Editor’s note: This is part of an ongoing series of stories examining police practices and policies, including officer-involved shootings, police misconduct, the influence of the police union and police reform efforts.

One day in 2013, Stephen Dinnan ate lunch with his girlfriend and their children at the Ala Moana Center. On the way home to Kaneohe, they stopped in Waimanalo so his girlfriend could check on repairs being made to her truck. 

The Police Files BadgeHe had just smoked a joint with some friends when a Honolulu police officer appeared. Dinnan didn’t know it, but the officer was investigating the theft of a truck that had been stolen that day from Makapuu Beach Park. The owner, a Honolulu firefighter, and his son had tracked an iPhone the son left in the truck to Waimanalo.

According to a later lawsuit, Dinnan was scared because he had just smoked a joint. He also had methamphetamine in his fanny pack, a report from the Honolulu Prosecutor’s Office said. When the officer, Eric Matsumoto, shouted “Haole boy, don’t run,” Dinnan did just that. Matsumoto chased him.

The man who believed Dinnan had stolen his truck jumped into the fray, choking the 35-year-old. Dinnan broke free, but Matsumoto caught up and knocked him to the ground, handcuffing him behind his back. Matsumoto planted his knee on Dinnan’s back between his shoulder blades, the lawsuit alleged, and pushed his face into the ground. 

Stephen Dinnan, in this photograph shortly before his death, obtained by Hawaii News Now, was one of 34 people killed by Honolulu police since 2010. Hawaii News Now

A friend at the scene reported that Dinnan told him, “I love you, brah.” Other officers arrived and noticed that Dinnan’s skin had turned blue and he was no longer breathing. A day later, he was dead. The Honolulu Medical Examiner concluded that he had died of asphyxia from being pressed to the ground. A neck injury caused by the truck owner’s choking was found to be a contributing factor.

It turned out Dinnan had nothing to do with the stolen truck. The case against the city, Matsumoto and others settled in 2017 for $1 million. Matsumoto remained on the force until 2018.

Dinnan was one of 34 people killed by HPD officers since 2010, according to a database compiled by Civil Beat. An analysis of the data shows that, by several measures, Honolulu is near the middle of the pack in the number of such killings and shootings per capita.

But one difference stands out. In a far higher percentage of cases than for the U.S. as a whole, those killed in encounters with HPD officers died from physical restraint or asphyxiation, as Dinnan did, according to a comparison to database of police killings nationwide.

The numbers also show a much higher percentage of cases in which the subject was unarmed, like Dinnan.

There is no definitive, government-sanctioned compilation of police killings, which researchers and advocates of police reforms say is a problem in and of itself.

So Civil Beat compared its list to a database compiled by Mapping Police Violence, “a research collaborative collecting comprehensive data on police killings nationwide to quantify the impact of police violence in communities.” 

Cofounder Sam Sinyangwe — a data scientist, racial justice advocate and podcaster — said two analyses by outside groups have found that the database captures 95% or more of police killings. Unlike a database compiled by the Washington Post, the one by Mapping Police Violence includes deaths not caused by gunshot wounds, such as physical restraint. 

Since 2013, the annual number of people per million killed by HPD added up to 3.6, compared to an average of 4.2 for all U.S. big cities, as calculated by Mapping Police Violence. 

The Washington Post data, starting in 2015, also shows HPD near the average — 3.4 annual shooting deaths per million, compared to a rate for the U.S. as a whole of 2.9.

In yet another measurement of violent police encounters, HPD recently presented data to the Honolulu Police Commission on officer-involved shootings — including non-fatal incidents — in Honolulu and 10 other departments of similar size. An analysis of the raw numbers HPD shared with the commission shows HPD exactly in the middle. 

Each of the 34 Honolulu deaths since 2010 offers a unique story. 

Rene Velleses was swinging a machete in a shopping plaza when he was shot by police in 2010. After Aaron Torres called the police in 2012, paranoid and hallucinating, police pinned him to the ground in a struggle and he died. In 2013, police responded to a domestic disturbance and confronted Victor Rivera, diagnosed with schizophrenia, wielding a mango picker. He was shot nine times.

After a seven-hour standoff in 2018, during which he stabbed a police dog with an arrow, Steven Allan Kaluahinui Hyer was shot and killed by an HPD officer.

Several of the cases involved subjects who were acting erratically and showing signs of mental distress. Many were under the influence of meth, cocaine, alcohol or other drugs. Some deaths occurred after long standoffs. In other cases, police said the subject was coming at them in a car. People have died at the hands of the police in every corner of Oahu.

The subjects in these cases are disproportionately Pacific Islanders, much as Black people on the continental U.S. die at the hands of police in numbers above their representation in the population. HPD has argued that the disproportionate use-of-force numbers for Pacific Islanders and Native Hawaiians mirrors their higher numbers for arrests overall. But at least one police commissioner says that raises the deeper question of why arrest rates are disproportionate.

Because each deadly force case is unique, it can be hard to discern trends and patterns. 

But the comparison to Mapping Police Violence data shows at least three ways Honolulu appears to be different — the number of deaths from physical restraint or asphyxiation, the number of unarmed subjects and — not surprisingly, given Hawaii’s demographics — the race and ethnicity of those who were killed.

The number of deaths since 2010, only 34, means that just a few cases of physical restraint or unarmed suspects could skew the data.

But the disparities, compared to national figures as measured by Mapping Police Violence, are large. About 20.6% of fatal encounters with HPD were caused by physical restraint or asphyxiation, according to Civil Beat’s data, compared to just 0.5% nationwide.

And 26% of subjects in HPD fatal encounters were unarmed, compared to just 13.3% in the Mapping Police Violence data.

The Police Files

In a written response to questions, HPD said that, in four of the eight cases in which someone died from physical restraint, family members or first responders had restrained or tried to restrain the person before officers got there.

“Officers did not initiate the physical interaction, but they became involved after the initial struggle,” an HPD spokesperson said.

In seven of eight of these cases, HPD said, the subjects were found to have had one or more illegal drugs in their systems.

HPD declined a request for an interview about the significance of these observations or deadly force in general.

There is a natural connection between these physical restraint deaths and unarmed subjects, Sinyangwe said, since an officer is far less likely to try to use a hold, say, on an armed suspect.

An unusually high number of deaths from physical restraint suggests that minor incidents, such as traffic stops or mental health checks in which the subject is unarmed, may be escalating to violent confrontations, he said.

Many such deaths occur when subjects struggle and police put them face down on the ground or in a police cruiser, said Darrel Stephens, former police chief of the Charlotte-Mecklenburg Police Department in North Carolina and executive director of the Major Cities Chiefs Association.

“In a lot of cases, the subjects are under the influence of alcohol or other substances,” most often methamphetamine, Stephens said. “In a lot of cases, they end up having a heart attack.” 

Unlike when a subject dies from bullets, he said, it can be hard to sort out the cause of death. 

In at least three of the HPD cases, the medical examiner found that the subject showed signs of “excited delirium,” a term that refers to people who are distressed or aggressive as the result of mental illness or drug use. The diagnosis has been controversial, with some citing it as a real medical condition and others dismissing it as a nebulous diagnosis used to exonerate police violence.

“One thing we have seen is how coroners and law enforcement in some types of cases use excited delirium or the presence of drugs or alcohol to justify the actions of police,” Sinyangwe said. “Those explanations often don’t hold water.”

These kinds of cases — involving subjects who are unarmed or carry weapons other than firearms —  should be the focus of police department efforts to reduce deaths, said Chuck Wexler, executive director of the Police Executive Research Forum, an organization of law enforcement officials with the goal of improving professionalism in policing.

“When police are confronted with someone with a gun, they have very few options,” Wexler said.

But in about 40% of cases nationwide, the subject does not have a gun, even though many carry weapons such as knives. Many are experiencing a crisis of some kind. PERF has developed protocols to defuse these tense situations — slowing things down, creating distance between the subject and police and giving it time.

“Those kinds of situations lend themselves more to a new kind of thinking,” he said.

Wexler and Stephens both agreed that departments should prohibit shooting at moving cars in most cases. Not only that, Stephens said — police should be trained not to put themselves at risk when a suspect is in a car.

“Don’t stand in front of a car and expect them to stop,” he said. “Your first responsibility is to get the hell out of the way, because you can get them later.”

But police often have a hard time giving up on an arrest, he said.

HPD implemented a new use of force policy, effective April 1, limiting the circumstances in which officers were allowed to shoot at moving vehicles. 

Just four days later, however, the police shot and killed 16-year-old Iremamber Sykap, the driver of a car they had been pursuing. Whether the circumstances justified the shooting under HPD’s new policy is still unclear.

Former Police Chief Susan Ballard said Sykap rammed two police cars but the department has released few other details. Body camera footage obtained by Hawaii News Now, meanwhile, shows an officer shooting at the car 10 times as it appeared to be standing still before it lurched through a fence and into a canal.

Iremamber Sykap, 16, was killed on April 5 in a police shooting in Honolulu. Courtesy: Sykap family

David Johnson, a sociology professor at the University of Hawaii Manoa, said it’s encouraging that HPD revised its use-of-force policy, including not only the provision on shooting into cars, but another limiting vascular neck restraints to cases in which deadly force is warranted and adding a new section on de-escalation techniques.

“I hasten to add that there’s often a big difference between what a policy seems to say on paper and how faithfully that policy is implemented in practice,” he said. “That’s where the challenge and the rub is going to be.”

Johnson said that the success of policy reforms depends on leaders convincing rank-and-file officers to change their ways.

“Without leadership,” he said, “nothing will change.”

In fact, Johnson said, HPD leadership has in recent years inadvertently revealed that its mindset remains mired in the status quo. He pointed to a 2020 Police Commission meeting in which Ballard downplayed the number of police killings in Honolulu, saying that “for any other large major police department, that’s nothing.” She also said she hoped police reforms sweeping the nation would bypass Hawaii because, “We’re kind of doing OK over here,” she said.

“With all due respect,” Johnson said, “that’s nonsense.”

He said he hoped the replacement for Ballard, whose resignation became effective today, would recognize that there are too many police killings in Honolulu. “Many of them are avoidable,” he said.

Assistant Chief Rade Vanic has been named acting chief by the Honolulu Police Commission.

Stephens said the deadly force cases should be judged by the officers’ experience of the incident rather than facts that emerge later. A weapon may turn out to be a replica gun, for instance, “but in the police officers’ mind, it’s a gun.” 

But it’s crucial for police departments to make strong statements about the sanctity of life as a guiding principle for their policies on using deadly force, Stephens and others said. 

HPD’s revised use-of-force policy begins by saying that HPD “is committed to ensuring the public’s safety, as well as preserving the lives and protecting the rights of all individuals without prejudice to anyone.” The new policy “is consistent with the national standards and recommendations of the Major Cities Chiefs Association,” HPD said.

In the end, Stephens said, these policies boil down to “having the expectation that they will do everything they can without having to use deadly force, or force of any kind.”

Civil Beat compiled a database of the past 10 years of officer-involved fatalities:

Before you go

Civil Beat is a small nonprofit newsroom that provides free content with no paywall. That means readership growth alone can’t sustain our journalism.

The truth is that less than 1% of our monthly readers are financial supporters. To remain a viable business model for local news, we need a higher percentage of readers-turned-donors.

Will you consider becoming a new donor today? 

About the Author