Oxford University Press's
Academic Insights for the Thinking World

Color blindness in the demographic death toll of the Civil War

By James Downs


An 2 April 2012 New York Times article, “New Estimate Raises Civil War Death Toll,” reports that a new study ratchets up the death toll from an estimated 650,000 to a staggering 850,000 people. As horrific as this new number is, it fails to reflect the mortality of former slaves during the war. If former slaves were included in this figure, the Civil War death toll would likely be over a million casualties.

Dr. J. David Hacker, the demographer who tabulated these numbers, relies on new microdata samples of the 1850, 1860, 1870, and 1880 censuses to estimate white deaths that occurred during the war. His findings challenge “the gospel” of statistics that two amateur 19th century historians used to count the Civil War dead. But in framing his analysis in response to 19th century historians, Hacker reproduces a 19th century problem. He does not consider the black population as part of his overall calculation.

In Hacker’s 37-page study, there is a passing reference to former slaves’ mortality. It is quickly pushed aside in favor of a more intensive analysis of the experience of white soldiers. According to both the 19th century counters and Hacker, former slaves were “civilians” and thus were not considered part of the larger military death toll.

Yet tens of thousands of slaves died during the Civil War. They were shot by Confederate soldiers, and according to new research conducted by Duke historian Thavolia Glymph, Union soldiers shot at former slaves. But many, many more died from the same illnesses that claimed the lives of white soldiers.

As many Civil War historians have documented, more soldiers died from the outbreaks of pneumonia, yellow fever, and smallpox that plagued Union and Confederate camps than died in combat. As ex-slaves liberated themselves from Southern plantations and took refuge in Union camps during the war, they came into direct contact with these illnesses, became infected, and died in staggering numbers. These former slaves and soldiers died in the same camps from the same diseases and therefore should considered be the same in terms of the war’s death toll.

Just like white soldiers, emancipated slaves died from an explosive smallpox epidemic that began in Washington in 1862 and spread across the South from the Carolinas to the Mississippi Valley in 1863 and 1864. By the war’s end in 1865, the epidemic had moved westward, where it claimed the lives of countless Native Americans. The rough 19th century estimate was that 60,000 former slaves died from the epidemic, but doctors treating black patients often claimed that they were unable to keep accurate records due to demands on their time and the lack of manpower and resources. The surviving records only include the number of black patients whom doctors encountered; tens of thousands of other slaves had no contact with army doctors, leaving no records of their deaths. This is to say nothing of the thousands of Native Americans who died during the war — and who are not referenced in Hacker’s study. By relying purely on the census as the primary document, Hacker follows in the 19th century illogic of seeing black people and Native Americans as outside of the purview of those who died during the Civil War.

Not counting the deaths of former slaves ignores the fact that the Civil War, which was intended to liberate bondspeople from chattel slavery, led to widespread sickness, suffering, and death. It also implies that former slaves’ experience is not part of the nation’s history.

James Downs is an associate professor of history at Connecticut College and author of Sick from Freedom: African-American Illness and Suffering during the Civil War and Reconstruction, and editor of Taking Back the Academy: History of Activism, History as Activism and Why We Write: The Politics and Practice of Writing for Social Change.

Subscribe to the OUPblog via email or RSS.
Subscribe to only American history articles on the OUPblog via email or RSS.
View more about this book on the

Recent Comments

  1. J. David Hacker

    According to James Downs, I am somehow guilty of implying that “former slaves’ experience is not part of the nation’s history.” What? As I clearly state in my article, the census-based method I rely on, which estimates male losses relative to female losses, cannot be used to estimate black mortality for the very reasons Downs enumerates:

    “The two-census approach described above cannot be used to estimate excess male deaths in the black population. One reason is that black civilian deaths, rather than being a negligible part of the total excess number of black deaths, likely approached or exceeded the number of military deaths. High mortality rates among both black males and females in contraband camps during the war, high mortality associated with the postwar migration of blacks from rural to urban areas, the transition from slave to free labor, and postwar violence directed at the black population strongly suggests a high number of civilian deaths. Because black women also suffered elevated mortality, excess male deaths cannot be inferred from the age pattern of sex differentials in survival. Even if excess black mortality in the 1860s could be estimated, it is unclear what proportion should be attributed to the Civil War.”

    Simply put, it couldn’t be done. Perhaps Downs would have been happier if I had wrung my hands for a few paragraphs more. But I am implying that former slaves’ experience is not part of nation’s history? Please.

    Downs also seems upset that I dared call slaves and former slaves “civilians” because they were sometimes shot at by soldiers. So what do we call non-enlisted men and women who are targeted by armies? Downs doesn’t say it, but perhaps he is implying that slaves were valid targets under the rules of engagement? He criticizes me for not counting black deaths as part of the _military_ death toll, after all. I very much doubt if he considers slaves valid military targets, however. Right? I’ll extend to Downs a courtesy that he did not extend to me and assume that he was not implying anything nefarious in the blank spaces at the end of his post.

    Downs is correct, of courses, that civilian (that term again) losses should be part of the story and black losses a major part. I wish that the census was a fine enough of an instrument to measure those losses. Alas, it is not. I could perhaps string together some stray qualitative evidence and pick a number out the air—tens of thousands?—but then I’d probably be accused of reproducing nineteenth-century historians’ methods and mistakes. Perhaps Downs’ book will be a good start on that project. Certainly it is a great topic and I find the title promising. Let’s hope that its contents are not filled with the straw-man arguments and uncharitable readings of historiography that he displays in this short blog post.

  2. Michael DeGruccio

    I hope this debate doesn’t get any more personal than it already has. Layered beneath the steam, though, are some interesting points. Could I summarize what seems to me to be the thrust of the debate?
    Professor Downs read over Professor Hacker’s exceptionally refreshing article only to find some old wine in new wineskins. Focusing on the tools at his disposal (nineteenth century records which largely erased black and native American experience), Hacker seems to have let the erasures and elisions stand. Hacker claims that he can’t simply fill in the void with conjecture (that’s not what demographers do!)–and even admits as much in his article. Downs senses that this kind of reasoning is a bedrock problem that if not addressed with some imagination, sussing new kinds of sources, and risk (and more than passing acknowledgements), will replicate the same story about white men fighting for a white America. Fair enough.
    Hacker suggests that Downs failed to read his article closely enough. Or at least that read it with jaundiced eyes. But Downs seems to suggest that Americans shouldn’t have to read the fine print (or the passing apologies) in order to know how central marginalized peoples were to the war and its destructions. Fair enough again.

    At the risk of sounding spineless I think there is some important middle ground to be had here. Hacker poses a rhetorical solution of “wringing his hands” for a few more paragraphs. My sense is that if Hacker could go back, he would be more than willing to have done so-especially after hearing some of the criticisms from Downs. (I think a larger question is: “why in all the writing, sharing, collaborating, editing, re-editing, anonymous readings, and publishing, Hacker’s network of scholars didn’t prod him to “wring his hands” a little more where his sources leave him high and dry?” (Assuming that he wasn’t prodded in some way, which he might have been.)
    But I refuse to assign any nefarious intentions to anybody. And I think Hacker is right that sometimes Downs’ blog borders on such aspersions.
    Still, in response to Hackers rhetorical solution of wringing his hands for a few more paragraphs my first gut response was, why not “wring away”?– it is our national story at stake here, after all. And some hand wringing might have only enhanced or underscored the power of Hacker’s bold work. Even more, maybe Downs has something vital to tell us: attempts at assessing the costs of a national war might require more than a sincere shrug of the shoulders where our sources are mute.

    I do think, though, that Professor Downs could have been a little more measured in his critique by giving some of the benefits of doubt that Hacker deserves. At times Downs’ blog post suggests some sort of conscious effort by Hacker to paper over black and native America experience. I wish Hacker had wrung his hands more. But I wish Downs had been more surgical in avoiding the suggestion of personal blame and instead focused on how erasures, silences, and parenthetical apologies still leave our scholarship–even the best scholarship–wanting.

    I hope the ad hominem thrust of this exchange quickly gives way to a discussion of how it is we can use Hacker’s work, or at least how we can enhance his findings, by locating ways to measure the pain and sorrow of the war that extended across racial lines.
    I’m excited to read Downs’ work and see where some of this fruitful conversation might lead. I am grateful that Hacker has given us some new things to chew on. I have already incorporated some of his findings and questions into the classroom. All of this energizes me because it reminds me that after all that has been written by this overwritten war, there is so much at the core that we still do not know.

    Michael DeGruccio
    St. Peter’s College

  3. Jim Downs

    Dear Professor Hacker,

    I agree with your view that we must revisit the subject of the number of people who died during the Civil War and I applaud you for starting this much-needed conversation.

    My concern is that by using the census as the organizing principle for your analysis, you are privileging a 19th century system of categorization. Using 19th century categories fundamentally excludes former slaves and unwittingly propagates a narrative about the Civil War that suggests only white deaths matter. If black deaths matter, then we need to at least try and count them. And by counting them, it is not just the number who served in the military, but the hundreds of thousands, who were displaced, victimized, killed, and became sick and died during the war. In sum, we need to rethink who were the casualties during the war.

    When I made a fuss over your distinction about civilians, I was trying to explain that many of white soldiers died from dysentery, smallpox, malaria, and pneumonia, not from battle. My point is that in the same exact Union camps, black people often died of the same diseases; therefore, they must be included in the count and not segregated off in a separate category.

    Further, since the turn of the 20th century, writers from Suzie King Taylor to W.E.B. Dubois to contemporary historians, such as Barbara J. Fields, have been documenting the black experience during the war to avoid this problem. By marshaling extensive evidence about the presence of black people on battlefields, in the camps, and in the military, they have established a framework of analysis that avoids making a distinction based on an empirical justification, like the census, that summarily omits the black experience. Moreover, the political force of the historiography of the last century or more has been to challenge the very ways that 19th century people discussed, conceptualized, and framed the war experience. For instance, the recent publication of Stephanie McCurry’s award-winning book, Confederate Reckoning, reframes the category of “slave” and “women,” which invariably calls for a reexamination of how women and former slaves appear in the archives, the census and by extension the death toll.

    That said, I am not arguing that the census data should not be used, I am simply arguing that using it as a sole primary source (which may be a disciplinary difference) will invariably make former slaves invisible during the war. Finally, I want to reiterate that I think this is the beginning of an important scholarly discussion, and I want to think more rigorously about what is at stake in making certain claims about the number of people who died during the war.

  4. [...] came across this article by James Downs today in the Oxford University Press Blog assessing Dr. J. David Hacker’s New [...]

  5. [...] article by a Connecticut College professor says a revised Civil War death toll would be even higher – [...]

  6. J. David Hacker

    I sadly agree with DeGruccio that my response was too steamed and personal. I wish I had taken a break before hitting that submit response button. My apologies, Professor Downs. In my defense, being accused of implying “that former slaves’ experience is not part of the nation’s history” stings. It’s a substantial allegation to level toward a historian. I certainly don’t believe that to be true and thought my article clearly and explicitly stated that mortality among former slaves and freed people was substantial.

    For the sake of argument, let’s skip my explicit statement and agree that my sources and methods implicitly make black civilian deaths invisible. But one must also grant that they render white civilian deaths invisible. White civilians also died from dysentery, smallpox, malaria, and pneumonia. My estimate of 750,000 excess male deaths included *both* black and white military deaths and excluded *both* black and white civilian deaths. So if the article’s sources, methods & conclusions privilege one category over another, it’s military over civilian, not white over black. I would have been less steamed—and may have even agreed with you—if you had argued that the article largely sidestepped the important question of civilian deaths (black and white).

    Finally, I’ll repeat that I am looking forward to your book and what you’ve uncovered about this relatively unexamined part of the Civil War and its aftermath.

  7. [...] on the impact of disease on African Americans during the war and in the decade after, argues that even these newer numbers are too low, because they overlook the deaths of many tens of thousands of black laborers, slaves and [...]

  8. [...] “If former slaves were included in this figure, the Civil War death toll would likely be over a million casualties,” he writes at OUPBlog.  [...]

  9. [...] the myths of emancipation permalink buy this book read more Posted on Thursday, July 26th, 2012 at 8:30 am [...]

Leave a Comment

Your email address will not be published. Required fields are marked *