How we can make racism a solvable problem -- and improve policing

When we define racism as behaviors instead of feelings, we can measure it -- and transform it from an impossible problem into a solvable one, says justice scientist Dr. Phillip Atiba Goff. In an actionable talk, he shares his work at the Center for Policing Equity, an organization that helps police departments diagnose and track racial gaps in policing in order to eliminate them. Learn more about their data-driven approach -- and how you can get involved with the work that still needs to be done. (This ambitious plan is part of the Audacious Project, TED's initiative to inspire and fund global change.)

When people meet me
for the first time on my job,

they often feel inspired to share
a revelation they've had about me,

and it kind of goes something like this.

"Hey, I know why police chiefs

like to share their deep,
dark secrets with you.

Phil, with your PhD in psychology,

and your shiny bald head,

you're basically
the Black Dr. Phil, right?"

(Laughter)

And for each and every person
who's ever said that to me

I do want to say thank you

because that was the first time
I ever heard that joke.

(Laughter)

But for everybody else,
I really hope you'll believe me

when I tell you no police chief
likes talking to me

because they think
I'm a clinical psychologist.

And also I'm not.

I have no idea what your mother
did to you, and I can't help.

(Laughter)

Police chiefs like talking to me

because I'm an expert on a problem
that feels impossible for them to solve:

racism in their profession.

Now my expertise
comes from being a scientist

who studies how our minds learn
to associate Blackness and crime

and misperceive Black children
as older than they actually are.

It also comes from studying
actual police behavior,

which is how I know that every year,

about one in five adults
in the United States

has contact with law enforcement.

Out of those, about a million
are targeted for police use of force.

And if you're Black,

you're two to four times more likely
to be targeted for that force

than if you're white.

But it also comes from knowing
what those statistics feel like.

I've experienced the fear
of seeing an officer unclip their gun

and the panic of realizing that someone
might mistake my 13-year-old godson

as old enough to be a threat.

So when a police chief,

or a pastor,

or an imam, or a mother --

when they call me after an officer
shoots another unarmed Black child,

I understand a bit
of the pain in their voice.

It's the pain of a heart breaking
when it fails to solve a deadly problem.

Breaking from trying to do something

that feels simultaneously
necessary and impossible.

The way trying to fix
racism usually feels.

Necessary and impossible.

So, police chiefs like talking to me
because I'm an expert,

but I doubt they'd be lining up
to lie down on Dr. Phil's couch

if I told them all their
problems were hopeless.

All of my research,

and the decade of work
I've done with my center --

the Center for Policing Equity --

actually leads me to a hopeful conclusion

amidst all the heartbreak
of race in America,

which is this:

trying to solve racism feels impossible

because our definition of racism
makes it impossible --

but it doesn't have to be that way.

So here's what I mean.

The most common definition of racism

is that racist behaviors are the product
of contaminated hearts and minds.

When you listen to the way we talk
about trying to cure racism,

you'll hear it.

"We need to stamp out hatred.

We need to combat ignorance," right?

It's hearts and minds.

Now the only problem with that definition
is that it's completely wrong --

both scientifically and otherwise.

One of the foundational insights
of social psychology

is that attitudes are
very weak predictors of behaviors,

but more importantly than that,

no Black community
has ever taken to the streets

to demand that white people
would love us more.

Communities march to stop the killing,

because racism
is about behaviors, not feelings.

And even when civil rights leaders

like King and Fannie Lou Hamer
used the language of love,

the racism they fought,

that was segregation and brutality.

It's actions over feelings.

And every one of
those leaders would agree,

if a definition of racism
makes it harder to see

the injuries racism causes,

that's not just wrong.

A definition that cares
about the intentions of abusers

more than the harms to the abused --

that definition of racism is racist.

But when we change the definition
of racism from attitudes to behaviors,

we transform that problem
from impossible to solvable.

Because you can measure behaviors.

And when you can measure a problem,

you can tap into one of the only
universal rules of organizational success.

You've got a problem or a goal,
you measure it,

you hold yourself accountable
to that metric.

So if every other organization
measures success this way,

why can't we do that in policing?

It turns out we actually already do.

Police departments already practice
data-driven accountability,

it's just for crime.

The vast majority of police departments
across the United States

use a system called CompStat.

It's a process that,
when you use it right,

it identifies crime data,

it tracks it and identifies patterns,

and then it allows departments
to hold themselves accountable

to public safety goals.

It usually works either by directing
police attentions and police resources,

or changing police behavior
once they show up.

So if I see a string of muggings
in that neighborhood,

I'm going to want to increase
patrols in that neighborhood.

If I see a spike in homicides,

I'm going to want to talk
to the community to find out why

and collaborate on changes on police
behavior to tamp down the violence.

Now when you define racism
in terms of measurable behaviors,

you can do the same thing.

You can create a CompStat for justice.

That's exactly what the Center
for Policing Equity has been doing.

So let me tell you how that works.

After a police department invites us in,

we handle the legal stuff,
we engage with the community,

our next step is to analyze their data.

The goal of these analyses is to determine

how much do crime, poverty,
neighborhood demographics

predict, let's say, police use of force?

Let's say that those factors predict

police will use force
on this many Black people.

There?

So our next question is,

how many Black people
actually are targeted

for police use of force?

Let's say it's this many.

So what's up with the gap?

Well, a big portion of the gap
is the difference

between what's predicted
by things police can't control

and what's predicted
by things police can control --

their policies and their behaviors.

And what we're looking for
are the types of contact

or the areas in the city

where that gap is biggest,

because then we can tell our partners,

"Look here. Solve this problem first."

It's actually the kind of therapy
police chiefs can get behind,

because there is nothing so inspiring
in the face of our history of racism

as a solvable problem.

Look, if the community in Minneapolis
asked their police department

to remedy the moral failings
of race in policing,

I'm not sure they know how to do that.

But if instead the community says,

"Hey, you're data say you're beating up
a lot of homeless folks.

You want to knock that off?"

That's something police
can learn how to do.

And they did.

So in 2015, the Minneapolis PD let us know

their community was concerned
they were using force too often.

So we showed them
how to leverage their own data

to identify situations
where force could be avoided.

And when you look at those data,

you'll see that a disproportionate number
of their use-of-force incidents,

they involved somebody who's homeless,
in mental distress,

has a substance abuse issue
or some combination of all three --

more than you expect

based on those factors
I was just telling you about.

So right there's the gap.

Next question is why.

Well, it turns out homeless folks
often need services.

And when those services are unavailable,
when they can't get their meds,

they lose their spot in the shelter,

they're more likely to engage in behaviors
that end up with folks calling the cops.

And when the cops show up,

they're more likely
to resist intervention,

oftentimes because they haven't
actually done anything illegal,

they're literally just living outside.

The problem wasn't a need to train
officers differently in Minneapolis.

The problem was the fact
that folks were using the cops

to "treat" substance abuse
and homelessness in the first place.

So the city of Minneapolis found a way
to deliver social services

and city resources

to the homeless community
before anybody ever called the cops.

(Applause)

Now the problem isn't
always homelessness, right?

Sometimes the problem is
fear of immigration enforcement,

like it was in Salt Lake City,
or it is in Houston,

where the chiefs had to come forward

and say, "We're not going
to deport you just for calling 911."

Or the problem is foot pursuits,

like it was in Las Vegas,

where they had to train their officers
to slow down and take a breath

instead of allowing the adrenaline
in that situation to escalate it.

It's searches in Oakland;

it's pulling folks
out of cars in San Jose;

it's the way that they patrol
the neighborhoods

that make up Zone 3 in Pittsburgh

and the Black neighborhoods
closest to the waterfront in Baltimore.

But in each city,

if we can give them a solvable problem,

they get busy solving it.

And together our partners have seen
an average of 25 percent fewer arrests,

fewer use-of-force incidents

and 13 percent fewer
officer-related injuries.

Essentially, by identifying
the biggest gaps

and directing police
attentions to solving it,

we can deliver a data-driven vaccine
against racial disparities in policing.

Right now, we have the capacity
to partner with about 40 cities at a time.

That means if we want the United States
to stop feeling exhausted

from trying to solve
an impossible problem,

we're going to need
a lot more infrastructure.

Because our goal is to have
our tools be able to scale

the brilliance of dedicated organizers

and reform-minded chiefs.

So to get there we're going to need
the kind of collective will

that desegregated schools

and won the franchise for the sons
and daughters of former slaves

so that we can build
a kind of health care system

capable of delivering our vaccine
across the country.

Because our audacious idea

is to deliver a CompStat for justice

to departments serving 100 million people
across the United States

in the next five years.

(Applause and cheers)

Doing that would mean arming
about a third of the United States

with tools to reduce racial disparities
in police stops, arrests and use of force,

but also tools to reduce
predatory cash bail

and mass incarceration,

family instability

and chronic mental health
and substance abuse issues,

and every other ill that our broken
criminal-legal systems aggravate.

Because every unnecessary
arrest we can prevent

saves a family from the terrifying journey
through each one of those systems.

Just like every gun we can leave holstered

saves an entire community
from a lifetime of grief.

Look, each and every one of us,

we measure the things that matter to us.

Businesses measure profit;

good students keep track of their grades;

families chart the growth
of their children

with pencil markings in doorframes.

We all measure the things
that matter most to us,

which is why we feel the neglect

when nobody's bothering
to measure anything at all.

For the past quarter millennium,

we've defined the problems
of race and policing

in a way that's functionally
impossible to measure.

But now the science says
we can just change that definition.

And the folks at the Center
for Policing Equity,

I actually think we may have measured

more police behavior
than any one in human history.

And that means that once we have the will

and the resources to do it,

this could be the generation

that stops feeling like racism
is an unsolvable problem

and instead sees

that what's been necessary
for far too long is possible.

Thank you.

(Applause and cheers)