Page:Memoirs of Henry Villard, volume 2.djvu/257

 Longstreet pronounced in favor of it, provided it could be executed secretly and with great rapidity and with no less than 20,000 men, and provided further that the remainder of the army would be withdrawn to a strong position behind the Chickamauga until the return of the expeditionary force. (This is Longstreet's version of what occurred at the council, but General Hardee, when requested in writing, some months later, to confirm it, replied that he did not recollect the suggestion of the temporary withdrawal behind the Chickamauga.) It was decided to try a coup against Burnside. Longstreet was to undertake it with two divisions, against which he claims to have protested as too small a force for quick success, but he was overruled and yielded. Marching orders were issued, and the two divisions, with an extra complement of artillery, numbering not much over 16,000 men, were under way by the 5th. But, owing to all sorts of unforeseen impediments and unexpected delays, the march was not rapid, but very slow, so that the rebels did not appear near Knoxville, behind whose fortifications Burnside had concentrated most of his command, until the 18th, and actually delivered the famous unsuccessful attack upon the place only on the 29th.

A decided improvement had taken place in our life since the raising of the blockade of the river. Mails and newspapers arrived again daily and banished the oppressive feeling of isolation from the rest of the world. While the mass of the army was, of course, ignorant of the plans of the commanding generals, the presence of Grant and the visit of Sherman were generally looked upon as sure indications that something was up, and that action would soon supersede our passiveness. Aware as I was of the reconnoitring of General Smith and his excursions with the commanding generals to the north of the town, I could guess their intentions without asking any questions. The dread of being shut up maybe for the winter months in Chattanooga, and the long continuance of the prevailing