What’s In a Number?

There’s been a lot of talk lately about numbers, specifically in assessing the number of Civil War dead. J. David Hacker’s essay “A Census-Based Count of the Civil War Dead” appeared in a 2011 issue of Civil War History. The essay received a great deal of attention and a shorter version appeared in the New York Times Disunion blog. Keith Harris offers some thoughts here. More recently, Nicholas Marshall published an essay in The Journal of the Civil War Era that can be seen, in part, as a response to Hacker. Whether Marshall’s essay garners as much traction has yet to be seen, but it did receive a thorough critique from fellow blogger Vince Slaugh at Lancaster at War.

The National Park Service at Andersonville has also been counting the Civil War dead. Here is the latest update for 150 years ago this May 8.

2 comments… add one
  • J. David Hacker Oct 20, 2014 @ 8:35

    Fox and Livermore’s figures were indeed speculative, based as they were on incomplete sources, guesswork, and several questionable assumptions. These omissions and assumptions, especially Livermore’s assumption of equal rates of non-combat (disease) mortality in both armies, biases their estimates downward. It matters not how exhaustive they were in their research—clearly both men were dedicated, hard-working historians—no amount of effort in the archives will count a death that went unrecorded in the sources. And it’s unreasonable to expect that an estimation procedure based on faulty assumptions and guesswork will return a precise and accurate figure. Unfortunately, Fox and Livermore did not suggest upper or lower limits to their estimates. Livermore did indicate that his figures were likely to be conservative, but didn’t suggest a range. I’d argue that this failure to consider the possible error is unhelpful rather than helpful, but understand that not everyone is comfortable with historical uncertainty. But we are uncertain (or should be). So perhaps it is unhelpful of me to suggest, somewhat tongue in cheek, that Livermore’s estimate should be 620,000 +30,000 to +230,000. If a researcher wants to cite a figure free of assumptions and uncertainties, he or she should cite the figures counted by the War Department. In my opinion, the only certainty about Livermore’s estimates is that it’s too low.

  • Garry Adelman May 9, 2014 @ 2:32

    I did not jump on the Hacker bandwagon when his research was published and therefore did not change the Civil War Trust casualty figures. His conclusions in what he included and the broad and therefore wildly unhelpful range he provides does nothing in my opinion to unseat the admittedly also largely speculative figures assembled by Fox and Livermore long ago. http://www.civilwar.org//education/higher-number.html. The unhelpful range reminds me of a snow forecast in places where people don’t get much snow–2″ to 36″ expected overnight…buy milk, bread and toilet paper!

Leave a Reply

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