Atlanta 1864: Last Chance for the Confederacy

APPENDIX TWO Numbers and Losses

  Numbers and Losses Readers do not need much exposure to Civil War history to realize that numerical data concerning the strength of opposing armies can generate much dispute. The same is true for the casualties the armies suffered. Each general, it seems, tried to convince readers that his forces were outnumbered, but, even so, they managed to inflict very heavy casualties on the enemy—thanks, of course, to his superior ability as a commander. Regulations governing the opposing armies from April to August  required them periodically to submit strength reports (“strength returns ”). These data are the “official” information we have as to how many men served under a particular general’s command at a given time. Like all statistical data, however, they can be manipulated, and they should be used with great care. Fortunately, in most cases, it is not important to ascertain the exact numbers of the opposing forces or the casualties they suffered. Returns from Sherman’s armies from April to August  show his strength as follows: Apr.  May  June  July  Aug.  , , , , , These numbers are called the “effective strength” of Sherman’s forces, but as used in these returns that is an unofficial term. The context as well as the numbers themselves make it clear that they include officers. Thus the term as used with these data does not have the same meaning it does when used with Confederate strength returns (see below). For purposes of comparing the two forces, it is clear, these numbers should be called Sherman’s “present for duty” strength. Numbers and Losses  Confederate “present for duty” strength returns show the following totals: Apr.  June  June  July  Aug.  , , , , , The Rebels reported their “effective strength” as follows: Apr.  June  June  July  Aug.  , , , , , In Confederate terminology an army’s “effective strength” excluded all enlisted men not “present for duty, equipped” and all officers. Thus, any lieutenant, a private detailed for the day as orderly at regimental headquarters, or a cavalry sergeant whose horse had thrown a shoe would not be counted. Confederate “effective strength” in the Atlanta campaign ran at about  percent of the Rebels’ “present for duty” strength. Confederate generals liked to use their “effective strength” in discussing their operations because that number made their forces seem smaller, thereby magnifying any successes they chanced to win and helping to explain any defeats they suffered. In comparing opposing forces, we must be careful not to compare Sherman’s “present for duty” strength with the Secessionists’ “effective strength,” as is often done. To get comparable figures, we must convert “effective strength” to “present for duty” strength or vice versa. To do so, multiply the total “effective strength” by ., or the “present for duty” strength by . (ratios derived from Confederate strength returns above). Confederate “present for duty” strength as a percentage of Sherman’s “present for duty” strength (these numbers give us by far the best comparison of the opposing forces): Apr.  May – June  July  Aug.  .% .% .% .% .% Confederate “effective strength” as a percentage of Sherman’s “effective strength” (former as reported by Johnston and Hood; latter determined by conversion formula above): Apr.  May –June  June  July  Aug.  .% .% .% .% .% (These Confederate data should be used with extreme care. For one thing, they are incomplete. They do not include the Georgia militia troops serving with the army in June, July, and August. For another, the August  Appendix Two  numbers use Wheeler’s cavalry strength as of August  because he submitted no later return.) The most interesting of these numbers is that for August . The Rebels were then stronger relative to Sherman than they had been at any other time in the campaign. This unexpected result was owing to the drop in Sherman’s strength in August (, men) as some of his units completed their three-year term of service and went home for discharge.) The data we have on casualties in the campaign are also very imprecise and equally frustrating. Although many commanders on both sides reported their individual unit’s casualties in this or that engagement or day’s operations, meaningful overall figures for particular battles are difficult to find. In most cases they do not exist. Federal losses for May, June, July, and August (and September – for the Armies of the Tennessee and of the Ohio) were reported as follows: Killed Wounded Missing Aggregate Officers Enlisted Officers Enlisted Officers Enlisted Army of the Cumberland  ,  ,  , , Army of the Tennessee  ,  ,  , , Army...