During the days after Christmas in 1864, residents of Savannah were just becoming acquainted to life under Federal occupation. Even the troops in Savannah were still setting up, having transitioned over the span of barely two weeks from a light order field army, to siege operations, and finally to winter quarters of sorts. Yet Major-General William T. Sherman was already looking to the next campaign. The details of such were still in the air somewhat, but clearly the armies which had crossed Georgia would not remain garrisoned around Savannah. Such a large body of healthy, veteran troops would move to press the Confederates in other sectors.
But Savannah was too good a port for the Federals to abandon entirely. Sherman’s plan was to leave the minimum garrison possible in Savannah as his forces moved out on the next expedition. Good, well positioned defenses were the key to reducing the manpower allocated to the garrison to a minimum. Toward that end, Sherman asked his chief engineer, Captain Orlando Poe, to survey the area in and around the city. On December 26, 1864, Poe offered his brief report:
In accordance with your instructions, I have the honor to submit the accompanying rough sketch of plan for the defense of this city. I have reduced the garrison to the lowest probable limit; a smaller one would render it difficult to use any part of it for such offensive operations as might be desirable. The proposed line will be so close to the city that some of the buildings will have to be torn down, and in case of attack all parts of the city will be under artillery fire. Still, the presence of the women and children of the enemy within our lines will render such a fire extremely improbable; and should it be decided by the enemy that they ought to bombard the city, all stores and other valuable property will be quite secure at or near the levee. It is proposed to hold Fort Jackson only because a temporary occupation of it by the enemy would cause us serious inconvenience; to destroy it would require much labor, and even then its site would remain, which would be as detrimental to our interests as the fort itself. Fort Boggs should be dismantled, and so much of it as can give a fire upon the city should be destroyed, because, being an inclosed work, an enemy might effect a lodgment and hold it for a limited time, much to our annoyance. All the remainder of the enemy’s old line, being open to the rear, can do us no injury, and can therefore stand as it is. It is a good line, but too extensive for any garrison that will probably be left in the city; it would require 15,000 men to man it completely. The accompanying sketch does not show the character of the works proposed, but merely the approximate position of the line. The line of works should consist of a system of detached redoubts, in defensive relations, which could be connected by infantry parapet at our leisure.
Poe’s map is lost somewhere between 1865 and today. But the line of works he mentions here does appear on the map submitted for the Official Records (that surveyed by Poe to accompany Sherman’s report):
The inner blue line, well back of the old Confederate works, is that built under Poe’s supervision. As Poe suggested, the line is very close to the city itself, much closer than the Confederate line. It also covers the rear of those former works rendering them useless for any attacker. Most of the ground between the new Federal line and the old works was open field… rice fields with defensive qualities well known to the Federals at that time.
As Poe mentioned, the exception to all this arrangement was Fort Boggs, and would need be dismantled. I would offer that the work towards that goal was not completed with vigor. Remains of the fort appeared on maps right up into the 20th century, and traces of the line there are visible today.
Fort Jackson… Old Fort Jackson to all… remained as an outpost. Far too much effort to tear down the brickwork. The fort also served as a communication link and control point downriver to Fort Pulaski.
Poe’s line would require less than 10,000 men. Sherman sought to pull troops from the Department of the South, but Foster was already too thin. Further he hoped to allocate invalid troops (who were not capable of field duty) from the Department of Mississippi, for this duty. Thus a call back to Nashville for troops originally from the Armies of Tennessee and Georgia (Fourteenth, Fifteenth, Seventeenth, and Twentieth Corps) for the task. But this was both impractical and would provide far too troops for the task. The ultimate solution was to pull troops from Virginia – Second Division, Nineteenth Corps under Brigadier-General Cuvier Grover from the Army of the Shenandoah. Grover received orders for transit to Savannah in early January, and was able to relieve Brigadier-General John Geary on January 19, 1865. (The movement took place in just over two weeks. A fine example of the operational mobility possessed by the Federals in the later stages of the war.)
One portion of Poe’s report that I would call out for thought: “Still, the presence of the women and children of the enemy within our lines will render such a fire extremely improbable….” Somewhat a twist of circumstances here with a Federal officer offering the presence of civilians would preclude indiscriminate firing. Was this to say the Confederates would never fire upon the civilians? I would not go that far. The precise military tone of Poe’s assessment continued in the second half of the sentence, “… and should it be decided by the enemy that they ought to bombard the city, all stores and other valuable property will be quite secure at or near the levee.” So it was not so much that the Confederates would not fire on a city filled with civilians, but that their guns could not hit anything of importance. In short, bombarding the city under those circumstances was not considered worthwhile. Such, of course, was very apparent to any Confederate threatening Savannah, as their generals had shrugged off a similar threat from Sherman just over a week earlier.
(Citation from OR, Series I, Volume 44, Serial 92, pages 811-12.)