John M. Schofield, Forty-six years in the Army

[123] army was composed of three separate armies, or such portions of them as could be spared from their several departments, united for that campaign. General Thomas was, naturally enough, disinclined to part with any of his troops, and the troops did not wish to be separated from the old army in which they had won so much honor, nor from the commander whom they revered. Besides, General Thomas had had much greater experience in the command of troops in the field than I, and General Sherman, if he thought of it at all, may well have doubted the wisdom of diminishing the command of the one to increase that of the other. I do not know whether this matter was discussed at all before the opening of the campaign, certainly not by me, who would have been restrained by motives of delicacy, if by no other, from mentioning it. But in fact my ambition was then limited to fighting well and successfully with the single corps under my command. It was only after experience had drawn attention more pointedly to the evils resulting from faulty organization, and success had inspired legitimate confidence, that this subject became matter of much thought and some discussion.

But this faulty organization continued to the end of the Atlanta campaign, and was, as I think will clearly appear, one of the causes of many of the partial failures or imperfect successes that characterized our operations. General Thomas's command often proved unwieldy and slow from being larger than one man could handle in a rough and in many places densely wooded country, while the others were frequently too small for the work to be done. It was often attempted to remedy this defect by ordering a division or corps of the Army of the Cumberland to ‘cooperate with’ or ‘support’ one of the others in making an attack; but military experience has shown that ‘cooperate’ and ‘support’ mean, in general, to do nothing effective. The corps commanders,