The 160-Year Shadow of Slavery: How America’s Past Still Predicts Who Lives and Dies

In 1860, U.S. census takers traveled across the country recording a statistic that now feels almost incomprehensible.

Nearly four million human beings were legally enslaved.

Their distribution was not uniform. In some Southern counties, enslaved people made up a relatively small share of the population. In others, they represented half—or even a majority—of everyone living there.

More than 160 years later, researchers have compared that old geographic map with something very different:

Who dies, and at what rate, in those same places today.

The result is disturbing.

A 2026 study published in the Proceedings of the National Academy of Sciences found that counties where enslaved people made up a larger share of the population in 1860 tend to have substantially wider mortality gaps between Black and white Americans today.

The researchers estimated that for every 10-percentage-point increase in a county's enslaved population in 1860, there were roughly 22 additional deaths per 100,000 Black residents relative to white residents during 2010–2020.

The finding does not mean slavery is somehow directly killing people 160 years later.

It does not mean every resident of a historically slave-dependent county experiences the same health outcomes.

And because the study is based on county-level historical and contemporary data, it cannot follow individual families across six generations and prove a simple one-step causal chain.

What it does suggest is arguably more consequential.

Institutions can outlive the laws that created them.

The study found that the historical relationship was statistically connected to modern racial segregation, economic inequality, and Black-white disparities in poverty and education. Those contemporary conditions, in turn, help explain why the mortality gap remains wider in places where the antebellum economy depended most heavily on enslaved labor.

Slavery ended.

The geography it helped build did not disappear with it.

The Study at the Center of the New Findings

The paper is titled:

Historical slavery predicts contemporary disparities in mortality between Black and White Americans.

It was published online in August 2026 in Proceedings of the National Academy of Sciences, volume 123, issue 33, article e2508037123.

The authors are:

• Manuel J. Galvan
• B. Keith Payne
• Keely A. Muscatell
• Julian M. Rucker
• Paschal Sheeran

Its DOI is 10.1073/pnas.2508037123.

The researchers linked three periods of American history.

First came slavery.

They used county-level information on the proportion of the population recorded as enslaved in the 1860 U.S. Census, the final federal census conducted before the Civil War and abolition.

Then came measures of structural inequality, primarily from 1990 through 2000.

Finally, they examined Black and white mortality outcomes from 2010 through 2020.

The purpose was not simply to ask whether historically slave-dependent counties have worse health.

It was more specific:

Does the local intensity of slavery in 1860 predict the size of the Black-white mortality disparity approximately 150 years later?

The answer was yes.

The 1860 Census Captured a Country on the Edge of Civil War

The choice of 1860 is important.

The federal census began on June 1 that year, only months before Abraham Lincoln's election and the secession crisis that led to the Civil War.

The Census Bureau reports that the United States contained 31,443,321 people, including 3,953,760 enslaved people—about 12.6% of the total population.

But the national percentage hides enormous local variation.

The federal government maintained a separate Slave Schedule.

It recorded information such as:

• The slaveholder's name
• Number of enslaved people
• Age
• Sex
• Racial classification
• Fugitive status
• Manumission
• Certain disabilities

The enslaved people themselves were generally not recorded by name.

That administrative reality captures something fundamental about the institution.

The census counted enslaved human beings as demographic and economic units while often preserving the identity of the person legally claiming ownership rather than the identities of the people being owned.

The Government Literally Mapped the Geography of Slavery

In 1861, the U.S. Coast Survey turned census data into an extraordinary map.

County by county, darker shading represented a larger proportion of enslaved residents.

The result showed the geography of slavery across the South with remarkable clarity.

The Library of Congress notes that this was an early landmark in American statistical cartography and that Abraham Lincoln is believed to have consulted the map during the Civil War.

More than a century and a half later, that kind of geographic variation gives researchers something powerful.

If slavery had existed everywhere at exactly the same intensity, separating its local legacy statistically would be extremely difficult.

Instead, neighboring counties could have dramatically different levels of dependence on enslaved labor.

Researchers can ask whether those historical differences correspond with differences that remain today.

What Did the 2026 Study Find?

The headline result was substantial.

The model estimated approximately:

22 additional Black deaths relative to white deaths per 100,000 residents for each 10-percentage-point increase in the enslaved share of a county's 1860 population.

Imagine two otherwise comparable historical counties.

In one, enslaved people made up 10% of the population in 1860.

In another, they made up 50%.

That is a 40-percentage-point difference.

Using the study's estimated relationship as a simplified illustration, the second county would be associated with a Black-white mortality disparity roughly 88 deaths per 100,000 larger.

That does not mean slavery mechanically produced exactly 88 deaths in that county.

The coefficient is a statistical estimate across counties, not an individual prediction.

But it gives a sense of the magnitude.

The Gap Was Not Simply Because Everyone in Those Counties Dies Younger

One of the study's more striking findings makes the result harder to explain as generic Southern poverty or generally poor regional health.

The authors report that counties with a greater historical reliance on slavery have higher contemporary mortality among Black Americans but lower mortality among white Americans.

In other words, the historical association does not simply look like:

High-slavery county → everyone has worse health.

Instead, it looks more like a divergence.

The same local history is associated with disadvantage for one racial group and relative advantage for another.

That pattern is one reason the researchers focused on institutions that distribute resources and opportunities unequally rather than treating slavery merely as a marker for poor counties.

How Large Is the Black-White Mortality Problem Nationally?

The 2026 paper sits within a much larger public-health problem.

A major 2023 JAMA analysis used CDC mortality data to calculate excess deaths among Black Americans relative to white Americans from 1999 through 2020.

Its conclusion was staggering.

Over those 22 years, the researchers estimated approximately:

1.63 million excess Black deaths

and more than:

80 million years of potential life lost.

Those numbers do not mean 1.63 million deaths had one single cause called “racial inequality.”

Rather, they quantify what would have happened if Black Americans had experienced age-specific mortality rates comparable with white Americans during the period studied.

The difference accumulated into more than a million lives.

Geography Matters More Than It Might Seem

National averages can make racial health disparities appear abstract.

They are not evenly distributed.

A 2024 study examining the geography of the Black-white life expectancy gap concluded that more than 90% of the national disparity could be attributed to differences occurring within the same counties, rather than simply Black Americans and white Americans living in completely different parts of the country.

That matters for understanding the slavery study.

The question is not merely whether Southern states have worse health than Northern states.

It is whether people living in the same local institutional environments experience systematically different outcomes by race.

Those environments include housing, schools, employment, wealth, exposure to violence, access to medical care, environmental conditions and political institutions.

How Could Something From 1860 Affect Mortality in 2020?

This is the obvious question.

No one alive in 2020 was legally enslaved in the United States in 1860.

Most institutions operating today did not exist in their current form in 1860.

Medicine changed.

Cities changed.

People migrated.

Governments changed.

Jim Crow was dismantled legally.

Civil-rights legislation transformed American law.

So what exactly could persist?

The answer proposed by the researchers is not biological inheritance from slavery.

It is institutional persistence.

Places develop pathways.

Economic systems shape land ownership.

Land ownership shapes wealth.

Wealth shapes political power.

Political power shapes schools, taxation, infrastructure and public investment.

Residential systems influence where people can live.

Where people live influences school access, pollution exposure, employment, neighborhood wealth and healthcare access.

Those conditions can reproduce themselves even after the original legal institution disappears.

Think of History as a Path Rather Than a Scar

A scar suggests something bad happened in the past and left a mark.

The mechanism described by this research is more dynamic.

Think of it as a path.

Once a community begins moving down one institutional path, later choices are constrained by what already exists.

A railroad built in 1850 can determine where a city develops in 1900.

A highway constructed in 1960 can shape neighborhoods in 2020.

A school-district boundary can reproduce residential inequality for generations.

A concentration of property ownership can influence wealth accumulation long after the original property owners die.

Social scientists call related processes path dependence.

The past does not dictate the future.

But it changes the menu of futures that become easy or difficult.

Slavery Was Not Just a Labor System

This is essential to understanding the study.

American slavery was not simply unpaid agricultural work.

It was a comprehensive political and economic institution.

It determined:

• Who could own property
• Who could legally marry
• Who could move freely
• Who could testify
• Who could be educated
• Who controlled labor
• Who accumulated wealth
• Who possessed political power
• Who could pass property to descendants

Counties with extremely high enslaved populations were therefore not merely places with more plantations.

Their governments, economies and social hierarchies developed around racial domination at an unusually intense level.

Abolition removed legal ownership of human beings.

It did not instantly equalize land, wealth, education or political power.

Then Came Reconstruction—and Its Destruction

The period after the Civil War briefly produced profound changes.

The 13th Amendment abolished slavery, with its exception for punishment after conviction.

The 14th Amendment established citizenship and equal protection.

The 15th Amendment prohibited denying voting rights based on race.

Black Americans built schools, churches, businesses and political institutions.

Black men held public office throughout the South.

But Reconstruction was violently resisted.

White supremacist organizations emerged.

Black political participation was attacked.

By the late 19th and early 20th centuries, Southern states constructed Jim Crow systems that enforced racial hierarchy through law and violence.

The point is not that slavery remained unchanged.

It did not.

The institutional form evolved.

The Study Found Four Major Modern Pathways

The 2026 researchers tested whether several contemporary structural inequalities statistically mediated the relationship between historical slavery and mortality.

Four stood out:

Racial segregation

Economic inequality

Racial disparities in poverty

Racial disparities in education

“Mediated” is an important statistical term.

It does not mean the researchers proved every individual death passed through one of those four variables.

It means the statistical relationship between slavery and mortality became explainable in significant part through those intervening conditions.

Segregation Is Not Merely About Who Lives Next Door

Residential segregation can influence nearly every major determinant of health.

Where someone lives affects exposure to:

• Air pollution
• Lead
• Heat
• Traffic
• Crime
• Food access
• Parks
• Safe exercise space
• Hospitals
• Pharmacies
• Emergency response
• Schools
• Employment
• Transportation

Neighborhood wealth also determines the local tax base in many areas, influencing schools and public services.

A 2024 demographic analysis found residential segregation to be strongly related to the Black-white life expectancy gap and estimated that the overwhelming majority of the national disparity originates from within-county inequality.

The word segregation therefore describes much more than physical separation.

It can determine which institutions surround a person from childhood onward.

Poverty Becomes Biology Without Being Genetic

Economic inequality may seem separate from medicine.

It is not.

Chronic financial insecurity affects:

• Nutrition
• Housing
• Stress
• Environmental exposure
• Preventive healthcare
• Prescription affordability
• Transportation
• Employment conditions
• Sleep
• Ability to take time off work
• Access to safe neighborhoods

These experiences eventually become measurable in bodies.

Blood pressure rises.

Diabetes risk changes.

Exposure accumulates.

Preventive conditions go untreated.

This is one of the most important concepts in social epidemiology:

Social conditions can become biological outcomes without those outcomes being genetically predetermined.

Education Is Also a Health Variable

Educational inequality affects health through multiple routes.

Education is associated with:

• Employment
• Income
• Health literacy
• Insurance coverage
• Workplace hazards
• Ability to navigate medical systems

But schools themselves are also local institutions shaped by historical property patterns, residential segregation and public investment.

A prior study of the American South found that counties with greater historical dependence on slavery continued to show racial differences in public-school enrollment patterns, reinforcing the broader literature suggesting that slavery's geography predicts modern institutional differences beyond health alone.

Earlier Research Found Political Legacies Too

The new mortality findings did not emerge from nowhere.

In 2016, political scientists Avidit Acharya, Matthew Blackwell and Maya Sen published influential research examining slavery's political legacy across the American South.

They found that white residents of counties with larger enslaved populations in 1860 were, generations later, more likely to express racially conservative attitudes and opposition to affirmative action.

Their proposed explanation was institutional and cultural persistence after emancipation: local elites had incentives to reinforce racial hierarchy after slavery ended, and those norms and institutions could then persist across generations.

The 2026 health study extends this general historical-persistence framework into mortality.

But Can a Study Like This Really Prove Causation?

Not completely.

This is the most important scientific caveat.

Researchers cannot randomly assign counties to have slavery in 1860 and compare them 160 years later.

This is observational historical research.

That creates a major problem called confounding.

Perhaps counties with more slavery differed in some other important way:

• Soil
• Climate
• Agriculture
• Urbanization
• Wealth
• Disease environment
• Population density
• Geography

Maybe that other characteristic—not slavery itself—created the modern mortality pattern.

The authors attempted to address this problem aggressively.

The Researchers Controlled for Historical and Modern Differences

According to the paper, the association remained after accounting for a broad suite of historical and contemporary covariates.

They also performed multiple supplementary robustness analyses.

This helps answer objections such as:

“Maybe historically high-slavery counties are simply poorer places today.”

If controlling for contemporary poverty eliminates the entire association, the historical variable may be doing little independent work.

Instead, the researchers found persistent relationships and then investigated whether structural inequality formed the pathway through which history operated.

Still, statistical controls cannot measure every conceivable unobserved factor.

The Study Also Used an Instrumental-Variable Strategy

The authors went further by using instrumental-variable analysis to test the plausibility of a causal pathway.

Instrumental variables are an attempt to isolate variation in an exposure—in this case historical slavery—that comes from factors related to the exposure but not independently to the modern outcome through another pathway.

This is a sophisticated method frequently used when randomized experiments are impossible.

But instrumental-variable estimates depend heavily on assumptions.

They strengthen causal inference.

They do not transform historical observational data into a randomized clinical trial.

That is why the authors themselves use cautious language:

Their findings are consistent with a model in which slavery's legacy shapes modern mortality through enduring structural inequalities.

That wording matters.

What the Study Does Not Say

The findings are powerful enough without exaggeration.

The study does not show that:

• Every Black person in a historically slave-dependent county has poor health.
• Every white resident benefits personally from slavery's legacy.
• A person's ancestry determines their life expectancy.
• Modern medical treatment is irrelevant.
• Personal behavior never affects health.
• Migration does not matter.
• Slavery explains every racial mortality disparity.
• History makes inequality permanent.

Mortality is produced by many interacting factors.

The lead author has publicly noted that the full path model explains only part of the variance in mortality disparities, which is unsurprising for an outcome as complex as death.

A historical legacy can be important without being the only thing that matters.

Migration Does Not Automatically Erase Place-Based Institutions

One intuitive objection is:

“People moved. How could counties from 1860 still matter?”

Millions of Black Americans did move, most famously during the Great Migration.

Millions of white Americans moved too.

But institutions can persist even when populations change.

A county's:

• Property system
• Road network
• School structure
• political institutions
• land-use pattern
• economic base

can survive extensive population turnover.

Meanwhile, many descendants of historical Southern populations remained within the region, and more than half of contemporary Black Americans still live in the broader former slave-state geography, according to the lead author's discussion of the research.

The study is therefore primarily about places, not about tracing each resident's genealogy to 1860.

Race in This Research Is Not a Biological Cause

Another misconception must be avoided.

The paper does not imply that Black Americans have higher mortality because of an inherent biological characteristic called race.

Race is a social classification with real historical consequences.

The study's proposed pathways are explicitly structural:

• Segregation
• Poverty
• Education
• Economic inequality

In other words, the relevant mechanism is not:

Black identity → biological vulnerability.

It is closer to:

historically racialized institutions → unequal environments and resources → unequal health risks.

That distinction is fundamental.

White Mortality Makes the Finding Especially Interesting

The finding that historical slavery was associated not only with higher Black mortality but with lower white mortality deserves more attention.

It suggests that inequality can operate relationally.

Institutions that disadvantage one group may simultaneously direct resources, opportunity or security toward another.

That does not mean every white person in those counties is wealthy or healthy.

Southern white poverty is real.

Many predominantly white rural communities experience severe mortality problems.

Population-level advantage and individual hardship can exist at the same time.

The study is measuring average racial differences within historical geographic structures.

Could Rural Health Explain Everything?

The rural South has major contemporary health challenges.

Hospital closures, physician shortages, poverty, chronic disease and long travel distances can all raise mortality.

So perhaps the slavery variable is merely identifying rural counties.

That is exactly the type of alternative explanation historical-persistence studies must test.

The new paper reports that its central association survived extensive controls and alternative analyses intended to address historical and contemporary confounding.

That does not make rurality irrelevant.

It means rurality alone does not appear sufficient to explain the historical pattern the researchers identified.

COVID-19 Also Complicates the 2010–2020 Mortality Window

The study's mortality period ends in 2020, the first year of the COVID-19 pandemic.

That year profoundly widened some racial mortality disparities.

Separate research has shown that racially and socioeconomically segregated counties experienced significantly higher COVID-19 death rates, demonstrating how structural conditions can amplify the effect of a new disease.

However, the PNAS analysis spans an entire decade rather than relying only on pandemic mortality.

The broader relationship therefore cannot be reduced simply to COVID.

Still, the pandemic illustrates how historical inequality can interact with a completely new biological threat.

The virus was new.

The unequal exposure environment was not.

History Does Not Need to Cause the Disease to Shape Who Dies From It

This is perhaps the most useful way to understand the research.

Slavery does not cause a heart attack in 2020.

It does not create a cancer cell.

It does not infect someone with SARS-CoV-2.

But historical institutions can influence:

• Whether someone lives near pollution
• Whether preventive care is available
• Whether hypertension is controlled
• Whether a hospital is nearby
• Whether employment includes insurance
• Whether housing is crowded
• Whether accumulated wealth provides a buffer during illness

Those conditions alter the probability that disease becomes disability or death.

History operates through the present.

Why the 22-Deaths Figure Should Not Be Misquoted

The study's headline number is likely to travel widely online.

It needs context.

The result is not:

“Every 10% more slavery caused 22 Black people to die today.”

The actual interpretation is approximately:

For every 10-percentage-point increase in a county's enslaved population share in 1860, the statistical model estimated a contemporary Black-white mortality disparity about 22 deaths per 100,000 larger.

That is a difference in a racial mortality gap.

It is not a count of individual deaths proven to have been caused by slavery.

That distinction preserves both the significance and the scientific accuracy of the result.

The 20-Fold Comparison Is Even More Striking

Reporting on the paper highlighted another striking comparison.

Between 2010 and 2020, the Black-white mortality gap was reported to be approximately 20 times larger in counties with the greatest historical reliance on slavery than in counties where the 1860 census recorded no slavery.

That comparison helps translate the regression coefficient into something intuitive.

Historical geography does not appear merely as a tiny statistical trace detectable only with enormous datasets.

At the extremes, the difference is large.

Does This Mean America Has Made No Progress Since 1865?

No.

That interpretation would be historically false.

The United States of 2026 is fundamentally different from the slave society of 1860.

Enslavement was abolished.

Legal segregation was dismantled.

Voting rights expanded.

Black educational attainment rose dramatically.

Black political representation expanded.

Black wealth, professional opportunity and life expectancy all changed enormously.

The existence of persistent inequality does not erase progress.

Likewise, progress does not prove historical inequality has vanished.

Both statements can be true.

Abolishing a Law Is Faster Than Rebuilding a Society

This may be the central lesson.

A legislature can change a law in one day.

Property takes generations to redistribute through markets and inheritance.

School systems persist.

Neighborhoods persist.

Infrastructure persists.

Social networks persist.

Political institutions persist.

Wealth compounds.

Debt compounds.

Environmental exposures compound.

That is why an institution can disappear legally while its downstream effects remain measurable.

There Is Nothing Inevitable About the Result

Historical persistence does not mean historical destiny.

If segregation, poverty, education and economic inequality help mediate the mortality relationship, that implies something hopeful as well.

Those are not immutable features of geography.

They can change.

Investment can change.

Healthcare can change.

School quality can change.

Housing patterns can change.

Environmental exposure can change.

The implication of the study is therefore not simply:

“Slavery still matters.”

It is:

The pathways through which slavery appears to matter are contemporary conditions that societies can still alter.

The Study Changes How We Think About “Past” and “Present”

Americans often divide history into clean chapters.

Slavery.

Civil War.

Reconstruction.

Jim Crow.

Civil Rights Movement.

Modern America.

The chronology is useful.

But institutions do not respect textbook chapter breaks.

The end of one legal regime becomes the starting condition for the next.

A plantation county in 1860 becomes a Reconstruction county in 1870.

Then a Jim Crow county.

Then a New Deal county.

Then a desegregating county.

Then a modern county.

The geographic unit stays while laws, industries and populations evolve around it.

The 2026 study essentially asks whether the statistical memory of those transitions is still visible in mortality.

It is.

A Census Map From 1861 Is Predicting a Health Map From the 21st Century

That may be the most unsettling image produced by the research.

Take the old 1861 Coast Survey map.

Look at the counties shaded most darkly because enslaved people made up the largest shares of their populations.

Now move forward more than a century and a half.

Change the variable.

Instead of enslaved population, map racial mortality disparities.

The maps are not identical.

History is not that simple.

But they remain statistically related.

The political institution disappeared.

The geographic pattern did not disappear completely.

The Research Is Really About Time

Most public-health studies search for causes relatively close to outcomes.

Smoking now.

Cancer later.

Pollution exposure now.

Cardiovascular risk later.

The slavery study stretches the causal horizon across generations.

That makes it difficult.

But it also asks a profound scientific question:

How long can social institutions remain visible in human health?

The answer appears to be:

Much longer than a human lifetime.

Frequently Asked Questions About Slavery and Modern Black-White Mortality

Did a 2026 study really link slavery to modern mortality?

Yes.

A 2026 paper in Proceedings of the National Academy of Sciences found that the proportion of enslaved people in a county in 1860 was strongly associated with the Black-white mortality gap measured during 2010–2020.

Who conducted the study?

The authors were Manuel J. Galvan, B. Keith Payne, Keely A. Muscatell, Julian M. Rucker and Paschal Sheeran.

What is the DOI of the study?

The DOI is 10.1073/pnas.2508037123.

What did the researchers actually find?

Their model estimated approximately 22 additional deaths per 100,000 Black residents relative to white residents for every 10-percentage-point increase in the county's enslaved population share in 1860.

Does that mean slavery directly caused those specific deaths?

No.

The figure is a population-level statistical estimate.

The study shows a robust historical association and tests plausible causal pathways, but it does not identify individual deaths and prove that slavery was their direct cause.

What years of modern mortality did the researchers examine?

The study linked historical data to mortality outcomes from 2010 through 2020.

What historical year did the researchers use?

They used the 1860 U.S. Census, the final census before the Civil War and abolition of slavery.

How many enslaved people lived in the United States in 1860?

The Census Bureau reports 3,953,760 enslaved people in 1860, out of a national population of approximately 31.4 million.

Why use the percentage enslaved rather than simply whether slavery existed?

Slaveholding counties differed enormously in how dependent their populations and economies were on slavery.

Using the enslaved share allowed researchers to compare the intensity of local exposure to the institution rather than using a simple yes-or-no classification.

What modern factors connected slavery to mortality?

The study identified four important statistical mediators:

• Racial segregation
• Economic inequality
• Racial disparities in poverty
• Racial disparities in education

What does “mediated” mean?

A mediator is an intervening factor that helps explain how one variable is related to another.

In this study, historical slavery predicted modern structural inequality, which in turn predicted racial mortality disparities.

It does not mean every individual death can be traced through a single mediator.

Are Black mortality rates simply higher everywhere in the South?

No.

Mortality varies substantially across counties.

The study's contribution is showing that within former slaveholding geography, variation in the historical concentration of slavery predicts variation in contemporary racial mortality disparities.

Did historically high-slavery counties have worse mortality for white residents too?

The paper reports a particularly striking pattern: greater historical reliance on slavery was associated with higher mortality for Black Americans but lower mortality for white Americans.

Why is that important?

It makes the result harder to explain as merely a general disadvantage affecting everyone in historically poor Southern counties.

Instead, it suggests that local historical institutions may have produced racially unequal distributions of advantage and disadvantage.

Is this proof that racism causes every Black-white health difference?

No.

Mortality is affected by many factors, including age, disease, behavior, income, healthcare, environment, violence and geography.

The study identifies one important historical relationship, not a complete explanation of racial health disparities.

How large is the national Black-white mortality disparity?

A separate 2023 JAMA analysis estimated 1.63 million excess deaths among Black Americans relative to white Americans from 1999 through 2020, corresponding to more than 80 million years of potential life lost.

Does this mean race itself biologically causes mortality differences?

No.

The mechanisms examined by the researchers are structural and socioeconomic rather than an inherent biological property of racial identity.

The study focuses on segregation, education, poverty and economic inequality.

Could genetics explain the county pattern?

Genetic explanations would have difficulty accounting for why a specific county's antebellum economic institution predicts a modern racial mortality gap through contemporary structural conditions.

The paper's hypothesis concerns institutions and environments rather than biological racial categories.

Could modern poverty alone explain the findings?

The authors tested a broad set of contemporary and historical covariates and found that the association remained robust. Poverty was also one of the pathways helping explain how historical slavery relates to current mortality.

Could migration have erased the effect?

Migration undoubtedly changed county populations substantially.

But local institutions, property structures, schools, infrastructure and political patterns can persist despite population turnover.

The research is primarily a study of place-based institutional persistence, not individual ancestry.

Did the study prove causality?

It provides evidence consistent with a causal model but cannot offer the certainty of a randomized experiment.

The authors used path analysis, extensive robustness tests and an instrumental-variable analysis to strengthen causal inference.

What is instrumental-variable analysis?

It is a statistical method used to estimate causal relationships when randomized experiments are impossible.

It relies on finding sources of variation related to the exposure of interest but assumed not to affect the outcome independently through other channels.

Its conclusions depend on the validity of those assumptions.

Why can't researchers run a randomized experiment?

Historical slavery obviously cannot be experimentally assigned.

Researchers must therefore use historical variation and statistical techniques to rule out competing explanations as carefully as possible.

Has previous research found other modern effects associated with historical slavery?

Yes.

A well-known 2016 study found that slavery's county-level prevalence was associated with contemporary political and racial attitudes among white Southerners.

Research has also linked slavery's local legacy to racial inequality in public-school enrollment and other institutional outcomes.

Why would segregation affect mortality?

Residential segregation can influence exposure to pollution, neighborhood wealth, healthcare access, school quality, transportation, food access, violence and numerous other determinants of health.

A 2024 study found segregation strongly associated with the geographic Black-white life expectancy gap.

Could COVID-19 have influenced the results?

The mortality period includes 2020, the first pandemic year.

COVID-19 contributed substantially to racial mortality disparities, and separate research has linked racial and socioeconomic segregation with higher county-level COVID mortality.

However, the PNAS analysis covers 2010–2020 rather than relying only on COVID-era deaths.

Was slavery abolished more than 160 years ago?

Yes.

The 13th Amendment was ratified in December 1865, so the 2026 study appears more than 160 years after legal chattel slavery was abolished nationally.

Does the study mean nothing has improved since slavery?

No.

American racial institutions and health outcomes have changed dramatically since the 19th century.

The finding concerns the persistence of measurable geographic inequalities despite that progress.

Why did researchers use counties?

Counties provide a useful geographic unit because historical census data, modern mortality data and many structural indicators can be linked geographically.

They also capture meaningful differences in local institutions, economies and population patterns.

Was every enslaved person listed by name in the 1860 Census?

No.

The 1860 Slave Schedule generally named the slaveholder while recording enslaved people by characteristics such as age and sex rather than by individual name.

Did the U.S. government make a map of slavery in 1861?

Yes.

The U.S. Coast Survey produced a county-level map showing the proportion of the population that was enslaved across the Southern states using 1860 Census data.

Did Abraham Lincoln use that map?

The Library of Congress notes accounts that Lincoln consulted the 1861 slavery-distribution map during the Civil War.

What is the most important limitation of the 2026 study?

It is an observational county-level analysis spanning an enormous historical period.

Researchers can test alternative explanations and use sophisticated causal-inference techniques, but they cannot directly observe every institutional process linking 1860 to 2020.

Its findings should therefore be understood as strong evidence consistent with historical structural persistence rather than a perfectly identified individual causal chain.

What is the most important finding?

The most important result may not be the 22-deaths-per-100,000 estimate itself.

It is that a geographic measure taken before the Civil War still predicts racial differences in mortality roughly a century and a half later—and that modern segregation, education, poverty and economic inequality help statistically connect the two periods.

Does this mean historical inequality can never be overcome?

No.

In fact, the study points toward the opposite conclusion.

If the relationship operates through contemporary institutions, then changing those institutions can change future health outcomes.

History creates conditions.

It does not remove human agency.

What does this research ultimately tell us?

It tells us that saying an institution is “over” can mean two very different things.

Slavery as a legal system of human ownership ended in the United States in 1865.

But societies do not reset when laws change.

Land remains where it was.

Wealth remains distributed.

Cities inherit infrastructure.

Schools inherit boundaries.

Political institutions inherit constituencies.

Neighborhoods inherit investment and neglect.

Families inherit assets—or the absence of them.

The 2026 PNAS study does not argue that nothing has changed since 1860.

It demonstrates something more difficult to dismiss:

Places remember.

Not consciously.

Not mystically.

They remember through roads, property, schools, wealth, segregation, political power and opportunity.

And eventually those abstract structures become physical.

A hospital is farther away.

Hypertension goes untreated.

Pollution exposure lasts longer.

Educational opportunity narrows.

Wealth provides less protection from crisis.

Years accumulate.

Then mortality statistics record the final outcome.

More than 160 years after slavery's abolition, an 1860 census map can still help predict where the racial mortality gap is widest.

That does not mean the past is alive because Americans failed to move forward.

It means moving forward does not automatically erase the path already traveled.

And if the structures connecting that past to today's mortality can be identified, they can also be changed.