Military Intelligence 1861-63: Part II
APPROVED FOR RELEASE 1994
CIA HISTORICAL REVIEW PROGRAM
18 SEPT 95
OFFICIAL USE ONLY
A study of the information that opposing commanders acted on substantially modifies historical evaluations of the Civil War.
MILITARY INTELLIGENCE 1861-63
Edwin C. Fishel
PART II. CHANCELLORSVILLE AND GETTYSBURG1
The few historians who have thought seriously about Civil War intelligence have generally concluded that it left much to be desired. They are near the truth, though not necessarily for the right reason. The belief held by some that intelligence in that war was more than ordinarily laden with error is an example of a wrong reason. Another is the assumption that it was seldom of much influence in battlefield decisions. The valid reason for deprecating Civil War intelligence is the limited scope of both the Northern and the Southern effort. Intelligence was not pursued on a scale that seems commensurate with the size and desperateness of the conflict. What was pursued was almost altogether military, and even the military sector was not fully covered: strategic intelligence was severely subordinated to the tactical.
The reason for the neglect of political and economic intelligence was simply the common ethos and ethnos of the civilian leaders of the North and South. Having known each other for much of their political lives, they had little motivation for investigating what the other side was fighting for and how far it would go. As they knew the enemy's geography and his language, they also knew the extent of his economic wherewithal.
This is not to say that opportunities for political intelligence combined with subversion were entirely ignored. The Confederates in particular engaged in such attempts, their effort being directed toward turning the Copperhead conspiracy into an effective peace movement. But this was a mere forlorn hope, and its outcome was pathetic; when the point of action was reached at the time of the Democratic convention of 1864, the Southern agents saw their collaborators melt away.2
The same inbred knowledge of the enemy limited military leaders' inquisitiveness, accounting in some measure for the lack of interest in strategic and even in tactical intelligence. The generals and admirals knew each other even better than the politicians did; they had gone to school together, lived together, fought together. This familiarity enabled many an officer to foresee how his adversary would act under given conditions. Thus, for example, the Confederates could predict that McClellan would try to "engineer" his way to Richmond; and they were right, for he jumped at the opportunity to build siege lines when he encountered thinly held fortifications half way up the Peninsula.
And even tactical information was not always felt to be a requirement for planning an action. When Lee sent Jackson around Pope's flank to Manassas, he was without information as to where the Federals' front might be vulnerable; so he picked a point where there could be no question about vulnerability-directly in their rear. When McClellan landed on the Peninsula to begin the long-awaited campaign that was supposed to end the war, he had woefully little knowledge of what particular obstacles he would face; it came as a surprise to find in his path the Warwick River, a stream of no mean size. Burnside made so little effort to obtain information before Fredericksburg that he misplaced an enemy concentration that was directly across the Rappahannock, only a mile or so from his own pickets.
These cases, and others about to be seen, were manifestations of the Civil War commander's habit of thinking that possession of the initiative greatly reduced his need for information, that it was up to the threatened forces to find out what was going on. But in only one of these cases-Lee's attack on Pope-was the initiative an adequate substitute for information, and it was adequate then only because of the impotence of Pope's cavalry.
To a considerable extent, however, the smallness of Union and Confederate intelligence activity is an optical illusion resulting from the tendency to measure it on a modern scale-not only the scope of the activity but more particularly the size of the bureaucracies associated with it. Only a handful of organized intelligence entities have surfaced, none of them giving any appearance of great size or authority. The largest had a peak strength of 70, counting cooks and teamsters along with operatives. Several important commandsLee's army, for one appear to have done without any separate intelligence organization whatever.
As one of the chief misconceptions about Civil War intelligence touches this question of size and organization, it will be worth whileto pause and examine the matter. The literature reflects a widespread belief that the Union had a "Secret Service"-presumably a nationwide or army-wide organization. This belief has existed side by side with the contradictory one that intelligence bureaus were few in number, small in size, and of limited influence.
What the North had was a lower-case secret service comprising a hodgepodge of unintegrated, uncoordinated intelligence, counterintelligence, and military police activities. The service in the field was improvised pretty much according to the taste of the individual commander. It was most often under the direction of the provost marshal, but it was sometimes assigned to the adjutant of the command or to the signal officer or the chief of staff. In some commands, particularly those with the more sizable positive-intelligence efforts, a special staff position was created for the purpose. This was the case with Pinkerton's bureau, though nominally it was under McClellan's provost marshal. Some commanders took a direct hand themselves in supervising their spies and detectives. The variety was rendered complete by the not uncommon practice of assigning intelligence to one staff officer and counterintelligence to another.
It goes almost without saying that under these conditions the exchange of information between neighboring commands was usually haphazard, and that nothing but good luck or geographic separation could prevent duplication of effort on the one hand or working at cross purposes on the other.
This unsystematic system seems inevitable when it is realized that there was no superior intelligence and counterintelligence agency at Washington. Two bureaus operated there at overlapping periods, but with authority that was more local than central. And even their local operations were so ill coordinated that members of one were occasionally shadowed or arrested by the other.
One of the two was Pinkerton's bureau, attached not to the War Department but to the Army of the Potomac. Its positive-intelligence element was at the capital for the eight months in 1861-62 when the army was headquartered there. When \McClellan took the field in the spring of 1862, Pinkerton's spies went along, leaving behind a small counterespionage element that remained until Pinkerton resigned at the end of the year.
The other bureau, the counterespionage and police organization3 headed by the notorious Lafayette C. Baker, was initially under the State Department, but from early 1862 until it passed from view five or six years later it operated under the War Department. Baker's position as Department provost marshal did not, however, give him wide authority. Although his activities often ranged afield (for example, he sent detectives on the trail of Confederate agents in Canada, and for a time he kept a small office in New York City for liaison with the local police), special orders were required for him to invade the realms of the field armies or geographic commands, whose provost marshals operated independently (and often outranked him). Washington and environs remained his main field of operations.
It was Pinkerton and Baker themselves who fathered the "Secret Service" myth, not unwittingly. They published memoirs in which each named himself chief of the "United States Secret Service"-in each case an ex post facto title self-conferred. Then historians whose purpose it served to accept the existence of a Secret Service during the war took it from there, each solving the problem of the chiefship by naming whichever of the two men he happened to be writing about.The Confederacy's intelligence system was both more and less systematic than the Union's-more so in that there was a departmentlevel bureau in Richmond, less so in that field armies tended to do without a distinct and recognizable intelligence organization. (This apparent lack, however, may be partly due to the comparative scarcity of Confederate intelligence records.) At least there was some uniformity in the Confederate commander's practice of making intelligence a routine part of the mission of subordinate generals and letting it go at that. The high competence of Southern cavalry, attained early in the war, probably had something to do with this. Jeb Stuart, for example, was an intelligence collector and evaluator par excellence (though he preferred to employ his cavalry in more bellicose pursuits whenever he could).
The intelligence bureau in Richmond, though highly placed, was not a complete intelligence service. An alter ego of the Confederate Signal Corps headquarters, it was concerned more with communicating intelligence than with any other aspect of the job. It was generally known as the Signal Bureau, and its overt activities consisted of issuing army and State Department ciphers and enciphering and deciphering the correspondence of the Richmond authorities. But it also operated courier lines to the Potomac and beyond, and it was involved at least incidentally in obtaining and directing agents at the far end of these lines. In this capacity, and in various projects concerned with sabotage, ship captures, development of infernal machines, and collaboration with the Copperhead secret societies in the North, it was known as the Secret Service Bureau.4 With these activities it had a far broader mission than any or all of the Northern bureaus, but it would have served the Confederacy better if it had had a less ambitious set of functions and had done a bigger and better job as a straightforward information service. The bureau was in a position to develop an army-wide Secret Service; it does not appear that anything on that scale was even attempted, if indeed conceived.
The revelation that the Confederacy did not go all-out to obtain intelligence will not disrupt popular belief in the intrepidity, cunning, and invariable success of the Southern secret agent. But as we are about to see, the scenario theme that Reb ran rings around Yank in scouting, spying, and all related matters is another myth. Had it not been for Joe Hooker, that might not have been a myth.
New Sharpe Look
Hooker succeeded Burnside in command of the Army of the Potomac (headquartered still at Falmouth, Virginia) in January 1863. Among many improvements wrought by Fightin' Joe (whose forte was actually administration) was the founding of an organization called the Bureau of Military Information.5 This bureau was an improvement over its predecessors because any real effort to get and report the facts would have been an improvement; but happily there is more to the story than that.
Two principal factors in making the bureau a success were these: (1) Hooker was as strongly committed to the idea that Rebels were only about six feet tall as McClellan had been to the vastness and invulnerability of the enemy legions. With Hooker in command, the Confederates' strength ceased to be an all-absorbing question. Their positions (especially of fortifications and batteries) and movements became the main question to be answered-as it should have been all along. (2) Hooker supplied the bureau with real talent, especially at the top. Its chief was George H. Sharpe, a 35-year-old colonel of volunteers, an upstate New York aristocrat, a lawyer, linguist, and former diplomat. Sharpe, having refused brigade command in order to stay with his regiment, naturally took on the staff assignment with misgiving. A return to line duty was always in his mind, but he was too effective an intelligence officer to be spared.
John C. Babcock, the young civilian who had been Burnside's oneman secret service, readily accepted a stepdown to the No. 2 position in the bureau. Sharpe brought in a fellow townsman, Captain John McEntee, as No. 3 and recruited a dozen or so spies from among Unionist citizens of the locality and from the enlisted ranks of the army.6 Like Pinkerton, Sharpe was placed under the army's provost marshal, but-again like Pinkerton-only nominally. He sent his reports directly to, and obtained most of his guidance from, Hooker's chief of staff, Major General Daniel Butterfield, heir of the American Express empire.
Sharpe's bureau was not only the most highly developed intelligence activity on either side; it had a modernity about it that parallels the war's numerous other military innovations. It was a complete intelligence operation. Sharpe established detachments in neighboring commands. He had his own scouting facilities independent of, but cooperating with, the cavalry. When there was a cavalry expedition to distant country his men went along whether or not the main purpose of the raid was to probe for information. Interrogation of prisoners, ' deserters, and refugees became routine and -thorough; the commander who forwarded prisoners to Washington instead of to Headquarters heard about his error in short order.
Sharpe was the army's spymaster (a status Pinkerton had never fully achieved); though independent operations did not entirely disappear, they were few and limited in scope. An example of his authority is the fact that when the army invested Richmond in 1864 he took over direction of a ring of resident spies in the city who had previously been controlled, well enough it appears, by a local Federal commander long in the area .7 Finally, his bureau in one way or another managed to get its hands on information reaching headquarters from all sources-which (besides those under his direct control) consisted of newspapers, scouts working for corps and division commanders, a few spies in similar status, the cavalry, the balloons, the Signal Corps observers and cryptanalysts, and dispatches from distant commands and from Washington. Though the army commander and his chief of staff and adjutant had the first look at much of this material, Sharpe gathered it all, added it up with his own information, and produced digested, semi-evaluative reports.
Sharpe's reputation soon spread west; from the Mississippi theater came a request for his cooperation. Finally he became intelligence officer for Grant as commander of all the Union armies, but since Grant elected to take the field in Virginia, Sharpe's activities remained essentially what they had been with the Army of the Potomac. (Had Grant based himself in Washington with Sharpe at his side, the Union might have acquired some semblance of an army-wide intelligence service.)
Chancellorsville: the Intelligence Base
Sharpe quickly got results. One of his spies—another Indiana cavalry sergeant like Pope's savior on the Rapidan8—was so well received by the Rebels that he was permitted to make a 10-day, 120mile guided tour of their front and rear lines9 and then to get aNvav across the Rappahannock unchallenged. Back at the Falmouth headquarters, he gave Sharpe the locations of troop concentrations, fortifications, and artillery positions, in many cases with pinpoint exactness. A Northern-born farmer living in Lee's lines west of Fredericksburg sent information that corroborated the sergeant's picture of low enemy strength in that locality and added the compelling point that around Chancellorsville there was a six-mile vacuum between the positions on the river and the nearest troops to the rear.
Babcock, a ready-made O/B expert, derived after several weeks' work an organization chart that was considerably better than Pinkerton's chart of the previous year. Devoting particular effort to getting brigade and regimental averages, he added up Lee's strength to 61,500 plus artillery personnel-a total that was as close as the Confederate returns could have given it.10 Although Hooker could not have known how good a piece of work this was, he could not have failed to see that Babcock's service under Pinkerton had not afflicted him with any hallucinatory disease.11
While this reformation was going on, the Confederates' chronically haphazard intelligence did not improve. Lee's dispatches at this period show a surprisingly poor understanding of Army of the Potomac organization; when well-informed prisoners were taken the Confederates often could not fit the most basic O/ B data into place. An effort to get a spy into Washington failed. Lee complained that his scouts could not get past Hooker's pickets. When a full. and correct statement of Hooker's strength reached Richmond through a medical return printed in a Washington newspaper,12 its relay to Lee was so poorly handled that he did not appreciate its authoritative character and instead relied on his own estimate, which was 25 percent below the mark.
Hooker, appointed in midwinter, had three months to ready a campaign. The farmer-spy's report gave him the opening he was looking for: weakness on the enemy left, and a gap that opened directly on the rear of their main force. The gap could be reached, however, only if the Federals could evade discovery in a long march through country that Stuart's cavalrymen were watching.
Hooker's march did evade effective discovery. Part of this success was due to security measures far more stringent than any the army had ever seen.13 To these Hooker added signal deception. His plans for his cavalry occasioned this piece of trickery. Intending to send the horsemen to raid toward Richmond when the infantry marched, he placed them 30 miles upriver, poised to jump off southward. His signalmen planted, with what seems a singular lack of subtlety, the word that the cavalry was headed for the Shenandoah Valley. Confederate flagmen copied and deciphered this message (and the Federal interceptors ascertained that they had done so); Lee alerted his forces in the Valley and kept Stuart upriver, ready to follow the Federal horse.
This left a 20-mile stretch of front so lightly patrolled that Hooker marched his main striking force of 55,000 men upcountry, across the Rappahannock and the Rapidan, and back downriver to Chancellorsville-an average distance of 50 miles-before Lee realized what was happening. By that time Hooker had within his reach a position only three or four miles in the rear of Lee's center. Once this position was taken, the Confederates could save themselves from wholesale loss only by retreating, and even in retreat they would be punished severely.
But Hooker held back the coup de grace. A few miles short of the target position he inexplicably assumed the defensive, 14 and in a poor position-around Chancellorsville, in dense wilderness relieved onlv by an occasional farm.
Lee was quick to accept the initiative thus tendered him. Stuart's men found the enemy right flank unguarded, facing south; Lee sent Jackson on another end run to the west of that point. Jackson's attack-his last, for he was fatally ,vounded that night-rolled up a large part of one Federal corps. It was enough to spread demoralization through Hooker's entire force, or at least as far as his headquarters.
This battle was a classic in several ways; one of them was the Federals' misconstruction of Jackson's march, which they could see at several points. Thinking the underbrush much too thick to let even foot soldiers get through, they did not believe that what was happening could really happen. Commanders on the Federal right, receiving many reports from their own scouts, pickets, and signal posts that showed an attack to be imminent, rejected them because of Headquarters' confidence that the movement was a retreat. Headquarters was thus denied information from the front that might have changed its mind.
A sufficiency of cavalry could have caused the true character of the march to be reported to Hooker in time, but he had held only three cavalry regiments back from the raid to the south. He was almost as empty-handed on the field as Pope had been at Manassas, though as a result of his own deliberate choice.
After two more days of fighting and shifting, Hooker worked the army into a position where the attacking Confederates would have been at a disadvantage. But then he retired across the Rappahannock to his old camps, apparently in the belief that Lee had been heavily reinforced. This story, entirely false, came from a pair of deserters who reached the Federal rear headquarters at Falmouth. Hooker's chief of staff believed they were telling the truth, but Babcock, who also questioned the men, insisted for several days that they were not.15 Then he too accepted the story, though only overnight. It was during that night that Hooker decided to retreat. Hooker's state of mind, however, was such that it is difficult to assign any strong influence in his decision to the enemy's supposed reinforcements, which would still have left them at a considerable numeric disadvantage in any case.
Chancellorsville was a campaign based on flawless intelligence but lost to a defender who was caught flat-footed and who, even at the moment of striking the decisive blow, had only begun to recover from that condition. If it had been Hooker who attacked that day, be would have known what corps, what divisions, and probably what brigades he was striking. Lee had not identified in advance even the corps that became his victim, and if he had, the information would have been a new entry on his mental O/B chart (presumably he kept no other kind).16
As usual, the role of intelligence in these events has been appreciated only dimly. Hooker's plan is repeatedly characterized as the most brilliant of the war-but one searches the literature in vain for a trace of suspicion that intelligence might have had something to do with it. The possibility that a report of enemy reinforcements was of some influence in Hooker's decision to retreat also goes unnoticed. That Lee based Jackson's attack on knowledge of the vulnerability of the Federal right flank is well known, but his otherwise uninformed and misinformed state is not recognized, even though it makes the result he extracted from that one perceived fact all the more remarkable.
The usual impression of Confedyrate superiority in intelligence is largely owed to this faculty of Lee's for manufacturing a victorious stroke out of such fragments of information-or out of no information at all; in other words, out of imagination and prescience. Had he possessed less of this kind of genius, he would have been forced to provide himself with a better, more organized information service. He understood the art of war much better than the art of military intelligence.
The reverse was true of Hooker. He is remembered as the general who, bidding strongly to end the Army of the Potomac's history of frustration, led it to its most unnecessary defeat. But if the intelligence side of the Chancellorsville story makes his defeat seem all the more tragic, it at least reveals him as the creator of a hard-hitting, effective information service that persisted after he was gone, contributing substantially to the victories of the army he left behind.
Potomac Pas de Deux
The Battle of Gettysburg, together with the maneuvers that led to it, has an intelligence story that has suffered from a negligence and fantasy proportional to its prominence in the national heritage.
Early in June, a month after his battle with Hooker, Lee began drawing westward from Fredericksburg, leaving a third of his army temporarily behind as cover. His plan was to march northward through the Valley. A pause at Culpeper to receive reinforcements and supplies lengthened Sharpe's opportunity for discovering what was going on.
Though he had predicted a general movement of this kind, Sharpe soon afterward fell for a story that the stirrings across the Rappahannock were connected with an impending raid by Stuart. The story may not have been spread by the Confederate command, but if it was not it was the most successful non-plant of the war. Hooker ordered Pleasonton, his cavalry chief, to break up the raid.
Pleasonton caught Stuart by surprise and fought a gigantic cavalry battle with him, but drew off without a victory. After the Gettysburg campaign was over Pleasonton covered this failure by explaining that his attack was really just a reconnaissance in force and that it had succeeded in unmasking Lee's movement toward the Valley and Pennsylvania. The historians have trustfully repeated Pleasonton's improvement on the facts. Actually his reports after the cavalry battle, and papers captured during it, strengthened the Federals' inclination to believe that a big raid by Stuart, rather than a massive infantry move, was in the making.
It was another four days before Hooker began an all-out pursuit of Lee, and what sent him on his way was a routine interrogation. The interrogator was Captain McEntee of Sharpe's bureau, who had been upriver with Pleasonton for two weeks, trying without success to get some of the bureau's spies over to Culpeper. He chanced upon a Negro boy, one Charley Wright, who had "refugeed" from Culpeper in the wake of the cavalry fight. The boy had been a servant to Confederate officers for a year. He said he had seen two infantry corps pass through Culpeper en route to the Valley, and when his knowledge of Confederate organization turned out to be about as good as McEntee's own, the captain got off two quick telegrams to Sharpe.
These set the army into motion on June 13. Despite Lee's partially successful effort to confuse Hooker and so delay pursuit, the Federals' discovery of the movement came soon enough to put them on the Potomac in good time. When Hooker set up headquarters at Fairfax on the 17th, his army in front of him, the Confederate advance was about Hagerstown and had raided into Pennsylvania,17 but Lee was another week in assembling the rest of his army near Winchester.
His crossing into Maryland took place June 23-26. Hooker crossed, east of the Blue Ridge, on the 25th and 26th. By now Hooker was marching almost abreast of Lee, and on an inside track.
Because Hooker had had to take care not to cross prematurely,18 only intelligence or prescience could have made his move into Maryland so timely; and no one has ever accused Hooker of prescience. The intelligence that put his army over the river has been in published records for many decades; but its obvious effect on his actions-on" the whole outcome of the campaign-has not been seen. So the fact that it was substantially erroneous intelligence has of course also escaped notice.
Its chief author was John Babcock. Sharpe had sent him to Frederick on the 17th to organize local espionage. Crippled by the presence of Confederate soldiery and busy secessionist citizens who knew his purpose,19 Babcock had to hide out and operate at a survival level. The result was that when two men he had sent beyond Hagerstown returned on the 24th, their report was his first real news in four days. He jumped at it so eagerly that be oversold it. It put all of Lee's army at or over the Potomac; this was correct as to the leading half of the Confederate main body, but it placed the rear half two days' march ahead of where it actually was-still about Winchester.
By the time Babcock's emphatic telegram arrived on the 24th, a dozen other bits and pieces of information pointing to the same conclusion had filtered in. Some of these appear to have derived from the Confederate habit of sowing false leads; if that was their origin, the trick served not to confuse the enemy but to hasten him on his way. In any event, once Babcock said that no Confederates were lingering in Virginia, Hooker lost no time in issuing marching orders.
It was this action, founded on a misplacement of the enemy, that more than any other single factor caused the Federals to arrive first at the place of eventual collision and take possession of the famous fishhook-shaped ridge that gave them an insuperable advantage.
But thus told, Babcock's error makes a better story than it deserves. If one set about to list instances in which wrong intelligence pointed to the correct action more forcibly than correct intelligence would have, this one would surely stand near the head of the list. But the story is robbed of some of its beauty by the fact that Babcock was only half wrong, and the correct half of his findings was itself a compelling indication that this was not a raid or a feint. The story loses a bit more when it is realized that Hooker's action was such as to leave a good margin for error. Washington, after all, is on the Maryland side of the Potomac. If after Hooker crossed, Lee's rear elements had struck for Washington on the Virginia side, Hooker could have detached enough force to deal with them, in front of the city if not sooner.
Well Met at Gettysburg
The most general distortion in the standard version of Gettysburg is the theme that the battle was an accidental collision of two great armies groping about almost as if blindfolded. That is correct enough for the Confederates; it is false for the Federals.
The reason Lee was in the dark was that he had failed to appreciate the object lesson Hooker had presented for him on the Rappahannock: he had voluntarily parted with the bulk of his cavalry. Before crossing the Potomac he allowed Stuart, who was smarting under newspaper criticism of his surprise by Pleasonton, to ride off to the east and march on the Federals' right. There is an entire sub-literature on this march and its wisdom or unwisdom; it may be summarized here by saying that Stuart had an image-restoring trip, full of raids, captures, and skirmishes, but the country was so full of enemy soldiers that he did not find Lee's army again for ten days-at Gettysburg, fighting for its life.
Not until the night of the 28th did Lee learn that the Federals were in Maryland. The discovery came from a spy sent out weeks before, not by Lee but by Longstreet. This man-named Harrison, a civilian-had, the story goes, been sojourning in Washington. More probably he had been following Hooker's army, at least for some days. On his way to rejoin Longstreet he passed through or near several Federal corps. They were about Frederick, and comparatively stationary-for June 28 was the day Lincoln replaced Hooker with George Meade,20 and the Army of the Potomac was catching its breath while the new chief got his bearings.
The Confederates were strung out for sixty miles, from McConnellsburg to the Susquehanna, in an excellent position to be whipped one element at a time. Lee, after first doubting Harrison, sent out messengers with orders for a hurried concentration about Gettysburg, a point all his infantry forces could reach in a day or two.
Harrison had saved an army in the manner of Pope's spy on the Rapidan. His story has won a place in the literature, but its numerous retellings never capture a full sense of the absolute vitalness of his service; they do not depict an escape creditable to one man. Yet the Confederacy conceivably owed the last year and a half of its existence to Harrison, fnu.
While the Confederates were getting their key intelligence from a single spy; the Federals were reaping the benefits of team espionage. Most of the citizens of the invaded region remained behind shuttered windows, but a few dozen self-appointed patriots were out spying, scouting, and making courier runs to Harrisburg. From there their news was telegraphed to Washington and forwarded to the armythus making a circuit of nearly 360 degrees.
From these dispatches and from his own sources, Meade correctly placed the bulk of the enemy force on his left, about Chambersburg. But his marching orders sent the army fanning out from Frederick over a wide angle, with scarcely any more strength on the left than on the center and right. On the 30th, though he had comparatively little evidence, he predicted with phenomenal accuracy the movement of the main enemy force eastward toward Gettysburg; but he still did not reinforce his left xving, which was moving on that place. Thus his actions have given the impression that he did not know where to expect the enemy-hence the Federal half of the collision-of-two-blindarmies myth.
Luckily Meade's generals on the left were even better informed than their chief, by virtue of the advance of Union cavalry beyond Gettysburg. Federal horsemen detected enemy approaches from west, north, and east, and captured some revealing dispatches. The wing commander on July 1 hurried his infantry on to the town, where by now the cavalry, dismounted, was fighting against heavy odds. The infantry seized the ridge and adjacent hills and managed to hold on for the rest of the day.
Not until news of this battle reached Meade did he order up the rest of his army, some of which was still in Maryland, more than a day's march away.
Although the Confederates were surprised by this collision, believing all the Federal infantry to be well down in Maryland, they bad a big numeric advantage at the outset because they arrived from all directions at about the same time. While they had been concentrating, the Federals had been dispersing. Again Lee had seized the upper hand while being less well informed than his adversary; but this time the blow he struck did not quite turn the battle.
He attacked again on the 2nd and again inflicted heavy punishment. That night Meade called a council of war, asking whether the army should retreat. The vote was negative and he accepted it as his decision.
Every One But Piekett's
Though the demoralizing effect of such a retreat had weighed heavily in the council, some writers give intelligence a share of the credit for this decision. These, however, repeat an error already seen in the case of Pope's escape on the Rapidan. The decisive intelligence at Gettysburg, they say, was a captured dispatch-this one from President Davis addressed to Lee, which Union cavalry scouts took from a courier party on that July 2 over near Hagerstown. It contained the important information that the Government could not send the Army of Northern Virginia any more reinforcements, and in explanation of the shortage Davis gave a tremendously revealing summary of military conditions throughout the Confederacy.
That Lee was getting no new troops was good news to Meade, but the chief value of the captured document was strategic. It was of more significance in Washington than in the field. It does not look like the kind of information that would have swayed a roomful of generals who had been in a desperate fight for two days; their concern was with the forces Lee then bad in hand. If Davis had said he was putting 20,000 men on the road for Pennsylvania the next day, Meade and his generals might only have shrugged their shoulders, knowing it would be two or three weeks before the reinforcement could arrive. Almost certainly, moreover, this dispatch, like the one supposed to have saved Pope, arrived after the decision it pointed to had been made.
To the extent that intelligence influenced the decision, it was intelligence obtained on the field, but in a manner that will never make a TV script. This intelligence was a simple tabulation, compiled from prisoner interrogations, of Confederate regiments and brigades that had been in the battle. Sharpe and Babcock had set to work on this as soon as they reached the field, of course; by afternoon of July 2 it was a most revealing compilation. It showed that the entire Army of Northern Virginia infantry was present except one division, and that every brigade present had been in the fight. By evening, wheb Sharpe reported to Meade just before the council of war opened, he was able to add that this one division, Pickett's, was now on the field and could be expected to be used heavily on the morrow. As Meade's rear elements had arrived only a few hours before and he could count on having an advantage of about 6 to 1 in fresh troops, the decision to stand fast cannot have been as hard a one as has been supposed.
Here in all probability was the war's biggest payoff for the bookish side of intelligence, the headquarters staff work that was then almost a novelty. Despite a drastic reorganization Lee had carried out just before marching north, the O/B chart that accompanied Babcock everywhere (except, presumably, to Frederick) was in so good a condition that the enemy situation could be reconstructed in a matter of hours 21 by the jigsaw method. This was a precise kind of battlefield intelligence that the Confederates could not have produced even if their chief intelligence arm, Stuart's excellent brigades, had been put to proper use.
Not only was Pickett's Charge foreseen, but the sector at which it was to be delivered was divined by Meade. It came against the center of his line. Two other divisions were with Pickett's; the three were smashed-thrown back in a state of complete disorganizationby about 10,000 Federals, two-thirds the number of the attacking force.
The Army of Northern Virginia, badly spent, fell back to home soil. Vicksburg surrendered to Grant July 4, the day after the battle ended in Pennsylvania; the Federals now held the Mississippi to its mouth. From then on the most the Confederates could hope for was to keep armies in the field until the North should tire of the war.
Though the events covered here represent only about a quarter of the major campaigns of the war, several instructive points can be drawn from them:
The importance of the contingent factor,22 and of having the intelligence resources to deal with it. This is an unsurprising discovery, the more so when one is aware of the Civil War commander's preference for tactical over strategic intelligence. Yet it is striking to see the principle at work. Two battles-First Bull Run and C:hancellorsville-were decided, so far as intelligence decided them, by information obtained during the action, and arising out of it. The principle was also at work at Gettysburg (though the intelligence obtained on the battlefield was of less profound influence than the advance information that put the Federals on Cemetery Ridge ahead of the enemy).
Equally striking is what happened when the principle was not observed. A whole series of examples is available, for not until the Pennsylvania campaign did the Federals consistently keep their intelligence abreast of the action. Lee often divided his army-so often as to make it seem a habit; at the beginning of the Seven bays, before Second Bull Run, Antietam, Fredericksburg, and Gettysburg, and during the fight at Chancellorsville. He took these risky actions, and won battles by them, in the knowledge that Union generals would probably lack either the facilities or the aggressiveness to discover his dispersion in good time, or would be unlikely to call him fully to account if they did make a timely discovery. Even when dispersion led him into a losing battle at Antietam, the loss was not due to tactical reconnaissance by the enemy. Not until Gettysburg did the Northern army display enough vigor in discovering his positions to inspire any great fear of the consequences of dispersion.
"It takes all kinds." Each type of intelligence-espionage, intercept, interrogation, etc.-sometimes produced about as expected and sometimes fell short. Each also produced windfalls-discoveries and results of great impact, and of a kind not ordinarily expected. In espionage, the shortfall is exemplified by Pinkerton's evidently barren penetration of Richmond, normal good performance by the farmerspy who told Hooker the way to the enemy rear, and surprise successes by Harrison and by Pope's spy, neither of whom could have been expected to move with enough speed and luck to save an army in the way they did. Signal interception, after a slow start, leaped to an extraordinarily fruitful performance in the deception that opened the way for Hooker's march to the enemy rear (for all that the plant seems to have been thrust at Lee with too obvious a generosity).
The bread-and-butter sources-interrogation, cavalry, etc.-followed the same pattern. Interrogation, at least as a source of O/B information, certainly fell short under Lee. It achieved good standard results,. not only for Babcock but even for Pinkerton. Two of its greatest successes, however, were in much more vital matters-uncovering the Confederates' movement to Pennsylvania and showing, at Gettysburg, what the result would almost certainly be if Lee continued attacking, as he did. Cavalry, ordinarily the main reliance on the march and in close action, often did produce according to expectations, as for Lee at Chancellorsville and for Pope until he wore out his mounts; but it performed over its head in numerous captures of important dispatches, most notably Stuart's theft of Pope's entire files; the frequency and importance of these captures was abnormal. And newspaper intelligence, though ordinarily a yawn-producing activity on account of the masses of error, bluster, and planted information that it had to deal with, had its day in that one authoritative disclosure about Hooker's strength, though the validity of this was not apparent to Lee until their main encounter was over.
Each different type of intelligence thus enjoyed at least one fine hour; all were indispensable. Conclusion: it took all kinds, and the greater the total effort, the greater the scope for serendipity-which was the most beneficient supplier of all.But to say that all types were important is not to say that their importance was fully appreciated. Had these commanders been asked to name the one information source they could least afford to do without, the cavalry would have been their choice. Their willingness to employ it on other tasks even at critical times is perhaps the best single indication of the Civil War general's lack of passion for intelligence.
How a big built-tin advantage in intelligence can be overcome. In the contest for tactical information the Confederates held the upper hand by virtue of fighting on home ground 95 percent of the time. It seems to have been habitual with Federal commanders, when operating in the Confederacy, to concede the enemy this advantage.24 Pope resisted this tendency and by force of effort succeeded in getting an even break or better as long as his cavalry held out. Like Pope, Hooker was too stubborn to be resigned to coming in second in the information contest. Other Union commanders were liberal in regard to newspaper and flag-of-truce exchanges, probably because those were situations in which the Northerners for once stood to get a quid pro quo. Hooker clamped down hard on both. By these and other strong security measures, and by his insistence on vigorous and competent intelligence work, he marched to Chancellorsville with well nigh perfect information while Lee, surrounded by a friendly population, suffered from information that was as bad as Hooker's was good until the armies had been at close grips for two days. And the advantage Hooker seized early in his command did not prove transient; Sharpe maintained it, apparently, all the way to Appomattox.
A characterization of commanders as getters and users of intelligence. Three rather sharp classifications emerge from the performance of Hooker, Meade, and Lee respectively.
Hooker, administrator par excellence, saw the value of intelligence and knew how to get the job done right. He also did an excellent job of translating his intelligence into a campaign plan. And an excellent job of security. But in the pinch he did not trust the plan that be must have admired as much as history does. It is hard to put one's finger exactly on his flaw, but this much is clear: he could comprehend a big picture (before the battle), and he could see a big plan, but he could not see it through.
There is no particular reason to believe that if Meade had inherited the army without a going intelligence outfit, he would have set up as good a one as Hooker did. But he had one ability in which Hooker did not especially distinguish himself: as an evaluator of intelligence evidence he could hit the bull's eye as surely as someone else could find the target. On the morning he left the Fifth Corps and rode over to army headquarters to take command, he was completely uninformed of the enemy situation (Hooker had carried security too far) ; by evening he had picked out the correct information from a great welter of conflicting reports.25 In the succeeding days he continued this performance, though not against such great odds. It is hard to believe that if he had been in command at Chancellorsville a flanking movement would have been read as a retreat or a planted story of enemy reinforcements would have been entertained for several days.
But Meade was far from an ideal applier of intelligence. While in Maryland he read the evidence correctly and then acted as if the erroneous reports were as sound as the correct ones, and the army would therefore have to go out and look for enemy concentrations all over south-central Pennyslvania. It is reasonable to question whether, if he had not had his generals' views to rely on in the council of July 2, he would have made the decision that his information pointed to.
Lee, as has been shown, did not do anything like the job Hooker did in providing himself with intelligence. Evidently he also lacked Meade's flair for evaluation; for example, despite his own habitual use—over-use-of deception, he accepted a planted signal message that should have seemed suspiciously gratuitous. But Lee excelled in putting information to work. Give him a scrap of it and he knew what action to take, and he took it, and saw it through.
Thus each of these men seems to have excelled in but one of the three skills-getting intelligence, evaluating it, and applying it. The second of these skills is of a higher order than the first, and the third is higher than the second, but the higher orders do not seem to require any degree of excellence in the lower ones. This stratification, though it looks a little over-sharp, may be exactly what a psychologist would expect to find; but it is something that the abundant legacy of Civil War history has not previously been made to demonstrate.26
Clearly, Civil War intelligence was not the pale, irrelevant stuff that the literature reflects (and here we are speaking of the whole literature, not merely that of the horseflesh and magnolia blossom school). Yet one may fairly ask whether its story is more valuable than any other piece of antique intelligence history.
It is in one way-in what might be called Intelligence's public relations. The Civil War remains our most profound national experience. It is disturbing that a collection of myths 27 has been permitted to usurp the place of intelligence in the history of a struggle so important and so well documented-to usurp without even filling it, leaving millions of words of campaign narrative that explain critical decisions weakly if at all.
Will the story, once set right, necessarily establish that intelligence contributed substantially to the result? At this distance, Northern might looks so overwhelming that one is tempted to believe the end would have been the same, and would have come as quickly, if the Union armies had made no organized intelligence effort at all. Yet the fact remains that a rebellion which holds at the start a big and integral territory is quite likely to prevail, even in the face of greatly superior might. The American rebellion of 1861 was such a one. And the outcome was touch and go up to the autumn of 1864; the Confederacy's defeat could not be foreseen until the Northern anti-war element lost at the polls that November, thanks largely to the Federals' battlefield successes of the months immediately preceding.
The contribution of intelligence to the 1864 victories is missing here, and until it is supplied we cannot represent that this was another American war in which intelligence had as much to do with the outcome as in, say, the Second World War. But there were at least two earlier points at which a Confederate victory might have won Southern independence. These were Lee's invasions of Maryland in 1862 and Pennsylvania in 1863. Had Antietam or Gettysburg gone the other way, the North might have given up. At Antietam decisive intelligence came to the Federals merely by virtue of their being in the vicinity of the enemy. But in the next year's campaign they won an intelligence contest that ran from the Rappahannock to Cemetery Ridge; they won every round of it, and by organized effort in each case. And the product of this effort affected the battlefield result as profoundly as the Lost Order affected the decision at Antietam. Not a bad heritage, all by itself.
2 The literature contains hints that political intelligence was actually practiced on a far greater scale than this, through spies whom each side succeeded in placing high in enemy councils. While this remains a possibility, these references to high-level spies omit the evidence, if indeed there is any; they are usually mere innuendo, the wishful figments of "popular" historians.
3 Generally thought, mistakenly, to have been also a major positive-intelligence organization. Baker did make one or two trips to Richmond early in the war, and later on some of his men occasionally engaged in minor espionage projects. Although Baker's account of his Richmond adventures is hard to swallow, it is partially supported by expense accounts and a credential made out in the name he claims to have used there and signed by the Confederate Secretary of War; this latter paper is in the Walter Pforzheimer Collection on Intelligence Service, Washington, D. C.
5 "Information" was at that time the full-fledged equivalent of "intelligence," which was still about three decades short of acquiring its present place in the military lexicon. Alone, "intelligence" carried no implication that it was information about an enemy or foreign power.
7 This is the ring in which the chief figures were Elizabeth Van Lew, a wealthy spinster, and Samuel Ruth, manager of the Richmond, Fredericksburg and Potomac Railroad. There is every reason to believe that it was far more successful-at least more continuously successful-than the more highly publicized Confederate espionage in Washington. The Richmond ring's story is omitted here because nearly all of its known activities belong to a period later than the campaigns covered in this article.
8 Names of Ohioans, Indianans, Illinoisans, and Iowans abound in the records of Sharpe's bureau and other Federal intelligence organizations: presumably Easterners were excluded because of the dialect problem. On the Confederate side, some of the leading spies were from the border region but no sharp geographic pattern is evident. Apparently it was easier for a deep Southerner to drop his drawl than for a New Englander or New Yorker to assume one.
9 He was in Confederate uniform, but if he had represented himself as belonging to some regiment or battery of the Southern army he would in all probability have been sent promptly to that unit. It is a reasonable conjecture that he claimed membership in one of the partisan organizations operating in northern and western Virginia, and that his trip under escort was a pretended effort to make contact with his company.
10 It was within 1/4 of I% of the figure now most generally accepted, though the last few percentage points of accuracy were accidental: Babcock was decidedly low on two of the infantry divisions and too high on the cavalry, and the errors canceled out.
11 Babcock had served under Pinkerton on the Peninsula and seems to have had a hand in the O/B work of that period, which (as noted in Part I) produced adequate basic data. But he deplored the ponderous record-keeping enforced by Pinkerton; he also deplored, it would seem from his own later work;, the detective's practice of making strength estimates that had little or no relation to his unit-byunit findings on the composition of the Southern army.
13 Hookers security actions are a story in themselves; there is room here for only an outline: (1) He not only tightened picketing but on the eve of his march placed officers within earshot of all pickets to prevent leaks through across-the-river fraternization, to which the armies on the Rappahannock front were much addicted. (2) Early in his command he stopped civilian visits to the army, so that this telltale action would not coincide with his march. (3) Also early in his command, he canceled newspaper exchanges with the enemy and sharply curtailed flags of truce. (4) He punished newspaper conjectures about army movements without regard to their truth or falsity, and he attempted-though probably without much success-to crack down on the soldiers' habit of putting military secrets in their letters home. (5) On the eve of his march he stopped all outgoing soldier mail at the Washington post office. (6) He had a shipment of pontoons brought by a circuitous back route to avoid Confederate observation on the Potomac. (7) Synchronized with the upriver march of his main body was a sizable secondary action against the heights below Fredericksburg. He took pains to see that ordinary security measures protected this latter movement; anything less risked exposing its diversionary nature. (8) He protected the main movement by an unusually thorough sealing-off of the countryside, which included stationing guards at farmhouses to prevent local residents from spreading word of their observations. Although most of these measures seem obvious, they were none the less departures from previous practice.
17 En route, this force almost 25,000-surprised the Federal garrison at Winchester, capturing 3,500. It had marched the 75 miles from Fredericksburg without being discovered. Hooker had neglected for a full day to report to Washington that the enemy was moving north; it was during that day that Winchester was invested.
18 His orders required him to protect Washington as well as to find and fight Lee. After Jackson's stolen marches of the previous year, the Federals constantly feared an attack on Washington from the Virginia side of the Potomac.
21 The Federals did not have a good cross-section of prisoners until the battle of July 2 was well advanced. Babcock may have been on the field the first day, but Sharpe probably did not arrive until the night of July 1-2. McEntee's activities at this stage of the campaign are unknown.
23 Flag signaling made it possible for enemy correspondence to be intercepted in far greater quantities and with much more ease than ever before in the history of warfare. (It was overwhelmingly the main medium of Civil War intercept, though in the general literature of the War virtually all cited incidents of interception and cryptanalysis involve telegraph or courier dispatches.) Little capital appears to have been made of these opportunities in the earlier campaigns, not because ciphers were unusually secure but precisely because they were insecure: commanders were cautious about putting valuable secrets on the air.
24 In the Mississippi theater, at least, Federal spies were instructed to tell the Confederates the truth about Federal forces, on the assumption that the Rebels would have correct information anyway and that the spy who made misstatements would only betray himself. At least one spy who ventured to interpret this rule to suit himself wound up by having to be reassigned east.
26 One commander who may have shown two or all three skills is Grant. He seems to have been well informed from his Vicksburg campaign onward, and his ability to use intelligence sometimes appears to belong in the same class with Lee's. But this study has not yet touched his history in detail.
OFFICIAL USE ONLY