In December 2011, J. David Hacker, a professor of history at SUNY–Binghamton, made headlines with a study published in the journal Civil War History contending that the longstanding estimate of 620,000 Civil War dead could be 750,000, or even as high as 850,000. He makes his case using census data from 1850 to 1880 to reveal soldier deaths from disease or war-related injuries that were never officially counted.
Why did you suspect the estimate of 620,000 deaths was too low?
Shocking as it may seem, neither the federal government nor the CSA government felt it was their obligation to keep careful personnel records of the men participating in the war. The records they had were geared toward determining effective field strength, not counting the dead. Both armies lacked the bureaucracies, personal identification numbers and dogtags needed to identify and count the dead. After the war, the War Department tried to get a direct count of the dead from muster-out rolls. Quartermaster Montgomery Meigs came up with 300,000 Union soldiers who had died in the war. But after the federal government established a pension system, and widows and orphans started applying, the toll climbed to 360,000 Union soldiers.
Were there estimates for Confederate dead?
Many of the Confederate records were destroyed in the war. Several decades later William Fox combed through the battlefield reports and came up with an estimate of 94,000 Confederate combat deaths. But Fox knew that the tallies of the Union dead were going up, and he suspected that the Confederate deaths would as well.
Who came up with the total of 620,000?
At the turn of the 20th century, Thomas Livermore estimated noncombat deaths in the Confederate forces by assuming that Confederate men died at the same rate of disease and other noncombat causes as Union men did. The combined total was 620,000.
Did anyone doubt that number?
In 1870 Francis Amasa Walker, the census superintendent, realized that the population returns were much lower than expected. He suspected 850,000 direct losses from the war, but he didn’t have the tools to verify it. Back then they had to do hand tallies of the census returns, which took years and millions of taxpayer dollars. They didn’t cross-tabulate the population by age, sex and birthplace, which is essential in trying to estimate Civil War deaths with the census.
How did you come up with your estimate?
We have created new census samples by examining the original manuscript returns, typing them into a database and exploiting modern computing capacity. Samples of the 1850, 1860, 1870 and 1880 censuses created at the University of Minnesota allow me to cross-tabulate the population in the necessary ways—by birthplace, sex and age—to estimate the number of men missing from the 1870 and 1880 census.
What did you look for?
I follow age groups across censuses. Individuals 0-9 years old in 1850 are 10-19 years old in 1860, 20-29 in 1870 and 30-39 in 1880. So you can follow the number of men and women in age groups across time. Now, immigration makes that difficult, because so many people are coming into the country. What is essential to my results is that I restricted my numbers to the native-born population, and I assumed that the mortality rates of the foreign-born and native-born did not vary much.
How did you know the census is accurate for those turbulent years?
No census is perfect. I’m looking at the survival rates of men across four censuses relative to survival rates for women in the same censuses. If one census was more poorly conducted than another, it is likely to miss approximately the same number of men and women. Sex differences in survival will not be affected.
So you want to spot males counted in 1860 who don’t show up in 1870?
Essentially, yes. That means I may be counting men that Walker was concerned about—men who contracted a disease in the war, came home and succumbed to it. I can’t tell why they are dying. They’re just missing. I don’t know what proportion of the 750,000 died of their wounds or disease or on the battlefield. I’m really careful to say that it’s an estimate of excess men who died as a result of the war. It’s not the pure number of deaths of enlisted soldiers that people might be using for other wars like, say, Vietnam.
How did you come up with this approach?
I get a phone call about once a year from somebody asking whether there is any way to use the census to estimate civilian deaths in the war, and the answer I always gave was no— and, sadly, it’s still no. But the last time I took that phone call, I thought, “You know, I could estimate male losses just from that central assumption that the female losses were negligible relative to the number of the male losses.” I did the calculation and got the result very quickly
Can you compare Union and Confederate losses?
No. The problem is that people don’t stay put. Men born in Virginia, for example, might have moved North and fought for the Union. But I have looked at death rates for men born in the slave states, including the border states, relative to men born in northern states, and found that their death rate was twice as high.
How did Civil War deaths affect the country?
When you look at the impact of the war, the death rate—how many people died as a percentage of the population—is a better statistic than the absolute death toll. If we had the same death rate today as we had during the Civil War, we would be expecting 7.5 million deaths, which is staggering. And the South suffered disproportionately. We’ve long known that. My estimates indicate that more than 20 percent of the men ages 20–24 in 1860 who were born in the South died as a result of the war. A huge proportion of Southern women remained widowed in 1880. The loss of men in their prime affects more than marriage. We know that the Southern economy struggled for a long time.
Has your research changed how you view the war?
I’m of two minds. The war touched communities and families across the country more deeply than we’ve long believed. That’s important. On the other hand, as a demographer, I know that far more people are dying from other causes than the war, even in the war years. It’s not the most important thing that is happening in demography in the 19th century. In my opinion, that is fertility decline: Women go from having 7 children to 3.5 children over the course of the 19th century. There’s a deficit of more than one million babies during the war, and it’s much deeper in the South. In terms of the human cost of war, the birth deficit is larger than the direct losses. So the implications of the war echo into the next generation demographically.
Originally published in the August 2012 issue of Civil War Times. To subscribe, click here.