, these forms of errors may occur in large quantities, entirely devoid of examiner awareness. As Professor Michael J. Saks explains, “Indeed, such distortions will be more ubiquitous and more insidious precisely because they are not intended and their presence goes unnoticed.”[8]

As indicated above, observer effects or expectation bias are governed by the fundamental principle of cognitive psychology that asserts that the needs and expectations individuals possess shape their perceptions and interpretations of what they observe, or in the forensic science context, what they examine. To fall prey to such bias, examiners typically must (a) confront an ambiguous stimulus capable of producing varying interpretations and (b) be made aware of an expected or desired outcome.[9] In short, examiner bias is the “tendency to resolve ambiguous stimuli in a manner consistent with expectations.”[10]

1. Sources of Subjectivity and Uncertainty in Forensic Examinations

With respect to ambiguity, the individualizing forensic sciences (e.g., fingerprints, bite marks, hair, toolmarks, lip prints, etc.) are top-heavy with subjectivity and ambiguity.[11] In the forensic science context, subjective tests are those where identifications and interpretations rest entirely on the examiner’s experience or conviction.[12] As we will come to see, many, if not all, forensic identifications are premised on an examiner’s unyielding belief that his or her experience is all that is required to render an absolute identification.

Forensic examinations represent two layers or phases of subjectivity. The first, and most obvious, instance of subjectivity occurs when examiners are determining whether two pieces of physical evidence are ‘consistent with’ or ‘match’ one another. Forensic examiners accomplish this by identifying an unspecified number of corresponding points of similarity. The majority of forensic fields do not require their examiners to isolate a specific number of matching points before they can claim an absolute identification.[13] Likewise, examiners regularly utilize divergent criteria that are typically not published or even articulated.[14] Interwoven into this subjective decision making is the probabilistic determination that the match is not a coincidental match.[15] Given that many of the individualizing forensic sciences are armed with no serviceable probabilistic models and no base rate data (e.g., no research), forensic examiners rely entirely on intuition, instincts, impressions, and subjective probability estimations (termed “experience” or “judgment”) when determining the likelihood of a coincidental match.[16] Consequently, as mentioned beforehand, identifications are simply manifestations of an examiner’s experience rather than empirical research.[17]

For instance, consider how one forensic scientist defined the individualizing process. According to the forensic scientist, “An item is said to be individualized when it is matched to only one source and the forensic scientist is morally certain that another matching item could not occur by chance.”[18] This definition clearly indicates the subjective nature of the individualization process because morality is an entirely personalized concept. Morality differs from person to person and from examiner to examiner. Why can’t the examiner be scientifically or empirically certain that another matching item could not occur by chance? The answer, as alluded to above, is that the forensic science community has conducted an inadequate amount of research over the past century in order to provide such answers.

In many forensic sectors, a second layer of subjectivity enters into the equation. For example, once fingerprint examiners intuitively determine the improbability of a coincidental match, they must then instinctively decide whether all other fingerprint examiners would reach the same conclusion before they are permitted to claim an absolute identification. Put simply, “fingerprint examiners must draw subjective impressions about other people’s subjective impressions.”[19]

(removed section)

The term subjective, it should be noted, does not signify that a particular manner of examination is invalid. Rather, subjective evaluations can, in certain instances, provide reliable evidence. This reliability, nonetheless, must be objectively and empirically evaluated and established (e.g., through blind proficiency testing). Subjectivity merely implies that room for disagreement exists. It’s when like minds are able to disagree, however, that the likelihood of error increases. Moreover, as the opportunity or room for disagreement swells so does the chance for error.[23]

2. Expectation Bias & Forensic Science:

Forensic and Law Enforcement Practices that Generate Expectation

In the forensic science sector, examiners are confronted daily with many opportunities that cultivate certain expectations. The most common expectation nurtured in the forensic science environment is one concerning the suspect or defendant’s guilt (i.e., namely that the defendant is guilty). This section details certain forensic science and law enforcement practices that are capable of inducing expectation biases on the part of forensic practitioners.

(a) Single Sample Testing

For the most part, when criminal investigators turn over evidence to forensic examiners the evidence typically falls into two groups—(1) samples taken from the crime scene; and (2) samples provided by the suspect. Single sample testing undoubtedly affects whether an examiner’s report will associate the suspect to the crime scene or the victim. For instance, in one study researchers found that fewer than 10% of forensic reports failed to associate a suspect to the crime scene or the victim.[24] Stated conversely, 90% of forensic reports inculpated the suspect in some fashion. As Professor D. Michael Risinger and his colleagues explain,

“This high rate of inculpation comes from the fact that each piece of evidence connected with any suspect has a heightened likelihood of being inculpatory, since investigators do not select suspects or evidence at random, but only those they have some reason to think were connected to the crime. Thus, forensic scientists have a continuing expectation that the evidence before them is inculpatory…”[25]

(removed section)

(b) Communication Between Investigators and Examiners:

Domain Irrelevant Information

When forensic examiners are presented with a new case they typically receive the case information in one of two ways; they either meet directly with the lead investigator[s] or the key case information is forwarded to them via mail, facsimile or electronically. Both circumstances are easily capable of inducing certain expectations on part of the forensic examiner.

Before discussing how the content of investigative reports and discussions can bias an examination, the forensic scientist’s proper role must be delineated. The forensic scientist’s purpose is to provide answers to questions that pertain solely to his or her forensic discipline. Moreover, when answering these inquiries the forensic scientist is ethically and legally obligated to employ or make use of only those applications and domain-relevant bits of information that are sanctioned by their designated area of expertise. For instance, if the question presented is outside the forensic scientist’s area of specialized knowledge he or she theoretically is ethically and legally barred from offering any answer to the question, even if he or she is capable of providing an honest an accurate response. Likewise, when forensic scientists interweave domain-irrelevant information into their evaluation and ultimate conclusion[s] they are abusing their special stature bestowed upon them by the courts. Expert testimony regarding a litany of forensic science issues is permissible not because forensic practitioners are more superior at deducing inferences about the meaning of ordinary pertinent evidentiary information than detectives or jurors; rather, admissibility is warranted because the law has conceded that forensic experts are better equipped, than unaided juries and detectives, at producing reliable and justifiable answers to specialized questions.

(removed section)

This brings us to the question of where and how forensic examiners become cognizant of domain-irrelevant information. Forensic examiners, for the most part, simply do not receive the physical evidence when assigned a new case. Rather, the lead investigator[s] frequently supplements his or her forensic examination requests with detailed crime scene and investigative reports. Likewise, investigators customarily provide forensic examiners with extraneous domain-irrelevant information in their transmittal correspondences (e.g., fax cover letters). For instance,

(removed section)

For instance, suppose a fingerprint examiner is having difficulty identifying an adequate number of corresponding points between two prints. If the examiner is already aware of domain-irrelevant information (i.e., the defendant’s DNA was discovered on the victim’s undergarments), an expectation has been covertly or overtly planted into his or her conscious. The expectation being that the defendant must be the culprit because the crime scene DNA corresponds with the defendant’s DNA. Couple this expectation with fingerprinting’s subjective nature and the stage is set for a subconscious and inadvertent bias to negatively impact a competent and ethical fingerprint examiner’s identification.

(removed section)

B. Recommendations that Can Minimize Observer Effects in Forensic Science Practice

To decrease personal bias, conscious or unconscious, the forensic science community must utilize many of the same checks and balances being utilized in other scientific communities. For example, retesting within the same laboratory or another laboratory, having two or more examiners re-confirm the original examiner’s interpretation, and identifying samples so the examiner cannot foresee a particular test’s outcome. The ensuing discussion will focus on two of the more important procedural reforms that would curtail observer effects in forensic science—blind testing and evidence line ups.

1. Evidence Line-Ups

A procedural reform that can minimize or counteract subconscious biasing influences involves employing ‘evidence line-ups.’[40] In an evidence lineup, multiple samplings are presented to the forensic examiner. Some samples, though, are “foils.”[41] Examiners would be blind to which samples constitute the foils and which samples constitute the true questioned evidence. For instance,

“[A] firearms examiner might be presented with a crime scene bullet and five questioned bullets labeled merely ‘A’ through ‘E.’ Four of those bullets will have been prepared for examination by having been fired through the same make and model of firearm as the crime scene bullet and the suspect’s bullet had been. The task for the examiner would then be to choose which, if any, of the questioned bullets was fired through the same weapon as the crime scene bullet had been.”[42]

As the example illustrates, forensic examiners and eyewitnesses perform comparable evaluations. Not surprisingly, then, both have the potential to suffer from similar methodological shortcomings when they are viewing their respective forms of evidence—be it a fiber or an alleged assailant.[43] Consequently, evidence lineups would serve many of the same purposes as a properly structured eyewitness lineup. Currently, forensic examinations are equivalent to eyewitness identification show-ups.[44] In both situations the single-suspect or sample configuration implies that the assumption of guilt is correct. Evidence line-ups would resolve these methodological deficiencies. The Justice Department’s recent report concerning eyewitness identification procedures discusses many guidelines that are similarly relevant to physical evidence line-ups.[45] In the end, the Justice Department’s objective in publishing its eyewitness identification procedures—reducing the frequency of false positive errors without diminishing the occurrence of true positive identifications—is equally applicable and just as easily achievable for the forensic science community.

2. Blind Testing

As Barry Scheck points out, “forensic laboratories have historically resisted external blind proficiency testing and other efforts to assess laboratory error rates.”[46] This must change to minimize rates of error, especially covertly caused mistakes such as observer effect errors. All forensic testing, be it proficiency or case examinations, should be conducted blindly.[47] According to Professor Saks and his colleagues,

“The simplest, most powerful, and most useful procedure to protect against the distorting effects of unstated assumptions, collateral information, and improper expectations and motivations is blind testing. An examiner who has no domain-irrelevant information cannot be influenced by it. An examiner who does not know what conclusion is hoped for or expected of her cannot be affected by those considerations.”[48]

The key, as the passage indicates, is to create an impenetrable wall between forensic examiners and any examination-irrelevant information.[49] This can be accomplished by providing examiners with the information they require to carry out the tests, and only that information.[50] With respect to proficiency testing, although many contend that the “logistics of full-blind proficiency tests are formidable,”[51] its practical implementation is by no means impossible.[52] Likewise, developing a system that filters out all unnecessary domain-irrelevant data before examiners perform their evaluations is also achievable.[53]

C. Eliciting Evidence of Potential Observer Effects:

What to Ask and What to Ask For

Whether the abovementioned reforms are implemented is for the forensic science community to determine. Lawyers, for the most part, especially those in criminal defense, will only have a peripheral role in determining whether these reforms are incorporated if they remain uncritical of forensic evidence and crime labs. On the other hand, if defense attorneys become more judicious and cautious in how they view crime labs and forensic evidence they can have a more profound impact on forensic science reform than initially perceived. Why, some may ask? As the wrongful conviction cases continually demonstrate, faulty science and fraudulent misfits have typically been identified where critical thinking defense attorneys have shunned their blind faith belief in science and questioned the findings of so-called forensic scientists. With an increasing number of defense attorneys critically evaluating crime lab and examiner reports there has been a corresponding increase in the number of cases where faulty forensics played a substantial role in generating a wrongful conviction or accusation.[54] As the number of identified mishaps increase,[55] so to does the notion that forensic science reform is essential to ensure that victims of crime and criminal defendants are afforded justice and liberty. Accordingly, to facilitate change and reform the criminal defense bar must intensify its scrutiny of crime labs and forensic evidence.

As mentioned, procedural safeguards (e.g., blind testing & evidence line-ups) have not been implemented in the overwhelming majority of crime labs. Likewise, given the defense bar’s limited, yet increasing, ammunition to bring about these reforms, defense attorneys must rely on their advocacy skills to ensure that defendants are not unjustly accused or convicted because of observer effect errors. Advocacy, as many experienced defenses litigators know, is more about knowing what to look and ask for than anything. With respect to observer effect issues, the key is to elicit any evidence that may call into question the forensic examiner’s alleged objectivity. The fact that the examiner is employed by a prosecutorial or law enforcement agency is not the principal issue here. For the most part, thanks to the numerous forensic science T.V. shows, jurors are already cognizant of this affiliation.[56] Rather, attention should be directed at obtaining evidence that emphasizes the (1) subjective (or ambiguous) nature of the examination and (2) the examiner’s predisposed expectations. The objective is to implant into the jurors’ minds that the forensic examiner, though highly skilled and well-intentioned, may have been adversely affected by unnoticed context effects. Consequently, the following sections will briefly identify various questions and bits of information that defense attorneys should ask for or obtain when mounting an attack against a forensic identification examiner.

(1) Subjectivity & Ambiguity Questions and Inquiries

These questioned are intended to expose the subjective nature behind many forms of identifications. For the most part, examiners will answer that many of their conclusions are based on their [subjective] experience and specialized training.

1. What was the condition of the evidence?

As previously discussed, physical evidence identified and collected at the crime scene is rarely in pristine condition when forwarded to the examiner. Often times, physical evidence can be exposed to contaminating environments or, worse yet, altered or damaged by post-collection circumstances. These alterations can be a by-product of incompetence (e.g., improperly trained evidence technicians) or environmental or situational factors (e.g., when removing a bullet from a wall the bullet may be innocently damaged). During cross examination question the examiner whether he is capable of differentiating between original crime scene markings and post-crime scene artifacts & alterations. Unless the examiner is privy to evidence collection photographs that were taken immediately after the evidence was collected and processed, it will be extremely difficult for him to distinguish between original markings and post-crime artifacts. Likewise, if the evidence, for instance, is a smudged partial fingerprint, you may want to consider enlarging the print so the jury can witness firsthand its non-pristine nature.

2. Can the evidence deteriorate or change over time?

This question is primarily directed at handwriting, bite mark, toolmark, firearms, and shoe print evidence. As mentioned, two fundamental assumptions underlie all forensic identifications—individuality and permanency. Here, we are attacking the permanency assumption because it is possible to alter the individualistic features created by these forms of evidence. For instance, “[H]andwriting may change substantially in both systematic and random ways over both the short and long term, depending on the writer’s health, the speed of writing, the positioning of the writing surface, maturation, and so on.”[57] With respect to bite marks, an individual’s bite pattern can easily be altered if he or she loses a tooth, chips a tooth, has braces removed, has braces installed, or receives dentures. Likewise, the surfaces of tools and the inter-workings of firearms will undoubtedly deteriorate and change over time through repeated usage. Lastly, everyone at one time or another has worn a pair of shoe so frequently that they eventually, and literally, deteriorate right before your eyes.

The point being emphasized here is that while investigators or examiners maybe in possession of a crime scene shoe print [toolmark, bullet, etc.], they typically are not able to immediately compare this print with an actual shoe until a suspect has been identified. Depending on the type of crime, apprehending or identifying a suspect maybe a lengthy process. It’s this interval of time between when a print is collected and when it is ultimately evaluated against an actual shoe that is of crucial importance here. When this gap represents a protracted period of time, examiners must intuitively factor into their evaluation the “wear and tear” factor. As the word intuitively suggests, the “wear and tear” consideration makes the subjective identification process even more subjective.

3. Has it been empirically established that the evidence can be individualized?

Forensic scientists have dealt with induction and deduction rather haphazardly. Induction is a form of inference that advances from a set of specific observations to a generalization, called a premise. This premise is an operational hypothesis, but its validity may not always surface. Conversely, a deduction is a type of conclusion that progresses from a generalization to a specific case. For the most part, deduction is the preferred type of inference within the forensic science community. Providing that the generalizations or premises that buttress the inference are valid, the examiner’s conclusion will be valid. As John Thornton and Joseph Peterson explain however,

“knowing whether the premise is valid is the name of the game here; it is not difficult to be fooled into thinking that one’s premises are valid when they are not.”[58]

Over the past century, forensic examiners have been ignorant of the fact that hypothesis testing’s equivalent is not deduction but instead induction. Once scientists have inductively developed a working assumption or premise, they verify the validity of their premise through testing. Unfortunately, forensic examiners often incorrectly confuse a hypothesis with a deduction. For a deduction to be considered scientifically legitimate its supporting premises must be validated through testing. If, however, the examiner’s supporting premises have never undergone such testing then all that we have is an assertion awaiting verification through testing.[59]

Consider, for example, forensic science’s fundamental principle of individuality. Many forensic examiner’s claim that they can individualize a particular form of evidence (e.g., fingerprints, lip prints, shoe prints, bite marks, etc.). More specifically, when testifying in court they assert that they were able to deduce that the defendant deposited the crime scene fingerprint. For this deduction to be correct, its supporting premises must be validated through testing.

“ Individuality’s forensic context is supported by three premises. First, numerous forms of biological, physical, and psychological entities exist in unique, one-of-a-kind form. Second, these entities leave equally distinctive traces of themselves in every environment they encounter. Third, the methods of observation, measurement, and inference employed by forensic science are adequate to link these traces back to the one and only object and/or individual that produced them.”[60]

Unfortunately, numerous forensic science sectors, in particular the identification fields, have conducted little, if any, systematic research geared toward validating the basic premises that constitute their continued existence.[61] Moreover, when research has been performed the findings seriously weaken one, two or all three of the aforementioned premises.[62] Consequently, the examiner’s deduction is not in fact a deduction; rather, it is an assertion, supported by ‘specialized training’ and/or ‘experience’, which is pending confirmation through testing. Because examiners’ experience will differ considerably from one another, experience is a subjective criterion supporting an identification.

4. Is the examiner required to identify a certain number of corresponding points of similarity before an identification can be rendered?

This question, more than any other, will expose individuality’s subjective nature. As mentioned, many forensic professions do not require their examiner’s to isolate a predetermined number of corresponding points before an identification can be made.[63] If an examiner states that his agency or certifying organization requires a specified number he should then be asked whether this number was derived at through testing or whether it was simply agreed upon by the examiners in that agency or organization through informal discussions. For the most part, if a specified number is required it is so because of the later rather than the former.

If a minimum number of matching points is not required, how does the examiner go about rendering his conclusion? For the most part, as alluded to earlier, identifications typically rest exclusively on intuition, instincts, and probabilistic estimations. (i.e., the examiner’s experience).

5. Is the identification premised on a statistics?

Two issues surface with respect to the statistical determination of individuality. First, is the examiner appropriately trained in statistics? Second, what statistical database was relied on?

As Professor Peterson and Thornton note, “Behind every opinion rendered by a forensic scientist there is statistical basis.”[64] The statistical basis “provides… an evaluation of the likelihood that his testimony reflects the truth, rather than his personal belief or bias.”[65] Consequently, besides understanding science and the scientific method, forensic scientists must be competent consumers of statistics.[66]

Unfortunately, forensic scientists are rarely forced to take one statistics course, let alone a series of classes, during their undergraduate or graduate education. As a result, practicing forensic scientists are poor consumers of statistics.[67] Many recent examples clearly illustrate this problem.[68] Forensic practitioners seem to bungle even the most basic statistical rule—the “multiplication” rule.[69] For instance, even though “[i]t should be common knowledge to criminalists that properties must be statistically independent before the probability of a conjunction of these properties can be derived from the multiplication rule,”[70] forensic scientists routinely fail to consider the “independence” issue.[71]

Nonetheless, forensic practitioners who are statistically ignorant routinely testify as if they are knowledgeable consumers of statistics.[72] Such conduct is inexcusable because,

“Without a firm grasp of the principles involved, the unwary witness can be lead into making statements that he cannot properly uphold, especially in the matter of claiming inordinately high probability figures.”[73]

Similarly, experts have testified in numerous cases to specific probabilities based on statistical studies of unexplained origin.[74] Couple this incompetence with the fact that no functional databases exist for many of the identifications fields. Once more, when push comes to shove, the forensic examiner’s probabilistic determination is more likely to be a by-product of his subjective and personalized experience and training rather than his statistical acuity.

6. Documentation?

Three questions must be answered regarding documentation. First, did the examiner document any of her supporting data? Second, if documented, in what manner was the data documented? Last and most importantly, can the laboratory’s activities, observations, and results be reconstructed exclusively on the basis of the available records? In many cases, it is imperative that you look outside the examination report to ascertain how the examiner documented her work. The court-intended forensic report should be corroborate by a case file that includes all the notes, worksheets, printouts, charts, and other data or records used by the examiner to support her conclusions.[75] More importantly, as forensic scientist Janine Arvizu stresses,

“A laboratory case file is the repository for records generated during the analysis of evidence from a case. It should be an internally consistent, unbroken chain of records that document all activities, observations, measurements, and results relating directly to evidence from a given case. It should provide sufficient detail, so that someone who is versed in the technique, but not involved in the laboratory’s work, can understand what was done and the basis for the reported conclusions.”[76]

As these questions make clear, defense attorneys must obtain a copy of the case file (see infra next §). It is crucial that the entire case file is requested to determine whether the laboratory’s reported results are technically valid and whether the quality and uncertainty of the reported results can be supported based on the laboratory’s records. If the laboratory records and underlying documentation are noticeably absent from the case file, any results reached by are no more reliable, or for that matter verifiable, than eyewitness testimony. Science is premised on reproducibility; not on the forensic examiner’s ever fading memory. Consequently, the analysts and the lab must be able to produce any supporting documentation that verifies the quality and validity of its reported results.[77]

Moreover, if the lab’s results are to be reproducible the supporting documentation must be comprehensively and legibly written so that an independent expert is capable of retesting the original analysts hypotheses and conclusions. Again, as Janine Arvizu explains,

“The required information includes documentation regarding the integrity of the evidence sample(s), the procedures used during testing, the qualifications of the responsible analyst(s), the traceability of standards and measurements, instrument operating conditions and maintenance, results obtained for unknown samples and known controls, and the assumptions and basis for any statistical analyses and interpretation of results.”[78]

Quite often, especially with respect to forensic identifications, the supporting information is lacking given the subjective nature of the identification process. Customarily, forensic examiners cloak their conclusions in terms of their experience or specialized knowledge rather than any verifiable documentation. The fact that an identification is buttressed by no supporting documentation should make the obvious even more obvious—that forensic identifications are purely subjective decisions.

(2) Expectation Bias Questions & Inquires

These question and inquiries are intended to illicit evidence that calls into question whether the examiner was blind to certain expectations. Given the manner in which evidence is typically processed and the high rate of interaction between investigators, prosecutors and the examiner, it would be a shocking discovery to find that a state employed forensic examiner was not made privy to certain inculpatory expectations—conscious or unconscious.

1. The Case File & Domain-Irrelevant Information

As mentioned previously, in order to mount a legitimate challenge concerning questionable scientific conclusions or results by a state-employed forensic examiner, it is imperative that the defense obtain the underlying raw data (aka. the case file).[79] Obtaining the underlying data will permit an independent expert to evaluate the legitimacy of the forensic examiner’s conclusions. More importantly, however, the case file information can prove invaluable in determining whether the examiner was made privy to domain-irrelevant data or other examiners’ conclusions—information that is easily capable of cultivating a negative expectation against your client. For instance, did the examiner’s case file have all the preliminary crime scene reports and investigative reports? If so, what examination-irrelevant data was contained in these reports? For instance, a firearm expert’s case file would not require a crime scene report indicating that twelve eyewitnesses saw a man matching the description of your client running from the murder scene. This information has absolutely no bearing on the expert’s examination. Likewise, does the odontonlogist’s case file contain the victim’s statement that the defendant bit her right breast? Again, considering the nature of the odontologist’s examination, this superfluous data is not only irrelevant but it is entirely prejudicial given its expectation inducing power. Similarly, does the case file possess other experts’ reports? For example, given the nature of a toolmark expert’s examination, his case file need not contain the results of inculpatory DNA tests. Correspondingly, a fingerprint examiner’s case file would not require the odontologist’s report that inculpates your client. Once more, these bits of information are not relevant to the examiner’s ultimate responsibility. His duty is to answer one question—whether the defendant deposited the crime scene print. To answer this question, the examiner need only to focus his attention on the two prints—and only the two prints. Any superfluous data can only call into question the examiner’s conclusion[s].

2. Written Correspondence and Notes of Communications Between the Investigating Officers and the Examiner

Defense counsel must obtain any documented communications [e.g., handwritten, fax, e-mail] between the forensic examiner and the investigators. Such documentation can indicate whether the examiner was made aware of domain-irrelevant data or the investigators’ expectations with respect to the defendant’s guilt. For example, as previously discussed, transmittal correspondences that supplement a submission to a crime lab frequently convey more about the case than is required to carry out the required examination[s]. This information at times advises examiners about other inculpatory evidence and may include what the submitting investigator expects or hopes the requested tests will conclude.

George Castelle provides an excellent example of such a letter.[80] The example concerns a Fred Zain case.[81] Stephan Casey was arrested and charged with sexually abusing a five-year-old child. Prior to the West Virginia crime lab obtaining the physical evidence, Zain resigned from the lab and accepted a position with the Bexar County Medical Examiners Office in San Antonio, Texas. Once the West Virginia crime lab received the physical evidence, technicians who examined a carpet sample could not identify any semen stains on the carpet. Investigators had hoped that the carpet sample would contain the offender’s semen [aka. Mr. Casey’s semen]. Undisturbed by the crime lab’s failure to discover inculpatory evidence, investigators sent the carpet sample to Zain in San Antonio. The carpet sample was accompanied by the following letter:

“Mr. Zain:

This is the carpet that we discussed via Public Service. The W.Va. State Police Lab was unable to show any evidence of sperm or blood being present on it.

The suspect was arrested for 1st Degree Sexual Abuse on a 5-year-old female. Any evidence you can find pertaining to this crime will greatly increase our chances of conviction.

Thank you,

Det. R.R. Byard

Huntington Police Department”[82]

The letter contains both domain-irrelevant information and expectation cues. The irrelevant data being the fact that the West Virginia lab failed to identify any semen; and the fact that the defendant was being charged with sexually abusing a 5-year old. Both bits of information are completely irrelevant to whether Zain can identify a semen stain and whether that semen stain is consistent with the defendant’s semen. The expectation cue is quite obvious, “Any evidence you can find pertaining to this crime will greatly increase our chances of conviction.” When both of these factors are intertwined and spoon-fed to an individual like Zain, it is quite clear what the expected outcome must be (i.e., discover semen and then match this semen to the defendant); regardless of whether he can actually and truthfully accomplish such a task. As might be expected, Zain found what his forerunners failed to find.

Likewise, examiners may become deeply involved with investigators as the evidence in a case develops. This may encourage an increasing number of phone calls, emails or facsimiles between the examiner and investigators. These communications, like the initial transmittal correspondences, typically involve superfluous information that can only bias the examiner’s evaluations and ultimate conclusions. If the correspondences were via fax or email they should be easily obtainable and decipherable. Likewise, given the nature of the case, examiners and/or investigators may document, in handwriting, each conversation or visit with one another. If this is the case, it is imperative to obtain this documentation so you can ascertain whether the examiner was subjected to any irrelevant and potentially biasing information. Again, such information may include other inculpatory evidence, the defendant’s post-crime behavior, his jail behavior [if he is incarcerated pending trial], whether the defendant committed similar crimes or other offenses, etc.

One way to ascertain whether the examiner worked closely with investigators is to obtain copies of the crime lab’s attendance log. Most labs, especially those that are accredited, keep a detailed list of who enters and exits the crime lab and who they are visiting—even if those who are entering the lab are law enforcement or other examiners from a neighboring crime lab. Once in possession of the log, identify all instances where investigators met with the examiner. During cross-examination or a deposition, defense counsel is encouraged to ask the examiner to explain the contents of each visit. If he documented these visits, then his documentations should be turned over for inspection. If he failed to document when, why and what they discussed, scrutinize why a particular meeting was not worthy of documentation.

3. Peer Review—Exactly How and Why Was it Performed?

The fact that an examiner’s results were peer reviewed by another co-worker means absolutely nothing in certain contexts. Peer review, like the original examination it is reviewing, can be affected by context effects. Defense counsel must make a concerted effort at identifying and distinguishing between a “formalistic review” and an “independent confirmation.”

“Formalistic reviewing” is the type of “peer review” advocated by ASCLD standard 1.4.2.16.[83] Under this form of review, the peer reviewer merely acts as a process check on the procedures utilized by the initial examiner. His role is to make certain that the report satisfactorily documents and justifies its findings and conclusion[s]. ASCLD asserts that this form of “peer review” is designed “to ensure that the conclusions of its examiners are reasonable and within the constraints of scientific knowledge.”[84] Despite whether “formalistic” peer reviewers are exposed to the contaminating data that an initial examiner was exposed to, the reviewer normally knows the original examiner’s conclusions, itself a strong form of contamination. Nonetheless, if the reviewer’s role is simply to ensure that the report is satisfactorily documented and its conclusion adequately supported, then the fact that the reviewer is aware of the initial examiner’s outcome is arguably necessary. More importantly, however, if the reviewer’s sole purpose is to make certain that the initial examiner dotted his “i’s” and crossed his “t’s” the question becomes what is a review like this supposed to accomplish?

By no means does this form of peer review “independently confirm” the original conclusions’ correctness. All the peer reviewer did was read the final report and determine whether the initial examiner’s conclusions were reasonable and supported by the appropriate documentation. Reasonableness does not mean correctness. Given the subjective nature of many forensic assessments, examiner reasonableness may vary according to the nature of the examination. Take two fingerprint examiners for example. Fingerprint examiner A is able to identify fifteen corresponding points between two prints. Fingerprint examiner B is able to identify only eight corresponding points between two other prints. From this scenario, it’s obvious that the more reasonable identification would be examiner A’s given that he was able to identify twice as many corresponding points. However, no matter how reasonable a particular identification or conclusion may seem, the intensity of its reasonableness says nothing about whether the conclusion is correct. For instance, examiner A’s evaluation skills may not be as cautiously refined as examiner B’s. Consequently, examiner A may incorrectly identified various points on both prints. Again, correctness cannot be verified under these circumstances because the results have not been “independently confirmed” by another expert.

“Independent confirmation” involves having another examiner who is entirely blind or ignorant to the facts of the case and the original examiner’s conclusions. Under these circumstances, the reviewing examiner will not be persuaded by any extraneous information or expectations. This is a far cry from the cross-contamination of expectation inducing information that surfaces in “formalistic reviewing.” (e.g., the original examiner is aware of expectations; so is the reviewing examiner). Independent confirmation essentially involves evaluating all of the underlying documentation to determine whether the reviewing examiner can recreate the initial examiner’s conclusions. If the reviewing examiner cannot replicate the initial conclusions then the validity of the original examiner’s conclusions must be called into question. Again, it needs to be stressed that independent confirmation is a blind-testing procedure, in that the reviewing examiner has no active knowledge with respect to any superfluous data or anticipated outcomes.

In short, if confronted by an examiner who asserts that his conclusions were peer reviewed, defense counsel must dig deeper to ascertain the specific manner of peer review. This essential inquiry should be whether the independent reviewer was made privy to any information that could have potentially biased his review.

D. Conclusion

Crime labs and forensic science have increasingly played roles in numerous wrongful convictions.[85] As one criminal defense attorney recently commented,

“Although the law enforcement community and the courts bear the heaviest responsibility for convictions driven by police and prosecutorial misconduct, a substantial number of these cases involved faulty forensic evidence that should have been exposed and vigorously challenged by the defense.”[86]

Forcefully challenging forensic evidence involves not only attacking the legitimacy of the science that supports the alleged area of expertise, but also identifying context cues that could potentially skew the examiner’s evaluation. As this Article has repeatedly emphasized, “The most obvious danger in forensic science is that an examiner’s observations and conclusions will be influenced by extraneous, potentially biasing information.”[87] More importantly, however, while this Article has identified the more common forms of observer effect errors within the forensic science context, “there are other potentially error-producing sources of expectation beyond those induced by intentional or unintentional suggestion” that this Article has not covered.[88] Accordingly, it is incumbent upon defense attorneys to become more knowledgeable with respect to these covert forms of unintentional bias. Once aware that these unobserverable biases permeate crime labs and many forensic examinations, defense attorneys will be better equipped at eliciting evidence and testimony that can identify whether an examiner’s conclusions may have potentially been contaminated by observer effect errors. These inquiries will typically involve ascertaining whether the examination is subjectively top heavy and whether the examiner was made aware of certain expected or desired outcomes.

--------------------------------------------------------------------------------

[1] Investigator, Illinois Appellate Defenders Office, Death Penalty Trial Assistance Unit. J.D. candidate (2004), Northwestern University School of Law; M.S. (Forensic Science), University of New Haven; B.S. (Psychology), University of Pittsburgh. Mr. Cooley can be contacted via his website: www.law-forensic.com; or by email: c-cooley@law.northwestern.edu or Craig.Cooley@osad.state.il.us

[2] Irving Stone, Capabilities of Modern Scientific Laboratories, 25 Wm. & Mary L. Rev. 659, 659 (1984) (emphasis added).

[3] Paul L. Kirk, Crime Investigation 2 (2d. John I. Thornton ed. 1974) (emphasis added).

[4] Robert Rosenthal, Experimenter Effects in Behavioral Research 3 (1966).

[5] See Paul C. Giannelli, The Abuse of Scientific Evidence in Criminal Cases: The Need for Independent Crime Laboratories, 4 Va. J. Soc. Pol’y & L. 439 (1997) (discussing how the forensic community’s structural configuration has created many pro-prosecution forensic scientists).

[6] See Scott Bales, Turning the Microscope Back on Forensic Scientists, 26 Litigation 51 (2000) (discussing many instances concerning FBI examiners).

[7] See D. Michael Risinger et al., The Daubert/Kumho Implications of Observer Effects in Forensic Science: Hidden Problems of Expectation and Suggestion, 90 Cal. L. Rev. 1, 12-27 (2002).

[8] Michael J. Saks, Ethics in Forensic Science: Professional Standards for the Practice of Criminalistics, 43 Jurimetrics J. 359, 363 (2003) (book review).

[9] See Ulric Neisser, Cognition and Reality: Principles and Implications of Cognitive Psychology 43-45 (1976).

[10] David C. Thompson, DNA Evidence in the O.J. Simpson Trial, 67 U. Colo. L. Rev. 827, 845 (1996),

[11] See Victoria L. Phillips et al., The Application of Signal Detection Theory to Decision-Making in Forensic Science, 46 J. Forensic Science 294, 298 (2001) (“[F]orensic scientists often encounter ambiguous and murky decision-making situations.”).

[12] See John I. Thornton & Joseph L. Peterson, The General Assumptions and Rationale of Forensic Identification, in Science in the Law: Forensic Science Issues § 1-8.0 at 26-27 (David L. Faigman et al. 2d eds., 2002).

[13] Consider the wide ranging point ‘system’ in fingerprint, see Commonwealth v. Hunter, 338 A.2d 623, 624 (Pa. Super. Ct. 1975) (fourteen points); United States v. Durant, 545 F.2d 823, 825 (2d Cir. 1976) (fourteen points); Alexander, 571 N.E.2d 1075, 1078 (Ill. App. Ct. 1991) (eleven and fourteen points); State v. Starks, 471 So.2d 1029, 1032 (La. Ct. App. 1985) (twelve points); People v. People v. Garlin, 428 N.E.2d 697, 700 (Ill. App. Ct. 1981) (twelve points); Garrison v. Smith, 413 F. Supp. 747, 761 (N.D. Miss. 1976) (twelve points); State v. Murdock, 689 P.2d 814, 819 (Kan. 1984) (twelve points); Magwood v. State, 494 So.2d 124, 145 (Ala. Crim. App. 1985) (eleven points); State v. Cepec, 1991 WL 57237, at *1 (Ohio Ct. App. 1991) (eleven points); Ramirez v. State, 542 So.2d 352, 353 (Fla. 1989) (ten points); People v. Jones, 344 N.W.2d 46, 46 (Mich. Ct. App. 1983) (ten points); State v. Jones, 368 S.E.2d 844, 846 (N.C. 1988) (ten points); Commonwealth v. Ware, 329 A.2d 258, 276 (Pa. 1974) (nine points); State v. Awiis, 1999 WL 391372, at *7 (Wash. Ct. App. 1999) (eight points); Commonwealth v. Walker, 116 A.2d 230, 234 (Pa. Super. Ct. 1955) (four points). See also Robert Epstein, Fingerprints Meet Daubert: The Myth of Fingerprint “Science” Is Revealed, 75 S. Cal. L. Rev. 605 (2002).

[14] See Victoria L. Phillips et al., The Application of Signal Detection Theory to Decision-Making in Forensic Science, 46 J. Forensic Science 294, 299 (2001).

[15] Determining whether the match is a coincidental match—examiners are essentially asking how probable is it to find a match by pure chance.

[16] See id. (“[W]ith the exception of such areas as biological fluids… the forensic sciences possess little empirical data to assist examiners in interpreting the meaning of their test results and affixing a probability or confidence to their findings.”).

[17] See id. at 299 (“Ordinarily, the examiner does not have access to a database that assists in quantifying the rarity of the marks, or which even records them, but must rely on memory of other samples viewed in the past.”).

[18] Barry Gaudette, Basic Principles of Forensic Science, in 1 Encyclopedia of Forensic Science 300 (Jay A. Siegel et al. 2000) (emphasis added).

[19] Michael J. Saks, Banishing Ipse Dixit: The Impact of Kumho Tire on Forensic Identification Science, 57 Wash. & Lee L. Rev. 879, 882 (2000).

[20] Joan Griffin & David J. LaMagna, Daubert Challenges to Forensic Evidence: Ballistics Next on the Firing Line, Champ. (Sept.-Oct. 2002), at 20, 58.

[21] See Lynn C. Hartfield, Daubert/Kumho Challenges to Handwriting Analysis, Champ. (Nov. 2002), at 24 (discussing various manners to challenge handwriting).

[22] See Joan Griffin & David J. LaMagna, Daubert Challenges to Forensic Evidence: Ballistics Next on the Firing Line, Champ. (Sept.-Oct. 2002), at 20, 58.

[23] See Paul C. Giannelli , The Twenty-First Annual Kenneth J. Hodson Lecture: Scientific Evidence in Criminal Prosecutions, 137 Mil. L. Rev. 167, 184-85 (1992).

[24] See Joseph L. Peterson, Steven Mihajlovic & Michael Gilliland, Forensic Evidence and the Police 117 (National Institute of Justice Research Report, 1984).

[25] See D. Michael Risinger et al., The Daubert/Kumho Implications of Observer Effects in Forensic Science: Hidden Problems of Expectation and Suggestion, 90 Cal. L. Rev. 1, 47 (2002).

[26] See Technical Working Group for Eyewitness Evidence, United States Dep’t of Justice, Eyewitness Evidence: A Guide for Law Enforcement (1999) (discussing the various methods of eyewitness identification).

[27] See Gary L. Wells et al., Eyewitness Identification Procedures: Recommendations for Lineups and Photospreads, 22 Law & Hum. Behav. 603 (1998).

[28] See D. Michael Risinger et al., The Daubert/Kumho Implications of Observer Effects in Forensic Science: Hidden Problems of Expectation and Suggestion, 90 Cal. L. Rev. 1, 47-50 (2002).

[29] D. Michael Risinger et al., The Daubert/Kumho Implications of Observer Effect in Forensic Science: Hidden Problems of Expectation and Suggestion, 90 Cal. L. Rev. 1, 29 (2002). Consider, for example, the following quote by a forensic scientist advocating the teaching of forensic science in high school.

“Forensic science appeals to the detective in all of us as evidenced by the growth in popularity of media including TV, best-selling novels, and movies that portray some aspect of crime solving.”

Editorial, Forensics: The Thrill Is The Detective Work, Wall St. J., Mar. 5, 2002, at A17, available at 2002 WL WSJ3387690 (emphasis added). I would argue that for aspiring scientists, forensic science would appeal to the scientist in all of them. It is comments like this that create the incorrect notion that forensic scientists are suppose to solve the case by acting like investigators. When forensic practitioners act and think like detectives, wrongful convictions are bound to surface.

[30] D. Michael Risinger et al., The Daubert/Kumho Implications of Observer Effect in Forensic Science: Hidden Problems of Expectation and Suggestion, 90 Cal. L. Rev. 1, 28 (2002).

[31] Clive A. Stafford Smith & Patrick D. Goodman, Forensic Hair Comparison Analysis: Nineteenth Century Science or Twentieth Century Snake Oil, 27 Colum. Hum. Rts. L. Rev. 227, 259 (1996) (emphasis added).

[32] Larry S. Miller, Procedural Bias in Forensic Science Examinations of Human Hair, 11 Law & Hum. Behav. 157, 158 (1987) (emphasis added).

[33] Michael J. Saks, Banishing Ipse Dixit: The Impact of Kumho Tire on Forensic Identification Science, 57 Wash. & Lee L. Rev. 879, 886 (2000).

[34] See Office of Inspector General, U.S. Dep’t of Justice, The FBI Laboratory: Investigation Into Laboratory Practices and Alleged Misconduct in Explosive-Related and Other Cases (April 1997).

[35] Scott Bales, Turning the Microscope Back on Forensic Scientists, 26 Litigation 51 (2000) (discussing the OIG’s findings and recommendations).

[36] See John I. Thornton & Joseph L. Peterson, The General Assumptions and Rationale of Forensic Identification, in Science in the Law: Forensic Science Issues § 1-1.1 at 2 (David L. Faigman et al. 2d eds., 2002) (“Most forensic examinations are conducted in government-funded laboratories, usually located within law enforcement agencies, and typically for the purpose of building a case for the prosecution.”).

[37] See Maurice Possley, New tests requested on victim’s bite marks, Chi. Trib., July 25, 2003, at 4.

[38] See id. See also Chase Squires, Dentists cleared in wrongful arrest suit, St. Petersburg Times, Dec. 22, 2000, at 1 (discussing how Dale Morris was jailed for four month for murdering a nine year old neighbor because a forensic dentist mistakenly concluded that Morris’ bite pattern matched those on the victim; Morris was exonerated through DNA evidence); Katherine Ramsland, Whose Bite Mark is it, Anyway?, at http://www.crimelibrary.com/criminal_mind/forensics/bitemarks/5.html?sect=21 (discussing Ricky Amolsch’s case); Ellen O’Brien, From DNA to Police Dogs, Evidence Theories Abound, Boston Globe, Jan. 22, 1999, at B1 (discussing Edmund Burke’s case).

[39] See generally David W. Peterson & John M. Conley, Of Cherries, Fudge and Onions: Science and its Courtroom Perversion, 64 Law & Contemp. Probs. 213, 227-32 (2001) (discussing ‘cherrypicking” in the legal and scientific context).

[40] D. Michael Risinger et al., The Daubert/Kumho Implications of Observer Effect in Forensic Science: Hidden Problems of Expectation and Suggestion, 90 Cal. L. Rev. 1, 47-50 (2002) (advocating the use of evidence lineups); see also Larry S. Miller, Procedural Bias in Forensic Science Examinations of Human Hair, 11 Law & Hum. Behav. 157, 159 (1987) (same).

[41] Professor Saks and his colleagues employ this word, see D. Michael Risinger et al., The Daubert/Kumho Implications of Observer Effect in Forensic Science: Hidden Problems of Expectation and Suggestion, 90 Cal. L. Rev. 1, 48 (2002).

[42] Id.

[43] See Anne Constable, Eyewitness Identification Has Shortcomings, Some Say, Sante Fe New Mexican, Dec. 8, 2002,a t A-1 (discussing the various shortcomings concerning ‘old-school’ eyewitness identification methods); Amy L. Bradfield et al., The Damaging Effect of Confirming Feedback on the Relation Between Eyewitness Certainty and Identification Accuracy, 87 J. Applied Psych. 112 (2002) (highlighting more weaknesses regarding certain methods of eyewitness identification).

[44] A ‘show-up’ is an identification procedure where the witness is presented with a single suspect for identification.

[45] See Technical Working Group for Eyewitness Evidence, United States Dep’t of Justice, Eyewitness Evidence: A Guide for Law Enforcement (1999).

[46] Barry C. Scheck, DNA and Daubert, 15 Cardozo L. Rev. 1959, 1997 (1994).

[47] See Peter Straton & Nicky Hayes, A Student’s Dictionary of Psychology 82 (3d ed. 1999). Defining double-blind control as,

“An experimental control in which neither the person conducting the experiment nor the research participants in the study are aware of the experimental hypothesis or conditions. Double blind controls are necessary as precautions against experimenter effects, and are considered essential in tests on new drugs or assessments of therapeutic procedures.”

[48] D. Michael Risinger et al., The Daubert/Kumho Implications of Observer Effect in Forensic Science: Hidden Problems of Expectation and Suggestion, 90 Cal. L. Rev. 1, 45 (2002) (emphasis added).

[49] See id. at 31 (“forensic examiners should be insulated form all information about an inquiry except necessary domain specific information…”); Steve Selvin & B.W. Grunbaum, Genetic Marker Determination in Evidence Bloodstains: The Effect of Classification Errors on Probability of Non-discrimination and Probability of Concordance, 27 J. Forensic Sci. Soc’y 57, 61 (1986) (“The possibility of bias is minimized for any testing procedure if all analysts are unacquainted with circumstances of the alleged crime and unaware of any previous results.”); Janet C. Hoeffel, The Dark Side of DNA Profiling: Unreliable Scientific Evidence Meets the Criminal Defendant, 42 Stan. L. Rev. 465, 486 (1990) (“[E]xaminers should be told neither the origin of the samples nor the prosecution’s theory of the case.”).

[50] This notion is by no means novel. Rather, in 1894 Hagan expressed the fact that,

“The [document] examiner must depend wholly upon what is seen, leaving out of consideration all suggestions or hints from interested parties… Where the expert has no knowledge of the moral evidence or aspects of the case… there is nothing to mislead him…”

William E. Hagan, Disputed Handwriting 82 (1894).

[51] National Research Council, The Evaluation of Forensic DNA Evidence 79 (1996).

[52] Considering that the forensic community “deals mostly with inanimate objects, the blinding procedure will be simpler than in fields that work with humans and animals, such as biomedical research and psychology. Those fields must construct double-blind studies, while forensic science needs only single-blind procedures.” D. Michael Risinger et al., The Daubert/Kumho Implications of Observer Effect in Forensic Science: Hidden Problems of Expectation and Suggestion, 90 Cal. L. Rev. 1, 45 n. 205 (2002).

[53] See R. Cook, et al., A Model for Case Assessment and Interpretation, 38 Sci. & Just. 151 (1998) (discussing the ‘filtering’ process system that has been developed in the United Kingdom’s Forensic Science Service).

[54] For a list of cases see http://www.law-forensic.com/cfr_science_myth.htm (website maintained by author).

[55] See Associated Press, A Year of Scandals With Forensic Evidence, Wash. Post, July 27, 2003, at A05 (discussing the various forensic mishaps over the past year; many would not have been identified without the assistance of skeptical defense attorneys).

[56] See Dan Cray et al., How Science Solves Crimes From ballistics to DNA, forensic scientists are revolutionizing police work—on TV and in reality. And just in time, Time Mag., Oct. 21, 2002, at 36.

“TV viewers can tune into a forensics drama almost every night of the week, starting with the trend setting CSI on CBS; its first-season spawn, CSI: Miami, also on CBS; and Crossing Jordan on NBC. On cable, The Forensics Files is Court TV’s biggest prime-time show ever, while Autopsy is wooing—and spooking—viewers on HBO.”

[57] See Victoria L. Phillips et al., The Application of Signal Detection Theory to Decision-Making in Forensic Science, 46 J. Forensic Science 294, 299 (2001).

[58] John I. Thornton & Joseph L. Peterson, The General Assumptions and Rationale of Forensic Identification, in Science in the Law: Forensic Science Issues § 1-5.3 at 14 (David L. Faigman et al. 2002).

[59] See id.

[60] Craig M. Cooley, Forensic Individualization Sciences and the Capital Jury: Are Witherspoon Jurors More Deferential to Suspect Science than Non-Witherspoon Jurors?, 27 S. Ill. U. L. J. 221, 250 (2003) (citing to Michael J. Saks, Implications of the Daubert Test for Forensic Identification Science, 1 Shepard’s Expert & Sci. Evidence 427 (1997)).

[61] By no means is this a recent realization. Various commentators have noted the community’s lack of research for nearly half a century, see James Osterburg, A Commentary on Issues of Importance in the Study of Investigation and Criminalistics, 11 J. Forensic Sci. 261, 261 (1966) (“In criminalistics, however… [published data] is almost nonexistent. Testimony reported in the hearings emphasizes unintentionally the scarcity of published data through failure to mention any journals in which such vital information is available.”); Paul Kirk, The Ontogeny of Criminalistics, 54 J. Crim. L. & Criminology 235, 238 (1963) (“Its unfortunate that the great body of knowledge which exists in this field is largely uncoordinated and has not yet been codified in clear and simple terms.”); Alfred Biasotti, The Principles of Evidence Evaluation as Applied to Firearms and Tool Mark Identification, 9 J. Forensic Sci. 428, 428 (1964) (“[Forensic] authors have given many theoretical and an few practical applications of statistical probability to criminalistics problems and have pointed out the serious lack of fundamental data which would allow broader practical applications.”).

[62] See e.g., John J. Harris, How Much Do People Write Alike?, 48 J. Crim. L. & Criminology 637 (1958) (finding that, contrary to apparent belief of handwriting experts, it is not true that no two people write indistinguishably alike).

[63] See e.g., Simon Cole, Suspect Identities: A History of Fingerprinting and Criminal Identification (2001) (discussing how the American fingerprint examiners are not required to identify a certain number of matching points).

[64] John I. Thornton & Joseph L. Peterson, The General Assumptions and Rationale of Forensic Identification, in Science in the Law: Forensic Science Issues § 1-7.2 at 24 (David L. Faigman et al. 2002); Christophe Champod & Ian W. Evett, A Probabilistic Approach to Fingerprinting Evidence, 51 J. Forensic Identification 101, 103 (2001) (“[T]he process of identification… is essentially inductive and… probabilistic.”).

[65] Paul L. Kirk & Charles R. Kingston, Evidence Evaluation and Problems in General Criminalistics, 9 J. Forensic Sci. 434, 437 (1964) (emphasis added).

[66] See Norah Rudin & Keith Inman, Principles and Practice of Criminalistics: The Profession of Forensic Science 302 (2001) (“Now, more than ever, the onslaught of technology obligates the criminalist to draw on a strong background in the physical sciences, including an understanding of statistics and logic.”).

[67] See Paul L. Kirk & Charles R. Kingston, Evidence Evaluation and Problems in General Criminalistics, 9 J. Forensic Sci. 434, 435 (1964) (noting that “criminalists… do not understand statistics, and do not know how to use them constructively.”).

[68] David Derbyshire, Misleading statistics were presented as facts in Sally Clark trial, Daily Telegraph (London), June 12, 2003, at 04 (discussing how erroneous statistics lead to Sally Clark’s wrongful conviction); Carlos Miller, Phoenix police lab errs on DNA, Arizona Republic, May 6, 2003, at 1B (discussing how Phoenix crime lab technicians miscalculated the likelihood that a person’s DNA was present on evidence in nine cases; the cases ranged from aggravated assaults to rapes and murders); Melody McDonald, DNA tests sways prosecutor, Star-Telegram (Ft. Worth, TX), Oct. 10, 2002, at 1 (discussing how a Ft. Worth DNA analyst’s miscalculations forced Ft. Worth prosecutors to drop the death penalty against Jamien Demon Nickerson). Probably the most glaring example of statistical incompetence is Arnold Melnikoff. Melnikoff’s statistical errors helped erroneously convict Jimmy Ray Bromgard. See Wrong conviction brings scientist’s work into question, San Diego Union-Trib., Jan. 5, 2003, at A4 (discussing Bromgard’s wrongful conviction). According to Walter Rowe, a forensic chemistry professor at George Washington University,

“Everything about the estimates in Melnikoff’s [statistical] testimony is just bullshit… It is nonsense on stilts. The most glaring is this idea that microscopic features of head hair and pubic hair are not correlated, that those are independent probabilities… This kind of theorizing flies in the face of every adult’s common knowledge. Anyone can look at that and rightly ask what in the world he was talking about.”

Charlie Gillis, Scandal in the forensic labs: Hundreds of cases undergoing review in Montana, Nat’l Post, Feb. 1, 2003, at B01. Josiah Sutton’s wrongful conviction, can be attributed to a Houston crime lab technician’s statistical miscalculation concerning DNA evidence. See Roma Khanna, DNA from conviction of teen will be retested, Hous. Chron., Feb. 8, 2003, at 33. Likewise, in Gary Dotson’s 1979 trial for the rape of Cathleen Crowell Webb,

“a state forensic scientist testified that according to a genetic marker test he performed, Dotson was one of only 10% of Caucasian men who could have been the source of the semen found on her panties. Years after Dotson had gone to prison for the rape, Webb recanted her story. To bolster their claim that Dotson was guilty, authorities retested the semen, using the same test, but a different operator. The retest showed that two-thirds of the white male population could have been the source of the semen.”

Anthony Pearsall, DNA Printing: The Unexamined ‘Witness’ In Criminal Trials, 77 Cal. L. Rev. 665, 674 (1989).

[69] The multiplication rule states that the “chance that two things will both happen equals the chance that the first will happen, multiplied by the chance that the second will happen given that the first has happened.” David Freedman et al., Statistics 229 (3d ed. 1998); see also Michael J. Saks, Merlin and Solomon: Lessons From the Law’s Formative Encounters With Forensic Identification Science, 49 Hastings L.J. 1069, 1086 (1998) (“The essential idea of this concept is that if objects vary on a number of independent (i.e., uncorrelated) dimensions, the probability of occurrence of any one combination of characteristics is found by multiplying together the probabilities associated with each dimension.”).

[70] Charles R. Kingston & Paul L. Kirk, The Use of Statistics in Criminalistics, 55 J. Crim. L. & Criminology 514, 516 (1964).

[71] See Charlie Gillis, Scandal in the forensic labs: Hundreds of cases undergoing review in Montana, Nat’l Post, Feb. 1, 2003, at B01 (discussing how Arnold Melnikoff misapplied the multiplication rule by failing to correctly consider the independence of variables). See also David Freedman et al., Statistics 229 (3d ed. 1998) (“Two things are independent if the chances for the second given the first are the same, no matter how the first one turns out. Otherwise, the two things are dependent) (emphasis in original).

[72] According to Professor Andrea A. Moenssens,

“Experts use statistics compiled by other experts without any appreciation of whether the data base upon which the statistics were formulated fits their own local experience, or how the statistics were compiled. Sometimes these experts, trained in one forensic discipline, have little or no knowledge of the study of probabilities, and never even had a college level course in statistics.”

Andrea A. Moenssens, Novel Scientific Evidence in Criminal Cases: Some Words of Caution, 85 J. Crim. L. & Criminology 1, 18 (1993).

[73] Paul L. Kirk & Charles R. Kingston, Evidence Evaluation and Problems in General Criminalistics, 9 J. Forensic Sci. 434, 437 (1964).

[74] See Clive A. Stafford Smith & Patrick D. Goodman, Forensic Hair Comparison Analysis: Nineteenth Century Science or Twentieth Century Snake Oil, 27 Colum. Hum. Rts. L. Rev. 227, 257-58 (1996) (discussing various instances).

[75] See Scott Bales, Turning the Microscope Back on Forensic Scientists, 26 Litigation 51 (2000)

[76] Janine Arvizu, Shattering the Myth: Forensic Laboratories, CHAMP. (May, 2000), at 18,23.

[77] See id.

[78] Id.

[79] See George Castelle, Lab Fraud: Lessons Learned from the ‘Fred Zain Affair’, Champ., May 1999, at 12; Janine Arvizu, Shattering the Myth: Forensic Laboratories, CHAMP. (May, 2000), at 18.

[80] See George Castelle, Lab Fraud: Lessons Learned from the ‘Fred Zain Affair’, Champ., May 1999, at 12.

[81] See Matter of Investigation of West Virginia State Police Crime Laboratory, Serology Div., 438 S.E.2d 501 (W.Va.1993) (discussing, at length, Zain’s fraudulent conduct and incompetence).

[82] George Castelle, Lab Fraud: Lessons Learned from the ‘Fred Zain Affair’, Champ., May 1999, at 12.

[83] See American Society of Crime Laboratory Directors, Laboratory Accreditation Board Manual (2000).

[84] Id. at § 1.4.2.16.

[85] See Robert Tanner, Crime Labs Placed Under a Microscope: Miscues Lead to Calls for Changes in Forensic Labs, Wash. Post, July 27, 2003, at A05 (discussing recent forensic mishaps and wrongful convictions attributable to crime labs or suspect forensic science).

[86] Michele Nethercott, Indigent Defense, Champ. (June 2003), at 61 (emphasis added)

[87] D. Michael Risinger et al., The Daubert/Kumho Implications of Observer Effect in Forensic Science: Hidden Problems of Expectation and Suggestion, 90 Cal. L. Rev. 1, 9(2002).

[88] Id. at 9-10.


______________________________________________________________________
Feel free to pass The Detail along to other examiners.  This is a free newsletter FOR latent print examiners, BY latent print examiners. There are no copyrights on The Detail (except in unique cases such as this week's article), and the website is open for all to visit.

If you have not yet signed up to receive the Weekly Detail in YOUR e-mail inbox, go ahead and join the list now so you don't miss out!  (To join this free e-mail newsletter, send a blank e-mail from the e-mail address you wish to subscribe, to: theweeklydetail-subscribe@topica.email-publisher.com)  If you have problems receiving the Detail from a work e-mail address, there have been past issues with department e-mail filters considering the Detail as potential unsolicited e-mail.  Try subscribing from a home e-mail address or contact your IT department to allow e-mails from Topica.  Members may unsubscribe at any time.  If you have difficulties with the sign-up process or have been inadvertently removed from the list, e-mail me personally at kaseywertheim@aol.com and I will try to work things out.

Until next Monday morning, don't work too hard or too little.

Have a GREAT week!