The bloodlust of the Civil War historian

Having put in a few bad words for Mark Neely's Civil War and the Limits of Destruction, let me now share a few moments of enjoyment.

First there is a general tone of exasperation with Civil War history which suits me very well. Not all the complaints rock my own hobbyhorses, of course.

For instance, Neely is unhappy with ACW historians' readings of the Mexican War; they have not incorporated the social lessons within the regimental histories that have been issued since the Centennial. Myself, I think this is missing the point of "What the Civil War historian needs to learn from the Mexican War." I'll save that for a series of posts.

More convivial to me is Neely repeatedly mining James McPherson for examples of ahistoricisms and analytic errors. I need to do more of that here.

Neely's last chapter is the starting point for an altogether new book, an inquiry into why Civil War historians avoid numerical analysis whilst they grossly abuse the few numbers they deploy.

Here we have a theme dear to this blog: you've seen it in posts on the surgeon's morning reports, Lincoln's ongoing sub rosa dialog with McClellan on desertion, on the use of round numbers in battle estimates, and in the ignorant modern criticism of contemporary strength estimates.

Neely starts off noting that no one has ever bothered updating Fox or Livermore - a huge tell on Civil War historians, I think - and thus"we can add to our 'to-do' list a future sophisticated statistical assessment of the traditional figures given for losses in the Civil War."

The current crop of deeply misunderstood casualty numbers, Neely notes,
"serve Civil War historians themselves, for example. The emphasis on unequaled bloodiness has become a way for those of us who write on the war to impress our readers with the importance of our subject. The number killed, usually put at around 620,000, exceeded the number of American soldiers killed in all our wars put together... It is a horrifyingly impressive figure..."
He then proceeds to take it apart, noting it is used "to sensationalize Civil War history."

He notes the figure totals friendly and enemy dead - producing a unique, one-of-a-kind aggregate.
If we consider the Civil War casualties one "country" at a time, then the 360,000 Union dead do not equal even the 407,000 Americans killed in World War II.
He then explores why people use the combined totals.
If the casualty figures are meant as a measure of tragedy, then they certainly are tragic. But if they are meant as a measure of the intensity of fighting, destructiveness, mercilessness, and hardness, then to combine the two [figure sets, Union and CSA] is unfairly to have doubled the intensity ...
(Emphasis in the original.)

He notes this is "a simple statistical fallacy." He also notes the tendency to take the total of 620,000 and extrapolate the losses against the current population base. "That escalation of numbers typifies the tendency ... toward sensationalizing results."

As a further deflator, Neely uses readily available statistics to identify death figures from disease: 225,000 for the North and 194,000 for the South. Suddenly, we see combat deaths combined for North and South total 201,000, not (he notes) 620,000. If we disaggregate blue from grey, we find 135,000 combat deaths for the North and 66,000 for the South. "In absolute numbers, then, the North suffered deaths equal to about two and a half times the number of Americans killed in Vietnam." [...] The point, put another way, is to show that the claims of "bloodiest conflict" can be qualified so as not to make the Civil War exist in some unfathomably violent category by itself."

Neely, looking for a relevant comparison, turns to casualties in the Crimean War, seeing an aggregated death toll of 640,000 over a two year period. "... the Crimean War generated as many soldiers' deaths as the American Civil War did in about half the time..."

He reaches a conclusion that we have previously seen in Nosworthy's Bloody Crucible of Courage:
Repeated assertion of the destructive nature of the Civil War may, in fact, only serve to remind readers of the provincial nature of American history-writing.
In a look back at his own work in this volume, Neely imagines he has put paid to the idea that the ACW was a "total war" and muses, "Destructiveness, shorn of the greater overarching concept of 'total war,' now leads nowhere. It is not clear what the cult of violence in writing about the Civil War now serves."
It cannot possibly be true that we are now at risk of underestimating the destructiveness of that war. Instead, we consistently underestimate many other important features of the conflict because of the fixation on violence.