What Really Counts

Dimitri Rotov has been blogging recently about how military historians use/misuse/abuse estimates of military strength and losses.  Rather than summarize his position, I point you here for a summary of the discussion:

http://cwbn.blogspot.com/2006/10/historians-as-numerologists-recap.html

Once you are there, you can make your way back to his previous commentary on the subject.

I think Dimitri has a point.  Over the last sixty or so years, American historians have been engaged in producing massive documentary editions of the writings of various people, movements, and events.  And we are talking large-scale enterprises: The Papers of Ulysses S. Grant, which published volume one in 1967, is still going strong twenty-seven volumes later.  See

http://twister.lib.siu.edu/projects/usgrant/paper1.html.

Such enterprises suggest the importance of the documentary record, and yet we have no modern version of such an enterprise to update, even replace, books such as Thomas Livermore’s Numbers and Losses in the Civil War or William Fox’s Regimental Losses in the American Civil War.  By the way, Livermore is now online:

http://www.carlisle.army.mil/cgi-bin/usamhi/DL/showdoc.pl?docnum=447

as is Fox:

http://www.civilwarhome.com/foxspref.htm

Recent research has called into question traditional estimates of strengths and losses in several battles and campaigns where traditional estimates have become entrenched in the literature.  There’s new information that calls into question strength estimates during the Seven Days (good news for McClellan fans) and numbers and losses during the Overland Campaign of 1864, especially the June 3 assault at Cold Harbor.  The last finding is key to Gordon Rhea’s recent account of the battle, and no one who studies Rhea’s account will ever again recite the “7,000 down in 30 minutes”  (or variations of time thereof) line of Civil War lore. 

It would be a useful endeavor to ascertain estimates that are as accurate as possible, as such estimates are at the core of many assessments of combat performance and leadership.  But I want to go beyond this to offer a more intriguing observation, and that is that there is often something misleading about assessing the outcome of a battle or campaign or determining good generals and bad generals based primarily on some sort of numbers game.  One could half the losses on the June 3 assaults at Cold Harbor, for example, without having to reassess the poor performance of the Union high command that day (a close examination of Meade and his corps commanders shows that they have to share more than some of the blame usually unloaded primarily on Grant).

In some cases, of course, adjusting estimates in light of new information changes the story.  If Grant inflicted more losses on Lee during the spring of 1864, and if some of Grant’s losses (I speak here of the “lightly wounded”) can be attributed to the impact of impending expiring enlistments (short-timer’s syndrome), then reassessments are in order, since the original assessments hinged so much on comparing numbers and losses.  That said, however, would that really cause us to see the campaign as more successful for Grant, or would it cause us to reassess what constitutes victory and defeat and how we as scholars make such judgments?  Let’s say that Lee’s strength was at least equal to or perhaps exceeded McClellan’s strength on June 25, 1862.  How would that change an account of the ensuing campaign?  Oh, sure, we’d have to toss out notions that McClellan was inept because he “lost” despite outnumbering Lee, but, once we set that aside, how much else would change?  How much would it change?  Or would it force us to look at assessing outcomes in some way that placed force strengths and body counts in appropriate context (the definition of which would depend upon the context)? 

Just some food for thought and discussion on how we do what we do, why we do it, and whether that’s how and why we ought to do it.            

Comments (4) to “What Really Counts”

  1. The problem with Dmitri’s analysis is that it is almost always designed to make
    McClellen look better. So I confess I don’t pay a lot of attention to Dmitiri.

    JFE

  2. Let’s set that aside for the moment, because one of my points is that while “correct” numbers may knock down certain explanations, would an account of the Seven Days otherwise be all that different?

    After all, we often assess and explain victory and defeat without reference to numbers and losses. Take the Crater for example. Very few historians could recite the numbers engaged and the losses off the top of their heads.

  3. Brooks, — Interesting post. I wonder to what extent our assessments of generals are based “primarily” on numbers. McClellan is the obvious example, but I agree that a reassessment of Cold Harbor would probably not bring about a signficant change in our judgement of the Union high command. In the case of the Crater the numbers pale in contrast with judgements about Burnside, Meade, and Grant. I simply don’t know enough, but are there additional examples where the numbers are driving our judgements?

    Dimitri’s problem is that his generalizations about the Civil War tend to be about McClellan and raising his stature. I should also say that a better grasp of how armies recovered from campaigns may also give us a clearer view of the high command. Here I am thinking of Kent Brown’s massive study of Lee’s retreat from Gettysburg.

  4. I agree with Dmitri that the numbers game is the most mis-understood aspect of the ACW, and the aspect most likely to be abused – be it intentional abuse or otherwise.

    Too many figures are accepted at face value, in both the strength and losses categories. The more careful studies – those that tend to work from the ground up, starting with the regiments and battalions – have shown some startling differences from the ‘stock’ numbers in campaigns like Antietam, the Seven Days, etc. The most common trend is to under-estimate CSA engaged strengths and losses. I do think that the use of these traditional under-estimates is a problem, and has significantly affected the way most people view the Civil War and the men who fought it.

    Antietam is the classic example. If we assume that McClellan had a better than 2-1 advantage in strength and threw that advantage away due to timidity, this is a sharply divergent view than if we know that McClellan probably had no more than 70,000 men at that battle (a very different number than the accepted 87,000) and that something like 20% of that 70,000 had been in the army less than a month and were so raw that they were hardly a match for veteran troops; this would all greatly affect our preceptions of the battle.

    On the other hand, the above stats can quickly become bricks in the wall of apology someone might build to excuse McClellan from ALL blame.

    Numbers are important. Accurate numbers are indespensible for understanding a battle in detail. However, they do not replace other analysis, or always demand a re-assessment (of Cold Harbor, for instance, as per Brooks’ comments.)Understanding the nature of Bragg’s personality can be as critical in evaluating the performance of the Army of Tennessee as any rigorous accounting exercise.

    Dave Powell