What race can tell us about homicide, and what it cannot
By Michael Smith | Reflective MVS
October 11 has never needed a reminder on my calendar.
It was a Sunday in 1992.
My mother was murdered that day.
I was 11 years old.
The boy who killed her was 14.
About three years older than me.
The next day was Columbus Day. I remember the television news covering my mother's killing alongside the usual Columbus Day stories, and those two things somehow fused together in my head. To this day, the holiday can pull me backward without asking permission.
Youngstown.
1992.
My mother was 29.
I was 11.
And she was gone.
While fact-checking this piece, I learned something I had never known as a child. President George H.W. Bush had proclaimed that Sunday, October 11, National Children's Day.
There is probably some cruel poetry in that.
An 11-year-old lost his mother.
A 14-year-old had killed her.
The next morning America moved on to Columbus Day. Source
At 11, I didn't have words like deindustrialization, concentrated poverty, residential segregation, structural inequality or ecological fallacy.
I had grief.
I had adults.
And I had whatever explanations the people around me gave me.
One of those explanations was race.
The explanations children inherit
I am biracial.
After my mother's murder, there were people in my family who made some version of the same observation:
A Black man killed your mother.
Technically, the person who killed her was a Black boy. He was 14.
But the implication was larger than him.
Black men are violent.
Black people are violent.
This is something they do.
Nobody had to sit me down and teach a formal doctrine of racial determinism. Children learn plenty without anyone calling it a lesson.
And I am not interested in dragging people I love through an article for things they said while grieving more than three decades ago.
Grief makes people reach for explanations.
Sometimes those explanations are wise.
Sometimes they are just whatever society has already placed within reach.
But I absorbed some of it.
For a while, I probably believed some of it.
Then I left Youngstown.
And the story started breaking apart.
Leaving Youngstown changed the evidence in front of me
Once I moved away, I met a much broader Black America.
Not Black people as a category.
Black people as people.
Professionals.
Parents.
College students.
Working-class families.
People with money.
People without it.
People who were strict.
People who were gentle.
People who were conservative.
People who were radical.
People who were brilliant.
People who were a mess.
People who were boring as hell.
Normal human variation has a funny way of ruining stereotypes.
The category became too large for the story I had inherited.
Being biracial also meant I had spent years moving between different expectations.
On one side of my family, there were rules that could feel harsh to a child. Watch your tone. Respect adults. Say yes ma'am and no ma'am. Know when you've pushed far enough.
On another side, I could sometimes say things or behave in ways that would never have survived five minutes in a Black household.
I don't say that to declare one family structure superior to another.
Families are complicated.
But as I got older, I began understanding some of what I once experienced simply as strictness differently.
Preparation.
Survival.
Different families prepare children for the world they believe those children will have to survive.
The older I became, the easier it was to see contradictions.
And explanations based on race alone began sounding less like explanations and more like shortcuts.
Including the explanation for my mother's murder.
My mother was more than the way she died
My mother was 18 when she had me.
That number means much more to me now.
She was raising a child while she was barely finished being one herself.
I loved her deeply.
At different points, I have called her my best friend.
At others, maybe more honestly, my frenemy.
She could be loving.
She could be funny.
She could be difficult.
She could make choices no child should have been expected to understand.
She struggled with addiction.
She had been involved with drugs.
She knew the streets.
I've written about that before because I don't believe love requires turning dead people into saints.
My mother deserves to be remembered as a whole person.
That means remembering the good without hiding the hard parts.
By the time she was killed, I believe she may have been trying to get clean.
There are still things about that day I do not know.
I know she was shot through the window of her car.
I know the person who shot her was 14.
As an adult, knowing my mother's history and personality, I have sometimes tried to fill in the missing pieces.
Maybe it involved drugs.
Maybe there was an argument.
Maybe she saw a 14-year-old and underestimated him.
Maybe she tried to pull away with something that belonged to him.
I can picture scenarios.
But I cannot call those scenarios facts.
That distinction matters.
People once filled the unknown parts of my mother's death with race.
I don't want to replace their speculation with mine.
What I know is simpler.
Saying the boy who killed my mother was Black describes something about him.
It does not explain why my mother died.
My mother and me, before October 11, 1992.
He was 14
That number bothers me more now than it did when I was 11.
Fourteen.
Not because his age excuses what he did.
It doesn't.
My mother died.
I grew up without her.
Understanding someone's circumstances does not erase responsibility for what they did.
But 14 is still 14.
And once I got old enough to understand that, the question changed.
Instead of only asking:
What was wrong with him?
I started wondering:
What was happening around him?
What kind of world puts a gun into a 14-year-old's hand?
What was happening in his home?
His neighborhood?
His school?
His city?
What had he already seen?
Who had already failed him?
I don't know those answers.
I am careful about that.
But those questions are more useful than asking what was wrong with Black people.
And once I started looking at Youngstown itself, the wider picture was hard to ignore.
Youngstown was already coming apart
Youngstown did not wake up violent one morning in 1992.
The city had been hemorrhaging for years.
In 1970, Youngstown had roughly 140,000 residents. Then came the collapse of the steel economy.
When Youngstown Sheet and Tube closed its Campbell Works in 1977, roughly 5,000 workers lost their jobs. By 1980, mill closures had eliminated approximately 10,000 steel jobs in the area. Census source
By 1990, Youngstown's population had fallen below 96,000.
The city's poverty rate, measured using 1989 income, was about 29 percent.
These were not abstract economic indicators.
Jobs disappearing means households losing income.
Households losing income means people leaving.
People leaving means empty houses.
Empty houses mean declining property values and a shrinking tax base.
A shrinking tax base means weaker public institutions.
Meanwhile drugs, guns, addiction and illegal markets do not wait politely for a city's economy to recover.
Youngstown's 1990s violence unfolded after decades of industrial collapse, population loss and economic distress.
Then came the homicide crisis.
The Youngstown Police Department's official 1992 annual report recorded 53 criminal homicides, with a homicide rate of 55.38 per 100,000 people.
The year before, there had been 59.
Interestingly, the same report says Youngstown's broader Part I crime total actually declined in 1992.
That matters.
Because crime and homicide are not synonyms, even though political arguments often treat them as interchangeable. 1992 Youngstown Police Department Annual Report
Youngstown Police Department's 1992 annual report recorded 53 criminal homicides, a rate of 55.38 per 100,000.
Youngstown homicide counts rose sharply in the early 1990s and reached 68 in 1995 in the historical series used here.
Later in the decade, Youngstown recorded even more killings.
So I need to correct something from my own memory too.
I remember hearing as a child that Youngstown had the highest murder rate in America.
Maybe somebody said it.
Maybe television said something close enough that my 11-year-old brain filed it that way.
What I can document is that Youngstown was experiencing extraordinary levels of homicide.
I cannot responsibly tell you it was definitively number one in America in 1992.
Memory is evidence that something happened to you.
It is not always a spreadsheet.
I went looking for numbers
This article started because I kept seeing the same racial argument dressed up as statistical literacy.
Look at Chicago.
Look at Detroit.
Look at Baltimore.
Look at Black population percentages.
Look at homicide.
Then comes the little smirk:
Explain that.
The implication is that the racial demographic is the explanation.
So I went looking.
I did not start with the assumption that the numbers would prove poverty explains everything.
That would be another form of cherry-picking.
I wanted to know what the data actually said.
The first dataset I assembled contained a smaller set of cities chosen because they offered interesting comparisons.
Heavily Black cities.
Heavily White cities.
Poor cities.
Affluent cities.
High-homicide cities.
Low-homicide cities.
In that sample, higher Black population share was associated with higher homicide.
Poverty was associated with homicide too.
When poverty was added to the model, the statistical relationship between Black population share and homicide dropped by about one-third.
That would have made for a very convenient article.
There was one problem.
I had selected the cities because they were interesting.
That made them good for comparison.
It did not make them a representative national sample.
So I rebuilt it.
The better data ruined the easier argument
For the second pass, I used an objective selection rule.
I assembled 112 larger U.S. cities meeting fixed population, FBI reporting and geography-matching requirements.
I used 2024 FBI homicide figures and city-level structural and demographic data, with additional segregation measures where comparable data were available.
And the result became much less tidy.
Cities with larger Black population shares did, on average, have higher homicide rates in that sample.
That association did not disappear when I added the structural variables I had available.
Concentrated poverty reduced it some.
Income reduced it some.
Adding employment, education, segregation and regional controls did not make the relationship vanish.
In some model specifications, the coefficient increased.
That was not the result a simplistic version of this article would have wanted.
So I kept it.
Because if I am accusing other people of abusing statistics to tell a racial story, I don't get to hide statistics when they complicate mine.
I rebuilt the analysis after the first sample produced a cleaner story than the methodology could support.
The correct takeaway is not:
Poverty explains the racial homicide gap.
The data I sourced does not establish that.
But it also does not establish:
Black people are inherently violent.
The data cannot establish that either.
What it establishes is an association between the racial composition of cities and homicide rates.
And an association is not a mechanism.
That difference is where this entire conversation tends to go off the rails.
A racial disparity is not a racial explanation
Suppose a city with a larger Black population has a higher homicide rate.
That is a fact about two variables measured at the city level.
It does not tell you that a randomly selected Black resident of that city is violent.
It does not tell you which residents committed the homicides.
It does not tell you why they committed them.
It does not tell you that Blackness caused anything.
A city is not a person.
This is the ecological fallacy at work: taking characteristics of places or groups and pretending they establish characteristics of individuals.
Race also does not float freely through American geography.
Black Americans did not land randomly across cities with randomly distributed wealth, schools, housing, policing, employment, environmental conditions and political histories.
Residential segregation happened.
Disinvestment happened.
Deindustrialization happened.
Redlining happened.
Mass incarceration happened.
Unequal wealth accumulation happened.
And those things don't disappear from a model simply because somebody added a poverty column to Excel.
Peer-reviewed research has repeatedly found strong relationships between violent crime and concentrated structural disadvantage. One well-known study by Lauren Krivo and Ruth Peterson found substantial evidence that extreme neighborhood disadvantage influences violent crime in both Black and White neighborhoods. Another ten-city homicide study found that social disadvantage reduced the statistical relationship between racial concentration and homicide, although the degree varied considerably from city to city.
That is a much messier answer than race.
Reality has terrible branding.
The cities refuse to behave like stereotypes
Some of the most useful evidence I found wasn't in a regression table.
It was in the cities themselves.
Take Southfield, Michigan and Boise, Idaho.
Southfield is heavily Black.
Boise is not.
Yet in the comparison data I sourced, their poverty rates and 2024 homicide rates were remarkably similar.
Now look at St. Louis and Newark.
Their Black population shares are much closer.
Their homicide rates are nowhere near each other.
Or Detroit and Johnstown, Pennsylvania.
Their racial compositions are radically different.
Their poverty and homicide figures in the year examined were surprisingly close.
Illustrative city comparisons complicate attempts to treat racial composition as a sufficient explanation for homicide. These are comparisons, not causal tests.
These examples do not prove some alternate theory.
That would be another statistical parlor trick.
They prove something narrower.
Black population share by itself is not enough.
If Blackness were the mechanism, cities with similar Black population shares should behave much more similarly than they do.
They don't.
Cities are systems.
Systems are inconvenient because they require more than one number.
Now they call it "Black fatigue"
This is where the conversation happening right now begins to sound very familiar.
A phrase has been circulating online: Black fatigue.
I need to be precise about what I mean, because the term originally meant almost the opposite.
Mary-Frances Winters used Black Fatigue to describe the cumulative physical and psychological toll that racism takes on Black people.
Then the phrase got hijacked.
In 2025, Winters herself wrote that people in social-media spaces had co-opted the term to describe being "fatigued" by Black people.
Commentary around the trend documented the phrase appearing alongside videos of Black people fighting, behaving badly or committing acts of violence, followed by sweeping complaints about Black culture or Black people generally.
That is the version I am talking about.
The version where somebody scrolls through a feed engineered to maximize outrage, watches three videos of Black people acting badly, and emerges with a racial theory.
The version where somebody says they are tired of hearing Black people talk about slavery, racism, inequality or discrimination.
Tired of the excuses.
Tired of the victimhood.
Tired of us bringing up history.
Tired of us explaining.
And somewhere in that same conversation comes:
Look at the crime.
Look at the violence.
Look at Black people.
There is an irony here so obvious it practically needs its own laugh track.
You are tired of Black people talking about race.
But you are perfectly willing to use race as the explanation for everything you think is wrong with Black people.
Apparently race is exhausting only when Black people are discussing what happened to us.
When somebody wants to explain what is supposedly wrong with us, suddenly everybody has plenty of energy.
The phrase is new.
The move isn't.
I heard an earlier version of it at 11 years old.
A Black boy killed your mother.
Therefore, here is what you need to understand about Black people.
Thirty-four years later, social media has added algorithms and better lighting.
The logic hasn't improved much.
What gets called violence depends on who is doing it
There is another reason I have trouble accepting racial explanations of violence at face value.
America has never exactly been a pacifist experiment.
That does not mean White people are inherently violent.
That would simply be racial determinism with the paint reversed.
It means America understands perfectly well how to explain violence structurally when it wants to.
The federal government and state militias forcibly removed more than 16,000 Cherokee people from their homelands during the Trail of Tears.
Nobody serious explains that history by saying White DNA caused removal.
They discuss land.
Power.
Law.
Expansion.
Economics.
Government policy.
Thousands of Black Americans were murdered in racial terror lynchings after slavery.
The Equal Justice Initiative has documented more than 4,400 such killings between 1877 and 1950 in its core study of racial terror.
Nobody needs a genetic theory of Whiteness to explain that violence either.
History gives us institutions, ideology, economic interests, racial hierarchy, political power and social permission.
Equal Justice Initiative source
That distinction is important.
Forced removal, lynching, war, state violence and interpersonal homicide are not the same phenomena.
I am not pretending they belong in one statistical column.
I am pointing out how quickly context becomes available when America wants it.
Violence committed by states gets history.
Violence committed by mobs gets ideology.
Violence committed during war gets geopolitics.
Violence in poor Black neighborhoods gets race.
That difference in explanation is worth examining.
My mother's killer still pulled the trigger
Structural analysis can become its own kind of avoidance if you are not careful.
The boy who killed my mother made a choice.
My mother made choices too.
Conditions do not pull triggers.
Human beings do.
My mother was living with addiction and whatever else she was carrying into that car that Sunday.
The boy who killed her was living with whatever he carried at 14.
I do not know what happened in his home.
I do not know who raised him.
I do not know how he got a gun.
I do not know what he had witnessed.
I do not know whether he had opportunities that he ignored or opportunities that never existed.
And I refuse to invent a tragic childhood for him simply because it would fit my argument nicely.
He killed my mother.
That belongs to him.
But the decision to turn his race into an explanation for millions of other people belongs to everybody else.
There is nothing contradictory about holding both ideas at once.
People are responsible for what they do.
And the world in which people make choices matters.
We understand that everywhere else.
Addiction affects behavior without making every addict violent.
Economic desperation affects behavior without turning every poor person into a criminal.
War affects behavior without making every person in a war zone a murderer.
Context is not absolution.
It is context.
October always brings her back
I tend to write about my mother in October.
I don't plan it that way every year.
The month just knows where to find me.
October 11 arrives.
Then Columbus Day.
Then memories that are somehow both 34 years old and five minutes ago.
But I do not think about my mother only as a murder victim anymore.
I think about her life.
How young she was.
How young I was.
How much fun she could be.
How difficult she could be.
Her addiction.
Her mistakes.
Her love.
The things she never had enough time to fix.
She was my mother.
Sometimes my best friend.
Sometimes my frenemy.
She was not a homicide statistic.
And the boy who killed her was not a racial statistic either.
He committed an irreversible act.
That belongs to him.
For years, I thought the question was:
Why did a Black boy kill my mother?
I don't think that is the right question anymore.
I think the harder question is:
What created the world in which both of them were living?
Race may help locate parts of that world on a map.
Race may correlate with what happened there.
Race may tell us something about how American history distributed opportunity, deprivation and exposure to violence.
But correlation is where an investigation begins.
It is not where an explanation ends.
And if somebody is already too afflicted with "Black fatigue" to make that distinction, maybe the problem was never that Black people kept talking about race.
Maybe it was that some people had already decided what they wanted race to mean.
READ OFFLINE
Prefer a PDF?
Download the complete illustrated Reflective MVS edition of this essay.
Receipts & Methodology
The Youngstown historical section relies heavily on the 1992 Youngstown Police Department Annual Report, archived by the U.S. Department of Justice. It records 53 criminal homicides in 1992 and a homicide rate of 55.38 per 100,000. Read the report.
The industrial-history figures draw on U.S. Census records and historical research documenting Youngstown's population decline and the collapse of the Mahoning Valley steel economy.
The national analysis described in this essay uses the objectively selected 112-city dataset I assembled for this project. It is a cross-sectional, city-level analysis. Its results show associations, not individual racial propensity and not proof of racial causation.
The broader criminology literature cited here includes research on concentrated disadvantage and homicide showing that neighborhood structural conditions matter substantially, while also demonstrating that the relationship is not identical across every city. Review the research.
The contemporary discussion of "Black fatigue" in this essay refers specifically to the co-opted use of Mary-Frances Winters' term to describe supposed fatigue with Black people. Winters' original use of the phrase described the accumulated toll of racism on Black people.
Reflective MVS
Reflection with receipts.
0 Comments