On July 16, 2009, Harvard University Professor Henry Louis Gates Jr. was arrested by Sgt. James Crowley of the Cambridge, Ma., police for “exhibiting loud and tumultuous behavior in a public place.”
The “public place” was Gates’s own front porch.
Gates had just returned home from a trip abroad and found his front door jammed shut. With the help of his driver, he shoved the door open. Moments later, Sgt. Crowley was sent to the address in response to a 911 call about a possible in-progress breaking and entering.
After a tense verbal encounter, Crowley arrested one of the foremost African-American scholars in the United States and, in so doing, reignited national outrage over the role of race in policing.
A study I published recently offers a fresh perspective on the Gates incident by applying a wider frame to examining police behavior, one that includes the actions of 911 call-takers and dispatchers. By analyzing the 911 call and radio transmission, I show how a cautious and uncertain call escalated into a high-priority dispatch that resulted in Gates’s arrest.
The call began with the 911 caller saying that she did not know what was happening. Instead, she provided the call-taker with a list of seemingly factual observations: An older woman was standing outside the address; that woman (not herself) noticed two gentlemen trying to get in a house; they had suitcases. She cautiously concluded her account by saying that the men might live at the address. Later in the call, she even suggested that the men might be having trouble with the key.
Despite the caller’s own attempt to tone down her report, the call-taker and dispatcher made several escalating decisions.
First, the call-taker pressed the caller to consider the incident in criminal terms by asking multiple times whether she thought the men were breaking in.
Second, the call-taker began referring to the men as having “broke into” the house despite the caller’s refusal to agree with that characterization.
Third, the dispatcher made no mention of the caller’s proposition that the men were simply locked out, nor the caller’s vast uncertainties. Instead, the dispatcher’s message to police was:
Respond to Seventeen Ware Street for a possible B and E in progress. Two SPs (suspects) barged their way into the home. They have suitcases.
Unpacking the decisions of the 911 call-taker and dispatcher is an important exercise because their actions can affect policing in the field.
In Cambridge, an in-progress breaking and entering is a serious, high-priority call accompanied by a lights-and-sirens response. So serious that Sgt. Crowley drove the wrong way down a one-way street to respond as quickly as possible. The information from dispatch primed Crowley to interpret the incident as a serious crime and no doubt contributed to his aggressive demeanor.
Research on anchoring bias suggests that if police are primed for a high-priority encounter, then they will be more likely to perceive of the incident in those terms upon arrival. Early work by police journalist Jonathan Rubinstein found that police responses are shaped by information from the dispatcher.
More recently, Paul Taylor has shown in a firearm training simulator that officers are ten times less likely to shoot a suspect if they are accurately told by dispatch that the suspect has a cell phone, rather than inaccurately told that the suspect has a gun.
My own dissertation research has revealed that 911 call-takers vary in how they prioritize the same types of calls, and police are more likely to make an arrest when responding to a call handled by a more “alarmist” call-taker.
Although evidence suggests that information from dispatch shapes officer action, the official attempt to figure out what went wrong in the Gates incident ignored the role of the 911 call-taker and dispatcher.
Indeed, reviewers were so laser-focused on the moment of interaction between Crowley and Gates that they overlooked other key actors’ decisions that preceded the police interaction. The Cambridge Review Committee—a group of academics, law enforcement leaders, and lawyers tasked with reviewing the event—concluded that “Sergeant Crowley and Professor Gates each missed opportunities to ‘ratchet down’ the situation and end it peacefully.”
I would add that so too did the 911 call-taker and dispatcher.
What lessons can we learn from this decade old incident to help inform the current police reform debate?
One, this analysis underscores the importance of including the actions of call-takers and dispatchers in incident reviews. Interactions that take place before the police arrive at a scene matter. Including the 911 call is not enough. Reviewers should interrogate how 911 call-takers extract, interpret and classify caller information; and how dispatchers relay information to the police.
Two, the Gates incident illuminates a critical 911 call-taking function—risk appraisal—and the challenges associated with it. Call-takers must transform amorphous caller requests into more manageable categories for the police. These categories serve as proxies for risk. The call-taker in this incident faced hard choices—classify the call as a higher priority “breaking and entering,” or a lower priority “suspicious,” or even a “lock out.”
Based on my experiences as a former 911 call-taker, albeit at a different center, when callers are uncertain and their reports straddle categories, internal rules and norms encourage call-takers to upgrade incidents.
Similar to how police have been found to engage in worst-case scenario thinking when faced with uncertainty, I write in my article that “Call-takers are often trained that ‘being wrong’ by sending an over-response is far better than ‘being disastrously wrong’ by sending an under-response.”
The Gates incident indicates that the risks associated with an over-response can be equally, if not more, consequential than an under-response.
Three, reforms aimed at training call-takers to de-escalate situations, such as by following standardized question protocols when extracting information, considering the risks from over-response when classifying information, and preserving callers’ uncertainties in their dispatch requests, may improve future interactions between the police and public.
These lessons are critical to apply to the current moment in policing.
Much of the nation is demanding massive police reform in response to egregious cases of police brutality. Political leaders and police officials in places such as Denver, Albuquerque, and San Francisco are considering sending alternative responders to low-level incidents to reduce troubling interactions between the police and public.
Yet even these alternative responder initiatives implicitly assume that 911 call-takers can distinguish between high-risk incidents that will require police attention and low-risk incidents that will not. The Gates incident highlights just how difficult it can be for call-takers to make judgements about risk.
In addition to the reform ideas mentioned above, disaggregating broad call categories—such as “check welfare,” “suspicious,” “emotionally disturbed person,” and so on—into more nuanced ones that better capture whether or not a police presence is needed will be a critical to the success of diversion efforts.
“How the 911 Calls System Turns Police into ‘Proxies’ for Racial Bias” by Rebecca Neusteter, The Crime Report, Jan 22, 2020
“Why I’ll Never Dial 911″ by Abraham Gutman, The Crime Report, Feb. 6, 2018
Jessica W. Gillooly, Ph.D., is a postdoctoral Policing Project fellow at the New York University School of Law