previous next

The opening of the Atlanta campaign.

by W. P. C. Breckinridge, Colonel, C. S. A.
In his paper “Opposing Sherman's advance to Atlanta,” General Joseph E. Johnston--clarum et venerabile nomen--writes [see p. 263]:
Cantey with his division arrived at Resaca that evening (7th), and was charged with the defense of the place. During the day our cavalry was driven from the ground west of Rocky-face through the gap. Grigsby's brigade was placed near Dug Gap,--the remainder in front of our right. About 4 o'clock p. M. of the 8th, Geary's division of Hooker's corps attacked two regiments of Reynolds's Arkansas brigade who were guarding Dug Gap, and who were soon joined by Grigsby's brigade on foot. The increased sound of musketry indicated so sharp a conflict that Lieutenant-General Hardee was requested to send Granbury's Texan brigade to the help of our people, and to take command there himself. These accessions soon decided the contest, and the enemy was driven down the hill. . . .

Information had been received of the arrival of the Army of the Tennessee in Snake Creek Gap on the 8th. At night on the 9th General Cantey reported that he had been engaged with those troops until dark. Lieutenant-General Hood was dispatched to Resaca with three divisions immediately.

It so happened that the brigade of Kentucky cavalry was present at Dug Gap and Snake Creek Gap, and that the regiment I commanded — the 9th Kentucky Cavalry--was in front at both places; and it may not be improper to put on record an [278] account of those affairs, and thereby correct the unintentional mistakes in the meager statements given above.

The winter having ended, and all possible preparations having been made, the operations known as the Dalton-Atlanta campaign opened on May 5th, 1864, by the advance of General Thomas on Tunnel Hill, and on May 7th the withdrawal of our forces within Mill Creek Gap marked the beginning of the long retreat. Including the corps of General Polk, then under orders to join him, General Johnston had under his command, available for strategic purposes, between 65,000 and 70,000 men of all arms. It was a superb army of veterans, with implicit confidence in its general, and capable of great achievements. Deficient to a certain extent in supplies, it had enough for any possible movement its commander might order. Being a Confederate army, it necessarily was inferior to the army before it in numbers, equipment, and supplies. This was generally the case. It was necessarily so. With a white population of 5,000,000 to over 20,000,000; with no market, no ships, no factories, no credit; against a people commanding the sea, rich in all resources, and with all the world to buy from,--it was the fate of the Southern armies to confront armies larger, better equipped, and admirably supplied. Unless we could by activity, audacity, aggressiveness, and skill overcome these advantages it was a mere matter of time as to the certain result. It was therefore the first requisite of a Confederate general that he should be willing to meet his antagonist on these unequal terms, and on such terms make fight. He must of necessity take great risks and assume grave responsibilities. While these differences between the two armies that confronted each other in the mountains of North Georgia existed, they were no greater than usually existed, and for which every Confederate general must be presumed to have prepared. I repeat, ours was a superb army. While it had met defeat, and knew what retreat meant, it had fought battles which were and are among the bloodiest in all the annals of war; and it felt that under Johnston it could parallel Chickamauga and renew the glories of Shiloh.

The army lay behind an impassable ridge, through which, on its left flank, were only two accessible gaps,--Dug Gap, less than four miles south-west from Dalton, on the main road from Dalton to Lafayette, and perhaps six miles from Mill Creek Gap; and Snake Creek Gap, some eighteen miles south from Mill Creek Gap. [See map, p. 251.] With these gaps fortified, the left flank and rear of that army were absolutely safe; for while the Rocky-face and Chattooga ridges protected our flank, through these gaps we had access to attack the flank of the enemy if he attempted to make a march so far to the left and rear as to threaten our communication south of the Oostenaula or Coosa. These gaps were capable of easy and impregnable fortification. Dug Gap was a mere road cut out of the mountain-side and really needed no breastworks, for the natural palisades and contour of the mountain rendered easy its defense by resolute men. Snake Creek Gap was a gorge apparently cut through the mountains by the creek that ran through it. It was a narrow defile between Milk Mountain and Horn Mountain, which are merely a prolongation of Chattooga Mountains, and capable of impregnable defense.

These gaps were well known to both armies. Through them ran public roads, and soldiers of both armies had marched through both. Late in February Dug Gap had been seized by an Indiana regiment and held until Cleburne retook it. As early as February General Thomas, knowing that at that time Snake Creek Gap was unguarded, proposed a campaign, the plan being to attract General Johnston's attention by a demonstration on Buzzard Roost, and to throw the main body of the army through Snake Creek Gap, and cut his communications between Dalton and the Oostenaula.

Neither of these gaps was fortified, and on May 5th, when the campaign opened, Dug Gap was guarded by a small command of Arkansas troops under Colonel Williamson, numbering perhaps 250, while Snake Creek Gap was left wholly unprotected. At Resaca, where the railroad crosses the Oostenaula, Cantey's brigade was held on the evening of the 7th of May, on its way from Rome to Dalton.

General Sherman had in hand for attack nearly 100,000 men and 254 guns, divided into three armies — the Army of the Cumberland, commanded by General Thomas, numbering 60,773; the Army of the Tennessee, General McPherson, 24,465; the Army of the Ohio, General Schofield, 13,559. It was a superb army, admirably equipped, abundantly supplied, excellently led. It was veteran, and had known victory. It had pushed its antagonist out of Kentucky with the surrender of Donelson; had captured Tennessee; captured Vicksburg; repossessed the Mississippi River; driven its foe over Missionary Ridge in flight. It knew how to fight, and was willing to fight.

On May 7th our cavalry was driven through Mill Creek Gap. On that night, after we had gone into camp, Colonel Grigsby, who commanded the Kentucky cavalry brigade, was ordered to send a regiment to the front of Dug Gap, to guard the approaches to it. In obedience to that order the 9th Kentucky Cavalry passed over Rocky-face Ridge, and near midnight bivouacked on Mill Creek, about a mile from, and in front of, Dug Gap. Heavy picket lines were thrown out on all the roads leading down the valley. There were several of these roads, and scouts were sent out to ascertain the movements of the enemy. By daylight it was discovered that very large bodies of troops were moving down the valley on all the roads leading to the south. General McPherson had marched from Chattanooga to Rossville, thence west of Chickamauga Mountain to Shipp's Gap and to Villanow, where the road forks--one branch leading down the east foot of Taylor's Ridge, the other leading across toward Rocky-face ; this road again forks--one branch leading through Dug Gap, the other down the valley to Snake Creek Gap. Until McPherson reached Villanow it was only a conjecture as to his course, and until the head of his column turned toward Snake Creek Gap his destination was uncertain. His march [279] was concealed by Hooker's corps of the Army of the Cumberland, which corps, forming Thomas's right, marching from Ringgold via Nickajack Gap and Trickum, hid the flank movement of McPherson. The plan was for Hooker to seize Dug Gap and push forward sufficiently to protect the flank of McPherson, and strike the flank of Johnston if he turned on McPherson; while McPherson, marching through Snake Creek Gap to Resaca, should not only destroy but hold the only railroad tributary to Johnston. The possession of Dug Gap by Hooker not only would render Dalton untenable, but would make a retreat from Dalton by the line of the railroad extremely hazardous, and completely protect McPherson from attack on his left flank . With Hooker descending from Rocky-face on our left flank and rear, McPherson holding Resaca, Thomas, with the corps of Howard and Palmer, pushing to D)alton, and Schofield to his left, our army would have been in a perilous situation.

The march of Hooker and McPherson was discovered early on the morning of May 8th by the scouts of the 9th Kentucky Cavalry, and timely information was given that at least an attack on Dug Gap was certain, and that the columns on the march were very heavy and their movements were guarded by forces too large to be either resisted or developed by the detachments sent out by the 9th Kentucky. On this information the remainder of Grigsby's brigade was ordered to Dug Gap, and reached there none too soon. All possible delay to the march of Hooker's corps was made, but about 2 P. M. Geary's division of that corps drove the 9th Kentucky across the creek and slowly up the mountain-side, until the regiment fell back in its proper position in the gap, where it found the brigade drawn in mere skirmish-line along the edge of the mountain-side. As one-fourth of cavalry soldiers are detailed to hold the horses, I presume that we had about 800 of our brigade in the fight and 250 Arkansas troops; and this handful of men held that gap until nightfall, repelling every assault. After nightfall Granbury's Texas brigade relieved us, but the assault was over. Hooker had failed in his part of the mission. That flank of our army was safe.

The importance of holding that gap was so manifest that Generals Hardee and Cleburne, with their staffs, galloped to the scene to encourage us by their presence and to aid Colonel Grigsby by their suggestions; and though the fight was made under their eye, that command needed no encouragement, and its officers and men knew that they were holding one of the doors to Dalton.

I hold in my hand the official report of General Geary, by whom that attack was made, and on the whole it is a fair and soldierly report. But he is mistaken in his belief that we had two lines of intrenchments, or that we were ever driven from our first position. Our loss was very small — in killed and wounded not a score. He reports that he made that attack with two brigades of infantry and two batteries, being an aggregate of perhaps 4500 men, or about four to one, besides the batteries. Assault after assault was made from 3 o'clock until after dark, and each assault was repulsed with loss. At first, in a mere spirit of exuberant fun, some of the men rolled stones down the mountain-side; but when the effect was noticed they were directed to use these means as part of our defense; great stones were rolled down on the supporting lines on the mountain-sides or at its foot; and as these bowlders would go leaping, crashing, breaking off limbs, crushing down saplings, we fancied we could see the effect of the unexpected missiles. It also proved a valuable resource to us, for without them our ammunition would have given out; indeed it was about exhausted when the attack ceased.

General Geary reports an aggregate loss of 357 officers and men, of whom some 50 were the adventurous advance, who actually reached the crest, only to be made prisoners. After dark our brigade, being relieved by the Texas brigade of Granbury, was ordered to the foot of the mountain to feed and to obtain ammunition.

While this attack had been going on, McPherson had steadily marched toward Snake Creek Gap, to protect which no steps had been taken. Undoubtedly if a cavalry force had been started to Snake Creek Gap at the moment Grigsby was ordered to Dug Gap, it would have reached there before McPherson, and held it during the night of the 8th, during which time infantry support could have reached there. I do not wish to be understood as offering any criticism on these facts; I am merely stating the facts as I believe them to be. Why these gaps were left unguarded, why a prompt effort was not made to hold Snake Creek Gap, I neither pretend to know nor venture to guess; nor do I offer any criticism. That they were not guarded, and that this gave Sherman the easy means of causing the evacuation of Dalton and the retreat to Resaca, is undoubtedly true. That we could have held Dalton or made an attack on Sherman if these gaps had been held is a problem over which military men may differ. Whatever may have been the reason or cause, the fact is that the provision made to hold Snake Creek Gap was an order to Grigsby during the night of the 8th to move his brigade to its mouth. The 9th Kentucky had been on duty continuously for over twenty-four hours; the whole brigade for over twelve hours, and under fire all the afternoon. But with cheerful alacrity the command began its march as soon as it could feed, after being relieved by Granbury — possibly about 10 o'clock. The night was dark, the road rough and unfamiliar, and it was difficult to find guides. But just at dawn we came in sight of the eastern mouth of the gap, and, contrary to our information, found it in possession of the enemy. Colonel Grigsby had been informed that a company of Georgia troops was on picket on the road to the gap, and at or near its eastern outlet. We had not seen that company, and Colonel Grigsby naturally concluded that the troops we saw a few hundred yards before us were those. The usual confusion of an all-night march and the halt of the head of the column had jammed the different organizations somewhat together in a narrow lane. The advanced vidette reported the troops to be Federals. Colonel Grigsby, still supposing [280]

Part of the Confederate intrenchments at Resaca. From a photograph.

them to be Georgians, ordered a small scout to the front. In these few minutes the enemy, having discovered us and being concealed by the character of the ground and the forest, had formed line of battle, while our column had become more confused by many of the men dismounting to rest. Between us and the foot of the mountain was a fallow cotton-field, on the near edge of which was a row of deserted cabins. The road ran along this field a few hundred yards with a gradual descent until it passed through a fringe of willows and underbrush, beyond which there were other open fields, and then on both sides of these open fields were also thick woods.

Suddenly a long skirmish-line broke from the woods, ran to the fringe of willows, and directly through toward the row of cabins, keeping up a brisk fire as they ran. Behind the skirmish-line was developed a line of infantry. For a moment the fire staggered the head of the column, and the order to fall back and form could not be executed. The 9th Kentucky was in front, and very quickly its front companies were dismounted and a dash made for the cabins. Fortunately our men reached them first and drove the Federal skirmishers back. This gave breathing time, of which immediate and brilliant advantage was taken by Major J. Q. Chenowith, who led a portion of the 1st Kentucky, on horseback, on a detour to the right through the woods until he reached the fringe of willows, when at full run he charged the skirmish-line on the left, and the dismounted men of the 9th Kentucky charged on foot through the open field. The audacity of this sudden and unexpected dash caused the skirmish-line to run at breakneck speed, and the line of infantry to halt and to await reenforcements. This gave ample time to form the brigade for its day's work of retreating fight.

The immediate result of this was a delay to the Federal column of several hours, increased caution on the part of McPherson in his march during the day, and prompt information of his movement to our army headquarters.

The force under McPherson was so large that our small brigade of cavalry could not force it to develop its line. All that was possible was to cause the march to be as slow as that of a skirmish-line. This was done. It was late in the afternoon when McPherson drove us into the works before Resaca, which were defended only by Cantey's brigade and ours. It was a gloomy prospect. We knew that McPherson had a force of from 15,000 to 20,000, and that there was no possibility of our receiving any reinforcements that afternoon or night. One serious attack by McPherson, and Resaca must have been captured.

Fortunately McPherson knew that Hooker had failed in his attempt to seize Dug Gap, and that consequently the road from Dalton was free to any Confederate column moving on him. The intrenchments at Resaca were formidable, and when McPherson felt the lines, the response was resolute and spirited. As Hardee came to reinforce us at Dug Gap, so here Hood joined us. He and part of his staff came to share our fate. Calmly we waited for the inevitable assault. We did not doubt that it would be made. McPherson was young, ambitious, and able. In our ranks he was [281] accounted the equal, perhaps the superior, of Sherman. Here was an opportunity that Sherman might well say “does not occur twice in a single life” ; and not for a moment did we doubt that such a soldier, with such an army, would seize such an opportunity.

I recall the scene, as a group stood on a knoll and watched the skirmishers advance. As the puffs of smoke arose in the distance, as the sharp-shooters paid compliments to this group, General Hood rode up, and after a few moments' gaze turned the head of his horse and rode a few feet, and by motion called Colonel Grigsby to him; in another moment Grigsby called me, and General Hood said in a cheery yet grave tone, “We must hold until night.”

Just at dusk the enemy began to fall back, and to our surprise the retrograde movement ended near to the point at which we had commenced our fight in the morning.

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 United States License.

An XML version of this text is available for download, with the additional restriction that you offer Perseus any modifications you make. Perseus provides credit for all accepted changes, storing new additions in a versioning system.

hide Dates (automatically extracted)
Sort dates alphabetically, as they appear on the page, by frequency
Click on a date to search for it in this document.
May 7th (3)
8th (3)
February (2)
May 5th, 1864 AD (1)
May 8th (1)
May 5th (1)
9th (1)
hide Display Preferences
Greek Display:
Arabic Display:
View by Default:
Browse Bar: