Revelations that a new type of junk science known as 911 call analysis has infiltrated the justice system have triggered calls by prosecutors, judges and defense attorneys nationwide to ban the use of the technique, review past convictions in which it was used and exact sanctions against prosecutors who snuck it into court despite knowing it was inadmissible.
The actions follow a two-part ProPublica investigation published last year that many judges and other court officers said took them by surprise. “I never anticipated that prosecutors — officers of the court — would engage in systematic organized frauds,” a judge in Ohio wrote in an email to ProPublica. She said she had alerted fellow judges to be on the lookout for 911 call analysis, which ProPublica found to be pervasive throughout the justice system: “I’m sure that some will care and share my outrage that innocent people are going to prison.”
The technique’s chief architect, Tracy Harpster, developed a program to spread his methods and says police and prosecutors who take his training will learn how to identify guilt and deception from the word choice, cadence and grammar of those calling 911. So far, researchers who have tried to corroborate Harpster’s claims have failed.
Last year, ProPublica documented more than 100 cases in 26 states where law enforcement has employed his methods. Those responsible for ensuring honest police work and fair trials — including the FBI — have instead helped 911 call analysis metastasize. The investigation revealed that some prosecutors knew 911 call analysis would not be recognized as scientific evidence but still disguised it in trial against unwitting defendants anyway.
During the reporting for the two stories, Harpster at first defended his program but then did not respond to repeated interview requests or detailed lists of questions. Supporters of his work in law enforcement have said 911 call analysis is a valuable investigative tool but not decisive evidence on which to base a conviction.
On Wednesday, he and Susan Adams, who co-authored the original study the technique is based on, sent a letter to ProPublica and argued its coverage had “presented an inaccurate narrative” and listed material they claimed to be omissions and misrepresentations. They asked that their letter be published. ProPublica is also publishing a point-by-point response.
Last October, ProPublica reported on the case of Jessica Logan, a young mother convicted of killing her baby after a detective trained by Harpster testified about his analysis of Logan’s 911 call. Shortly after the story was published, the Supreme Court of Illinois agreed to take another look at Logan’s case.
In addition, attorneys from the Exoneration Project and the Center for Integrity in Forensic Sciences offered to represent her. One of her new lawyers is Josh Tepfer, who was recently profiled by BuzzFeed News for helping exonerate 288 wrongfully convicted people, “making him among the most prolific exoneration attorneys since anyone began keeping track.”
Hope Bradford, who is a mother figure to Logan, said she’s encouraged by the recent developments but is reserving optimism. “I never thought it would even get this far,” she said. “I’m just waiting for her to come home — that’s all.”
Faulty scientific disciplines, including misleading testimony, are the second most common factor in wrongful convictions, according to the Innocence Project. “The criminal legal system can add 911 call analysis to the junk pile of fake scientific theories contributing to this statistic,” said Nellie King, president of the National Association of Criminal Defense Lawyers. She called the business of 911 call analysis “dangerous and insidious.”
“Most bothersome is the fact that justice agencies — police officers and prosecutors alike — are buying what he is selling,” King added. “Harpster’s impact on case outcomes is devastating to those falsely accused based on his claims.”
North Carolina’s Office of Indigent Defense Services issued a warning to attorneys across the state to be on the lookout for appearances of 911 call analysis. Fair and Just Prosecution, a network of elected prosecutors, called on its members to review past cases, as well: “Prosecutors must guard against these practices and correct the past injustices they’ve caused through post-conviction review processes.”
At a recent summit of prosecutors in Austin, Texas, Miriam Krinsky, the group’s president, presented ProPublica’s reporting to warn new district attorneys about 911 call analysis. In an email, Krinsky said her organization, in partnership with the Innocence Project, “is concerned about this issue and looking at where and how elected prosecutors can engage to guard against these and other practices that undermine the integrity of convictions.”
In recent weeks, dozens of readers, including defense attorneys and prosecutors, have reached out to ProPublica to inquire about the other jurisdictions where 911 call analysis has surfaced.
Reporters canvassed a sample of about 50 departments and training associations nationwide where records show Harpster’s methods appear to have surfaced over the past 10 years. These agencies, which range from Minnesota Bureau of Criminal Apprehension to the Tennessee Bureau of Investigation, have either hosted him for seminars, sent officers to attend, used his methods in actual cases or did a combination of all three.
Eleven of the 50 agencies responded, and the rest did not answer questions about whether or not they’d continue to support Harpster’s work. Those that did reply either distanced themselves from the program or minimized its role in past cases.
“We’ve never been of the opinion that 911 call analysis should be anything more than a potential investigative lead,” Susan Medina, chief of staff with the Colorado Bureau of Investigation, said in an email, adding that the department “would never advocate for 911 call analysis to be a deciding factor of an arrest or conviction.”
Some of those agencies that responded to ProPublica’s survey didn’t know about their past involvement with the program.
For example, Melaney Arnold, a public information officer with the Illinois State Police, said that department leaders were unaware of 911 call analysis or anyone who’s been trained in it. ProPublica then sent her emails and attendance lists documenting personnel who have attended the program or consulted with Harpster.
“Let me circle back with staff,” Arnold replied in an email. “They may have only been looking at 911 call analysis and not the concept in general.”
Two notable agencies that did not respond to questions are the Westchester County, New York, and Orange County, California, district attorneys offices.
A prosecutor in Westchester once wrote to Harpster to thank him for his consultation, which, he said, “proved to be an invaluable aid in understanding the defendant’s 911 call and greatly assisted in the successful prosecution.”
In Orange County, the district attorney charged a woman with murdering her boyfriend 26 years ago. The arrest came last spring, after prosecutors and detectives consulted with Harpster. “It significantly helped our district attorney to realize the indicators of guilt in the phone calls,” the lead detective told Harpster in an email, “as well as suggestions on how to introduce the 911 calls to the jury during trial.”
Editor's note: Following the publication of this article by Propublica last March, Deputy Chief Tracy Harpster (ret.) wrote the following letter to the editor regarding some of the article's findings.
I am one of the subjects of the ProPublica articles on 911 Call Analysis and I do not believe that these articles represent journalistic excellence and truth. Mr. Murphy omitted critical facts and mischaracterized the research. The premise of Mr. Murphy’s series of articles is that 911 Call Analysis has been used to convict innocent people. In his December, 2022 article he stated:
“In reality, people have been wrongfully accused and convicted of murder after
someone misinterpreted their call for help.
I wrote to ProPublica, challenged their premise, and requested that they correct the record. I explained the following:
“The articles attempt to give the false impression that 911 call analysis caused
the conviction of innocent persons.”
ProPublica publicly responded with the following contraction:
“We never reported that 911 call analysis caused the conviction of innocent persons.”
Mr. Murphy claimed to have conducted an exhaustive review of our research. However, he failed to cite even one case where an innocent person was convicted with the assistance of 911 Call Analysis. I have researched the topic for almost two decades and examined over 2,500 equivocal death cases. I am unaware of any case in which an innocent person was convicted using this analysis. If fact, the analysis has been used to exonerate innocent persons.
Mr. Murphy cited six case examples in an effort to discredit the use of 911 call analysis, leading the reader to believe that the research has contributed to the conviction of innocent persons. One of these cases is ongoing and under seal. In all remaining cases cited, investigators or prosecutors working the cases have refuted the ProPublica claims. The following paragraphs regarding the Illinois and Colorado cases reveal how omissions were used to mislead readers.
In the Illinois case, which was specifically mentioned in the Polk Awards announcement, Jessica Logan was convicted of killing her 19-month-old son. She appealed the case to the Illinois Appellate Court and that court unanimously upheld the conviction and concluded that there was “overwhelming” evidence that she knowingly asphyxiated her son. In his effort to discredit 911 Call Analysis, Mr. Murphy’s articles neglected to share the following information:
1. Ms. Logan and the father of her son had been jointly convicted of Aggravated Assault and Drug Trafficking. While the father was still in prison, their phone conversations were recorded. Twenty-five days before her son’s death Ms. Logan told the child’s father that she “literally broke down” due to her unpaid bills. Eighteen days before her son’s death Ms. Logan searched the internet for information on a life insurance policy she had purchased for her 19-month-old son. Five days before her son’s death Ms. Logan complained to the father that due to their debts “we haven’t had TV in over four 4 days.” She added that she told her child that if she didn’t get to watch TV the child couldn’t get to play with his toy, so she threw the toy away. Despite their dire financial difficulties, the father instructed Ms. Logan that the only bill she needed to keep paying was the life insurance premium on their son.
2. One day before her son was killed, Ms. Logan Google searched for “How do you suffocate?” The coroner’s report indicated that the cause of death was “asphyxia due to smothering and compression of the neck.”
3. Ms. Logan told detectives that on the date of the son’s death, she got up at 3:00 a.m. to give medicine to her child. There was no evidence that she possessed the medicine at that time.
4. Ms. Logan claimed that her son’s doctor prescribed medicine for the child. However, the doctor advised that she hadn’t prescribed medicine for the child in nine months.
5. There was a seventeen-minute delay from the time Ms. Logan got up at 3:00 a.m. and the time she called her son’s grandmother. The grandmother directed Ms. Logan to call 911 and Ms. Logan finally made the call. She told the police officers that she did not immediately call 911 because the victim was “really stiff,” and she did not attempt CPR because “he was cold and hard.” Ms. Logan stated that she put her son to bed at approximately 8:00 p.m., seven hours before she called 911. According to homicide experts, it takes from 6-10 hours for a young child (indoors) to be in full rigor. When asked what she did between the time she woke up at 3:00 a.m. and the time she called her mother-in-law at 3:17 a.m. Ms. Logan replied “nothing.”
6. Ms. Logan told the 911 dispatcher that she was complying with the dispatcher’s CPR instructions while on the phone. She later confessed to detectives that she lied about giving aid to her son.
The Appellate Court’s full opinion can be accessed at the following link:
https://ilcourtsaudio.blob.core.windows.net/antilles-resources/resources/3ac6534e-bd87-4768-8cae-febe5e174d72/People%20v.%20Logan,%202022%20IL%20App%20(4th)%20210492.pdf
In the Colorado case cited in Mr. Murphy’s articles, Miriam Helmick, wife of a wealthy businessman, called 911 after she allegedly found her husband shot to death in her home. She was convicted of the murder and the Appellate Court upheld her conviction. Mr. Murphy reported that 911 call analysis was used as an investigative tool in the investigation, but not presented during the trial. However, Mr. Murphy failed to report the following facts presented in court that ultimately led to the conviction of the defendant:
1. Ms. Helmick depleted her husband’s bank accounts and tried to take out a life insurance policy before his murder. After the insurance agent advised that her husband would be notified of the policy, she withdrew her application.
2. The husband’s adult son reported that his father had been uncharacteristically ill, frail and bedridden after his marriage to his wife. Prior to the murder, Ms. Helmick had searched the internet regarding her husband’s prescriptions and what amounts could cause an overdose.
3. The husband had been shot in the back of the head and killed. Prosecutors presented evidence that the wife had previously tried to kill her husband by blowing up his car.
4. Ms. Helmick told the investigators that her husband’s death was the result of “a robbery gone bad.” However, nothing was taken from the home.
5. After her husband’s murder, Ms. Helmick wrote an anonymous message indicating that she, too, would be killed. She later admitted to falsifying that message in an attempt to mislead investigators.
6. The Appellate Court noted that when Ms. Helmick became a suspect in her husband’s murder, she moved to Florida under an assumed name. While there, she frequented a dating site designed to match millionaires with romantic partners.
7. Ms. Helmick had been previously married and that husband also died by a gunshot to the head.
It should be noted that all of the above information was readily available to Mr. Murphy in the arrest reports, case files and court documents. Had he truly conducted a thorough and unbiased investigation into these cases he would have known and reported these facts. If he knew these facts and did not report them in his articles it speaks to his intent to mislead the reader.
Mr. Murphy alleged that he had “overwhelming evidence that the concept is baseless junk science.” However, he was unable to present evidence to support that assertion. Instead, he cited a few recent studies that were not directly applicable. Their use to refute the 911 Call Analysis research is misleading, as illustrated by the following examples:
1. In one study, untrained undergraduate students with a mean age of 19 years judged 911 calls for course credit. The undergraduate students examined different variables than the Harpster and Adams model, which was created for use by experienced law enforcement professionals after receiving training.
2. Another study focused solely on missing children cases; the Harpster and Adams research studied equivocal deaths in cases with a victim at the scene. 911 calls without the presence of a victim are very different from those with a victim.
3. One of the referenced studies examined suicides staged as homicides and actual suicides, which was not the focus of the Harpster and Adams research.
4. In one of the above studies, the audio recordings of 911 calls were examined without the corresponding transcripts. In another referenced study, 911 call transcripts were analyzed without audio recordings. The Harpster and Adams study examined both the audio recordings and the transcripts of every call.
Although the cited studies differed from the Harpster and Adams research, all studies supported the value of examining 911 calls as an investigative tool.
The ProPublica 911 Call Analysis articles attempted to deceive readers by wrongly reporting that Harpster “tries to keep his methods secret.” However, ProPublica failed to inform the reader that we published our findings on 911 call analysis in articles and books in the years 2006, 2008, 2009, 2010, 2015, 2016 and 2019. Additionally, local and national media have featured the study on their programs.
Respectfully,
Deputy Chief Tracy Harpster (ret.)
Related Articles
