Share This Article

Early morning on August 18, 1862, found Major General J.E.B. Stuart and his staff resting fitfully on the front porch and lawn of a house in the tiny community of Verdiersville, Virginia. They had spent the night there waiting for Brigadier General Fitzhugh Lee’s cavalry to arrive so that an attack on nearby Union forces could begin. Stuart was no doubt irritated by Lee’s tardiness–perhaps he was planning how he would greet Lee when he finally arrived.

When Stuart and his party made camp the night before, they could find no trace of Lee, who had been directed to have his troops in the area by that time. Residents had seen no cavalry, and Stuart and his men had settled in to wait. During the night, Stuart had sent his adjutant general, Major Norman Fitzhugh, to find Lee’s cavalry and hurry them on.

As Stuart lay on the porch in the early morning light, a group of cavalrymen approached. Thinking it was Lee’s force, he sent out two officers to greet them. In short order shots were fired, and the officers dashed back with the 5th New York and 1st Michigan Cavalry regiments close on their heels.

Leaving his coat, haversack and hat behind, Stuart ran to his horse and, along with his staff, scattered into nearby woods. The Union troopers broke off the pursuit, stopped to gather what they could at the house, including Stuart’s famous hat, and then rode back to the Union lines. Accompanying the Federals was Norman Fitzhugh, whom they had captured the night before. After reaching Union lines, they dispatched Fitzhugh and Stuart’s bag to Maj. Gen. John Pope, who, upon seeing a letter Fitzhugh was carrying that detailed Robert E. Lee’s plan of battle, decided to pull back his forces in time to save them from a crushing defeat. Anyway, that’s the oft-repeated story.

In truth, the captured letter had nothing to do with Pope’s decision to withdraw. At the earliest, the letter arrived in midafternoon on the 18th, long after the
decision had been made. Rather than luck, it was systematic intelligence gathering that saved Pope’s army and allowed it to escape the trap that Lee had set at Clark’s Mountain.

In the wake of the disastrous Battle of Cedar Mountain on August 9, 1862, Pope had taken up what he thought was a strong position in the triangle formed on the left by the Orange & Alexandria Railroad, on the right by the Rappahannock River and at the bottom by the Rapidan River. While awaiting reinforcements and pondering a move on Richmond, Pope separated his forces, positioning Maj. Gen. Franz Sigel’s division at the foot of Cedar Mountain, Brig. Gen. Irvin McDowell’s division north of Rapidan Station, Maj. Gen. Nathaniel Banks’ division near Culpeper and Brig. Gen. Jesse Reno’s division near Raccoon Ford.

In doing so, Pope unwittingly presented Robert E. Lee with an unparalleled opportunity to crush his army. On August 15, Lee met with his corps commanders, Maj. Gens. James Longstreet and Thomas J. ‘Stonewall’ Jackson, at Gordonsville and developed a plan to take full advantage of Pope’s bad planning. Using Clark’s Mountain as a screen, Lee would bring his infantry into place on the southern side of the mountain by August 17. In addition to blocking Pope’s view of his approach, Clark’s Mountain provided a perfect location for Jackson’s signal corps to observe Union positions north of the Rapidan. Once the infantry was in place, Fitzhugh Lee’s cavalry would cross the river at Raccoon Ford early on August 18 and burn the key railroad bridge over the Rappahannock River at Rappahannock Station, cutting Pope’s only supply line. After the bridge was destroyed, the infantry would ford the Rapidan, smash into the exposed left flank of the Union line, trap the Federals between the two rivers without supplies and dispose of them at will.

Jackson was enthusiastic and wanted to attack as soon as possible. Longstreet, somewhat more cautious, suggested delaying the attack until the 18th. He also wanted to strike the Union right, where his forces would be able to use the foothills of the Blue Ridge Mountains to their advantage. Lee accepted the one-day delay but remained firm in his desire to attack the Union left.

The orders were cut on August 16, and initial preparations and troop movements began. While Robert E. Lee’s forces began to move into position, Stuart ordered Fitzhugh Lee to have his cavalry in place near Raccoon Ford by Sunday night, August 17. The elder Lee knew that surprise would be the key to success in the attack, but unbeknown to him the secrecy of his movement had been compromised by a Union spy, Sergeant Thomas O. Harter of the 1st Indiana Cavalry, who had infiltrated the Confederate Army of Northern Virginia and marched with it toward Clark’s Mountain. Harter, dressed as a civilian, had been sent out in the direction of Staunton, Va., in late July but was arrested and sent to Richmond, where he gained his release by claiming that he had been looking for railroad work. Harter reached Gordonsville and fell in with the Confederate force on August 16, putting himself in a key position to learn the enemy’s plans.

On the morning of August 17, Stuart left Fitzhugh Lee’s cavalry, returned to Robert E. Lee’s headquarters near Orange Court House and then traveled to Verdiersville to await the cavalry, which would pass by on its way to Raccoon Ford. Meanwhile, Pope, hearing numerous reports that Confederate troops were moving up from Richmond, began worrying about an attack on his exposed left flank and took measures to determine what was occurring there. He temporarily placed Brig. Gen. John Buford’s cavalry brigade under the command of Jesse Reno and ordered him to ‘push his cavalry forward on the other side of the Rapidan’ and to ‘use spies and scouts, without regard to expense, to keep yourself constantly advised of everything in your front as far as possible.’

In response, Reno dispatched the 1st Michigan and 5th New York Cavalry regiments on a scouting mission in the direction of Raccoon Ford and Louisa Court House. Leaving their camp south of Stevensburg at midday on the 17th, the Union horsemen crossed the Rapidan at Raccoon Ford, which had been left unpicketed, and headed toward Verdiersville.

While awaiting the cavalry’s return (and no doubt spurred on by the skirmishing that had taken place earlier with Confederate cavalrymen along the Rapidan), Reno summoned Lt. Col. Jacob Eugene Duryee of the 2nd Maryland Infantry to his headquarters tent late in the afternoon of the 17th. He ordered Duryee to take 250 of his men and raid a Confederate signal operation atop Clark’s Mountain early the next morning.

‘A topographical engineer will accompany you and if possible find out the enemy’s position and strength,’ said Reno. As Duryee was leaving, Reno stepped from his tent, pointed to Clark’s Mountain and said: ‘Young man, when you reach the top you will be a damned sight nearer the rebel army than your own, so look out. The rebel pickets have been exchanging shots with our troops along the Rapidan this afternoon.’ Little did Reno realize the full truth of his words, for those Rebel pickets were in fact the cavalry screen for the Confederate army hidden just behind the mountain. Leaving camp at 1 a.m., the detachment from the 2nd Maryland slipped out of camp, crossed the Rapidan at Raccoon Ford and began making its way to the summit of the mountain.

The morning of August 18 dawned with the Union troops at rest in their camps, unaware that almost the entire Confederate army was less than five miles away. In the Confederate camps near Clark’s Mountain, the soldiers anticipated orders at any moment to cross the Rapidan and attack. At Verdiersville, Stuart and his staff waited for Fitzhugh Lee’s cavalry to arrive so that the attack could begin. On Clark’s Mountain, the 2nd Maryland detachment was about to attack the mounted troops manning the Confederate signal station.

The day’s events unfolded in a way that was vastly different from what has generally been portrayed by historians over the years. The first player to take the stage was Union spy Thomas Harter.

Realizing the importance of his knowledge of the Confederates’ plan, Harter left their camp on the morning of August 18, swam the Rapidan and found his way to Reno’s headquarters. Generals Pope, McDowell, Reno and others were present when he arrived. According to McDowell, Harter reported that ‘the enemy had
accumulated all his force, including several divisions just up from Richmond, behind the ridge [Clark’s Mountain] immediately beyond the river and opposite our extreme left.’ In addition, the spy reported that the Confederates’ artillery horses were harnessed and that the troops were ready to cross the Rapidan at Raccoon Ford to get in the rear of Pope’s army. Their movement into this position, according to McDowell, ‘had been completely hidden from our sight by the ridge, and even from that of our lookouts on the top of Thoroughfare Mountain, was one made in the direction which had been expected from the first, and had for its object the interposing of the whole of the enemy’s forces between our army and its re-enforcements.’ That information was timely and would be of use, said McDowell, ‘provided the enemy gave us the night and day the start.’ This was more than an understatement; Harter’s report to Pope may well have been the timeliest single product of espionage received by any Union commander during the entire war.

At sunrise on August 18, Colonel Duryee and 250 men of the 2nd Maryland Infantry reached the summit of Clark’s Mountain on their raid to disrupt the operations of the Confederate signal corpsmen, which was using the summit of the mountain as a signaling station. A short skirmish with the small Rebel force manning the post ensued, and two members of the 2nd Maryland were wounded. Two Confederates were taken prisoner. Also captured were several signal flags, code books and other papers. After securing the area, the Union detachment lingered long enough to allow a topographical engineer to make observations. The view from the mountain that morning presented an unpleasant surprise–Confederate troops so close that the raiders could hear drums in the camp below pounding out the ‘Long Roll.’

Observations completed, the detachment hurried back to their camp, spurred on by the knowledge that the Confederates would not be far behind. They stumbled on a shorter route via Somerville Ford that cut several miles off their march.

Recognizing the importance of the topographical engineer’s detailed observations, Duryee sent him ahead with an escort. According to Duryee, the following report was written and sent to Pope at about 7:30 a.m., after the engineer had reached Reno and before the remainder of the detachment, with prisoners in tow, had returned: ‘I sent, under command of Lieutenant Colonel Duryea [sic], Second Regiment Maryland Volunteers, a force of 250 men to break up the rebel signal station on Clark’s Mountain. The expedition left at 1 p.m. [a.m.] and arrived at the summit at day light next morning. A small mounted force was found there, and a slight skirmish took place in which several of the enemy were wounded and 2 captured. As soon as Colonel Duryea [sic] arrives I will send the prisoners to headquarters. They captured a signal flag and a memorandum book, from which it appears that Jackson’s Army is back of Clark’s Mountain, probably in the vicinity of Orange Court House. I send herewith the book and other papers. The cavalry has not yet returned.’

In addition to Harter, another Union spy, Richard Montgomery, had infiltrated the Confederate army. Leaving McDowell’s headquarters on the 17th, he spent the evening of the 17th and most of the 18th with the enemy force. He returned to the Union lines on the evening of the 18th. During his stay he learned that the Confederates had been reinforced and were about to make an attempt to cross the Rapidan upstream from the Union position.

Pope now found himself in a grave situation. What he and his superiors in Washington most feared had occurred. Freed from the necessity of engaging Maj. Gen. George McClellan on the peninsula, the Confederates had moved swiftly to reinforce Lee’s army. Worse than that, Lee’s force was at that moment less than five miles from Reno’s headquarters, and the attack was set to begin that very day. Pope needed no prodding. Harter’s information was clear. There was no alternative; he must pull back his troops immediately or face destruction. The decision to fall back was probably a hard one for Pope, particularly given his public comments about not turning his back to his enemies. But that was the choice he took, and it was the correct one, as was made abundantly clear by reports from the 2nd Maryland Infantry and Montgomery and by Lee’s captured order.

At about 10 a.m. on August 18, Pope ordered a full-scale retreat in the face of the enemy. He directed Reno to send his wagon trains toward Stevensburg by way of Kelly’s or Barnett’s fords. His whole corps would follow, and by night only cavalry would be left behind to screen the rear of the army.

At the same time that Pope’s orders for the withdrawal were sent out, Reno, drawing on Harter’s report as well as that of the 2nd Maryland, sent a dispatch to John Buford ordering him to make a cavalry scout: ‘The enemy are in strong force about 2 miles back of Clark’s Mountain, extending thence towards Raccoon Ford. I wish you to send a squadron of Cavalry near Raccoon Ford, and to scout from thence on the north side of the Rapidan as far as Germanna Ford. Let me hear as soon as your cavalry returns.’

The movement, ordered at 10 a.m., was underway by 1:30 p.m. After the withdrawal had begun, Pope informed his superior, General-in-Chief Henry Halleck, of his moves: ‘The enemy, heavily reinforced, is advancing on Raccoon Ford from Gordonsville, Louisa Court House and Hanover Junction. All the Richmond force has been thrown in this direction to turn my left….I have accordingly, in compliance with your instructions, started back all my trains to pass the Rappahannock tonight. My whole command will commence to fall back to that line.’

Meanwhile, Lee had been forced to postpone the attack from the 18th to the 20th because several of his units had not arrived at their assigned position. As Lee and his staff watched from atop Clark’s Mountain at midday on August 18, 1862, the Union camps looked quiet. But appearances were deceiving–the Union forces were preparing to withdraw. By midday on August 19 that truth was all too apparent.

Observing again from Clark’s Mountain as the last men and wagons of Pope’s Army of Virginia disappeared in ever shrinking clouds of dust into the Virginia countryside, Lee turned to Longstreet and said disappointedly, ‘General, we little thought that the enemy would turn his back upon us this early in the campaign.’

As the Union troops faded into the distance, the Confederate command had firm knowledge of only one reason for the withdrawal–the raid on Stuart’s signal station by the 2nd Maryland. J.K. Boswell, Jackson’s chief engineer, said of the raid, ‘On the morning of the 18th a body of the enemy drove our pickets from Clark Mountain, and found out the position of our troops, and on the 19th they commenced their retreat toward the Rappahannock.’

The race to Second Manassas was on. Clark’s Mountain would soon be forgotten, and the intelligence gathering that had served Pope so well would fail miserably. Nevertheless, the events surrounding Clark’s Mountain are instructive because they bring to light the enormous value of Union intelligence operations when conducted effectively, as well as providing a perfect example of why the writings of Civil War generals are not always to be trusted.

Taking Pope’s report at face value, it is easy to see how a misconception could arise. The report says that ‘the cavalry expeditions sent out on the 16th in the direction of Louisa Court House captured the adjutant-general of General Stuart, and was very near capturing that officer himself. Among the papers was an autograph letter of General Robert E. Lee to General Stuart, dated Gordonsville, August 13, which made manifest to me the position and force of the enemy and their determination to overwhelm the army under my command….’

It is this paragraph that historian Edwin C. Fishel, in his work The Secret War for the Union, describes as ‘the clearest example Civil War history ever produces of a general’s use of a cover story to protect a piece of espionage.’ According to Fishel, Pope himself admitted the deception in a postwar letter to Harter in which Pope said the former spy had been the first person to give him the vital information concerning Lee’s plan.

While Pope’s report was printed in the Official Records, both the records of Harter’s service and Pope’s letter to him remained unpublished and unexamined, as did the accounts of the 2nd Maryland’s raid. The absence of those accounts made the prominence of the captured order understandable. However, relying on the story of the captured order is troublesome, since it gives a false picture of Pope’s conduct in the Second Manassas campaign. It makes it seem as if he was blissfully going along when a sudden stroke of luck provided him with the information he needed to save his army. That could not be farther from the truth. While Pope may not have been aware of the precise location of the Confederate army until the morning of August 18, the fact that they were nearby probably did not surprise him too much. He had long suspected that the Confederates would seek to attack him on the left of his line, and he had been receiving reports, as he himself stated, since August 12 that Lee was being reinforced and was moving to confront him from that direction. Both Pope and his superiors in Washington rightly felt that with McClellan’s withdrawal from the Virginia peninsula, Lee would be reinforced and move against Pope’s left.

Pope’s order to Reno of August 17, which sparked the mission of the 2nd Maryland and the scout of the 1st Michigan and 5th New York, mentioned that exact scenario and was designed to try to avoid it if at all possible. Far from being ignorant of potential danger, Pope used every means at his disposal to keep watch on his vulnerable left flank: a cavalry scout, an infantry scout, spies and lookouts on Thoroughfare Mountain. While the lookouts failed to see the advancing Confederate army, the other three produced valuable intelligence.

A quick survey of current titles on the campaign and Battle of Second Manassas reveals how widespread the story of the captured order is, but this has not always been the case. Several historians came close to blowing Pope’s cover story before Fishel. One was Douglas Southall Freeman in his Pulitzer Prize­winning biography of Lee. He mentioned all three possible sources of intelligence, giving prominence to the captured order, then mentioning the 2nd Maryland’s raid: ‘To his [Lee’s] disappointment over his inability to strike Pope in his exposed position…there was added on the 18th a fear that the enemy had discovered his presence despite his efforts to conceal the army. He learned that at daylight the Federals had raided a signal station that Jackson had established on…Clark’s Mountain….There was no way of telling what the enemy had seen before he had been driven back, or what records he had found.’ Freeman also mentioned in passing the report of Thomas Harter, citing McDowell’s official report as his source.

Another historian, Charles F. Walcott, mentioned the 2nd Maryland’s report in his History of the 21st Massachusetts: ‘A strong cavalry expedition…which captured an important dispatch from General Lee to General Stuart, and a gallant reconnaissance by our 2nd Maryland regiment on the night of the 17th, disclosed not only General Lee’s determination to make short and decisive work with General Pope and his army, but also that a rebel force amply sufficient to crush us, masked by the hills across the river, was rapidly moving into position for an advance.’

Those two mentions of the 2nd Maryland’s raid are among the few accounts by historians that differ from the story of the captured order. Two additional accounts by members of the 2nd Maryland Infantry provide essential information about the timing of the arrival of the captured order and help establish approximate times for the report of the 2nd Maryland.

Benjamin F. Taylor, last commanding officer of the 2nd Maryland, wrote his own account of events, which makes a case for his regiment providing the information that saved Pope’s army. After telling the story of the raid, Taylor noted that ‘our Colonel [Duryee] reported to General Reno between seven and eight o’clock a.m. by courier and in person before 10 a.m.’ Drawing on Reno’s report of the unit’s action and Pope’s official report of the campaign, Taylor made a case for the importance of the raid. He presented first Reno’s report, then a lengthy portion of Pope’s report, the gist of which is that by the morning of August 18 Pope had become convinced that the newly reinforced Confederate army was assembling nearby.

Taylor continued with the rest of Pope’s rationale for the withdrawal: ‘On the 18th of August it became apparent…that this advanced position…was no longer tenable in the face of the overwhelming forces of the enemy. I determined, accordingly, to withdraw behind the Rappahannock….I directed Major General Reno to send back his trains on the morning of the 18th, by the way of Stevensburg, to Kelly’s or Barnette’s [sic] Ford, and…then follow with his whole corps.’

That passage makes it clear that Pope’s decision was made on the morning of the 18th, which is a key point, as it is unlikely that a large cavalry force traveling 13 or more miles deep into Confederate territory would have been able to return to Union lines before 10 a.m. According to Taylor, the captured order did not reach Pope until sometime after 3 p.m. on August 18.

As additional evidence, Taylor included a letter from A.N. Wood, a sergeant in the 6th New York Cavalry. Wood ‘was present when the report of the 2nd Maryland’s expedition was dictated and written, about ten a.m.,’ said Taylor. ‘Wood says the last sentence ‘The cavalry [Buford’s] has not yet returned’ will ring in his ears through life. The clerk became a little mixed and the general had to repeat it. He also says the cavalry returned in the afternoon.

‘This statement [Wood’s] taken with the reports of Reno and Pope…indicate clearly the information obtained by the Second decided the retrograde movements of the army, the wisdom of which was later confirmed by the cavalry when they returned with J.E.B. Stuart’s adjutant general and General Lee’s order for attack.’

In light of the available information on Harter and his report, Taylor was mistaken in his conclusion, but his account establishes the timing of the decision and the fact that the captured order did not arrive in time.

Another account, written by Jacob Eugene Duryee, provides additional details of the raid. According to Duryee, the detachment left camp at 1 a.m. on the morning of the 18th. ‘The night was cloudy and very dark,’ he wrote. ‘You could not see objects ten inches from you.’ After crossing the Rappahannock at Raccoon Ford, the men climbed over a fence and, avoiding a road near the river, headed up the side of Clark’s Mountain. ‘By avoiding the road we met with many obstacles, mostly consisting of fences, and it was with difficulty that we made the march up the side of the mountain,’ he wrote. The raiders had been ordered to attack the signal post at daylight, but it was sunrise when they captured it. By Duryee’s estimate, the time was 5:23 a.m. They spent about 20 minutes on the summit, and between 5:45 and 6 a.m. they began the march back to camp.

According to Duryee, their return journey went much more quickly than their march to the summit, since it was daylight and they found a ford that cut a mile off their march. ‘I am positive that the report of the engineer reached General Reno sometime before the detachment returned,’ wrote Duryee. ‘For shortly after leaving the signal station the great importance of the information he had obtained, I knew was being anxiously awaited for by Genl. Reno. I therefore sent him ahead with an escort to make all possible haste to the headquarters of the General….I feel sure that the engineer was present when Gen. Reno dictated this report and the time was about 7:30 a.m.’ He mentioned Taylor’s account and said that Taylor was incorrect in saying that he had reported to Reno by 10 a.m. ‘This should read 8 a.m. for about 10 a.m. the order from Gen. Pope had been issued for the retreat,’ wrote Duryee. In another letter he stated that he was sure that ‘the reports of Topographical Engineer and myself of the skirmish were in Gen. Pope’s hands before 8 a.m.’

When Taylor’s and Duryee’s accounts are merged with the reports and dispatches in the Official Records and with the facts of Harter and Montgomery’s reports, a completely different picture of Pope’s actions arises. The only workable chronology for the day’s events is that Harter provided the first intelligence of Lee’s army at an unknown time on the morning of August 18. Concurrently or soon after, news of the 2nd Maryland’s raid reached Reno’s headquarters at about 8 a.m., followed by Montgomery’s report on the evening of the 18th and the arrival of the captured order sometime between the afternoon of the 18th and August 22, which is when Pope reported to General Halleck that he had the captured letter.

For too long Thomas Harter’s and Richard Montgomery’s bravery in infiltrating the Confederate army and the story of the 2nd Maryland’s raid on Clark’s Mountain have been lost in the mists of history. Rather than a triumph of luck or the fortunes of war, it was instead a systematic use of intelligence-gathering through spies, signal corps operatives, cavalry and infantry reconnaissance that saved Pope at Clark’s Mountain. It was not blind luck, but skillful professionalism–an overriding factor in the entire outcome of the war.


This article was written by John Lam and originally appeared in the July 1998 issue of America’s Civil War magazine.

For more great articles be sure to subscribe to America’s Civil War magazine today!