Numbers & Losses in the Civil War in America, 1861-65

by Thomas L Livermore, Carlisle, John Kallmann, 1996

This is the most quoted reference work in Civil War historiography and for all that, it has rarely been in print since it came out in 1900. Finding this new edition shelved in an ordinary chain store was my nicest shopping surprise in 1998.

In this slim but complex volume, Livermore undertakes three major projects. First, to estimate the army strengths of the two sides overall (and in this he devotes considerable space to the Rebels, whose records are such a shambles). Second, he estimates the respective strengths of the opposing sides in every battle of the CW that produced more than minimal casualties. Third, for each battle he analyzes, he creates a ratio of damage received and damage inflicted by the respective sides, an analytic he calls "hit ratios."

It is with respect to these hit ratios that McClellan the tactician -- the battlefield commander -- emerges head and shoulders above all Civil War contemporaries. (This topic will be presented in a separate, future essay on this site.) However, since the hit ratios are contingent on Livermore's strength estimates, they are, ultimately, compounded hypotheticals and need to be treated with care.

Why do we need Livermore's estimates, when we have the Official Record (OR)? Livermore was a revisionist -- and his revisions have stood for a century. He was dissatisfied with the misleading picture presented by OR information. Typically, the most refined numerical strengths available from the OR are the official unit returns, showing troops present and accounted for (or present for duty), apportioned for each unit in a command. Careless historians, i.e. very many historians, used these numbers to determine respective strengths for campaigns and battles. Livermore makes two standard changes to OR figures. First, for a battle, he deducts the strength figures for units not present. Second, for units present he tries to arrive at an effective strength by deducting seven percent, which figure he allows for soldiers on detail, etc.

Is seven percent enough? Warren Hassler, the McClellan biographer, felt that even a 20 percent deduction from OR figures was conservative in arriving at battlefield strengths. Here we come to the stickiest points in the Livermore work: illness and AWOL (absent without leave). It is not hard to quantify Union illnesses: the Medical Record published by the Surgeon General in 1875 provided plenty of data that could easily have been used by Livermore to make better estimates. He didn't. AWOL is harder to quantify, but occasional data is available. Lee claimed about 50 percent AWOL before the battle of Antietam. Livermore scorns the figure, perhaps rightly. Figures in Wainwright's diary (not availble in 1900) suggest an 80 percent AWOL rate or higher for Burnside after Fredericksburg. These suggest the magnitude of the problem facing anyone proposing to make estimates of battlefield strengths.

Livermore is nonetheless a huge improvement on the OR and a reasonable starting point for a historian's own estimates using more recently uncovered information. The shame of modern popular CW historians is their general unwillingness to revise Livermore and their clear desire to steer clear of numbers issues as quickly and cleanly as possible. This takes the grotesque form of issuing no numbers, very very round numbers or even OR numbers. Now when an author, such as Stephen Sears, passes over Livermore entirely to reach back for discredited OR statistics, you can be sure that an agenda of some sort is in play. In Sears' case, he has publicly committed to the position, in book after book, that McClellan outnumbered his opponents on the Peninsula. In Livermore we see a different picture:






 Seven Pines






 Gaine's Mill



 Peach Orchard to Malvern Hill




Thus, in Livermore, McClellan not only fights slightly outnumbered (after Williamsburg) but produces a matchless record of casualties inflicted compared with casualties sustained (see McClellan's Statistics). In the prevailing anti-McClellanism of pop history, this is bitter medicine for too many.

A few words about this edition. Kallmann is to be commended for producing an attractive, clothbound hardback and resetting the type for clarity. There is a new introduction which is useful. What is lacking is a few essays on loss estimates since Livermore. It would not have been difficult to commission a few articles on "Livermore Today" or some such, but this publisher has not done that. Perhaps the next will.

So many of us now depend on our CD-versions of the OR for research. That OR comes with Fox's Regimental Losses but not with Livermore. Until these three indispensible works are bundled electronically, this edition is a necessary roundout to the set.


Amazon link

(c) 1998 The McClellan Society