Tag Archives: FBI

Cellphone, signals, wifi

Cellphone Experts Debate Validity of Location Evidence

In criminal cases across the country, FBI agents and local police detectives have testified as expert witnesses concerning the location of a suspect’s cellphone (and presumably the suspect) at the time a crime was committed. Whether law enforcement officers are qualified to give that testimony is a hotly contested question.

The Connecticut Supreme Court is the latest jurisdiction to consider whether cellphone location evidence is sufficiently reliable to be admitted at a criminal trial. The court’s decision will contribute to a continuing debate about the validity of the expert evidence, particularly when the expert witnesses are police officers with no background in engineering or telecommunications.

Eugene Edwards’ Appeal

Eugene Edwards Jr. was convicted of robbing Lieslotte Worysz, age 83, in Wethersfield, Connecticut. The prosecution argued that Edwards followed the woman home from the grocery store, entered her garage as she exited from her car, put his hand on her car door while brandishing a gun, stole her money and jewelry, and fled. The jury found Edwards guilty.

Edwards was also charged with committing a similar robbery of an elderly woman in Berlin, Connecticut three weeks earlier. The jury found Edwards not guilty of that crime.

Worysz described the robber as a black man between the ages of 35 and 40. She testified that the robber was holding a black gun. When the police showed her pictures of potential suspects, she said they all looked alike to her.

Worysz noticed a car behind her as she drove home. A surveillance video at the grocery store captured the car that Worysz described. Edwards was one of three people who, according to records of people with prior police contacts, owned a similar car. The police did not inspect the other cars. The car on the video had a broken brake light, but the brake lights on Edwards’ car were working.

Worysz said that the robber held her wallet and a garage door opener. A DNA analysis eliminated Edwards as a possible source of DNA found on the wallet and opener.

A fingerprint analyst could not match a partial print that was taken from Worysz’ car to Edwards, although a different analyst testified that the partial print came from Edwards’ left hand. Worysz testified that Edwards was holding the gun in his left hand. The defense argued that Edwards could not have opened the car door using a hand that was holding a gun. In any event, Connecticut law does not permit a conviction to be based on fingerprint evidence alone when the print was taken from a car that has been parked in a public place.

The police searched Edwards’ home and found a black BB gun with an orange tip. Worysz did not describe the gun she saw as having an orange tip. No jewelry or other crime proceeds were found in a search of Edwards’ car and residence about a week after the robbery.

Expert Testimony

Detective Christopher Morris testified as a cellphone expert based on his attendance at a three-day police training session in cellphone investigations. He opined that Edwards’ cellphone connected to towers near Worysz’ home and near the store where Worysz was shopping on the day of the robbery. He also testified that cellphones connect to the nearest tower, although he acknowledged that facts such as weather, network congestion, network maintenance, topology and foliage can all affect the cellphone’s connection to a particular tower.

The prosecution called a Verizon employee to authenticate the records upon which Morris relied, but the trial court ruled that the employee was not an engineer and was not qualified to offer any opinion as to whether Edwards’ phone would have connected to the nearest tower. However, the court held that Morris was qualified as a “police expert” to testify about cellphone locations based on his training and experience.

Morris gave no testimony about peer-reviewed studies that validated his methodology. He gave no testimony about error rates that occur when his methodology is used to determine cellphone locations. The prosecution offered no evidence that Morris was trained by scientists or engineers or that his training was based on accepted scientific principles. The Connecticut Supreme Court agreed to decide whether a police witness who is not an engineer and has no scientific training other than “police training” is qualified to give reliable testimony about cellphone locations.

Controversial Evidence

Connecticut joins a growing number of jurisdictions that are tackling the use of law enforcement officers to testify as expert witnesses regarding cellphone tracking. Some independent experts, like Michael Cherry, say that their testimony is based on “junk science.”

Cherry notes that police officers “tend to confuse the location of the cellphone with the location of the cell tower.” According to Cherry:

People like to say that the phone goes to the nearest tower. It goes to the clearest tower within range, not always the closest tower. You could be sitting on your living room couch and you could make four phone calls and each call would use a different tower.

Getting that point across to jurors may require defense attorneys to retain the services of independent experts. The initial question confronting courts, however, is whether police officers who have no background in science are able to give reliable testimony about cellphone locations based on brief training sessions that are taught by other police officers who have no background in science.

Judicial Decisions

As ExpertPages has noted in the past, different trial courts have reached different conclusions about the admissibility of expert testimony concerning cellphone locations. In Maryland, Adnan Syed was recently granted a new trial because his lawyer failed to cross-examine a state’s expert witness about the reliability of cell tower data that allegedly placed him near the site of the victim’s burial.

Lisa Marie Roberts, convicted of murder in Oregon, was also granted a new trial based on her attorney’s failure to challenge questionable expert cellphone location evidence. A judge in that case ruled that there were serious doubts about Roberts’ guilt.

A federal judge in Chicago refused to allow an FBI agent to testify about a cellphone location in a kidnapping case. The court concluded that the agent’s “chosen methodology has received no scrutiny outside the law enforcement community” and that the reliability of that methodology had therefore not been established.

Other courts have routinely admitted cellphone tracking evidence, perhaps because defense lawyers do not realize that they have grounds for challenging its admissibility. Evidence professor Edward J. Imwinkelried encourages lawyers to use their own experts to challenge law enforcement agents who claim that they can determine the location of a cellphone using cellphone tower records.

Criminal Forensics, word cloud concept 11

Expert Scientific Report Issues Scathing Review of Common Forensic Techniques

Recently, the President’s Council of Advisors on Science and Technology (PCAST) released an expert witness report  which sharply criticized a number of widely accepted forensic methods used in criminal trials. As expert witnesses become ubiquitously integrated into the American criminal justice culture, many forensic science techniques have become commonplace in trials, and the PCAST report represents an effort to trim so-called junk science from the courtroom.

Increased Scrutiny Raises Questions about Forensic Science Techniques

After experiencing a heyday of expansion in the 1990s, expert forensic science testimony used during criminal trials has seen a rise in scrutiny from the scientific community from the early-2000s through the present. Commonly used forensic analysis procedures such as bite-mark analysis, footwear examination, and microscopic hair comparisons have been widely criticized by scientific experts, and even widely accepted techniques such as DNA analysis and latent fingerprint identification have undergone rigorous scientific study in order to determine when and how such procedures are effective and valid. Among the number of studies conducted which analyze the validity of forensic evidence, a handful stand out:

  • In 2002, the FBI found that 11{d61575bddc780c1d4ab39ab904bf25755f3b8d1434703a303cf443ba00f43fa4} of its microscopic hair comparisons incorrectly matched hair samples, and in 2015 a report by the FBI and the Innocence Project found that hair matching analysis was incorrect a staggering 90{d61575bddc780c1d4ab39ab904bf25755f3b8d1434703a303cf443ba00f43fa4} of the time;
  • In 2004, a report commissioned by the FBI found that there was insufficient research and data to match bullets based on lead composition;
  • In 2005, a review of latent fingerprint testing used in the investigation of an alleged terrorist in Spain found that police bias allowed for the fingerprint evidence to be misinterpreted, and several studies throughout the mid to late 2000s have identified causes of error in fingerprinting which have refined the process;
  • Several studies since 2009 have found bite mark evidence to be unreliable in identifying defendants; and
  • Multiple studies and reports by the FBI and the National Institute of Justice have found that DNA testing has invalidated jail sentences for several defendants who were convicted on the strength of faulty bite mark, hair comparison, shoe tread comparison, and other common forensic methods.

Since the turn of the century, a cottage industry has arisen for scientists and legal organizations dedicated to re-evaluating convictions based on shaky forensic expert witness testimony, leading to several hundred exonerations across the United States. With widely accepted forensic methods facing questions from the scientific community, President Obama commissioned PCAST in 2015 to write a report which critically examined the state of forensic methods used in criminal trials.

PCAST Report Sharply Criticizes Common Forensic Methods

Last week’s PCAST report , which was written by high profile scientists who conducted thorough reviews of forensic techniques and methodology, issued a critical review of several types of evidence which have become common in criminal trials. Building on prior work and conducting further empirical research, the PCAST report critically examined 7 forensic methods: single source DNA testing, complex multi-source DNA testing, latent fingerprinting, bite mark, firearm analysis (which connects a bullet to a gun based on unique features of the weapon), footwear analysis, and hair matching analysis.

The report found that single source DNA testing is a highly valid forensic method and both multi-source DNA and latent fingerprint analysis have made strides in refining the process and there are reliable experts available to testify to each method in criminal trials. The scientific experts who generated the PCAST report concluded that bite mark identification, firearm analysis, footwear tread identification, and hair matching analysis are currently unreliable given the lack of empirical research supporting each field. The PCAST suggested opportunities for improvement, particularly in firearm analysis, but suggested that each of these forensic techniques must undergo more rigorous testing before being universally relied on in criminal trials.

Implications of the PCAST Forensic Expert Report

Since 1993’s Daubert v Merrell Dow decision, judges throughout the American legal system have been tasked with rigorous evaluation of expert witness testimony and are only allowed to admit experts who provide reliable testimony that is the product of sound theory and scientific methodology. Judges can consider many factors, but in reality asking the judiciary to critically evaluate scientific work imposes a duty beyond a judge’s qualifications. As a result, forensic methods based on junk science can gain widespread popularity if a collection of scientists convince a judge that the theory and methodology are sound enough to pass legal muster.

The PCAST report, which is written with an eye towards the legal standard for admitting forensic expert witness evidence, provides judges with the tools necessary to more closely evaluate many widely used forensic techniques. While the future use of these common forensic techniques in courtrooms is unlikely to change overnight, the long term effect of the PCAST work will likely see a decrease in reliance on unreliable forensic expert testimony. Regardless of its immediate effect, the PCAST report represents a change in how the scientific and legal community will interact in criminal investigation and prosecution as more rigorous scientific analysis influences when and how expert witnesses can testify.