Biometrics and the Courts

Reliability and Validity of Evidence in Court

Evidence, simply described as any material concerning a court proceeding, is one of the most crucial aspects in courts for both the prosecution and defendants. Both the prosecution and defendants have the right to present their evidence to the court, and the court must determine if the evidence will be admitted as part of the legal proceeding or if it will be excluded. Four groups of evidence include natural, demonstrative, testimonial, and documentary evidence (Keane &Mckeown, 2014). These groups can further be broken down into different types: pictures, audio, videos, witness testimony, scientific evidence like DNA results, fingerprints, written statements, and demonstrative evidence like graphs and digital evidence.

Several requirements must be met to consider evidence reliable and valid in court; the first is unfairly prejudicial evidence. Prejudicial evidence can arouse the jury’s emotions, especially rage, without adding any critical information. For instance, a picture of a crime scene with a mutilated body is considered prejudicial evidence and, hence, not valid (Keane &Mckeown, 2014). Secondly, the court determines whether the evidence is reliable and valid by examining the expert testimony. This is testimony that qualified experts can exclusively give. The law must establish such experts as qualified and can prove their qualifications in connection to what their testimonial is about. In addition, there are several standards that expert opinion has to meet to be considered valid, depending on the case. For instance, in a sexual abuse case, only a doctor can give a medical report because they are qualified to know the signs consistent with the crime, and the appropriate tests have been carried out and repeated to confirm the results (Keane &Mckeown, 2014). Further experts are most crucial where scientific evidence or digital evidence is involved.

Another crucial factor in determining the reliability of evidence is assessing if the source is reliable. For example, in witnesses with privilege, like attorney-client privilege, doctor-patient privilege, priest-penitent privilege, and spousal privilege, either party of these relationships has no right to testify against the other party in the relationship (Keane &Mckeown, 2014). However, there are exceptions to this rule depending on the relationship, and in some circumstances, the privilege right can be waived. In such cases, it’s the court’s responsibility to determine whether the evidence from such a witness is valid. Another factor is hearsay; hearsay is when someone gives testimony about what they heard someone else say outside the court. In most cases, hearsay is considered inadmissible—however, there are over forty exceptions to hearsay evidence, for example, in the dying declaration exception. Therefore, the court must analyze all these exceptions and determine whether the hearsay evidence presented is reliable.

Further, courts must determine if the evidence presented is misleading or relevant and necessary to make the case. For example, details about a defendant being of a particular race in a fraud case may seem unnecessary and a waste of time. The case is about proving the defendant committed fraud, making their race irrelevant and prejudicial. Therefore, the court has to determine if the evidence is valid or misleading in terms of such evidence. Nevertheless, when it comes to evidence reliability, there are rules, exceptions to the rules, and exceptions to the exceptions. Therefore, deciding whether to admit evidence or dismiss it must be carefully conducted.

The Role of the Frye Standard or Daubert Standard in Determining Whether or Not the Courts Will Accept Biometric Evidence

Biometric evidence has been retrieved using automated mechanisms that help identify individuals using their behavioral and biological characteristics. These characteristics include fingerprints, voice, face, handwriting, and eyes, to name a few. These mechanisms are based on different recognition techniques, which include feature matching, feature extraction, and feature robustness. Due to the complexity of these innovations, experts are the only ones who can analyze data using these innovations to try and solve a crime. Consequently, these experts are the only ones who can give expert testimonies in courts.

Even though there are factors to consider when vetting an expert to be a witness, the ultimate decision that the court has to make is whether that expert’s testimony is admissible. Subsequently, the courts have standards that determine whether expert testimony is admissible. The standards used are the Frye and Daubert standards. These standards differ between Federal and state courts; Federal courts exclusively follow the Daubert standards, while state courts are divided between the two (Benedict 2004).

The Frye standard was first set in 1923 in the case of Frye vs. the United States. Accordingly, the standard premise in this case was that an expert’s testimony is admissible if experts generally accept the scientific technique or biometric evidence. Therefore, in Frye’s standard, the general acceptance test is the main qualifier for admission of biometric evidence (Benedict 2004). Notably, most states abandoned the Frye standard in favor of the Daubert standard. The remaining states use the Frye standard.

The Daubert standard overruled the Frye standard in 1993 by the Supreme Court. In the case of Daubert vs. Merrell Dow Pharmaceuticals, the court ruled that the general acceptance test was inconsistent with Rule 702 in the Federal Rules of Evidence. Rule 702 has two standards, relevance and reliability, contradicting the Frye standard (Benedict 2004). Consequently, the Supreme Court recommended a more liberal approach towards expert testimony whereby the central aspect to consider was vigorously cross-examining the expert witness. Further, the court provided several factors to be considered in biometric evidence and expert testimonies (Jensen, 2002). They include;

Can the scientific knowledge provided by the expert witness be tested, or has it been tested and evaluated for reliability?
Has the scientific knowledge or technique been subjected to peer review and publication?
Does the technique, scientific knowledge, or theory have a known or potential error rate?
Are there any standards and controls for regulating and maintaining scientific knowledge?
Has the technique been “generally accepted” in its respective community?

The main difference between these two standards is that while Frye takes the broad approach by focusing on one question- whether the scientific evidence is generally accepted in the relevant community- the Daubert standard takes a more narrowed-down approach (Jensen 2002). Using a list of several factors, the Daubert standard can offer more factors that must be considered in expert testimony. In addition to federal courts, more than 40 states have adopted the Daubert standard.

Florida’s State Approach to Biometric Evidence

As of 2019, the state of Florida, my resident state, uses the Daubert standard when it comes to biometric evidence. This was after decades of using Frye’s and Daubert’s standards only as procedural. However 2018, in the case of DeLisle v. Crane Co, a legislature’s amendment was declared an unconstitutional breach of the court’s oversight of the procedural guidelines of evidence; consequently, in 2019, the court overturned its course and implemented the Daubert standard as a procedural rule.

Reliability of Fingerprint Evidence

Although fingerprint evidence was first used in the late 1800s in Europe, America’s first case was solved using fingerprint evidence in the early 1900s. This was after an armed robber, Thomas Jennings, killed a man who was the owner of the house Thomas was robbing. Even after being stopped half a mile away from the crime scene with a gun and a bloodied coat, his fingerprint at the crime scene is what was used to tie him to the crime scene and convict him of murder. Following this conviction, the use of fingerprint evidence has endured till today.

However, this type of evidence has been surrounded by controversy that mainly questions its reliability. This is primarily because most biometric techniques rely on the subjective judgment of an expert. According to a PCAST report, fingerprint evidence does have foundational reliability; however, a significant probability of error exists, which varies depending on how well individual examiners conduct their work, the objective proficiency of the individual examiner, and the quality of the evidence at issue (Garrett 2018). This probability exists because fingerprint analysis is a subjective technique, implying that it relies primarily on the judgment and experience of an individual examiner. Fingerprint evidence is only as good as the person who examined the prints. In addition, the expert may not be able to fully explain how they concluded that a fingerprint, ballistics, or other types of pattern evidence is a positive match, except to cite their judgment and experience. Furthermore, fingerprint evidence has been admitted into court for centuries without being questioned, and very few courts have assessed the reliability of this kind of evidence (Garrett 2018). Consequently, numerous dismissed or overturned cases have occurred due to unreliable fingerprint evidence.

One example of a case where fingerprint evidence was unreliable was the Mcphaul case. Accordingly, Juan Mcphaul, the defendant, was convicted of attempted murder, conspiracy to commit robbery, and assault. The conviction had been made due to, among other evidence, an expert testimony that described evidence retrieved from latent fingerprinting. However, using testimony, the expert could not meet Daubert’s standards of evidence. For example, she could not explain how the fingerprints were compared, what features matched, what standards were followed, what documentation proved that the fingerprints matched, or provide an impartial reason to support her evidence (Garrett 2018). Further, the expert could not provide any peer reviews or authority cited for her conclusion. As a result, the Court of Appeal concluded that the evidence was inadmissible.

Nevertheless, fingerprint evidence has been beneficial in the conviction of numerous criminals in history; however, it has been established that latent fingerprinting relies more on experts’ training and experience rather than a specific set of rules. This makes its reliability questionable.

Comparison of Biometrics to the Reliability and Validity of Fingerprint Evidence

Biometrics can be divided into two categories: behavioral and physiological. Behavioral measurements include voice, gait, and gestures, among others. On the other hand, physiological measurements incorporate biological analyses like DNA, semen, urine, and saliva and morphological identifiers like facial features, hand shapes, eye color, and iris and vein patterns, among others.

While the advancement of such biometric techniques has been impressive over the last decade, there is still room for error—besides, the more technology advances, the more the technology will counter the former advances. For example, in facial and voice recognition, the technique identifies one using facial and vocal features; in contrast, technology like a deep fake application challenges both techniques. Deep fakes are videos, audio, or pictures made using someone else’s face or voice. They are made using artificial intelligence. Such technology has made it difficult for biometric techniques to determine if the face or voice is the actual person or an artificially made imitation of the original person to frame the owner of the face or voice.

Furthermore, more risk factors have been discovered regarding biometric evidence. First, some techniques used in biometrics are either more or less suited for particular individuals. For example, a particular technology may work well for women and less for men, for dark skin, for light skin, for older people, and young ones (Singh & Prasad 2018). Research by the National Institute of Standards and Technology showed that facial recognition programs generated higher rates of false positives, especially for black faces (Leslie 2020). The results showed false positives of black faces, approximately 100 times more than Caucasian faces. This demonstrates an embedded bias in these technologies; therefore, the evidence is undoubtedly questionable.

Moreover, even though more than ten thousand police agencies exist in America, no specific and uniform regulations exist for developing facial recognition algorithms. Thus, the techniques used by these law enforcement agencies vary in how they use them but also in the levels of accuracy. This fact means the algorithms do not follow the Daubert or the Frye standards regarding expert evidence. There is no specific way to measure the accuracy of these algorithms because they all differ from each other.

Nevertheless, not all biometric evidence is questionable or has been proven to be biased. Biometric evidence obtained from physiological measurements can be pretty hard to disprove. For example, DNA, I believe that this is one of the most reliable and valid biometric evidence. DNA can be obtained from hair, body fluids, skin cells, and bones. Further, DNA profiling has been established as the gold standard in forensic and biometric science due to the many cases that DNA has solved, including identifying criminals or releasing innocent imprisoned people. In addition, unlike facial recognition and latent fingerprinting techniques, which have no regulations, DNA evidence has stringent and standardized rules (Walton 2006). These rules follow the two standards of biometric evidence. Therefore, physiological measurements are most reliable and valid when comparing the reliability of biometric evidence obtained using various biometric techniques. However, some biometric techniques like facial recognition are still questionable as much as latent fingerprint evidence is.

Challenges Associated with Lower Forms of Biometrics, Such as Facial Recognition and Acceptance as Evidence in Court

Some biometrics, considered the lower forms of biometrics, face several challenges regarding acceptance as evidence in court. These include voice recognition or facial recognition, among several others. For instance, facial recognition, a probabilistic technology, produces more or less likely results but is never definite. This leaves a lot of room for error. Secondly, not having established rules for the facial recognition software to follow makes it easy for cases to be challenged and dismissed. Thirdly, numerous factors can affect the accuracy of facial recognition software, for example, distance, light, camera quality, and a person’s gender or race. In addition, other challenges to accuracy include changes like when a person cuts or dyes their hair, camera angles, growing or shaving off a beard, and aging, among others.

Thirdly, facial recognition software raises the question of privacy and human rights violations. Over the last few years, various occurrences have brought about controversies regarding facial recognition software and human rights. One occurrence happened in the United Kingdom, whereby the Court of Appeal ruled that facial recognition for law enforcement was unlawful as it violated equality laws, human rights, and data protection laws. In addition, states like Massachusetts, California, San Francisco, and Cambridge have banned such technology due to privacy issues.

In addition, facial recognition using mass surveillance technology means that people are being watched in real-time; thus, law enforcement agencies can use these resources to run everyone through the algorithms to try and find a criminal amid the crowd. This puts everyone in a virtual criminal line-up without knowing about it, hence treating them like criminals without probable cause. Presenting evidence obtained from mass surveillance equipment could backfire on the law enforcement agencies using that technology. Furthermore, due to the complexity of these technologies, sometimes it gets challenging for the jurors to comprehend the details of the evidence. When presenting evidence using advanced methods, there is always a high chance of exacerbating the jurors’ stress. This can elicit two types of reaction: excessive jury skepticism or deference (Roberts 2015). Jury deference occurs when the jury does not understand the expert evidence, so they easily defer to the expert’s evidence. Contrastingly, jury skepticism occurs when the jury does not understand expert evidence and, consequently, fails to credit expert testimony with the probative value it truly merits (Roberts 2015).

Notably, most of the challenges that biometrics like facial recognition face mainly arise from its applicability. With few to no rules regulating how it’s built and used, this opens up more room for such biometric evidence to be quickly challenged, dismissed, or even brought up charges against the evidence presenting side for how it was obtained. Thus, there are low acceptance and usage rates of such evidence.

Role of The Expert Witness Related to Biometric Evidence in Court

An expert witness is an individual who provides their opinion on any admissible evidence that requires expertise in a particular field. Expert evidence is any evidence presented in court outside the jury’s or judge’s knowledge and experience. Therefore, an expert witness must present and clarify the evidence, how it was obtained, and its importance to the case. Further, the expert witness is available to assist the court, must have relevant expertise, be impartial, and provide reliable evidence.

Accordingly, an expert witness must follow Rule 702 of the Daubert standard; the expert testimony must be based on sufficient data and facts, the evidence must be a product of reliable methods and principles, and finally, the witness must apply these principles and techniques to the facts of the case. Moreover, the responsibility of an expert witness is to help the court accomplish the principal objective by providing objective and unbiased opinions regarding matters within their knowledge. This duty is owed to the court and supersedes any responsibility to the party from whom the expert is getting instruction.

References

Benedict, N. (2004). Fingerprints and the Daubert standard for admission of scientific evidence: Why fingerprints fail and a proposed remedy. Ariz. L. Rev., 46, 519.

Garrett, B. L. (2018). The reliable application of fingerprint evidence. UCLA L. Rev. Discourse, 66, 64.

Jensen, P. J. (2002). Frye versus Daubert: Practically the same. Minn. L. Rev., 87, 1579.

Keane, A., & Mckeown, P. (2014). The modern law of evidence. Oxford University Press, USA.

Leslie, D. (2020). Understanding bias in facial recognition technologies. Leslie, D.(2020). Understanding bias in facial recognition technologies: an explainer. The Alan Turing Institute. https://doi.org/10.5281/zenodo, 4050457.

Roberts, P. (2015). Paradigms of forensic science and legal process: a critical diagnosis. Philosophical Transactions of the Royal Society B: Biological Sciences, 370(1674), 20140256.

Singh, S., & Prasad, S. V. A. V. (2018). Techniques and challenges of face recognition: A critical review. Procedia computer science, 143, 536-543.

Walton, R. H. (2006). Combined DNA Index System (CODIS). In Cold Case Homicides (pp. 392-399). CRC Press.

ORDER A PLAGIARISM-FREE PAPER HERE

We’ll write everything from scratch

Question

Biometrics and the Courts

In an 11-page paper, identify and analyze the benefits and challenges of biometric evidence in the criminal justice system. Include at least three techniques in your paper and use at least 2 case studies to support your position. Consider the following questions when drafting your paper:
• How do courts determine if evidence is reliable and valid before allowing it into testimony?
• What is the role of the Frye or Daubert standards in determining whether the courts will accept biometric evidence?
• What rules does your state use in this regard?
• How reliable is fingerprint evidence? Consider examples of its use in criminal courts.
• How do other biometrics compare to the reliability and validity of fingerprint evidence?
• What are some of the challenges associated with lower forms of biometrics, such as facial recognition and acceptance as evidence in court?
• What is the role of the expert witness related to biometric evidence in court?

Biometrics and the Courts