Wherein we examine failures of intelligence
The first of two posts on mistakes by the North the first week of May 1862.
May 8, 1862, was the date of the Battle of McDowell, the beginning of Stonewall Jackson's legendary Valley Campaign. Much has justly been written about Jackson's operational-level brilliance and an equal amount has unjustly been written about his rather mundane tactics. Only in recent years, have reevaluations of the campaign been conducted, most notably in Peter Cozzens fantastic Shenandoah 1862 which acknowledges the immense personal cost to the soldiers of Jackson's army and, in the words of the friend who gifted the book to me, "bothers to examine the Yankee sources."
The Valley, though technically in the Potomac watershed, has simply been beyond the capacity of your blogger to post on, which means the area that drains directly into the Shenandoah has usually been beyond the scope of this blog. But Jackson's campaign deeply threatened the area we usually cover, and ended up being the strategic center of gravity for the entire Virginia theater. Without Jackson's operation, Joe Johnston, hundreds of miles away on the Virginia Peninsula, would have been unable to fend of McClellan's army, and Richmond would have been scrambling to find defenders in fall 1862 instead of Washington.
With the demotion of McClellan in March, Secretary of War Edwin Stanton became the Administration's principle war planner. Unlike his predecessor, Simon Cameron, Stanton believed the Secretary should be active in decisions about the prosecution of the war, rather than only concerned with supplying and administering the armies. Stanton signaled this when he added Peter H. Watson as a second Assistant Secretary of War as a condition of joining the Administration himself. Watson was appointed to straighten out and manage the contracts the Department made to supply the army, which was usually a tightly held responsibility of the Secretary in order to bestow political favors.
Stanton and Watson were old allies, and well known to Lincoln as well. When Stanton had been Attorney General to President Buchanan, he had used Watson to convey messages to then-Senator Seward that kept President-elect Lincoln appraised of the secret (and treasonous, in Stanton's mind) actions of the lame duck. But the connection between the three went back even further, to 1855 when Watson had traveled to Springfield to hire a local lawyer that might play better for an Illinois jury as part of a legal team defending the Manny Company from a patent infringement suit, and found a very game Abraham Lincoln. When the location of the trial was changed to Cincinnati, however, Watson's boss dumped Lincoln for a much hotter Ohio lawyer--Edwin Stanton. But a termination notice was never sent to Lincoln, so he traveled to Cincinnati and met a very put-off and scornful Stanton. At Stanton's insistence, Watson and the rest of the team took every opportunity to exclude Lincoln, but he refused to take the hint and stuck around for the whole (successful) trial.
Seven years later, it was Watson who was left out, albeit in a different context. Stanton and Lincoln had gone to Fort Monroe when the Confederate line at Yorktown collapsed to see first-hand the massive Army of the Potomac at last engaging in battle, and Watson was left in charge at the War Department. So when the Southern forces in Virginia appeared to be collapsing and it looked like Richmond would fall any day, it was Watson who sat in the 17th Street War Department building on the site of today's Eisenhower Executive Office Building and had to do the overall analysis of the situation. He failed; though--as we shall see--the hard-working Assistant Secretary is hardly at fault.
Watson had to answer two questions that have bedeviled military planners forever--what could his enemy do and what would his enemy do? Instead, he and the military leadership around him answered the question that military planners are never supposed to--what should his enemy do? In doing so, they arrived at a likely enemy course of action that proved disastrously different from the one that later occurred.
|Peter H. Watson|
Watson had four commanders with the ability to provide this reconnaissance to him. The first was Edwin Stanton himself, at Fort Monroe with the President and privy to (some) information coming from the headquarters of the Army of the Potomac, located near Williamsburg.
On May 8, without knowing of Jackson's attack in the western Virginia mountains, Stanton sent Watson a telegram from Fort Monroe warning of attack--but from the wrong source:
It is believed here that a considerable force has been sent toward the Rappahannock and Shenandoah to move on Washington. Jackson has been reinforced strongly... Telegraph to Generals McDowell, Banks, and Hartsuff, to keep a sharp lookout and report frequently. Tell General Hitchcock to see that the force around Washington is in proper condition.Stanton had seen the withdrawal of the force at McClellan's front and jumped to conclusions (probably not on his own) about their destination. Since December 1860 Stanton had wrestled with the Army leadership to make it recognize an enemy that he believed intended to capture Washington. He had been opposed to McClellan's plan to move the bulk of the army to the Peninsula, and now he saw the enemy's chance to accomplish its long-held intent by throwing everything to Jackson, who would march on Washington.
Watson dutifully passed on the warning to his other three intelligence gathering commanders, but a message from one of them written May 7 crossed paths with it. Maj. General Nathaniel P. Banks was commander of the Department of the Shenandoah, and had so far had the principal responsibility for chasing Jackson out of the Valley. With two divisions, he had pushed Jackson from Harper's Ferry, all the way south to beyond Harrisonburg. In a letter postmarked at 10:30 pm, the time Jackson was planning his battle near the mountain town of McDowell, Banks wrote Stanton via Watson:
My division is ready to move [out of the Valley and to Catlett's Station]... Our cavalry from near Harrisonburg report tonight that Jackson occupies that town, and that he has been largely reinforced. This refers probably to Ewell's division, an account of whose force I sent you last night. Deserters east of the mountains confirm reports of Jackson's move in this direction.Banks, who was not a military man, had been duped by Jackson's cavalry. The Confederate horsemen had effectively blocked his own (far too few in number) from riding as far as Harrisonburg and managed to convey the impression that they were screening a large force. Knowing that his own almost 20,000 men now at New Market drastically outnumbered Jackson's roughly 7,000, Banks assumed that only reinforcements would make the Confederate bold enough to make that move. Since a division of about 11,000 men under Confederate Maj. General Richard S. Ewell had left Culpeper Court-House [City of Culpeper] a few days before, it made sense that he would provide the reinforcements. Statements from deserters--who might not know or might lie for a loaf of bread--fit into this fact pattern, and so Jackson wound up "spotted" in Harrisonburg instead of about to inflict damage in McDowell.
Banks was not alone in misplacing Jackson. Almost on the other side of the theater, in Falmouth, Virginia, Maj. General Irvin McDowell also declared Jackson in front of him, instead of in the vicinity of the town that shared his name. Another of Watson's key intelligence gatherers, McDowell wrote Stanton through Watson on May 8, but after receiving the forwarded telegram from Fort Monroe. A lifelong military man, his reliance on deserter intelligence is less forgivable than Banks':
Two deserters from the enemy came in this afternoon. They are Irishmen, belonging to New York, intelligent, and I think truthful. They state that the [Confederate] forces under Anderson and Field [south of Fredericksburg] have been reinforced by Brig. Gen. Maxcy Gregg's South Carolina Brigade... The deserters say it was understood that Jackson's command was on the way to join Anderson, and that Jackson was to have chief command. The intelligence from Fort Monroe of the belief there that the enemy intends drawing in this line, and the fact that General McClellan is still before Williamsburg and not in a position to prevent an advance on me, have all caused me to believe that it is highly probable that they may attack me before I shall have force enough to attack them.McDowell was overly focused on his own planned operation. Convinced the war was ending before he could redeem himself for Bull Run, he had worked tirelessly to prepare the troops in the Department of the Rappahannock for a march south to Richmond in a pincer movement with McClellan. He had convinced Stanton to strip the capital of some of its defensive troops and transfer some of Banks' to form him two additional divisions and give him a force of about 40,000. Earlier in the day of May 8 he had sent one of his several-times-a-day updates to Stanton (through Watson, of course) on his progress toward moving:
The bridge across Potomac Creek is one-third done. All the timber for the one over the Rappahannock is cut, and will be on the ground this evening. I fear my forces will not be in hand by the time the bridges are done. There are reports of a heavy force of the enemy 30 miles this side of Richmond, on the railroad to this place.Both reports were self-serving, though it's doubtful that McDowell saw them that way. Instead he was blinded by his own responsibilities and saw everything his enemy (and friends) did in relation to its impact on his own planned operations. It was a similar myopia that let him march off to Manassas following a plan he had concocted, while relying on another army operating in the Valley and reporting to a different authority to understand and commit to his plan enough for it to be successful.
The final source of intelligence for Watson was Maj. General John C. Fremont. Known as the Great Pathfinder for his (overblown) exploits in mapping the Western United States, Fremont represented the original wing of the Republican Party and had been their first nominee for president back in 1856. He had won the approval of the stalwarts of the party when he had decreed all slaves in Missouri belonging to individuals in rebellion free, and Lincoln had incurred their ire by rescinding the order and recalling Fremont to Washington. After a winter of heavy pressure, he was back in command of troops, this time the new Mountain Department, charged with protecting the area we now know as West Virginia.
The problem with Fremont's reports, was that he was nowhere near his troops. Granted, they were scattered from Berkely Springs on the Potomac River to about 200 miles to the southwest near Bluefield, but Fremont had only just arrived in Petersburg [West Virginia], still over 50 miles from the site of his men's defeat near McDowell. On May 8, as the battle was beginning, he also sent a report to Stanton through Watson, in this case passing on the information forwarded to him by his two brigade commanders:
General Schenck, after a march of 34 miles in less than twenty-four hours, has effected a junction, as ordered, with General Milroy. He reports that he is just in time, as the enemy is approaching in two, and probably three directions, at distances from 4 to 7 miles--I think [Confederate Edward "Allegheny"] Johnson and Ewell, making together about 18,000 men--a force considerably greater than ours. Forseeing this emergency as the probably result of the rebel retreat from Yorktown, I have been endeavoring to meet it...Fremont, who both did have Jackson in front of him and attacking him in actuality on May 8, did not even give him thought in his comments on the situation. He also assumed the Confederate in front of him must have been reinforced, and so also plugged in the unaccounted for division of Dick Ewell into the situation he wanted to explain. Similar to Banks, McDowell, and Stanton, Fremont assumed that the Confederates must be reinforcing following their retreat from Yorktown, because that's what he would have done in their position. But Fremont would know his mistakes all too well in just a few hours.
As the beleaguered Assistant Secretary of War poured over these reports, along with the initial reports of the Battle of McDowell, he needed to synthesize them into a coherent picture of the likely enemy course of action in Virginia outside of the Peninsula as a whole, so he could pass it back to Stanton at Fort Monroe for decision on the Northern course of action. Stanton had not left Watson by himself, though, he had provided the grizzled Maj. General Ethan Allen Hitchcock.
Hitchcock had graduated from West Point in 1817, and as the grandson of the famous Vermonter and Revolution general Ethan Allen, was part of the Old Army elite. Winfield Scott had made him inspector general for the march on Mexico City, and he had commanded a geographic department. But a disagreement with then-Secretary of War Jefferson Davis had led to his resignation, and he turned his back on all things martial to devote himself to literature and philosophy (and also a fair amount of alchemy). When Stanton took over the War Department, he coaxed Hitchcock to Washington and back into the army to serve as an adviser to him, for situations just like the one they were in now.
On May 9, after analyzing the reports Watson had provided as well as an additional one from Ambrose Burnside in North Carolina, he sent Stanton a summary of the situation, as best he could tell:
General Burnside, by letter of the 5th instant, reports the belief that troops of the enemy have been sent from his front to Virginia. McDowell reports an increase of the enemy in his front. The movements of Jackson are uncertain. McDowell reports a rumor that he is to command in his front. Fremont reports that Milroy is threatened by Jackson [350 miles away]. Shields [a subordinate of Banks, who scouted on May 8 after Banks] reports a reconnaissance to the suburbs of Harrisonburg without discovering any enemy, which may show the departure of Jackson, but not his destination.And Hitchcock's conclusion:
An apprehension of the advance of McDowell upon Richmond may have originated some movement by the enemy which may be converted into an aggressive one, requiring high authority for the orders that may be necessary to meet it.It was ideal. He succinctly described the intelligence, noting when it conflicted, and drew the only conclusion that accounted for all the evidence. The modern J2 reporting intelligence to his commander would probably be chastised for the second clause of his conclusion, but Hitchcock was accurate that if some counteraction were to be taken, it would require orders from someone that could tell the various departments what to do, i.e., the Secretary of War himself (more about this next week).
Unfortunately the clarity began to disintegrate as soon as it reached Fort Monroe. "The movements of the enemy in your front are believed to have been defensive, founded on an apprehension of your advance through Richmond," Stanton wrote to McDowell (again, through Watson) later in the afternoon of May 9. The statement subtly but significantly changed Hitchcock's hypothesized link of causality into one of fact, which meant that McDowell could accept that the Confederate intent was to defend against his own planned advance. To both men it made sense, because that's what a commander in the Confederates' situation should have been worried about.
"A report from General Fremont," Stanton continued, "conveys his opinion that Jackson, with Johnson, and Ewell, are in his front threatening Milroy." Disregarding the work that Hitchcock had done untangling the various claims from the intelligence gatherers, Stanton passed on directly from Fremont the situation that was most likely to endanger Washington: all three divisions combined in front of a much smaller force on the flank.
Whether the Official Records covered a Stanton typo, or McDowell misread his message, or the error was made by a staffer not paying close enough attention, McDowell began his evening response to Stanton with a bit of nit-picky know-it-all-ism that indulged some of his worst vices.
I have heard of but one Johnston in the enemy's army--J.E. Johnston--and presume there can be no doubt as to his position, and that he was at Yorktown and is now before Richmond. As to Ewell, I have not conjectured as to his position, but have supposed him also at Yorktown. I cannot conceive of such a movement as the enemy's leaving Yorktown to go to Western Virginia, passing by both my force and that of General Banks, as General Fremont suggests.At least McDowell was finally turning his thoughts to what the enemy could do. It was impossible for Joe Johnston to go to McDowell. But it was not impossible for Ewell to go there, and it certainly wasn't impossible for Allegheny Johnson to be there already (since he was), both of which McDowell in his haste to trash Fremont's opinion doesn't even think worth considering. But McDowell was more on board with Stanton's thinking about the most likely enemy course of action based on what made sense to them:
I concur entirely in your opinion that the movements of the enemy in my front have been founded on an apprehension of my advance, and from all that I can gather from various sources I am inclined to believe this apprehension conduced to their evacuation of the Yorktown Peninsula. I think not only that they have been defensive, and are so at this date, but I also think that it is within the limits of possibilities, if not probabilities, that they may turn into offensive operations if they can throw forward troops enough for the purpose, and that it is the only thing they will do if they find they can resist General McClellan.There's a lot to unpack in those two sentences, even leaving aside McDowell's delusions of grandeur about his own role in the fall of Yorktown (which would be more convincing if he hadn't botched the geography). Bad: he ascribes a single, conclusive intent to the enemy, even those operating well outside of his area of responsibility. Good: he considers what the enemy is capable of, offensive operations, and what requirements are needed to make that happen. Bad: he rules out all alternative enemy courses of action, even when others clearly exist.
To Banks, Stanton sent a similar letter through Watson, but this one explicitly mentioned the confusion between McDowell and Fremont about Jackson's location. Banks responded at 4:30 pm on May 9, with considerably better intelligence gathering abilities than he had shown two days earlier:
General Fremont's opinions as to the position of Jackson is correct. He has not been in front of General McDowell unless within two or three days past. Ewell's division has been at Elk Run, between Blue Ridge and Shenandoah, on the road from Harrisonburg to Stanardsville, until now. Our scouts report the camp-fires seen yesterday... [Johnson, Jackson, and Ewell] are not more than 20 miles distant from each other unless Jackson has moved south recently. They will concentrate against any small force left in the valley... Hundreds of fugitives [escaped slaves] come through these places into our lines because there are no troops there.But the whole thing became muddied when Banks' overeager (and, as it turned out, mendacious) subordinate Brig. General James Shields wrote his own letter to Watson directly that night after reading a copy of Stanton's missive forwarded by Banks.
Sir: A copy of your dispatch of this date to General Banks has reached me. The probabilities are that Jackson took the cars at Waynesborough [sic] to join General Edward Johnson, who was retreating before Milroy, and that both united may attempt to strike a blow against him or check his advance.Advice that would have been useful 24 hours earlier. He continued:
Ewell, from all we can learn, is still east of the Shenandoah, with a force estimated at from 6,000 to 10,000. His object does not seem to be to fight, but to prevent the junction [of Shields' division] with McDowell. If he be within reach I hope to dispose of him on my way to the Department of the Rappahannock.Watson must have felt blessed to have yet another general who thought the enemy chose its course of action solely to provide a suitable punching bag for operations he already had planned. But Shields' act of insubordination and muddying of the waters wasn't enough for the diminutive man who had once challenged Lincoln to a duel:
I venture, merely an opinion, to say the Southern Army will never attempt an advance against Washington. If it makes the attempt the war will soon be over. They can never by any possibility reach the capital, and we can hem them in in such a way as to make their destruction inevitable. I still hold the opinion that they will fall down South, and that all these demonstrations are but feints.At least Shields would only be Banks' problem for a few more days, until he became McDowell's.
While Shields opined, Fremont had forwarded a similar letter from Stanton to the one received by Banks and McDowell out to the various isolated commands in his department. To one commander he explained:
Schenck and Milroy attacked yesterday by Jackson's, Johnson's, and Ewell's combined forces at McDowell. A sharp action this morning. Our forces retire on me [at Petersburg]. Blenker's division coming up. General Shields' division ordered to General McDowell. Banks retiring upon Strasburg. This all results from Yorktown evacuation. You must be vigilant and prompt.Stanton's acceptance of Fremont's original claim that Jackson, Johnson, and Ewell had consolidated had now come back to Fremont, solidifying the "fact" further, and now passed out to all the subordinate commanders as the enemy's force location. It was further solidified the next day, when a new letter from the Secretary of War to Fremont questioned whether the junction had really occurred, prompting Fremont's adjutant to write the enraged subordinate Robert Schenck "Do you still think Jackson's forces are opposed to you? Secretary of War says that General McDowell reports Jackson to be in his front." Schenck snapped back at 11:40 pm that "General McDowell must be mistaken."
But while Schenck's anger was confirming for Fremont that the forces were joined, Banks was sending Watson reports that Ewell was "still on this side of the Blue Ridge, apparently trending southward. One brigade is reported back to the Rapidan near Gordonsville. Jackson is believed to be near Stanton by our scouts and country people." It was a never ending carousel of mis-reporting or mis-valuing intelligence.
So, at the end of the day on May 10, here's what Watson's intelligence had revealed. Allegheny Johnson was at the town of McDowell, was reinforced, and had attacked two of Fremont's brigades. Two Confederate brigades stood opposite General McDowell at Fredericksburg, and had been becoming increasingly hostile. Jackson was either at McDowell reinforcing Johnson, or unlocated, possibly on his way to Fredericksburg. Ewell was either near Stanardsville, or with Johnson at McDowell, or on his way to Fredericksburg, or on his way to the Peninsula. Joe Johnston's army on the Virginia Peninsula was now unlocated, since McClellan had lost contact with it while licking his wounds after the Battle at Williamsburg, and may be on its way to reinforce the Confederates south of Fredericksburg or the Confederates at McDowell.
Believe it or not, that's not an unworkable intelligence report (with the exception of the ludicrous proposal that Joe Johnston was going to McDowell). Better intelligence gathering by the commanders in the field, and wiser valuation of the intelligence could have refined it further (e.g., it's safe to confirm that Jackson was at McDowell based on the report Shenck, who fought against him).
But a wise analysis would have followed the line of reasoning begun by Ethan Allen Hitchcock, formulating possible actions and intents, and rating them on scales of dangerousness and likelihood. In other words, formulating what the enemy could do and would do. From those, the analyst would present Stanton with several alternatives, highlighting the most dangerous enemy course of action and the most likely enemy course of action, and allowing Stanton to pick a course of action of his own that anticipated those.
Instead, Stanton and his commanders spent their time concerning themselves with what the enemy should do, and crafting a course of action for themselves that countered that one specific action. By doing so, they enabled Jackson to take the most dangerous course of action and reinvigorate the South.
Next Week: the second major failure that enabled Jackson's Valley Campaign, the failure of Union planning.