Sketch showing ford between Fredericksburg and Falmouth, February 27, 1863, drawn by John C. Babcock. Courtesy Huntington Library.
By mid-April, Hooker’s intelligence operation was working at a level of efficiency unprecedented in American military history. Hooker’s confidence grew in parallel with the revelations of Lee’s army set out by Sharpe and his men. One of Hooker’s staff officers, J. E. Hammond, wrote that not only was Hooker confident but “so are all who know anything of the situation here. We have a moral authority of all that it is necessary to know in regard the enemy, every regiment and brigade, division, etc., all their latest arrivals and departures, etc., all collated, compared from many sources and fully confirmed. The secret service of Gen. Hooker is far superior to anything that has ever been here before.”1
That so-called “Secret Service” (the name was still not fixed) had laid golden gifts before the commander of the Army of the Potomac in the seven weeks since its founding
Sharpe’s efforts quickly evolved into what would be called, by the modern term, the intelligence preparation of the battlefield (IPB), which is defined as “a systematic, continuous process of analyzing the threat and environment in a specific geographic area. It is designed to support staff estimates and military decision making. Applying the IPB process helps the commander selectively apply and maximize his combat power at critical points in time and space on the battlefield…”2
Sharpe understood clearly that this was a continuous process; the work of his organization would be unremitting. Even when the army rested between campaigns, they worked. He maneuvered constantly on the intelligence battlefield, which was one seamless, ongoing campaign. (See Appendices E and F.)
He quickly set himself to identify the enemy’s location, strength, and condition. Almost the first problem he encountered was to sort out the status and location of Lt. Gen. James Longstreet’s I Corps of the Army of Northern Virginia. There were all sorts of rumors circulating that he was going to Tennessee, South Carolina, or Georgia. Prisoners and deserters from Longstreet’s divisions seemed to dry up after February 25. Then he had a break in the contents of a captured mail bag. Letters revealed that a large part of Longstreet’s command had departed, beginning February 20, and passed through Richmond four days later. By the beginning of March, Major General Dix at Fort Monroe reported that at least two of Longstreet’s divisions, Pickett’s and Hood’s, had arrived. Major General John J. Peck commanding nearby at Suffolk operated a very aggressive intelligence collection effort that confirmed Longstreet’s presence. Longstreet had a third division, that of Ransom, as well, and settled into the siege of Suffolk.3
Sketch showing ford between Fredericksburg and Falmouth, February 27, 1863, drawn by John C. Babcock. Courtesy Huntington Library.
On the night of February 27, Sharpe unleashed the wily Sergeant Cline on the unsuspecting Confederates. Passing himself off as a Confederate soldier, Cline slipped across the Rappahannock at Port Royal, 5 miles below Fredericksburg. Presently, he encountered a Confederate captain, John W. Hungerford. The captain by strange coincidence commanded a company of the 9th Virginia Cavalry, the best scouts in the Army of Northern Virginia. They would be the counterparts of Sharpe’s scouts for the next two years. Cline appears to have concocted a story which allowed him to accompany Hungerford on his ride, the purpose of which was not identified, from the Confederate right to its left across the rear of the Army of Northern Virginia for a distance of 50 miles to Orange County Courthouse, identifying encampments and units along the way. At the courthouse, Cline spoke with several Georgians who told him of the large fictitious force at Gordonsville. Hungerford took his party back to the right by a different route, this time through the Wilderness to the Confederate center at Fredericksburg and then back to his headquarters on the right, passing through all of Jackson’s II Corps. On the night of March 4, Cline slipped away from Hungerford’s camp, his departure unnoticed by those engaged in a well-lubricated card game.4
Upon his return to Sharpe’s headquarters on the night of the 5th, he wrote a lengthy report from which Sharpe debriefed him in detail, as indicated by the margin notes in the colonel’s hand. Through the use of information obtained from prisoner/deserter interrogations, Sharpe was able to identify the units of which Cline had only seen their camps. Fishel points out, “Altogether, Cline’s written report and Sharpe’s notes mentioned, in terms definite enough for the locations to have been determined on a map a total of sixty-four major installations, twenty-four camps, twelve locations of batteries or heavy artillery, twenty-three fortifications, and five wagon and ambulance parks.”5
Of equal importance were Cline’s observations of the state of the enemy. He noted that their rations, when they could be had, were a pint of flour and a pound of bacon a day. Frequently even this was not to be had, and the men were told to obtain food on their own locally. Even officers needed new shoes. It was clear that Lee’s logistics were fragile and that few stocks were on hand. What little he obtained came up the railroad from Richmond. Sharpe did not know, but a reason for detaching Longstreet’s three divisions was to relieve some pressure on Lee’s starved commissary. Right after Cline’s debriefing, Sharpe sent out more scouts, who were unaccountably stopped by their own cavalry. Brigadier General Patrick was incensed: “The scouts sent out by Col. Sharpe were arrested & sent back by Averill [sic], notwithstanding they had my pass—It was a great piece of arrogance & stupidity combined, which caused ‘Fighting Joe Hooker’ to swear very wickedly & send for Averill [sic Averell] in a great hurry—What the said Averill [sic] caught is a question yet undecided…”6
Given Hooker’s unstinted support for Sharpe and his mission, Averell most likely left with a stern admonition to cooperate fully with the BMI. That same day the cavalry passed another one of Sharpe’s scouts, Dan Cole. This time the pass worked.
HEADQUARTERS, PROVOST MASHAL GENERAL
S. S. Department
March 5, 1863
Colonel TAYLOR,
Commanding Cavalry Brigade, Northern Neck:
COLONEL: The bearer Dan. R. Cole, 3rd Indiana Cavalry, is on peculiar duty, by proper authority.
In order that he may meet with no unnecessary detention, please furnish him with pass to outposts and return to this department, the pass he comes with, at earliest convenience.
Very respectfully, your obedient,
GEORGE H. SHARPE
Colonel &c.7
Cole chose not to rely on disguise but tried another method that involved just as much cool effrontery. He gave himself up to the pickets of the 9th Virginia Cavalry as a deserter. He was shunted from the headquarters of Fitzhugh Lee, Stuart, and Lee himself before being sent to Richmond where he was confined to Libby Prison. There he cleverly slipped in among the prisoners of war and got himself exchanged.8
Despite having a comprehensive understanding of the deployment of almost all of Lee’s forces, Sharpe was not confident yet in the numbers. He was loathe to report information in which he did not have confidence and said so. His scouts were careful to rate their confidence in their information, such as when Skinker reported on March 13 that “I am sorry to inform you that I have been unable to obtain any additional information in regard to the strength of the Rebel Army near Fredericksburg but have again to report it, indefinitely, from 28 to 35 thousand.”9
Everything now seemed to click into place, and on the 15th Sharpe delivered a summary report on the enemy’s strength and dispositions that for detail, accuracy, and timeliness it would not be amiss to compare it with a modern report produced with a vast array of technological tools. It is worth quoting in its entirety.
HEAD QUARTERS, ARMY OF THE POTOMAC BUREAU OF SECESH INFORMATION
Major General DANIEL BUTTERFIELD,
Chief of Staff :
March 15 1863
GENERAL: Having reported thirty days since for duty at these Head Quarters, I beg leave respectfully to submit the following brief summary as the monthly report of what has been done in this department.
Rebel Army of the Potomac
Prior to the middle of January the Rebel Army of the Potomac consisted of two Corps of Lieutenant Generals Longstreet and Jackson, the Cavalry Division of General Stuart, (a summary of which is herewith annexed) and the artillery, of which the most diligent labor has only enabled us to form the barest skeletons, on which no reliance can be placed.
Longstreets command before the middle of January
Lieut Genl. Longstreet had five divisions – Viz:
| Anderson’s. | Pickett’s. |
| McLaws’s. | Ransom’s. |
| Hood’s. |
Jacksons command before and since.
Lieut Genl Jackson had four divisions – Viz:
| A. P. Hill’s. | D. H. Hill’s |
| Early’s. | Trimble’s. |
And thus with their Cavalry their army consisted of ten divisions, which at their maximum ought to consist of five brigades each. This was not and is not now the case, as remarked below under the head of “Estimation.”
Divisions gone South.
About the middle of January the enemy began moving forces south, and we now know that General Longstreet has gone south, and has taken with him (or there had been sent south from his Corps d’ armée,) the divisions of Ransom, Pickett, and Hood.
It is also known that Gen. D. H. Hill has gone south, but it is not thought that any of his divisions has accompanied him.
Rebel Commanders.
Generals Lee, Stuart, and Jackson are now opposite us: Their Hd Qrs being located in the annexed “Extracts,” together with the following divisions of the rebel army.
Of Jackson’s Corps Of Longstreet’s Corps
| A. P. Hill’s | Anderson’s. |
| D. H. Hill’s | McLaws’s. |
| Early’s. | |
| Trimble’s. |
This is all that is believed to be opposite to us. (Infantry)
Artillery.
From the fact that most of our informants have seen little Artillery, we believe that a fair proportion of that arm has gone south with the divisions of Longstreet—but it should be remembered that the greater part of it was sent back some time ago to Hanover Junction, or near it, to obtain forage along Pole Cat Creek, and from Louisa C. H.
Location
It is believed that the location of each of the divisions of infantry, before mentioned, is fixed with very considerable exactness, and reference is made to exact language used by the latest informant, (for the personal judgment of the general commanding) selected on account of their strong corroboration by all previous information.
Estimations
Each division should consist of five brigades of five regiments each. There are however some divisions which have only four brigades, and there are many brigades that have much less than five regiments. Some have four regiments, and others have two regiments and two battalions, or three regiments and one battalion.
These battalions were originally independent organizations, and are composed of four, five, and six, and sometimes (though rarely) eight companies. Thus, though representing a regiment in a brigade, they detract materially from the brigade strength.
Strength of A. P. Hill’s Division
A. P. Hill’s Division, seems to be the favorite one, and is an exception, in the whole rebel army. Whenever we get information from a regiment in his division we find it strong, and we are continually told that all the conscripts go to him. It is estimated by themselves as from 1100 to 1300. This would show that his brigades run over 2000 each; and all admit that this is the strongest division in their service.
Estimate
We think the six divisions before mentioned, (as comprising the entire infantry force opposite us) have doubtless twenty five brigades in all. And we have many evidences to show that 1700 men (for duty) is a liberal estimate of the average of their brigades—By this we should have
15 × 1700 = 42,500 men.
Estimate by regiment.
Again, we have many evidences that 350 men is a liberal average of their regiments—and four and one half regiments to a brigade is rather over than under the proportion.
By this we have
4 ½ × 350 = 1575 × 25 = 39,375 men.
Lowest estimate.
On account of the numerous battalions which represent regiments in a brigade without their compounding strength, we think these figures are too large, and believe that a calculation of our own, from several scattering regiment is near the truth, viz—that the brigades will average about 1300 men for duty.
Mr Skinkers statement
Mr Jno. H. Skinker states that he understands from his relatives outside the lines (who sympathize entirely with the other side) that the rebel army is from 28000 to 35000 strong.
General Remarks
With the multiplied and laborious examinations which have led to the information contained herein, and in the exhibits forwarded herewith, an attempt has been made to establish communication with the other side of the river, which has been successful, and of which the General has been informed. Another has also been attempted which has not yet been fruitful, but from the short time which has elapsed since the departure of the agent it is hoped it may be—and if successful it will probably be productive of the highest results.
The enemy’s lines have been penetrated a number of times, notice given of the Cavalry raid of February 26th and of other movements of the enemy—the number of persons employed and the expenditures very inconsiderable.
It is considered fair to add that, from the best information, we learn that the rebel organization has never before been obtained in this army, until it was too late to use it; and that no previous time has any attempt been made to locate the enemy’s forces, that has proved anyway successful, or to estimate them within any reasonable number of men.
All of which is respectfully submitted in behalf of Mr. Babcock and myself –
Very respectfully your obedient servant
GEORGE H. SHARPE
Col & in charge10
This is an extraordinary document. From its submission can be marked the creation of professional U.S. all-source military intelligence. Sharpe makes it clear that all this has been done in the month since he was given the mission to set up his organization. He first lays down the organization of the Army of Northern Virginia from the corps to division levels and establishes the location of each division. Of vital importance is his statement that Lt. Gen. James Longstreet has taken three of the five divisions in his I Corps south of Richmond to the Petersburg area. That is fully one-third of Lee’s nine infantry divisions. Then he identifies the division of Major General A. P. Hill to be the strongest of the nine, the one to which replacements are sent first.
Next the report dissects the organization of the divisions and brigades, showing that there is a generally uniform number of brigades and within the brigades a uniform number (five) of regiments. However, he points out that there are exceptions to this rule, with some brigades having only four regiments or a mix of regiments and battalions.
From this base, he then shows the analytical process and formulas by which the strength of each unit was calculated and provides high, medium, and low estimates of Lee’s infantry strength. He appears to give greater confidence to the lower estimate but does not clearly commit himself.
This was the fruit of a body of cross-checked and cross-indexed information that had been accumulated over the preceding month and may have included information that Babcock had retrieved from the War Department files after he was hired by Burnside. It is a tutorial for the commander which Hooker no doubt appreciated, for it gave him confidence in Sharpe’s ability.
While the analytical formulas and techniques would later be refined to produce highly accurate numbers, Sharpe’s figures at even the upper levels (42,500) were wrong by only 15 percent. As his confidence grew so did the errors. The Army of Northern Virginia reported a “present for duty” strength as of the end of March of 60,298, with the six infantry divisions at the strength of 48,982 men. The medium estimate (39,275) is off by 20 percent. The lowest estimate (32,500) is off by 34 percent.11 Nevertheless, for an effort that began from a standing start, a fact Sharpe was at pains to remind Hooker of at the beginning of the report, it was a vast improvement over anything that had been done in the past. Its importance was in the establishment of a systemized collection and analytical process—a historical first. It was a system that would only improve with practice and experience. (See Appendix G: Abstracts from Field Returns of the Army of Northern Virginia.)
Sharpe ended his report with the phrase “submitted in behalf of Mr. Babcock and myself” to ensure proper credit to his invaluable deputy. Nothing will poison the harmony of a staff more than the chief taking all the credit. In another such gesture of good will, Babcock noted in an added memorandum to another report of the same day:
If not otherwise reported to the General it should also be stated, although the credit thereof is not due to this department, that Capt. Fisher (Chief Signal officer) is in possession of the full code of signals used by the enemy’s Signal Corps, with the exception of the numbers, and that their messages are read, daily, by his officers, whenever they can be observed from our stations.12
This report consisted of a breakdown of all of Lee’s infantry divisions by brigade and location. Importantly, Babcock made it clear that he was not sure of the full complement of each brigade. In still another memorandum of March 15, Sharpe provided an order-of-battle of Stuart’s Cavalry Division down to brigade and regiment. His only significant error was to identify Col. Thomas A. Munford as commanding a separate brigade, when he only commanded the 2nd Virginia Cavalry in Brig. Gen. Fitzhugh Lee’s Brigade, and to confuse that regiment with the 12th Virginia Cavalry.13
It did not take long for Sharpe to conclude that there is more to an enemy’s strength than is found in his order-of-battle. Hooker had been interested in the Confederate intelligence collection and raiding activities based in the Northern Neck and had directed Sharpe to send an agent to investigate. To Sharpe there were larger implications in these activities than Hooker had anticipated. In a report (through Butterfield) to Hooker on March 21, he gave the following reply:
In examination of prisoners and other agents furnishing information my attention has lately been drawn most strongly to the question of the subsistence of the Southern army, which seems to be nearing the point of total failure. From a mass of testimony, that in a court of record, would authorize the judge to rule, that no further evidence on that point would be received, we know that the ration of the army opposite us is as follows:
1 pint flour & ¼ lb bacon or pork per diem.
A month later he could report there was no improvement in the subsistence of the Army of Northern Virginia, as he describes in a letter to his uncle:
I have sent you two or three times Richmond papers and hope you will not think that I tear them in two. All southern papers are only published on a half sheet. By remarking the advertisements as well as the editorials you will notice what a state of destitution they are approaching $3000 are paid for substitutes and none are to be had at that price. $160 a pair for ordinary boots $150 a lb for pork and $15. a bushel for potatoes.
I have many prisoners brought out to me who have wretched shoes on for which they paid $20.
I had a man from Richmond yesterday & he paid $8. per day at the hotel. Desertion is now taking place from Geo & Miss Regts which have heretofore stood firm. All this of course is intended to be private.14
A few well-clad boastful exceptions proved the rule, “but the ragged and beggarly fellows who formed their rank and file on the other side did not so understand the ‘tea cup of flour and patch of bacon’ which was daily issued them by those who are interested in the success of the rebellion.” All other commissary stores, including candles, had simply been exhausted. The Confederate Government was forcing farmers to stop growing tobacco and cotton and grow wheat and corn instead. Government agents were essentially forcing the sale of meat and grain at below market prices. Sharpe described the sorry state of affairs:
The spring wheat crop in the south is not a regular one—their corn always comes in more or less abundance but their wheat is only an occasional success—and it is now known that this year it will fail. There is therefore nothing to carry the southern army through to the next crop, if we exercise the greatest vigilance in closing against them their outside granaries. And I am convinced that one of the greatest of these I am satisfied is the Northern Neck.
Sharpe then described from a number of sources the immense amounts of foodstuffs and stock that the Confederates were obtaining from the Northern Neck and that “Southern agents are now employed in buying it up and crossing it over the Rappahannock.” So robust was this traffic that Sharpe stated that even should all their boats be seized, it would hardly close down the operation since it was so lucrative. To emphasize its importance once again he stated clearly that four of the six Confederate divisions facing them had moved closer to the Northern Neck to be closer to their main source of supply. Finally, Sharpe concluded his report with the pointed hint, “Believing, General, that these facts are worthy [of] attention, they are respectfully submitted for such action as the judgment of the Commanding General may dictate.”15
Sharpe’s encouragement did prompt some attempts to stop the traffic, but lack of a sustained effort with anything more than small parties proved a failure. Sharpe, however, would continue to worry this bone of the Northern Neck even into 1865, when he finally convinced Grant to put a stop to it. Sharpe was clearly thinking beyond his normal tactical and operational purview. He had grasped how the enemy’s national logistics difficulties would affect the ability of his army to fight and that this analysis was appropriate for the army commander to take into serious consideration. Ironically, no one on the Union side was able to conceive how long and hard Johnny Reb could fight on a “tea cup of flour and patch of bacon.”
The next day Sharpe provided the report that Hooker had requested, showing that the Black Horse Cavalry was working closely with civilians in the Northern Neck. One of the civilians, already well known as a shady character by Brigadier General Patrick, was suborning desertions from the Union army, forcing his slaves to exchange clothes with them and then sending them back across the Potomac to head north ostensibly now civilians.16
That same day Sharpe reported on scout Dan Cole’s return and his observations. They confirmed his own analysis of the subsistence crisis in the Army of Northern Virginia. In particular, he had spoken to a quartermaster in Richmond who told him there were 90 days’ supplies stored there, but added, “God knows when we shall get any more.”17
Sharpe had thrown a wide net after his appointment. Not only was he concentrating on the enemy’s order-of-battle but also actively engaging in counterintelligence operations to disrupt the enemy’s intelligence collection against the Army of the Potomac. The army, in fact, lived in a security sieve created by its location amid a hostile population that actively sought and transmitted information to the Confederates. Sharpe’s reports led to the severe restriction on movement that would amount to house arrest for the local civilian population the closer the army got to striking.
In an effort to plug another source of leaks of information, and possibly at Sharpe’s recommendation, Hooker issued General Orders No. 48, on April 30: “The frequent transmission of false intelligence and the betrayal of the movements of the army to the enemy, by the publication of injudicious correspondence of an anonymous character, makes it necessary to require all newspaper correspondents to publish their communications over their own signatures.” Journalists who disregarded this requirement would be barred from the army and their publications banned from distribution. This had the unexpected effect of creating the byline in American newspaper journalism. Hooker went further to prevent information in the Northern press from being acquired by the Confederates by forbidding the exchange of newspapers. The stoppage of newspapers was quickly noted with annoyance by the Richmond Whig as one of the “new leaves turned over by Gen. Hooker.”18
The BMI was able to identify other operational security gaps, such as when scout John Skinker, also on March 11, reported that the scouts from Averell’s command completely neglected to cover an important area. Babcock immediately forwarded this report to Butterfield. This, however, could not have been popular with the cavalry, which resented the BMI’s trespassing on its traditional intelligence collection role.19
Skinker, at Babcock’s direction, established the BMI’s signal station, employing a “clothesline code” on the Confederate side of the river in Fredericksburg itself. On March 11, Babcock reported to Butterfield that “A clothesline with one piece denotes that the forces in the vicinity of Fredericksburg are on the move. An empty line denotes that they have all gone away. Two pieces shows that they are in force as they have been since the fight, three pieces that they are being reinforced.” Babcock’s obituary in 1908 mentioned a clothesline code being operated by a Negro woman. Butterfield, in his communications with Hooker, referred to this as “Sharpe’s signals.”20
Sharpe’s intelligence preparation of the battlefield included more than the enemy’s order-of-battle and poor logistics. A thorough knowledge of the enemy’s communications by road and rail would be vital in order to calculate the time required for enemy forces to move from one point to another and to sustain logistics support of those forces. That same information was equally vital for calculations necessary for planning friendly movement along these same routes. Without such knowledge, offensive operational planning was impossible.
Early in March Sharpe had written to the assistant adjutant general at Fort Monroe, trying to find a former employee of the Richmond, Fredericksburg & Potomac Railroad (the RF&P), Alexander L. Worrell, who had surveyed the line. Sharpe wanted either Worrell or the survey.21 The fruit of this survey in the form of numerous tables is found in the Hooker Papers at the Huntington Library. This collection contains the official documents Hooker took with him after his relief from command of the Army of the Potomac in June 1863. These are original documents and in many cases may well be the only copies to have survived. Worrell’s survey appears to have been the basis of a seriously expanded project. The various elements of this collection of tables shown below are a comprehensive analysis of the Confederate rail and road routes in Hooker’s intended area of operations.
Lee’s main supply route (MSR) ran along two railroads. The Richmond, Fredericksburg & Potomac Railroad (RF&P) and the Virginia Central Railroad. The RF&P ran 61 ½ miles from Richmond to Fredericksburg around which the Army of Northern Virginia was deployed. The Virginia Central ran roughly parallel to the RF&P until they both met at Hanover Junction, 24 miles from Richmond on the RF&P and 28 miles from Richmond on the Virginia Central. Hanover Junction lay 40 miles south of Fredericksburg. The Virginia Central then turned west through Gordonsville, Charlottesville, and into the Shenandoah Valley. Hanover Junction was identified as an especially critical target. Military supplies came up the RF&P to Lee from Richmond, while food supplies came through the Virginia Central largely from the granary of the Valley. Those supplies were stockpiled in warehouses at Hanover Junction for transfer to the RF&P. If there ever existed a vital communications/logistics node it was Hanover Junction. It was manifestly necessary to the survival of the Army of Northern Virginia.
These tables essentially were lists of the points of vulnerability of Lee’s logistics windpipe—bridges. For example, there is a description of the major bridge over the South Anna River just south of Hanover Junction: “600 ft long 73 ft high, stone piers and abutments same character of construction as the bridge over the Rappahannock at Fredericksburg & 3 or 5 ft shorter.”22
In addition to the railroad and road surveys, the Hooker Papers also yielded equally detailed analyses of the 19 Rappahannock and 14 Rapidan River fords. None of them show any evidence that Sharpe was involved in the creation or at least transmittal of these analyses other than the endorsement notation in his hand, “Memoir on the crossing places of the Rappahannock from Bealton to Kelly’s [Ford]” found among these reports. However, another analysis, identical in form, of the Rappahannock and Rapidan fords, found at the National Archives, was completed by the BMI and dated May 13.23 Sharpe’s scouts were constantly crossing the fords of both rivers to conduct operations behind Confederate lines and would have been ideal observers.
These reports are typical of the detailed and patient work the BMI was doing. That contention is supported by Stephen Sears in his superb account of the battle.24 The information in the ford analyses includes road access; condition of banks on both sides; ability of cavalry, artillery, and infantry to cross; and obstructions and water depth. Among all these tables there is only one date—April 22—an indication of how long it took to compile this information.
Table 4.1. Distances on the Richmond, Fredericksburg & Potomac and The Virginia Central Railroads
By the last week of March, Sharpe’s operation was working smoothly. The scouts continued to bring a constant stream of information that fitted more pieces into the mosaic of the enemy’s order-of-battle and dispositions as well as providing vital information on the enemy’s own scouting and clandestine activities that helped maintain a higher level of operational security. Deserters and the occasional prisoner continued to yield order-of-battle information to Sharpe and Babcock’s gentle guile. The number of subjects for interrogation had not been large. February yielded deserters and prisoners from only 20 regiments and four batteries; March provided only 12 regiments and April just 13.25
Unlike the large numbers of men captured during the heavy fighting of the Peninsula Campaign and Antietam, there had been little combat since the battle of Fredericksburg with resulting few prisoners. Most of the men interrogated had arrived by way of desertion, an overt act that severed loyalties and encouraged a desire to please and a willingness to be forthcoming. There may have been fewer deserters than prisoners, but the deserters more than made up the numbers by the quality of information they yielded, as Sharpe underlined when he wrote, “frequently deserters have been able to furnish us with complete organizations of their Brigades, Divisions &c.”26 The only problem with deserters was that they tended to come from the enemy units closest to Union lines. The further away a unit was, the less likely it was a man would risk such a long journey through an area thick with other units and suspicious provost guards. This was particularly true of the cavalry. (See Appendix H: Regiments/Battalions from which Deserters and Prisoners Were Interrogated.)
It was time, Sharpe decided, to fill a major remaining intelligence gap—lack of direct access to information from Richmond itself. Patrick knew the man who could help him.
James L. McPhail was the civilian provost marshal of Maryland. Formally that position entailed conducting conscription and apprehending deserters. Like Sharpe, McPhail saw that his duty lay beyond the official narrow confines of his job description. He had a nose for disloyalty and spies. He was a hard-headed Unionist who had achieved notoriety for his “sensational arrest of a pro-Southern country judge by force while court was in session.” He had been praised for arresting a number of Southern agents in the secessionist-minded city of Baltimore. That city was the natural focal point for agents and contraband smugglers traveling up and down the river to the Chesapeake and Virginia. Patrick’s and McPhail’s police duties overlapped enough for Patrick to recommend to Sharpe the provost marshal, whose knowledge of the contraband situation would be a good man to help him fill his intelligence gap.
By the last week in March Sharpe was in Baltimore conferring with McPhail, who, by happy coincidence, had been grooming an agent to be placed in the Rebel capital. Both men would refer to this agent—subsequently identified as Joseph H. Maddox, a Southern gentleman and slaveholder—as “our friend” in all their correspondence. His cover was as a purveyor of Northern goods in desperately short supply in Richmond. A sloop dropped him off quietly onto Virginia’s shore on April 11. Time would tell if McPhail’s investment would pay off. Hooker would have great need of it.27
His relationship with McPhail was not a one-way street. As deputy provost marshal, Sharpe was also aware of criminal activities in the smuggling of contraband supplies to the Confederacy which he provided to McPhail. In May he notified McPhail of the contraband business conducted on St. George Island in the Potomac River to the Virginia shore by Samuel G. Miles. That was a valuable piece of information to McPhail, who knew Miles as “an active Rebel” in Baltimore. Working together they ensured that Miles was unable to get his goods to Richmond. Eventually, this case would be linked to Maj. Gen. Benjamin Butler’s murky cooperation with smuggling contraband supplies in the following year.28
In early April a unique informant presented himself to Sharpe. James Craig, a Scot immigrant and Unionist, had fled Richmond with a wealth of detailed information on the transfer of Longstreet’s Corps from Fredericksburg through Richmond. That and other sources on Longstreet’s absence would be critical to the development of Hooker’s plan of campaign. Craig was also an eyewitness to the “bread riots” of starving women driven to desperation by unaffordable prices for food that had convulsed Richmond and had been reported in the National Republican on April 9. The report was a window into the social and economic tensions within Southern society that could have an eventual effect on military operations. For Sharpe, it was an eye-opener to the possibilities that lay with casting his intelligence net beyond the immediate military situation. There is nothing like the news of a starving family at home to destroy a soldier’s morale. He also saw the propaganda value of the eye-witness account and leaked it to the National Republican. Four days later the paper ran an initial story about the riot:
[The riot] caused the greatest consternation among the authorities. The women were the heads of families of the working classes, and were actually starving, many having been compelled to be on the street. A repetition of the demonstration is feared, and every precaution is being taken to avert it. The effect upon the troops was very demoralizing, the men being very clamorous, and demanding that their families be fed.
On April 25, the paper ran a fuller article, dated the day after Craig gave his statement and using some of the same phrasing such that the Richmond authorities had blamed the riots on “Irish and Yankee hags.”29
In quick succession, the BMI began receiving from its various sources the signs of a splendid opportunity to take Lee unawares. Lee’s army remained concentrated around Fredericksburg on the southern side of the Rappahannock River, and a wide gap in the forces guarding the upriver fords had opened up.
The intercepted deceptive flag signal that Lee had credited detailed an attack by the Union Cavalry Corps on the move up the river against the Confederates in the Shenandoah Valley. In response, Lee deployed Stuart’s cavalry opposite the Union cavalry, leaving a 20-mile gap in the Confederate defenses northwest from Fredericksburg along the river and leaving the vital Kelly’s Ford unguarded. Isaac Silver, living near Fredericksburg, reported that Lee’s army continued to be concentrated in that area. He also reported that although the U.S. Ford 10 miles up the Rappahannock from Fredericksburg was guarded, there were no other Confederate forces within 5 miles. A Confederate deserter, another Virginian, provided detailed information that confirmed Silver’s report. Sharpe’s scouts confirmed it a third time in this summary on April 28, forwarded by Butterfield to the commander of the Cavalry Corps:
One of Colonel Sharpe’s men just in from Kelly’s Ford says in his opinion no large body of infantry there. Held mostly by cavalry and artillery. Rebel sympathizers on this side believe enemy have fallen back beyond Rapidan, meaning to make that their line of defense. Says also that Fitzhugh [W. H. F.] Lee has taken place of Fitz. Lee, between Kelly’s and Culpeper. Latter gone to Valley, to join Hampton and Jones. Enemy’s artillery horses said not to be able to move their guns. They think our cavalry move a feint, and that the crossing will be made at United States Ford, where they are still at work.30
The BMI analysis of Kelly’s Ford called attention to it at this time as an excellent crossing point over the Rappahannock. It read:
Good road and easy of access
The banks are low on both sides
Cavalry, Artillery and Infantry can cross
The ford may be made less practicable by obstructions and an abattis, and defended by rifle pits.
The water is about two feet deep known from personal observation
Between Wheatly’s ford on the left, there are two places may be crossed by infantry only.31
Hooker determined to seize this priceless opportunity. His plan of operations would be judged the best devised in the entire war. It essentially had three parts. The first element would be the departure of the entire Cavalry Corps under Brigadier General Stoneman, to sever the RF&P, Lee’s main supply route to his logistics base at Richmond and beyond. Sharpe had identified the only two Confederate forces he was likely to encounter—Fitzhugh Lee’s cavalry brigade at Culpeper and a small provost guard at Gordonsville. “It is not in the power of the rebels to oppose you with more than 5,000 sabers, and those badly mounted…” These figures were close to the mark. Sharpe also supplied the BMI analyses of the railroad to assist him in the destruction of key elements:
Stoneman’s orders were clear. From Gordonsville it is expected that you will be able to push forward to the Aquia and Richmond Railroad [RF&P], somewhere in the vicinity of Saxton’s [Hanover] Junction, destroying along your whole route the railroad bridges, trains, cars, depots of provisions, lines of telegraphic communications, &c. The general directs that you go prepared with all the means necessary to accomplish this work effectually. As the line of the railroad from Aquia to Richmond presents the shortest one for the enemy to retire on, it is more than probable that the enemy may avail himself of it and the usually traveled highways on each side of it for this purpose…32
If this were the case, then Stoneman was to block or delay their retreat, and if that were not possible, to harry it aggressively.
Hanover Junction, with its surrounding bridges, was the primary target. As Sears observed, “Within a few miles of Hanover Junction were three good-sized wooden RF&P bridges … over the North Anna River, Reedy Swamp, and Pole Cat Creek, Burning any one of them would shut down the railroad instantly and for a substantial period.” These great, high, wooden structures were begging for the torch. He certainly had the force to do it, with 9,895 cavalrymen and 22 guns manned by 427 gunners. Even if he unexpectedly ran into Stuart’s entire command, he would outnumber it by two and a half times. Certainly, Sharpe had done everything within his grasp to make this a successful mission.33
Hooker’s Plan for the Chancellorsville Campaign based on the BMI’s Intelligence Preparation of the Battlefield (IPB), April 1863. (After Jespersen)
Second, Hooker would fix Lee’s attention on Fredericksburg, with three corps (58,000 men of I, III, VI Corps) under Maj. Gen. John Sedgwick, commander of VI Corps. Then, third, he would take three corps (41,000 men of V, XI, XII Corps) and cross the Rappahannock upstream of Fredericksburg at Kelly’s Ford and others, then march south quickly through the Wilderness to catch Lee between himself and Sedgwick. The choice of the three corps he would take on his encirclement was dictated not by their combat effectiveness but by the placement of their camps; they were nearer the fords, and their movement would be the most direct and shortest. Unfortunately, that meant that for this most critical maneuver, Hooker would be taking the newer, weaker, and less experienced XI and XII Corps. II Corps was left as a diversion east of Fredericksburg. As needed, Hooker would transfer Sedgwick’s I and III Corps to join his three. Either group of corps was equal in strength to Lee’s entire infantry force.34 It was a magnificent trap.
Consider how all the parts of that plan depended upon information provided by Sharpe’s BMI. His scouts had accurately identified the dispositions and encampments of all of Lee’s divisions and most of his brigades, allowing Hooker to pick the most unguarded and thus advantageous route to strike them. That route revealed itself as Kelly’s Ford by the combined work of the agent Silver, scouting, and the analysis of the suitability of the ford. The information that Longstreet had taken one-third of Lee’s divisions to the siege of Suffolk told him that he would be facing a weakened enemy. It also made it clear that they were out of range of intervention in the coming battle, for several days at least. Thus, Hooker would be facing only two-thirds of Lee’s force, which, combined with Babcock’s strength estimates, established that Hooker would have an advantage of more than two to one in manpower. Finally, the analysis of the railroads revealed vulnerabilities and established that the destruction of Hanover Junction and bridges would effectively cut Lee off from his supplies and block his retreat.
Sharpe had handed Hooker General Lee’s head on a silver platter. In a letter home on the 22nd, he expressed the optimism that had suffused not only the staff but the entire army. “I am not at liberty to say anything about our movements, but I beg you to rest strong in the faith that we are now commanded by a man who means to fight to win—and that when the blow is struck, it will be one of the heaviest ever felt on this continent—& with more than the results of a battle.”35
In the week before the army would begin moving, Hooker’s distrust of journalists appeared more than justified. In a letter of April 21 to Secretary of War Stanton, he described how the Morning Chronicle had published a letter by Dr. Jonathan Letterman, Medical Director of the Army of the Potomac, that revealed important information on the strength of the army. Hooker wrote, “The chief of my secret service department [Sharpe] would have willingly paid $1000 for such information in regard to the enemy at the commencement of his operations, and even now would give that sum for it to verify the statements which he has been at great labor and trouble to collect and systemize.” The letter strongly suggests that Hooker by this point was well pleased with Sharpe’s efforts.36
He was pleased enough to make Sharpe the president of a commission for the trial of a reporter who was charged with revealing vital information on the army in an article. An example needed to be made. E. F. Denyse had written an article for the New York Herald, published on March 14, that related all sorts of preparations for an advance of the army. Hooker had wanted a good example made, as Sharpe would write in the record of the proceedings:
While none is more desirous than the Major General Commanding, to extend all proper facilities to the press in its efforts to supply the public with reliable intelligence, it cannot be tolerated that newspaper correspondents should abuse the privilege of remaining with the army by the publication of intelligence certain to be of use to the Enemy and most likely soon to reach him.
Denyse pleaded guilty and was sentenced to six months’ hard labor, followed by permanent expulsion from the army. With a deft touch, Hooker commuted the sentence to expulsion, and Denyse was marched under guard and put on board a ship departing Acquia Creek for Washington. Sharpe and his staff would despise the New York Herald through the rest of the war for the way it fecklessly printed sensitive material.37
It was a responsible officer, Maj. Henry C. Jenckes, 2nd Rhode Island Regiment, who sat down after coming back from three days of picket duty and took pen in hand to alert his chain of command about the lax security he had become aware of. He wrote of his concerns on April 16 about how information was leaking across the river to the enemy despite orders forbidding any communication with the enemy by pickets. He wrote that a local civilian informed him of the move of the Union cavalry on the left on the 13th, the same day the Confederates called across the river that the cavalry had moved. On the 14th Confederate pickets announced that the paymasters had arrived in the Union camps only 15 minutes after Jenckes himself was officially informed. The Confederates also yelled across the river, “You need not be so still; we know all about it; you have got orders to move.” His report triggered an examination by Patrick on the 21st.
He already had a poor opinion of the pickets’ conduct, which he made a point to say he had complained of before. He found a lieutenant in command of the pickets actually engaged in sending little boats across the river with coffee, sugar, and newspapers. The young man was put out when Patrick smashed his boat. He then related a conversation between the pickets on the 15th.38
The part of the conversation was about rations. Secesh [the secessionist] then asked, “Any signs of a move?” Reply, “Yes, we have got eight days’ rations, and expect to move in a few days. We have three days’ rations in our haversacks and five in our knapsacks.” Secesh then asked, “Where is the move to be?” Reply, “Up to the right.” Secesh then asked how we were going to get transportation, or whether we would use the railroads. Our picket replied that he thought the trains would be kept up by pack-mules.39
It was enough to give Hooker, who had thrown up such strict operational security measures, an aneurism. Luckily the Confederates did not seem to make any use of the intelligence gems the careless pickets were throwing across the river. The picket had said the army was going to move in a few days and it did not. That may have convinced whoever was receiving the information that it was a false alarm and made no more of it.
About the same time a rumor began spreading through the army camps that the Confederates had run a telegraph wire under the Rappahannock to supply intelligence to Lee. On the 25th The Philadelphia Enquirer ran the story. It related how a guard stationed outside a house on the Rappahannock at Falmouth heard a clicking sound resembling a telegraph coming from within and reported it. He was told to investigate, and “on opening the door, he discovered a party of four or five persons, one of whom was seated at a telegraph instrument, sending messages by a submarine telegraph wire across the Rappahannock.”40
The New York Times published a similar story on April 28, with a header of “Headquarters Army of the Potomac, April 26, 1863.” Edwin Fishel states that it was an official dispatch from the army that the censor in Washington stopped but that the correspondent mailed to his editor. Apparently smarting from the Enquirer story, Major General Stoneman ordered the newspaper’s correspondent “immediately sent out of the lines of the army, never to return.” Sharpe’s hand can be seen in this since he perused not only Southern but Northern newspapers.41
The question remained whether there was any truth to the underwater telegraph story. The Confederates clearly had a rapid transmission system for intelligence from Falmouth to reach Lee’s headquarters at Culpeper. The Confederates had advance knowledge of the attack by Brigadier General Averell’s cavalry division across Kelly’s Ford on March 17. Brigadier General Fitzhugh Lee received a telegram from Culpeper the day before as the Union cavalry concentrated and prepared for the next day’s attack. The distance between Falmouth and Culpeper is over 30 miles. Averell’s regiments received their orders from the Headquarters, Cavalry Corps, on the 15th, and his entire command departed the camps of the army on the 16th. It appears that Fitzhugh Lee received a warning in record time, the very day Averell’s division departed its camps. That argues for some plausibility to the underwater telegraph story. On the other hand, there is no reference to this story in official records or in Patrick’s diary.42
Further evidence for a Confederate spy in Falmouth is a letter that Lee sent to Stuart on March 12. He wrote, “The information from Falmouth is that the enemy will, as soon as roads permit, cross at United States Ford, Falmouth, and some point below, the attempt at Falmouth to be a feint.” This is an astounding statement. Lee had summarized in broad strokes most of Hooker’s operational plan for the coming campaign. Then he stated, “This information comes from citizens, and especially from a lady, wife of one of our officers, and I do not know how true.”
Patrick frequently visited a Mrs. John Seddon, whom he thought of as friend, at her home in Falmouth. She was the wife of a Confederate officer and had been the specific object of an order “prohibiting against manipulation of window blinds or windows proper in a manner that could convey information.” Whether Lee was referring to Mrs. Seddon in his letter to Stuart as his source is unknown. Whether the source of information was Mrs. Seddon, another woman, or an underwater telegraph or a combination, it went dead just before the Army of the Potomac marched out of its camps for the collision at Chancellorsville. The posting of guards at every private home on the route of the army may have done as intended to prevent civilians from sending information. The proof would be found in the complete surprise of Lee. He may have learned of the outline of Hooker’s plan in March, but apparently never learned the vital “when.” His letter also indicated that he was not sure of the information. There was nothing subsequent that confirmed the information and no trigger that warned him of the sudden movement of the Army of the Potomac.43
Hooker put the army into motion on April 27. He had preceded the march with a ruthless exercise of local security to prevent the local population from sending news of the army’s movement immediately to Lee. The entire population along the army’s march was put under virtual house arrest. He also kept the full plan from his subordinates except for his second in command, Maj. Gen. Darius Couch, and Butterfield, his chief of staff. All this worked; not a word leaked to Lee that a giant trap had been sprung on him and the Army of Northern Virginia. The measure, however, by denying his corps commanders a thorough understanding of the commander’s intent, would have unforeseen consequences. Sharpe would accompany Hooker on the enveloping movement, while Babcock and McEntee would remain with Butterfield at Falmouth.
Command emphasis on intelligence collection at this crucial time was intense, as Butterfield on April 28 “Cautioned Sharpe, signal officers, and Lowe [Balloon Corps] to be vigilant and watchful; to get all information possible.” Sharpe also was able to report that deserters had revealed that as of April 26, three of the four divisions in Jackson’s II Corps had not moved and showed no signs of moving. Unfortunately, strong winds kept Lowe from ascending with his balloons Washington and Eagle. Hooker was not ready to write off the balloons and ordered Lowe to ascend at night “to find the enemy’s camp fires.” He specifically wanted “Some one acquainted with the position and location of the ground and of the enemy’s forces” to go up. He recognized that as intrepid as Lowe was, his reporting suffered from a lack of military training and experience. The winds abated, allowing the balloons to ascend in the day, but they were unable to see much due to dense ground fog. “What few camps that were visible, however, appeared to be occupied as usual.” Both Sharpe’s interrogations and Lowe’s balloons were confirming that Lee had not stirred.44