A Cable from Napoleon
APPROVED FOR RELEASE
CIA HISTORICAL REVIEW PROGRAM
22 SEPT 93
The story of a critical intelligence finding almost unrecorded in the history of French intervention in Mexico during and after the Civil War is reconstructed here from official records in the National Archives.
A CABLE FROM NAPOLEON
Edwin C. Fishel
The years 1864-67 saw the United States facing one of the severest international problems in its history: an Austrian prince ruled Mexico and a French army occupied the south bank of the Rio Grande. It was toward the end of this period that the Atlantic cable went into permanent operation. Thus the United States had both the motive and the means for what was almost certainly its first essay in peacetime communications intelligence.l
The nation had emerged from the Civil War possessing a respectable intelligence capability. Union espionage activities were generally successful, especially in the later stages of the war; Northern communications men read Confederate messages with considerable regularity (and received reciprocal treatment of their own traffic from the rebel signalmen); and there were intelligence staffs that developed a high degree of competence in digesting and reporting these findings.2
With the war over in 1865, this new capability was turned against Napoleon III and his puppet, Emperor Maximilian of Mexico. In the struggle to get the French army out of North America and Maximilian off his throne, this government had the use of an intelligence enterprise which, though conducted on a small scale, turned out to be very effective. Up to the last weeks this intelligence operation consisted of competent reporting on the part of espionage agents and diplomatic representatives; but when a crisis developed at that point, these sources were silent, and it was a cablegram from Napoleon to his commanders in Mexico that yielded the information needed by the nation's leaders.
As an intelligence coup the interception and reading of this message were hardly spectacular, for it passed over fifteen hundred miles of telegraph wire accessible to United States forces and, contrary to later assertions that it had to be deciphered, it appears to have been sent in the clear. Nevertheless, the event was an outstanding one in the history of United States intelligence operations, not simply because it represented a beginning in a new field but also because the message in question was of crucial importance.
State of the Union, 1861-65
The crisis in which America's intelligence capability asserted itself did not come until after the nation had spent five anxious years watching the European threat develop.
Napoleon had sent an army to Mexico late in 1861, assertedly to compel the payment of huge debts owed by the government of Mexico. His object, however, was not simply a financial one: a new commander whom he sent to Mexico in 1863 received instructions (which leaked into the press) to the effect that the Emperor's purpose was to establish a Mexican government strong enough to limit "the growth and prestige of the United States."3 At a time when the American Union appeared to be breaking up under pressure from its southern half, such a statement meant to American readers that Napoleon had no intention of stopping at the Rio Grande.
In June 1863 French arms swept the Liberal government of President Benito Juarez from Mexico City, and in the summer of 1864 Napoleon installed the Archduke Ferdinand Maximilian, thirty-two-year-old brother of Emperor Franz Joseph of Austria, on the new throne of Mexico. During this period the Northern people, their belligerence aroused by the Southern rebellion, were clamoring for action against France - action that might well bring disaster upon them. Aggressive behavior by the United States might give Napoleon the popular support he needed to join hands with the Confederacy in a declaration of war, a development that could provide Secession with enough extra strength to prevail.
While the Civil War lasted, Congress and the public were held in check largely through the prestige and political skill of the Federal Secretary of State, William H. Seward. But when the War was over - by which time the government had reason to believe that Napoleon had become disenchanted with his puppets in Mexico - Seward was ready to turn his people's aggressive demeanor to advantage, and he warned Napoleon that their will would sooner or later prevail. Before this statement reached Paris, however, the United States Minister there, John Bigelow, who had been mirroring Seward's new firmness for some months, had in September 1865 obtained a tentative statement from the French that they intended to withdraw from Mexico.4
While Bigelow was shaking an admonitory finger at the French Ministry of Foreign Affairs, an American military fist was being displayed before the French along the Rio Grande. Promptly upon the silencing of Confederate guns, General Grant sent Philip Sheridan, second only to William T. Sherman in the esteem of the General-in-Chief, to the command of the Department of the Gulf, with headquarters at New Orleans. A considerable force was posted along the Mexican frontier and designated an "army of observation."
Sheridan and Intelligence
Sheridan, thirty-four years old and the possessor of a reputation as a gamecock, adhered strongly to an opinion prevalent in the Army that a little forceful military action now would save a full-scale war later. The audacious statesman who was directing foreign policy at Washington was, to Sheridan, "slow and poky," and the general found ways of giving considerable covert aid to the Juarez government, then leading a nomadic existence in the north of Mexico.5 Sheridan and Seward, though the policy of each was anathema to the other, made an effective combination.
One of the ways in which Sheridan could exercise his relentless energy against the Imperialists without flouting Seward's policy was in collecting intelligence on what was going on below the border. There was an interregnum at the United States Legation in Mexico City, and all the official news reaching Washington from below the Rio Grande was that supplied by the Juarist Minister to the United States, Matias Romero, a scarcely unbiased source if a prolific one.6 Sheridan quickly undertook to fill the gap.
This task must have been decidedly to the general's taste, for he had been one of the most intelligence-conscious commanders in the Civil War.7 He had achieved something of an innovation in organizing intelligence activities when, during his 1864 campaign in the Shenandoah Valley, he established a group of intelligence operatives under military control. His previous sources of information, local citizens and Confederate deserters, had both proved unreliable. "Sheridan's Scouts" were a military organization in a day when it was customary to have civilians perform most of the intelligence-gathering tasks other than battle-zone reconnaissance. After the war, Major Henry Harrison Young, the Scouts' commander, and four of his best men went to the Gulf Department with Sheridan.
Sheridan also, in common with numerous other commanders North and South, had an acquaintance with communications intelligence as it was produced in the field command of that day. By the time the Civil War was well advanced, Signal Corpsmen in every theater had learned how to solve the enemy's visual-signaling alphabets, and they derived much information for the commanders by keeping their field glasses trained on enemy signal stations.8 There was not likely to be any opportunity for such methods along the Rio Grande, however, and no more likely was the possibility of tapping telegraph lines carrying useful information.
Young and his four men were dispatched to important points in northern Mexico to report on movements of the Imperial forces and the various projects of ex-Confederates who were joining Maximilian's forces and attempting to establish colonies under his flag.9 Judged by the accuracy of the reports reaching Sheridan and the strong tendency of the Southerners' projects to abort after coming under his notice, the work of these five men was most effective.10
1866, Year of Telegrams and Tension
The critical question - whether the French would tire of their venture and withdraw -was, however, one to which no intelligence service could divine an answer, for the French for a long time did not know the answer themselves. In 1865 Marshal François Achille Bazaine, now Napoleon's commander in Mexico, was informed by the Minister of War that he must bring the army home, and at about the same time he received word to the opposite effect from the Emperor himself.11 Napoleon's treaty with Maximilian by which the latter accepted the throne of Mexico contained a secret clause providing that French military forces to the number of 20,000 were to remain in Mexico until November 1867.12 As events were to prove, however, this compact was less likely to determine Napoleon's course of action than were the pressures on him represented by the United States' vigorous diplomacy and the rising military power of Prussia.
In April 1866 Minister Bigelow succeeded in pinning Napoleon down to a definite understanding, to the effect that the 28,000 French soldiers in Mexico would be brought home in three detachments, leaving in November 1866 and March and November 1867. Seward's reply to this promise was characteristic of his tone at this time: dwelling only briefly on the diplomatic niceties, he suggested that the remaining period of occupation be shortened if possible. The Secretary was in high feather; in the same month a protest by him induced the Austrian government to abandon an effort to send substantial reinforcements to the small Austrian force in Maximilian's army.13
In June Maximilian received a studiously insolent letter from Napoleon containing the stunning announcement that the French would withdraw. Attention now focused on whether he would attempt to hold his throne without French arms. The unhappy sovereign reacted first by dispatching his Empress, twenty-six-year-old Carlota, to Paris in a vain attempt to change Napoleon's mind. He soon decided to abdicate, then determined to remain on his throne, then wavered for many weeks between abdicating and remaining.14
Napoleon meanwhile had to contend not only with his protege's indecision but with some apparent recalcitrance on the part of Bazaine, who was variously suspected of having a secret agreement with Maximilian to remain in the latter's support, of being secretly in league with the Mexican Liberals, of profiting financially from his official position, and of having hopes of succeeding Maximilian. (There is evidence to support all these suspicions.)15 Soon Napoleon realized he had made a bad bargain with the United States; to attempt to bring the army home in three parts would risk the annihilation of the last third. Early in the autumn of 1866 the Emperor sent his military aide, General Castelnau, to Mexico with instructions to have the army ready to leave in one shipment in March, and to supersede Bazaine if necessary. Thus the evacuation was to begin four months later than Napoleon had promised, but to end eight months earlier.16
No word of this important about-face was, however, promptly passed to the United States government. At the beginning of November - supposedly the month for the first shipment - the best information this country's leaders possessed was a strong indication that Napoleon intended to rid himself of Maximilian. This was contained in a letter written to Maximilian by a confidential agent whom he had sent to Europe; it showed the failure of Carlota's visit to Napoleon. Somewhere between its point of origin, Brussels, and its destination, the office of Maximilian's consul in New York, it had fallen into the hands of a Juarist agent.17 Soon after Minister Romero placed it in Seward's hands, Napoleon's new Foreign Minister, the Marquis de Moustier, wrote his Minister in Washington, de Montholon, that the evacuation timetable was raising serious difficulties but that in no case would the November 1867 deadline for its completion be exceeded.18 This note should have reached Seward in early November (1866), but if it did, its strong hint that there would be no partial evacuation in that month was apparently lost on him.
When the French felt able to promise complete withdrawal in March, de Moustier revealed to Bigelow the abandonment of the three-stage plan. So alarmed was Bigelow by the prospect of a major outbreak of anti-French feeling in America that he refrained from sending the news to Seward until he had heard it from the Emperor himself, whom he saw on November 7. The November shipment had been cancelled for reasons purely military, the Emperor said, showing surprise that the United States had not known of the change. The order had been telegraphed to Bazaine and had been sent in the clear in order that "no secret might be made of its tenor in the United States."19 Undoubtedly the Emperor was perfectly sincere in implying that he expected the United States government to make itself a tacit "information addressee" on telegrams of foreign governments reaching its territory.
Receiving Bigelow's report of this interview, Seward struck off a peremptory cablegram to Paris: the United States "cannot acquiesce," he declared. The 774 words of this message unfolded before Bigelow on November 26 and 27, their transmission having cost the State Department some $13,000. On December 3 Bigelow telegraphed the Foreign Minister's assurance that military considerations alone were responsible for the change of plans and his promise, somewhat more definite than the previous one, that the French "corps of occupation is to embark in the month of March next."20
So strongly had this government relied on Napoleon's original promise that President Johnson had dispatched an important diplomatic mission to Mexico (republican Mexico, that is) - a mission that was already at sea, expecting, on arrival at Vera Cruz, to find the French leaving and Juarez resuming the reins of government. The mission consisted of ex-Senator Lewis D. Campbell, newly appointed Minister to Mexico, and General William T. Sherman, sent with Campbell to give the mission prestige, to advise Juarez in regard to the many military problems that would be plaguing him,21 and possibly to arrange for the use of small numbers of United States troops to assist the Liberal regime by temporarily occupying certain island forts.22
Evidence was accumulating that Maximilian and his European troops would soon be gone from Mexico,23 but it stood no chance of general acceptance in Washington. Such was the degree of trust now accorded Louis Napoleon that his promise to evacuate Mexico would be believed on the day when the last French soldier took ship at Vera Cruz.
At this juncture Sheridan's headquarters came into possession of a copy of a coded telegram to Napoleon from Bazaine and Castelnau. The message had left Mexico City by courier on December 3 and had been delivered to the French Consulate at New Orleans, from where it was telegraphed to Paris on the 9th. As will be explained below, there is every reason to believe that this message went unread by United States cryptographers. The possession of its contents would have been of great value, for the message (as translated from the version given by Castelnau's biographer) said:
New Orleans, 9 Dec 1866
To His Majesty the Emperor Napoleon at Paris.
Mexico, 3rd December.
Emperor Maximilian appears to wish to remain in Mexico, but we must not count on it. Since the evacuation is to be completed in March, it is urgent that the transports arrive. We think that the foreign regiment must also be embarked. As for the French officers and soldiers attached to the Mexican Corps, can they be allowed the option of returning?
The country is restless. The Campbell and Sherman mission, which arrived off Vera Cruz on November 29 and left December 3, seems disposed to a peaceful solution. Nevertheless it gives moral support to the Juarists through the statement of the Federal government.
Marshal Bazaine and General Castelnau24
As December wore on, rumblings from Capitol Hill indicated that Congress - the same Congress that was even then moving to impeach President Johnson - might attempt to take the management of the entire affair out of the Administration's hands. Word arrived from Bigelow that transports to bring the army home were ready to sail from French ports, but that information would by no means be convincing enough to reassure Washington. And that word was the last to be heard from Bigelow, as competent a reporter as he was a diplomatist. He was relieved as Minister by John Adams Dix, ex-senator, ex-general, who did not manage to turn his hand to report-writing until mid-February, after the crisis was past.25
Similarly, nothing that would clarify the situation was coming out of Mexico. General Grant received a report from Sherman, at Vera Cruz, containing two items of intelligence, highly significant and completely contradictory: two ships, waiting at Vera Cruz to take Maximilian home, had been loaded with tremendous quantities of royal baggage; and the Emperor had just issued a proclamation to the Mexican people announcing
First and last pages of the five-page message to Napoleon III from his commanders in Mexico, reporting on the situation there and asking instructions concerning the evacuation of the European forces. The French clear-text version, as repeated by General Castelnau in a letter to Napoleon on December 8, 1866 (and quoted by Castelnau's biographer), reads:
L'empereur Maximilien parait vouloir rester au Mexique, mais on ne peut y compter. L'evacuation devant etre terminee en mars, il est urgent que les transports arrivent. Nous pensons que le regiment etranger doit etre aussi embarque. Quant aux officiers et soldats frangais detaches aux corps mexicains, peut-on leur laisser la faculte de revenir? Le pays est inquiet. La mission Campbell et Sherman arrivee devant Vera Cruz le 29 Novembre et partie le 3 Decembre semble disposee a une solution pacifique. Elle Wen donne pas moins un appui moral aux Juaristes par la declaration du gouvernement federal.
His intention to remain. Sherman and Campbell were facing a dilemma, in that they could not reach Juarez without crossing territory held by the Imperialists, with whom they were supposed to have nothing to do. Sherman invited Grant to instruct him to go to Mexico City to see Bazaine, who, he was sure, would tell him the truth about French intentions, but nothing came of this suggestion. Wrote the general of the colorful pen and the fervid dislike of politics: "I am as anxious to find Juarez as Japhet was to find his father, that I may dispose of this mission."26
Tension mounted in Washington early in January as the Senate prepared for a debate on the Mexican question, and a wide variety of reports circulated, the most ominous being that half of the French forces were to remain in Mexico through the summer, and that Assistant Secretary of State Frederick W. Seward, who had sailed mysteriously from Annapolis on Christmas day, was on his way to see Napoleon. (He was en route to the West Indies on one of his father's projects for the purchase of territory.)27 But on January 12, before the Senate got around to the Mexican question, the War Department received a message from Sheridan at New Orleans transmitting the following telegram:
Paris Jan 10th
French Consul New Orleans
for General Cast[elnau] at Mexico.
Received your dispatch of the ninth December. Do not compel the Emperor to abdicate, but do not delay the departure of the troops; bring back all those who will not remain there. Most of the fleet has left.
Napoleon III's "Bring the army home" message, and the one by which General Sheridan transmitted it in translation to General Grant. The notation on the Sheridan-to-Grant message "Recd 230 PM In cipher" refers to its receipt and decipherment in the War Department, and so does not bear on Sheridan's later assertion that Napoleon's message was sent in cipher.
The phrase "will not remain there" was a translation error. It was corrected to "are not willing to remain" when Sheridan forwarded a confirmation copy of his telegram by mail later on January 12. "Most of the fleet has left" (referring to the departure of transports for Mexico) would have been better translated "Most of the ships have left."
Here now was a conclusive answer to both of the pressing questions, the French evacuation and Maximilian's future. The entire French force must be leaving; else there would scarcely be a question of compelling Maximilian to abdicate. And with the French gone, Maximilian, even if he remained firm in his decision to keep the throne, could hardly stand against the rising Liberals very long. The European threat to American soil could be considered virtually at an end.
How It Happened
Because of the historical importance attaching to the interception of this message and the Mexico-to-Paris message of a month earlier, the circumstances surrounding the interception are worth examining.
The two telegrams owed their existence to the successful installation of the Atlantic cable a few months before. The cable's own history went back to August 1857, when the first attempt to lay it ended in failure. A year later a connection was completed and the cable was operated for eleven weeks before it went dead, apparently because the use of a very high voltage had broken down the insulation. Renewal of the attempt awaited the development of better electrical techniques and the end of the Civil War. In 1865 a new cable was laid from Valentia, Ireland, but was lost six hundred miles short of Newfoundland. Another was started July 13, 1866, and brought ashore at Heart's Content, Newfoundland, on July 27. The ill-starred steamer Great Eastern, which laid it, then picked up the buried end of the 1865 cable and ran a second line to Newfoundland. Service to the public opened August 26.28
Thus Napoleon's September message to Bazaine passed after the permanent operation of a telegraph line across the Atlantic had been a reality for only a few weeks, and it must be conceded that the United States was reasonably prompt in availing itself of this source of intelligence -despite Napoleon's opinion to the contrary.
Although the first interception took place only a month after the French Emperor had virtually invited this government to read his mail, it appears that Napoleon's suggestion had nothing to do with it. The author of the intercept scheme, in all probability, was General Sheridan, and it is highly unlikely that Napoleon's remarks would have been communicated to him. In any case, no instructions for surveillance of the telegraph lines to obtain French messages appear in the correspondence to the Gulf Department from Army Headquarters.29
Years later Sheridan explained how the job was done: his telegraph operator and cipher clerk, Charles A. Keefer, one of the numerous Canadians who entered the Union and Confederate telegraph services, had succeeded in "getting possession of the telegraph and managing [a] secret line,"30 which presumably connected his office with the Western Union wires in New Orleans.
Keefer's "secret line" may not have been so remarkable a thing as Sheridan's cryptic account makes it seem, for there was a high degree of integration between the Military Telegraph system to which Keefer belonged and the commercial system over which the messages passed. Throughout the occupied areas of the South during and after the Civil War, the Military Telegraph service took over commercial and railroad telegraph facilities wherever they existed. These Military Telegraph offices accepted commercial as well as government business, and commercial offices of course sent and received thousands of military telegrams; many a telegraph circuit had a military office at one terminus and a commercial office at the other.
As the Reconstruction period advanced, this integration became even closer; when the wires were returned to the use of the companies that owned them, Military Telegraph officers remained on duty to take care of government business and exercise a loose kind of supervision over the commercial operations. At some places military and commercial operators worked side by side. The fact that Keefer's copies of the French telegrams were written on Western Union message blanks makes it appear that New Orleans was one of the cities where this arrangement was in effect. If it was not, and the Military Telegraph and Western Union offices there were located separately, they were nevertheless using the same wires for communication with distant points, which would have made it comparatively easy for Keefer to connect a "secret line."
This integration of operations went all the way to the top of the two telegraph systems. General Thomas T. Eckert, who had been the second-ranking member and active head of the Military Telegraph service, continued to be closely connected with it after becoming Assistant Secretary of War in 1866. In the period now under study Eckert was apparently occupying his War Department position and at the same time resuming his activities in the industry as Eastern Division superintendent for Western Union at New York.31
Sheridan also credited Keefer with having solved the French "cipher," 32 but there is strong evidence to the contrary:
(1) The amount of material Keefer could have had to work with was very small. The cable in its early years was used sparingly because of the very high tolls (note the $1,979.25 charge, in gold, that the French Consulate paid for the December 3/9 message). Thus Paris was still awaiting word from Castelnau at the end of November,33 although he had been in Mexico nearly two months. The only French messages referred to in any of the documents examined in the present study are the clear-text message that Napoleon said he sent Bazaine in September,34 the message of December 3/9, and the message of January 10. Accordingly, as the January message (to be discussed in detail below) was almost certainly sent in the clear, it is highly probable that the December 3/9 message from Bazaine and Castelnau to Napoleon was the only encrypted French telegram that passed between Mexico and France during the entire period of the French intervention.35 It is extremely unlikely that the code - for the message was in code and not cipher -could have been solved from this one message of eighty-eight groups.
(2) An examination of all available United States records that could reasonably be expected to contain such an item (if it existed) fails to uncover a decrypted version of the December 3/9 message or any other evidence that the government during the ensuing weeks had come into possession of the information it contained36
Somewhat surprising is the apparent fact that Sheridan did not send the message to the War Department cryptographers for study. On several occasions during the Civil War, these men had been able to read enemy messages referred to them. This experience (so far as it is recorded) was, however, limited to the solution of certain ciphers (some of which were relatively complex for that day),37 and the French code would have presented them with a strange and much more difficult problem. Union cryptographers at New Orleans had also once solved a captured message,38 a fact which may have induced Sheridan to rely on his own headquarters' capability and not turn to Washington.
It was the January 10 message from Napoleon, the only message mentioned in Sheridan's account of this episode, that the general said Keefer had solved. But there is every reason to believe that the French clear-text of this message is the message as received in New Orleans, and not a decoded version of that message. Note:
(1) The message heading. It is filled out in precisely the way that was standard procedure in telegraphic reception at that period. A considerably different format was used for the delivery of plain-text versions of friendly messages received in cipher, and since Keefer was also a Military Telegraph cipher clerk, he would probably have used that format or a similar one in writing up the plain text of a foreign cipher or code message. (This format is illustrated by the photostat of the deciphered version of Sheridan's January 12 message, of which Napoleon's message of the 10th was a part.)
(2) The difficulties that the copyist had with French spellings (Castelnau, dÃ©cembre, forcez, abdiquer, navires). These are the difficulties of a telegraph operator receiving in a strange language. A cryptographer in writing up a decoded message would scarcely have made so many false strokes and misspellings; and with such a poor knowledge of the French language, he could scarcely have solved a coded message in French.
In addition to the above evidence, there is the extreme unlikelihood that this message added to the earlier one would have given Keefer enough material to have solved the code. There is also reason to believe, from Napoleon's statement to Bigelow regarding the message he sent Bazaine in September, that political considerations might well have induced the Emperor to send this message through the United States in the clear.
Impact and Epilogue
Rare indeed is the single intelligence item that is at once so important and so unmistakable in meaning as the intercept of January 10. Its effect on events, however, can only be estimated, for no reference to it appears in the records of the developments that followed.
On the 17th the French Minister came to Seward proposing that France and the United States enter into an agreement for the governing of Mexico during the period that would follow the departure of the French troops. France's only stipulation was that the interim government exclude JuÃ¡rez. The United States, having consistently pursued a policy of recognition of JuÃ¡rez and nonrecognition of Maximilian, could never have voluntarily accepted such a proposal. And since southern Texas was well garrisoned with troops remaining from the magnificent army that had subdued the Confederacy, involuntary acceptance was likewise out of the question. But Seward might reasonably have entertained the proposal and then engaged in time-consuming negotiations, awaiting news from Mexico that the French were gone. Instead, he dismissed Napoleon's Minister with little ceremony;39 his firmness probably stemmed largely from knowledge that the French withdrawal was already well advanced and the Emperor's proposal could be only an effort to save face.
The effect that Sheridan's communications intelligence enterprise had on international affairs, then, was probably this: it did not induce a change in policy or any other positive action, but it materially helped the government ride out a dangerous situation simply by sitting tight.
The Administration's domestic position, however, was as weak as its international position was strong. When the Senate on the 15th got around to its foreign policy debate, an earnest effort was made to embarrass the Administration (although the threatened attempt to take foreign policy out of its hands did not materialize). The debate continued into the 16th, when Senator Charles Sumner, chairman of the Foreign Relations Committee, saw fit to announce that he had reliable information (including a copy of a dispatch to the State Department from the United States Consul at Vera Cruz) that the French were withdrawing. That ended the matter.40 Neither Seward nor the President seems to have said anything to counter the unfriendly speechmaking, having in Sumner a more direct means of silencing the opposition. Although the senator was no friend of the Administration, at least some of its intelligence information had been given to him for that purpose. From the conviction with which Sumner addressed his colleagues, one is tempted to believe that intelligence much more sensitive - and more convincing - than the consular dispatch had been confided to him.
Seward's ability to close out the Mexican affair with firmness and surehandedness must have substantially bolstered the Presidential prestige, which in that year was at the lowest ebb it has reached in the nation's history. Had the government's resistance to the French intervention been anything but a resounding success, Andrew Johnson might well have failed to muster the one-vote margin by which the impeachment proceedings against him were defeated.
Before January ended, the intelligence conveyed by Napoleon's cablegram was supported by details of the French withdrawal received from other sources, one of them an unnamed spy who was sent by Sheridan to the Vera Cruz area and returned with convincing evidence of preparations for the embarkation of the Army.41 Bazaine led the last remnants of the French force out of Mexico City on February 5. Two weeks later embarkation had begun at Vera Cruz, and by March 11 it was complete.
Maximilian's regime quickly collapsed. He foolishly bottled up his small army of Mexicans, Austrians, and Belgians in Querétaro, a hundred miles northwest of the capital. An agent of Sheridan, with this army by permission, late in February reported the Imperialists marching out of Querétaro and driving the enemy before them, but the offensive was shortlived. Soon Maximilian was back in Querétaro under siege, and on May 19, as a result of treachery by a Mexican Imperialist officer related by marriage to Bazaine, the garrison was captured.42
Seward had literally "scolded Napoleon out of Mexico," but if the final issue of l'affaire Maximilien was a triumph for American diplomacy, the fate of the unhappy sovereign himself was a sorry story of nonperformance of duty by an American diplomat. After Sherman had been excused from further participation in the mission, Minister Campbell stationed himself at New Orleans and determinedly resisted repeated efforts by Seward to get him into Mexico. In April, when it had become plain that the siege of Querétaro would end in the capture of Maximilian, Seward sent an urgent plea for Maximilian's life, instructing Campbell to find Juárez and deliver the message in person. It was delivered to the head of the Mexican government not by Campbell, ex-colonel, ex-senator, but by James White, sergeant. Such pleas delivered later on by a diplomatic Chief of Mission were heeded, but this one was of no avail, and Maximilian lost his life before a firing squad at Querétaro on June 19, 1867. Four days earlier, too late to affect the fate of the misguided prince, Seward had given Campbell a new title: ex-Minister.43
1 No earlier use of communications intelligence by the United States in peacetime is known to the writer. Any reader who knows of one is urged to present it.
2 At the beginning of the war the government's conception of military intelligence work was so limited that it employed Allan Pinkerton, by that time well known as the head of a successful detective agency, as the chief intelligence operative in Washington. Pinkerton proved effective in counterintelligence work, but his intelligence estimates so greatly exaggerated Confederate strength that he is commonly given a large share of the blame for the supercaution that caused his sponsor, General McClellan, to stay close to Washington with far superior forces. Pinkerton left the service with McClellan in 1862, however, and long before the end of the war competent intelligence staffs, entirely military in character though composed of men drawn from civil life, served the principal headquarters.
3 J. Fred Rippy, The United States and Mexico (New York, 1926), p. 261, citing Genaro y Carlos Pereya Garcia, Documentos ineditos o muy raros para la historia de Mejico (20 vols., Mexico City, 1903), XIV, pp. 8-20.
4 Rippy, op. cit., pp. 264-65 and 269-72; Seward to Bigelow, September 21, 1865. All diplomatic correspondence sent or received by United States officials that is cited herein will be found in the Papers Relating to Foreign Affairs Accompanying the Annual Message of the President to the First Session, Thirty-Ninth Congress (covering the year 1865), Second Session, Thirty-Ninth Congress (1866), and Second Session, Fortieth Congress (1867-68).
5 John M. Schofield, Forty-Six Years in the Army (New York, 1897), p. 381; Philip H. Sheridan, Personal Memoirs (2 vols., New York, 1888), II, pp. 215-19; Percy F. Martin, Maximilian in Mexico (London, 1914), p. 432.
7 When a division commander in 1862-63, Sheridan had exercised an initiative in intelligence collection that was more likely to be found in an army commander. His Memoirs reveal a constantly high interest in intelligence activities.
8 War of the Rebellion: Official Records of the Union and Confederate Armies (Washington, 1884-1901) contains hundreds of decipherments resulting from such interceptions, chiefly in the operations of 1863-65 in Tennessee and Georgia, the operations along the South Carolina coast beginning in 1863, and the Richmond-Petersburg siege of 1864-65.
10 See, for example, intelligence reports sent by Sheridan to Grant, March 27, May 7, June 24, July 3 and 13, 1866. All Army correspondence cited hereafter in this article will be found in the United States National Archives, except where otherwise indicated.
13 Seward to de Montholon, April 25, 1866; Seward to J. Lothrop Motley (United States Minister to Austria), April 6, 16, 30, May 3, 30, 1866; Motley to Seward, April 6, May 1, 6, 15, 21, 1866; James M. Callahan, American Foreign Policy in Mexican Relations (New York, 1932), p. 235.
15 Castelnau to Napoleon, December 8, 1866, quoted in Georges A. M. Girard, La Vie et les souvenirs du General Castelnau (Paris, 1930), pp. 112-124; Marcus Otterbourg (United States charge d'affaires in Mexico) to Seward, December 29, 1866; Martin, op. cit., pp. 298-99; Lewis D. Campbell (United States Minister to Mexico) to Seward, November 21, 1866.
16 De Moustier (Foreign Minister) to de Montholon (Minister to the United States), October 16, 1866, in Foreign Affairs; Bigelow to Seward, November 8, 1866; Martin, op. cit., pp. 56-57; Guedalla, op. cit., p. 133; Girard, op. cit., p. 122.
21 Seward's instructions to Campbell, dated October 25, 1866, are perhaps the most impressive of the numerous masterful documents produced by the Secretary in the Mexican affair. Grant was the President's first selection as the military member of the mission and was excused only after a number of urgent requests. Correspondence relating to the inception of the Sherman-Campbell mission includes: Andrew Johnson to E. M. Stanton, October 26 and 30; Grant to Sherman (at St. Louis), October 20 and 22; Grant to Johnson, October 20 and 21, and Grant to Stanton, October 27.
22 Sherman to Grant, November 3, 1866 (Sherman MSS, Library of Congress) ; Grant to Sheridan, November 4, 1866. Sheridan was directed to "comply with any request as to location of troops in your department that Lt. Gen. Sherman ... may make."
26 Sherman to Grant, December 1 and 7, 1866. Sherman, despite his reputation for hard-headedness, was not one of those who favored military action by the United States in Mexico. He wrote Grant, "I feel as bitter as you do about this meddling of Napoleon, but we can bide our time and not punish ourselves by picking up a burden [the French] can't afford to carry."
27 New York Herald, January 3, 1867; New York Evening Post, January 8, 1867; Frederick W. Seward, Reminiscences of a War-time Statesman and Diplomat (New York and London, 1916), pp. 348-55. Seward's project, a very closely kept secret, was the acquisition of a harbor in San Domingo. A treaty was later concluded but buried by the Senate.
28 Robert Luther Thompson, Wiring a Continent (Princeton, 1947), pp. 299-301, 319-20, 323, 433-34; S. A. Garnham and Robert L. Hadfield, The Submarine Cable (London, 1934), pp. 19-40. The cable laying was the only success in the long career of the leviathan Great Eastern, which bankrupted a succession of owners as a passenger and cargo ship, as an exhibition ship, and finally as a gigantic dismantling and salvage operation. Its history is told by James Dugan in The Great Iron Ship (New York, 1953).
30 Unaddressed official statement signed by Sheridan December 8, 1877 (sic). William R. Plum, The Military Telegraph During the Civil War in the United States (2 vols., Chicago, 1882), II, pp. 343 and 357, is authority for the information on Keefer's nationality.
31 Plum, op. cit., II, pp. 345-48. The War Department records for 1866 and 1867 contain frequent cipher telegrams to Secretary Stanton from Eckert in New York; some of these messages bear dates subsequent to Eckert's resignation from the Deparment.
35 This message and the French version of the January 10 message are filed in the National Archives with telegrams sent from the military headquarters at New Orleans during the years 1864-69. This filing is clearly in error, for the messages are foreign to the rest of the material in this file and they bear none of the marks that an operator would have placed on them had he transmitted them. War Department and Army Headquarters records do not show their receipt.
36 Besides the government records cited elsewhere, the following collections have been searched for such evidence: the Andrew Johnson MSS, Sheridan MSS, Grant MSS, Edwin M. Stanton MSS, all in the Manuscripts Division, Library of Congress, and the contemporary correspondence between the War Department and State Department in the National Archives. Despite the extreme improbability that the message contents were obtained by solving the French code, this search took account of the possibility that the developments reported in the message were learned by other means.
37 The Confederates used two kinds of cipher, both involving the substitution of one character for another. What appears to be a representative if not a complete account of the cryptanalytic experiences of the Washington cryptographers is given by David Homer Bates, Lincoln in the Telegraph Office (New York, 1907), pp. 66-85. Bates was in the War Department telegraph and cipher office throughout the Civil War. The infrequency of such activity was plainly the result of the difficulty in obtaining intercepts (except at the front, where the traffic intercepted was almost always visual). All the cryptanalytic episodes reported by Bates involved intercepted courier and mail dispatches rather than messages obtained by wiretapping.
41 Sheridan to J. A. Rawlins (Chief of Staff to Grant), January 4, 1867. The ordinary period for transmittal of mail would have caused this dispatch to arrive in Washington perhaps a week later than the January 10 telegram from Paris via New Orleans.
43 New York Herald, December 7, 1866; Seward to Campbell, December 25, 1866, January 2, 8, 23, April 6, June 1, 5, 8, 11, 15, 1867; Campbell to Seward, December 24, 1866, January 2, 7, February 9, March 12, and June 3, 6, 10, 15, and 16, 1867; Martin, op. cit., pp. 408, 411; Sheridan, op. cit., II, p. 227.