SECRETS OF THE CANUDOS WAR (1896-1897): ENCRYPTED TELEGRAMS ABOUT THE FAILURE OF MOREIRA CÉSAR’S EXPEDITION ARE FINALLY DECIPHERED¹

Abstract

In a rare example of how Humanities and Exact Science can come together in the same project, this study used technology to unravel the gaps in history and historical research to contextualize the content of confidential documents about the Canudos War (1896-1897). Cryptanalysis tools were used to decipher the text of four secret telegrams, held for over a century at the Army Historical Archives. In order to understand their content, it was necessary to consult documentary and bibliographic sources that allowed additional information to the encrypted reports about the failure of the third military expedition and what happened to the body of its leader, Colonel Moreira César.

Keywords
Canudos War; Euclides da Cunha; Moreira César; Cryptography; Militar History

Resumo

Num raro exemplo de como Ciências Humanas e Exatas podem se unir num mesmo projeto, este estudo se valeu da tecnologia para desvendar as lacunas da história e a pesquisa histórica para contextualizar documentos até então sigilosos sobre a Guerra de Canudos (1896-1897). Ferramentas de criptoanálise foram usadas para decifrar o texto de quatro telegramas altamente confidenciais, guardados há mais de um século no Arquivo Histórico do Exército. Compreender seu conteúdo exigiu ainda a consulta de fontes documentais e bibliográficas que permitissem complementar as informações contidas nos relatos criptografados sobre o fracasso da terceira expedição militar e o que aconteceu com o corpo de seu líder, o coronel Moreira César.

Palavras-chave
Guerra de Canudos; Euclides da Cunha; Moreira César; Criptografia; História Militar

A mystery remains until this day, more than 120 years after the end of the Canudos War (1896-1897): what happened to the body of Colonel Moreira César, leader of the second to last expedition sent to the backlands to attack the Antonio Conselheiro’s followers. It is known that one of his last pronouncements, shortly before he died wounded by two bullets, in the early hours of March 4, 1897, was to order a second attack on Canudos. According to him, despite the casualties on the day before, the hunger and thirst that gripped the camp, most of the soldiers were fit for combat and there was enough ammunition. Any step backward, in his words, “would disgrace the Army and bring serious dangers to the Republic” (CUNHA, 2009, p. 223). However, what followed was far worse than a strategic retreat.

“Each one takes care of himself”, ordered the new commander, Colonel Pedro Nunes Tamarindo, as soon as the day dawned, according to what Euclides da Cunha reported in Os Sertões [Rebellion in the Backlands] (CUNHA, 2018 [1902], p. 481). The huge troop that promised to take over the camp, made up of about 1.200 men, fled undeterred, leaving 13 officers unburied and hundreds of soldiers wounded along the way, as well as precious ammunition for Antonio Conselheiro’s followers (SAMPAIO NETO et al., 1986, p. 43). Bodies of the abandoned soldiers would be found, three months later, by the members of the fourth expedition, in one of the most macabre scenes described by Euclides da Cunha in Os Sertões. Surprisingly, Moreira César was not among them. Colonel Tamarind’s body was.

On the left bank of the road, erected on a log – like a coat hanger on which an old uniform was hung – the frame of Colonel Tamarindo, decapitated, arms hanging down, skeletal hands wearing black gloves...

His skull and boots lay at his feet.

And from the running from the edge of the path to the deepest of the prairie, other companions of misfortune: skeletons dressed in dusty and torn uniforms, stretched out on the ground, in a tragic graduation alignment; or unbalanced attached to the flexible bushes, which, oscillating in the face of the wind, gave them singular movements of specters – betrayed a demonic staging engineered by the gunmen

(CUNHA, 2018 [1902], p. 533).

The answer to what happened to Moreira César’s body seemed to be forever lost in one of the many folders of telegrams, four of them encrypted, kept in the Army Historical Archive on the sixth floor of the Duque de Caxias Palace in Rio de Janeiro. While the correspondence exchanged between the isolated soldiers on the Bahian front and the officers from the Ministry of War, in the federal capital, is still easily legible, at least four telegrams without date or signature held information so secret that they required a key to be decoded, in the possession of only the most senior officers of the time.

Access to its content has likely been lost over time, even for the Military. The incomprehensible telegrams were transcribed and then cataloged by Second Lieutenant José de Macedo Braga, who was commissioned by the Directorate of the Army’s Historical Archive to recover the documentation of the Canudos War for a Military History project in 1941 (AHEx, 1897a). In his report, the lieutenant shows that he doesn’t have has any information about the content of the telegrams, and only tells that everything indicated they were about the disaster of the third expedition.

With the digital tools currently available, would it be possible to crack an analog code used more than a century ago? That’s what we decided to do: find out what sensitive content the telegrams sent shortly after Moreira César’s death hid. Thus, we used technology to unravel the gaps in history and historical research to understand and contextualize these classified documents, comparing them with other documents and bibliographic sources.

Figure 1
1941 report, informing about the existence of encrypted telegrams, which would have been sent to the Minister of War in March 1987

War secrets

Although it allowed messages to be exchanged quickly between officers fighting in the backlands and the top commanders based in Salvador and even Rio de Janeiro, 1,400 kilometers away, the telegraph had a serious drawback for the military: it left the content of the message completely exposed, like a letter out of the envelope, allowing the information to be read and eventually leaked to their enemy. To ensure the confidentiality of sensitive information, it was necessary to make use of keys shared between the sender and the receiver. The problem is that these keys were not always found, generating confusion as demonstrated by the exchange of telegrams on March 24, 1897, just 20 days after Moreira César’s death, between two generals: Arthur Oscar de Andrada Guimarães, until then commander of the Second Military District, headquartered in Recife, hurriedly sent to the Bahian hinterland to lead a new expedition to Canudos, and General Bibiano Sérgio Macedo de Fontoura Costallat, adjutant general of the Army, based in Rio de Janeiro. The correspondence also mentions General João Tomás de Cantuária, at the time commander of the Third Military District, which encompassed Bahia, Sergipe and Alagoas, and later acting Minister of War, when the incumbent, Marshal Carlos Machado Bittencourt, went to Bahia to personally command the siege of Canudos (GUIMARÃES, 1965). After the assassination of the minister, during an attack in Rio de Janeiro, Cantuária took command of the Ministry of War until the inauguration of the new president, Campos Sales, the following year.

Figure 2
Telegram no. 883 from the Queimadas station

Figure 3
Telegram no. 886 from Quartel Rio station no. 588

Officially appointed as commander-in-chief of the fourth military expedition to Canudos, on March 6, General Arthur Oscar barely set foot in Salvador and left for Queimadas the next morning, on the 19^th. He only arrived at the first base of operations two days later (SAMPAIO NETO et al., 1986, p. 59). On 16 May he entered Monte Santo, the second base. At the end of the month, he began the siege, camping with the troops on Morro da Favela, from where it was possible to see the dreaded Belo Monte camp, as the village of 5,200 houses, according to an estimate probably exaggerated by the Army, was baptized by Conselheiro. What would have happened that was so serious on March 24, 1997, that no one, except the officers who knew the cipher used by the Moreira César expedition, could be informed? In the folders of the Army Historical Archive reserved for the War of Canudos, there is no record of the two encrypted telegrams sent by General Arthur Oscar to General Costallat. Nor who would be the mysterious Trovoada, the official responsible for “daring” the documents.

It is perceived that the encryption of the messages was a matter of national security, due to the fear of leaking classified information to secret supporters of Conselheiro and even another country.

In 1886, when the first military expedition was sent to Canudos, the Brazilian coast was already connected by 182 telegraph stations. The service was provided by the lines of the State Network RGT (General Telegraph Office), but most of the companies that operated in the national territory were English, such as the Brazilian Submarine Telegraph Company and the Amazon Telegraph Company. The most important of these, the Western and Brazilian Telegraph Company, was responsible for the main Latin America’s coastlines, as well as for the country’s direct connection with Europe and the United States, through submarine cables that went to Belém, São Luiz do Maranhão, Fortaleza, Recife, Salvador, Rio de Janeiro, Santos, Desterro (now Florianópolis) and Porto Alegre. It was reasonable to fear that the British would have access to the correspondence of the Republican Army, as suggested by documents found in the archive of military documents of the Canudos War (AHEx, 1897b).⁴

Although there was a division between the national flow – by the RGT and railways (the river network was inexpressive) – and the international flow (by the Western by the coastal cables), there was usually cooperation, not least because of the precariousness of the system. Often Western could not make a delivery on a stretch and asked RGT for help. Sometimes, RGT encountered some interruption in its extensions and asked for Western’s cooperation. That is, Western also operated the delivery of internal telegrams

(MATHEUS, 2014).

The problem is that England was pointed out as the epicenter of a plot held by European powers interested in the restoration of the monarchy in Brazil. At least according to the conspiracy theory put together by the Jacobin agitators, radical Florianists, most of them military and militant journalists of the republican cause. Widely disseminated by the national press, the restorative conspiracy theory created a smokescreen for the failure of the third expedition against Canudos and, especially, of the one who many believed to be the natural successor of Floriano Peixoto: the feared Colonel Moreira César. To do so, he lacked “only the generalship, which he had imagined taking away along with the head of Antonio Conselheiro” (PERNAMBUCANO DE MELLO, 2020 [1997], p. 133).

“Some explanation was needed for such great successes. They found it: the Canudos disturbances meant prodromes of a vast conspiracy against recent institutions”, Euclides da Cunha would recall, five years later, in Os Sertões (CUNHA, 2018 [1902], p. 498), minimizing his participation in it. In two articles for the newspaper O Estado de S.Paulo, the author compared Canudos with the Vendée, a region of Brittany where monarchists rebelled against the French Revolution, a century earlier, with the encouragement of the English crown (COSTA et al., 2022).

Based on the exploitation of a veritable wave of fake news by the newspapers, a delusional narrative gained strength. According to it, Antonio Conselheiro’s gunmen received support from European monarchies, which sent ultra-modern weapons, shipments of explosive bullets never seen in the country, and a lot of resources to destabilize the young Republic, in order to establish the Third Empire in Brazil. “It was guaranteed: one of the leaders of the stronghold was a very skilled Italian engineer, perhaps trained in the wild polygons of Abyssinia”, reports Euclides, mocking one of the widely disseminated fake news, also involving Austrian and French officers (CUNHA, 2018 [1902], p. 507). In an orchestrated action of disinformation, secret documents emerged that would link a certain International Union of Friends of Brazil’s Empire to the Conselheiro’s troops, allegedly sponsored by an Imperialist Committee based in Paris, with branches in Buenos Aires and New York.

It was astonishment at first; then a general madness of opinion; an intense agitation of conjectures to explain the inconceivability of the event or induce some reason for the crushing of a numerous, well-equipped force with a leader of such a carat. In the complete disorientation of the spirits, the idea that the turbulent country bumpkins did not act in isolation was soon raised, at first sparse in vague comments, then condensed in unshakable certainty. They were the vanguard of unknown phalanxes ready to burst forth, remanent, everywhere, converging on the new regime. And since in the federal and state capitals, for a long time, half a dozen Platonists, contemplative and meek revolutionaries, had been agitated sterile in the propaganda of the monarchical restoration, this circumstance was made the starting point for the most counterproductive of reactions

(CUNHA, 2018 [1902], p. 497-498).

By installing the panic that “invading hordes” of gunmen were prepared to plunder the cities towards the coast, the restorative conspiracy theory ended up securing the necessary public support to mobilize troops from Pará to Rio Grande do Sul (CUNHA, 2018, p. 508) against Conselheiro’s followers. If they were leaked at the time, the telegrams sent by the survivors of the Moreira César expedition could compromise this conspiratorial narrative, pointing out tactical errors that would explain in a less delusional way the failure of a legion originally formed by 1300 soldiers, of whom 253 lost their lives, 120 were wounded and 17 reported missing (SAMPAIO NETO et al., 1986, p. 43).

What could have caused such damage? In addition to Colonel Moreira César, no less than 12 officers died during the third expedition. Among those who said they had “miraculously” survived Conselheiro’s gunmen, at least one will be mentioned in unflattering terms in one of the encrypted telegrams: Major Rafael Augusto da Cunha Matos (apud MILTON, 2003, p. 82).

The art of deciphering secrets

Not surprisingly, sensitive documents from the Canudos War were protected by cryptography. From the Greek kyptós (to hide) and gráphein (writing), cryptography is a technique for concealing what is written that has its origin in ancient times and already appears in historical records of Ancient Greece. While today cryptography is part of everyday life, used both in banking transactions and in communication through cell phone applications, for a long time its main use was military and diplomatic, safeguarding strategic information.

Classical cryptography was based on two models: transposition, which shuffles the characters of the text according to a certain pattern that works as a key for the receiver to reorder them correctly, and substitution, which creates a correspondence table between the characters of the message and their substitutes in the encrypted text (LAREW; KAHN, 1968).

Morse code itself, used by electric telegraphs, can be thought of as a way of encrypting a message by transforming vowels and consonants into short (dots) and long (dashes) signs. But, because it was a widely known code, in order to maintain the secrecy of the messages it was necessary to deliver them already encrypted to the telegraph operator, protecting them from being intercepted during the process of transmission over the telegraph lines.

Figure 4
Morse Code

Without having equipment that would allow the use of more advanced techniques, it was to be expected that the encryption process used by the military at the time of the Canudos War would be relatively simple: a monoalphabetic substitution. In summary, monoalphabetic substitutions carry the same frequency distribution to the encrypted texts as the original texts. Thus, each symbol of the text is replaced by an equivalent, and always by the same one.

Since the frequency of certain letters in Portuguese texts is widely known, there is a test to see if the ciphertext, or cryptogram, has this same distribution – the Coincidence Index (CI) test, created by William Friedman in 1922. Since the value of this CI for texts in Portuguese is close to 0.075, monoalphabetic systems should have a CI close to this value. Thus, the content of the four telegrams was submitted to a tool that calculates the CI automatically, on the specialized website PlanetCalc. To be on the safe side, the result was confirmed on the dCode website, which also has tools capable of facilitating the work of cryptanalysis.

Table 1
Coincidence Index (CI) of the four telegrams analyzed

TEXTO	IC
1	0,0697
2	0,0883
3	0,0779
4	0,0695

• Source: PlanetCalc.

Combining the four texts into one, the CI was 0.0746. The conclusion was that all telegrams are monoalphabetic, despite the abnormal result of text 2. Therefore, effectively every letter was replaced by the same letter in all encrypted telegrams. For example, if the letter A was replaced by the letter Q, every time the A appeared in a word in the original text it was replaced by Q in the cryptogram.

The second hypothesis was that the encryption of the four telegrams had made use of the same key. For this purpose, four histograms were constructed, graphical representations of statistical data on the distribution of letters, one for each telegram. The intention was to verify how often a certain letter appears in each cryptogram. The result was interesting: they all have the frequency of letters distributed in a very similar way. Therefore, they would be using the same key or very similar keys.

Graphs 1, 2, 3 and 4
Frequency of letters distributed in the four telegrams analyzed

Graph 5
Four cryptograms gathered in one text

Graph 6
Frequency of use of vowels and consonants in Portuguese found in the analysis of the four telegrams

Another way of analyzing these parameters is through the elaboration of tables to measure the percentage of frequency of the letters and bigrams of the text. Bigrams are nothing more than a pair of consecutive letters, such as LH or CH. In a statistical analysis, the identification of the most frequent bigrams in a language is important to determine possible matches in the encrypted texts.

Through these tables, it was possible to compare the frequency of certain letters in a text of 11,607 characters, with the frequency with which they occurred in the four encrypted texts, which have a total of 419 characters. Excerpts chosen at random from the part of Os Sertões dedicated to the third expedition were used for this comparison.

The result was this, according to the tool of the Frequency Analysis – Enigmator website:

Tables 2 and 3
Frequencies of certain letters in a text of 11,607 characters and frequency that occur in the four encrypted texts

Língua Portuguesa		Criptogramas
%		%
OS	2,2%	KH	5,7
RA	2,1%	HZ	5,4
AS	2,0%	ZH	4,6
DO	1,9%	ZV	4,6
ES	1,9%	ZM	4,6
AR	1,8%	HK	4,6
DE	1,7%	FZ	4,1
AD	1,7%	FH	3,6
EM	1,6%	XZ	3,6
RE	1,6%	VZ	3,1
NA	1,6%	VX	3,1
SE	1.5%	XK	2,6

• Source: Frequency Analysis – Enigmator.

New tables were prepared using the same website to compare the frequency of the main bigrams that occur in the Portuguese language with those found in the four encrypted telegrams.

As with the previous tables, the same text with 11,607 characters, formed by excerpts from Os Sertões, and the 419 characters of cryptograms, was taken as a basis.

Note the prevalence of the letters Z, H, K, and you and, simultaneously, of the diagrams formed with H and Z. This suggests that these letters are replacing the vowels A, E, O, and perhaps the consonant R or S.

Tables 4 and 5
Frequencies of the main bigrams that occur in the Portuguese language with those found in the four encrypted telegrams

Língua Portuguesa		Criptogramas
%		%
OS	2,2%	KH	5,7
RA	2,1%	HZ	5,4
AS	2,0%	ZH	4,6
DO	1,9%	ZV	4,6
ES	1,9%	ZM	4,6
AR	1,8%	HK	4,6
DE	1,7%	FZ	4,1
AD	1,7%	FH	3,6
EM	1,6%	XZ	3,6
RE	1,6%	VZ	3,1
NA	1,6%	VX	3,1
SE	1.5%	XK	2,6

• Source: Frequency Analysis – Enigmator.

After this preliminary analysis, each text was examined separately in a cryptogram-solving tool, both automatically and iteratively, on the specialized website Cryptogram Solver. However, it is necessary to warn that automatic systems do not always achieve a perfect translation, especially when the texts are not long enough, the percentages of occurrence of letters are close, there is the use of abbreviations and even errors by the sender itself. For this reason, there is a need to analyze the truncated words, using knowledge of the language and context in which the message was transmitted. Even so, not everything will be decrypted properly. In the case of telegrams, the separation of words was maintained, which made the work easier, since they are usually gathered and transmitted in blocks of five characters, making it difficult to use the context to resolve doubts.

With this information, and looking at the originals of the telegrams, the results were as follows.

Telegram 1 review

Figure 5
Coded Telegram 1

Tkhxzg ulfhzhzm xzlevks juhguseqvzg tkskg myffkg muqk vqz fhua fhzdzlvk nexfz juggkzn xzgzg iezfhk rkhzg xkhk lun xuaz h tuhqvk hufqhke

The key used in Attempt 1:

Text 1 could be translated to:

FORCES ENTERED CANUDOS CHASED FIRES M_TTOS NOON THIRTEEN HALTING PERSONAL FIGHT HOUSES FOUR O’CLOCK COLONEL CEZAR WOUNDED WITHDRAWN

Note: M_TTOS appears to be a truncated version of “many” or “dead.” The time of the first attack was actually noon on March 3. In a move of excessive self-confidence, the commander stated that he intended to have lunch in Canudos, forcing the troop to invest immediately after an exhausting walk, dehydrated and without food since the day before. There was no plan B. Moreira César, “seemed to count less on the bravery of the soldier and the competence of loyal officers than on a dubious hypothesis: the astonishment and terror of the fleeing population, caught off the cuff by hundreds of bayonets” (CUNHA, 2018 [1902], p. 467).

In their haste, more than a thousand soldiers entered Canudos, as if entering a trap, Euclides da Cunha describes in Os Sertões.

And in this tumultuous pursuit, carried out in the very first minutes of the combat, the unique and most serious danger of that monstrous ditch began to be outlined; the platoons disbanded. They stuck in the narrow alleys, shoving the two of them deep inside, running over. They cheered hundreds of corners that followed one another from house to house; they folded them in disorder, with some weapons suspended, and others threw them haphazardly, aimlessly, forward; and they were divided, little by little, into sections for the whole band; and these, in turn, broke up into stunned groups, more and more dispersed and rarefied, dissolving in the end into isolated combatants...

(CUNHA, 2018 [1902], p. 470).

Moreira César ordered the cavalry to attack, but the animals retreated when they were met by a hail of gunfire as they crossed the Vaza-Barris River, knocking down their riders. Those who sought to cross on foot ended up slipping into the riverbed, offering a tragicomic scene for those who watched through binoculars, such as Moreira César himself. Angered, the commander decided to leave his protected position and spur his horse. “I’m going to give those people pride...”, he would have said, before being hit in the stomach (CUNHA, 2018 [1902], p. 475-476).

One of the first authors to publish a book about the military campaign that months later would destroy the stronghold of Antonio Conselheiro (Última expedição a Canudos [The last military expedition to Canudos], launched in 1898, revised, expanded and republished in 1912 under the title Destruição de Canudos [Destruction of Canudos]), Lieutenant-Colonel Emídio Dantas Barreto described in a novelistic tone a scene that, like Euclides, he did not get to witness. In the book he wrote about the whole conflict, Accidents of War (DANTAS BARRETO, 1905), he mentions a detail that has never been proven: the shot would have come from a modern rifle and not from the typical weapons used by the Conselheiro’s followers, which leaves room for the hypothesis of betrayal or friendly fire.

Colonel Moreira César was convinced that Antonio Conselheiro’s legionnaires would give him Canudos at the first moment of meeting, if they expected him, and he caressed with an inner smile the idea of returning soon to Rio de Janeiro, acclaimed with the same stunning noise with which the generals of Rome returned from distant lands to the capital of the world. That is why he did not stop at any convenience and marched straight to his grave.

Realizing the gravity of such a delicate moment, the colonel left the Army in the position they occupied in the eastern hills, re-examined the entire combat zone, and an immense sadness finally overcame him. And soon after, as if touched by a fatal breath, he was heading for the already disjointed line of his terrified legionaries, when a bullet from a modern rifle struck him. It was the supreme misfortune at the sharp moment of the battle, the final blow of fate upon twelve hundred vanquished men!

Changing course, the colonel then approached the artillery, with his countenance already broken, his gaze deadened, wandering uncertainly, and, seeing him thus, without fully understanding what was going on in this ruined organism, Lieutenants Severo and Avila went to meet him; they disembarked him from the animal patient and led him to the rear of the battery, where they pitched a tent to accommodate him.

Immediately summoned, Dr. Fortunato found that the bullet had penetrated his lower abdomen, and declared without further reticence that it was a fatal case: the commander was mortally wounded!

(DANTAS BARRETO, 1905, p. 128).

Writhing in pain, Moreira César would have sent his personal assistant, Captain Olímpio de Castro, and First Lieutenant Alfredo Severo, another name that will be mentioned in one of the encrypted telegrams, to warn Colonel Tamarindo that he should take over the lead of the troop and renew the attack. Amid the shooting, Tamarindo “received the heavy investiture with a gross blasphemy” (DANTAS BARRETO, 1905, p. 130). The order must have seemed unrealistic to those who were under a hail of bullets, such as Tamarindo and Major Cunha Matos, who had already been annoyed by Moreira César’s decision to attack that day, contrary to what was planned, to give rest to the troops exhausted by the 19-kilometer journey under inclement sun from Rancho do Vigário to Canudos, without food or water. Cunha Matos would have declared “categorically that it was impossible to renew the attack, since he had determined it to a certain tactical unit, without achieving a single movement in the commitment to claim his committed military pride” (DANTAS BARRETO, 1905, p. 130). This lack of discipline would lead to a much worse result than that of the two previous expeditions, led by Lieutenant Pires Ferreira and Major Febrônio de Brito, whose retreats took place without abandoning bodies and wounded soldiers. In the attack on the 3^rd expedition, many starving soldiers were killed as they left the fighting to forage for food.

Almost always, after storming the house, the hungry soldier did not wait to have lunch, after all, in Canudos. He scrutinized the suspended jiraus. There were meats dried in the sun; gourds full of paçoca, the sertanejo’s flour; baskets full of tasty urchins. In one corner the transuded bogós, moist with crystal clear and fresh water. There was no resistance. The soldier ate the meal in a minute. It was completed by a long draught of water. However, at times he had a very cruel and bitter aftertaste — a load of lead...

The gunmen at the door were assaulting him. And the roles were reversed, reliving the conflict, until he hit the ground – sewn with a knife and ground with clubs, trampled on by the hard shoe of the reckless fighter

(CUNHA, 2018 1902], p. 471).

After five hours of inglorious fighting, the troop naturally began to retreat, repelled by thugs hiding in a seemingly impassable maze of hovels and almost equal alleys. The fear of running out of ammunition and being buried in that anthill when night turned pitch black brought moments of panic, with soldiers running over each other to reach the other bank of the Vaza-Barris river. Trampled by the fleeing mass, some even drowned in its shallow waters, which were covered with bodies, in a dramatic scene described by Euclides da Cunha based on the accounts of soldiers who witnessed it.

Repelling each other; trampling the ill-wounded, who were falling; rudely driving away the strenuous stumbling ones; knocking them down, drowning them, the first groups struck against the right bank. There, anxious to avenge it, clinging to the scanty grasses, speculating themselves in arms, lining the legs of the fortunate ones who managed to overcome them, they shuffle again in noisy congery. It was a seething of bodies transudating in a stridulous, discordant, and long voice, giving the illusion of some sudden flood, in which the Vaza-Brassis river, thickened, jumped, suddenly, out of bed, bubbling, squeezing, roaring...

(CUNHA, 2018 1902], p. 477).

Telegram 2 review

Figure 6
Coded Telegram 2

fkhxy mkhhk iezfhk mzvhheszvz tznnuxuk zmqskg jzvqknz jhkxehzdzm neszh ulfuhhzh tkhzm zfzxzvhg

The same key from telegram 1 was used in the attempt in text 2:

The key was applied almost perfectly. We could translate text 2 to:

TORC_ MORRO QUATRO MADRUGADA FALECEO PADIOLA FRIENDS WERE LOOKING FOR A PLACE TO BURY WERE ATTACKED

In his book, Euclides da Cunha narrates what happened that morning in much more dramatic tones than the cold letter of the telegraphic communication. At the end of the afternoon, the soldiers camped in Alto do Mário, from where they listened to litanies and prayers that began religiously with the ringing of the bells of the Ave Maria (CUNHA, 2018 1902], p. 479).

In a ruined hut, the dying commander-in-chief learned from the officers that the next day they would begin the retreat. There was little point in ranting. At 4 a.m., his voice was no longer heard. The future Marshal Dantas Barreto, who would consecrate himself as Joaquim Nabuco’s successor in chair 27 of the Brazilian Academy of Letters, recovered in his book Acidentes da Guerra (DANTAS BARRETO, 1905) details of what followed:

Colonel Tamarindo, overwhelmed by such a situation born of the unforeseen; overcome by the fatality of his particular temperament; given over to the fainting of the adverse fate, which gave him the living expression of martyrdom; with his mind dulled by the lack of energetic intellectual stimulus, and then still more by the horrors of the journey which had ended with the disgrace of the whole expedition; reclining on a crate of ammunition, Indifferent to what was unfolding before his eyes, he had the contrite aspect, which a jet of light would betray, of the condemned already disillusioned with human pity and whose death is inexorably approaching. He did not venture a sentence, not a single thought about the situation that had so cruelly excited him, as the minutes and hours followed in their universal progress. One might even say that he had lost the notion of existence, and it was a pity to see him like this, crushed by misfortune! In the face of such discouragement, Lieutenant Severo, facing the danger in its just proportions, and wanting to indoctrinate him with a stroke of energy, approached the chief, drew on him with heavy paints all the disasters that were leading them to the hecatomb without example in the fasts of national history, and asked him to take an immediate resolution, leaked in the audacity of the supreme moments of the war, to avert, if it were still possible, the calamity that would manifest itself astonishingly would only lighten the day. A spark of life passed through his almost dormant soul and, regaining some hints of his old animation, the old soldier summoned the commanders of military units to listen to them about the what happened and in this council, attentive to the disorganization of all the expeditionary elements, he sat down, not without protest from lieutenants Domingos Alves and Alfredo do Nascimento, also consulted about the events, the withdrawal of the column, at the break of dawn

(DANTAS BARRETO, 1905, p. 135-137).

There was a plan for the troop withdrawal: the infantry would form a square, reinforced by artillery at the rear angles. Inside, the wounded would be protected. Weapons that could not be carried would be rendered useless (DANTAS BARRETO, 1905, p. 127). But the planning was soon replaced by a disorderly breakout, in which soldiers of all corps mingled in the rush along the roads and only one division with cannons, garrisoned by a few infantrymen, tried to secure the rear without success. Under boos and cries of disdain from the gunmen, the third expedition “did not withdraw, it fled” (CUNHA, 2018 1902], p. 485). Once the match was given, no one remembered the arrangement anymore, each one struggling to run more than the other, as portrayed by Euclides da Cunha.

Eight hundred men disappeared in flight, abandoning their rifles; pulling down the stretchers, in which the wounded were writhing: throwing away the pieces of equipment; disarming; unfastening the belts, for the unburdened career; and running, running at random, running in groups, in erratic bands, running along the roads and trails that cut through them, running to the recesses of the savana, dizzy, terrified, without leaders.

Among the bundles thrown by the roadside had remained, as soon as panic broke out — a very sad detail! — the commander’s body. They did not defend him. There was not a brief simulacrum of revulsion against the enemies, who did not see and guess in the stridulity of defiant cries and the cracks of an irregular and scanty firefight, like that of a hunt. At the first shots, the battalions were diluted

(CUNHA, 2018 1902], p. 487).

Three officers and a “soldier who had been his orderly and friend” (DANTAS BARRETO, 1905, p. 147) would have carried the litter with Moreira César’s body, planning to bury him with dignity in a deep grave that could be located later, in the vicinity of the Fazenda do Rosário. But the four did not make it three kilometers when they were assaulted by a gang of gunmen in an ambush.

Yielding, therefore, to the weight of increasingly implacable adversity, the brave soldiers lowered the litter to one side of the road, cast a look of nostalgia on the dead body they delivered to the desecration of the savage backwoodsmen, and hurried forward, without articulating a word, succumbing to this crushing ordeal. It is only by the audacity of which the bold man is capable in desperate situations that they have been able to overcome the obstacles that the victorious adversary opposed them. Then followed the slaughter along the entire line of retreat, and what happened then is not described in vivid light

(DANTAS BARRETO, 1905, p. 140-151).

From then on, a legend was created around Moreira César’s body. Not all those responsible for protecting him would have fled. In a heroic move, the soldier Arnaldo Roque would have paid with his own life for his loyalty to the commander-in-chief, protecting his body until the end. Unfortunately for those who invented the story, Roque himself tried to deconstruct the myth of the self-sacrificing hero by reappearing safely less than a week later.

The story inspired the playwright Dias Gomes to create Soldier Jorge, the protagonist of the play O Berço do Herói The Hero’s Cradle], in 1963. Although transposed in the play to the fighting in Italy in 1944, the story was so uncomfortable that it was banned by the military dictatorship that would be installed the following year. In 1975, the first version of the telenovela Roque Santeiro, in which Dias Gomes returned to the theme of the hero “who was without ever having been”, was also censored. Only a decade later, a third version would be aired, which became one of the biggest successes on Brazilian TV (RIBEIRO, 2019). It was a humorous critique of the manipulation of public opinion by appealing to real or imagined heroics.

The tragic death of Salomão da Rocha was a satisfaction to the national self-esteem. Later, more movingly, the legend of Soldier Roque was added, movingly shaking the popular soul. A humble soldier, transfigured by a rare stroke of courage, had marked the climactic event of the battle. Ordinance of Moreira César, when the troop had been dismantled, and the body of the former commander had been abandoned by the wayside, the loyal fighter had remained by his side, guarding the venerable relic abandoned by an army. Kneeling next to the commander’s body, he had fought to the last cartridge, finally falling, sacrificing himself for a dead man...

And the marvelous scene, strongly colored by the popular imagination, made almost compensation for the enormity of the setback. Patriotic subscriptions were opened; civic and solemn tributes were planned; and, in a triumphal chorus of vibrant articles and fervent odes, the obscure soldier transcended History when — victim of the misfortune of not having died, exchanging immortality for life, he appeared with the last superstitious stragglers in Queimadas village

(CUNHA, 2018 1902], p. 505).

Dantas Barreto narrates in detail the adventures that the hero who preferred to live rather than be immortalized by history, sacrificing himself hugged with Moreira César’s body. Together with the officers in charge of the body, Roque would have entered the savana to escape the gunmen. Presumed dead, he was one of the last soldiers to arrive at the Monte Santo’s base, frustrated to discover that the military commander himself had fled, leaving only a note: “Given the alarming news coming from the front, I retire to Queimadas, where all who arrive here should go” (DANTAS BARRETO, 1905, p. 155-156).

Telegram 3 review

Figure 7
Coded Telegram 3

rzdulvk nexfz xkhjk xkhjk zkmjnufz vughvum zhfqnvzhqz zjuazh fqhkg fkmzvz mkhhuh kttqxtzug zjhqgqhlzvz

Using the attempt key in text 3 you get:

THERE WAS A MELEE FIGHT, COMPLETE DISORDER, ARTILLERY, DESPITE SHOOTING, TAKEN, DIE, OFFICIALS IMPRISONED

While the troops fought over who ran the most, abandoning uniforms, weapons, ammunition, and supplies on the road to carry less weight, the artillery moved slowly in the rear, firing grenades at the advancing gunmen, under the firm command of Captain José Agostinho Salomão da Rocha, according to Euclides da Cunha. “The rest of the expedition could escape safely. That battery set then free. Against the four Krupps cannons of Salomão da Rocha, as if against a dam, it crashed, and stopped, it swelled, it swelled, and it retreated, and the roaring wave of the gunmen broke” (CUNHA, 2018 1902], p. 488).

But the German steel cannons and modern grenades were of little avail against a dragnet of gunmen “in a frenzy, screaming, running, firing trebuchets and pistols – astonished at this inexplicable resistance, hesitating in the assault with gunshots and knives of the small gang of indomitable brave men”. The siege closed. “One by one the soldiers of the stoic garrison fell” (ibid., p. 489), originally formed by Captain Salomão, First Lieutenant Alfredo Teixeira Severo, Lieutenant Paula Freitas, and a number of artillerymen of less than 10 soldiers, in addition to the animals’ drivers that pulled the cannons (ARARIPE, 1985, p. 70).

Colonel Tamarindo, who had returned to the rear, agitated fearlessly and indefatigably among the fugitives, heroically repenting, at the hour of the catastrophe, of the previous lukewarmness, when faced with that stupendous picture, tried in vain to help the only soldiers who had gone to Canudos. On this assumption, he ordered repeated “turn around, high!” rings. The convulsive trumpet notes emitted by the breathless buglers vibrated uselessly. Or rather — they hastened their escape. In that disorder, there was only one possible determination: “Disband!”

In vain some officers, indignant, cocked revolvers on the chest of the fugitives. There was no way to contain them. They passed; they ran; they ran madly; they ran from the officers; they ran from the gunmen; and when they saw those who were shot fall wounded, they were not moved. Captain Vilarim had fought valiantly almost alone, and when he thudded, dead, he did not find among those he commanded an arm to support him. Even the wounded and sick crippled went there, staggering, dragging themselves painfully, imprecating the most agile companions...

The notes of the horns vibrated on top of this tumult, imperceptible, useless... At last, they ceased. They had no one to call. The infantry had disappeared...

On the side of the road, one could see only scattered pieces of equipment, backpacks and rifles, belts and sabers, thrown haphazardly out there, like useless things.

Entirely alone, without a single soldier, Colonel Tamarindo threw himself desperately, his horse galloping along the now deserted road, as if he were still trying to contain the vanguard personally. And the artillery was at last entirely abandoned, before reaching Angico.

The gunmen then pounced on her.

That was the outcome. Captain Solomon had only about half a dozen loyal fighters. The blows converged upon him; and he fell, cut with stabs, by the cannon he had not abandoned.

The catastrophe had been consummated...

(CUNHA, 2018 1902], p. 489).

Dantas Barreto narrates in a very different way the dramatic moves in which the artillery was about to be sacrificed. Based on the testimony of Lieutenant Alfredo Severo, who will be directly quoted in one of the encrypted telegrams, he reports a scene that does not appear in Os Sertões. In desperation, Captain Salomão would have abandoned the cannons with Severo, galloping quickly in search of Major Cunha Matos.

And it was never known for sure what had become of the brave captain who, according to one of his traveling companions, died when he returned disillusioned from his generous objective, around the place when, four months later, the 25th battalion of vanguard infantry in the Arthur Oscar expedition found the disjointed body of Colonel Tamarindo. And he presumes that this was so, says the officer referred to, because fragments of the collar of a thin tunic were found with the number – two – and a tie containing the amount of four contos de réis, attached to a human shin, as Captain Salomão used to keep the same amount received in the military box that had accompanied the expedition, for the expenses of the personnel he commanded

(DANTAS BARRETO, 1905, p. 152-153).

If there are several testimonies of participants in the fourth expedition, such as the books written by the then Dantas Barreto himself and by Lieutenant Henrique Duque-Estrada Macedo Soares, as well as the medical students Alvim Martin Horcades and Francisco Mangabeira, and even that of the muleteer Manuel Bombinho, the same cannot be said of the third expedition. The memoirs of then-sergeant and future brigadier Marcos Evangelista da Costa Villela Jr., published only in 1988, are an exception. In it, the member of Captain Solomon’s artillery gave his version of what happened to the “stoic garrison” that would have given their lives to protect their comrades.

On the morning of March 4, after an hour of bombardment, he saw that the soldiers stationed on the sidelines, who were supposed to protect the artillery, were withdrawing, leaving it completely unguarded, as well as the wounded gathered in the blood hospital. Captain Solomon is said to have given the order to continue in constant fire, until the last wounded man was removed, and then meet him on the road. On the way, Villela Jr. says he saw the body of Colonel Moreira César lying on a stretcher abandoned in a field amid dying wounded and other dead soldiers, while the rest of the troop ran away. “I saw Colonel Tamarindo galloping and, taking the lead of the platoons, shouting: ‘High, turn around!’, and the platoons fled, sideways entering the savana to take the road forward” (VILLELA Jr., 1988, p. 27). Captain Solomon was also galloping when his horse was wounded, according to the gunner.

A little farther on, I found him limping, revolver in hand, and screaming words that it is not interesting to reproduce. I asked him if he was injured and he said yes (...) At this point we struck the three cannons that were caught in a narrow corridor, for in the hustle and bustle of the rescue, they had closed the passage completely. When we arrived, the captain leaned against one side of the gun, when a compact group of gunmen came screaming and shooting us mercilessly. While this was happening, I insisted on riding the captain on the mule, but he told me: “I’m dead and, where the battery is, your captain is, try to save yourself if you can

(VILLELA Jr., 1988, p 27-28).

When he looked back, the sergeant says he no longer saw the cannons, only the large group of Conselheiro’s gunmen who surrounded them. Captain Salomão, First Sergeant Júlio Edmundo Paes de Figueiredo, and some artillerymen whose wounds prevented them from fleeing would have been there. Like Villela Jr., Captain Salomão’s closest aides, Lieutenant Alfredo Severo and Lieutenant Paula Freitas, escaped.

Once the cannons were taken, the gunmen advanced on the rest of the troops. Colonel Tamarindo would have been one of the first to be hit. His last order was a desperate request to fetch Major Cunha Matos, who commanded the 7th Battalion and would be his natural successor in command of the expedition, but he was already far away.

Soon after, as he galloped across the Angico stream, Colonel Tamarindo was thrown from his horse by a bullet. The military engineer Alfredo do Nascimento reached it while still alive. Lying on the cliff, the old commander murmured to his companion who had come to him his last order:

- Look for Cunha Matos...

This order could hardly be complied with

(CUNHA, 2018 1902], p. 489-490).

Telegram 4 review

Figure 8
Coded Telegram 4

xklgfz fululfu guduhk dqz p zieq xruszhzm xzjqfzug znnuneqz fululfuq xzhjug tqseuqhz zdquz ieu tuhqvk fkrkg qlrq slzvk xklfhz mz pkh xelrz m ieu vuk dka vu xzvz em

The key was applied almost perfectly. The ending is evidently truncated.

Text 4 can be translated to:

ACCORDING TO LIEUTENANT SEVERO TRAVELED ARRIVED CAPTAINS ALLELUIA LIEUTENANT CARPES FIGUEIRA WARNED THAT INJURED ALL INDIGNANT AGAINST MAJOR CUNHA M WHO GAVE VOICE OF EACH ONE

Apparently, the final part of the telegram is missing. Was it Major Cunha Matos who uttered the famous phrase attributed to Tamarindo or simply repeated it to the troops, encouraging the “each one looks for himself” that led to the disorganization of the retreat? Another possibility is that the last word came truncated, or missing a piece, and the telegram says that the major who was to succeed Tamarindo on the third expedition gave the order to “disband”. The truth is that, as soon as he was safe, the agile Cunha Matos hurried to give explanations.

Figure 9
Front page of the newspaper O Paiz, edition of March 8, 1897, no. 4539

In the same edition of the newspaper O Paiz that, on March 8, announces the death of Moreira César and the failure of the third expedition, under the title “A CATASTROPHE” in capital letters, there is a telegram that exposes a report by the major, dated the 5th, which between the lines exempts itself from any responsibility. According to him, “all the commanders believed that a withdrawal should be made in order and in such a way as not to abandon a single wounded”. His account of Moreira César’s body coincides with the information that the soldiers responsible for the litter with the commander had fled:

“Conselheiro’s followers, with carbines, struck the unfortunate colonel with a bullet, an accident that was immediately reported to Colonel Tamarindo ...] Then Colonel Tamarindo was struck by a bullet, and the drivers of the wounded abandoned those who were sacrificed”

(O PAIZ, 1897, p. 1).

If Captain Salomão figured as the great hero of the third expedition, sacrificing himself resignedly amid the cannons, it is known that the entire support garrison based at the support base of Monte Santo fled to the base of Queimadas as soon as he learned of the disaster in Canudos from Lieutenant Atto Batista, the first to arrive, on the 5^th noon. The soldiers had their reasons, it was nothing more than “a platonic garrison of 80 sick people and 70 children” forcibly enlisted by Moreira César, so weak that they could not bear to continue their journey to Canudos, and there was effectively a serious risk of an invasion (CUNHA, 2018 1902], p. 431). After abandoning the support base in a hurry, its commander, Colonel Francisco Agostinho de Melo Souza Menezes, arrived at Queimadas on the 6th and from there immediately took a train to the capital. On the 7th, the Ministry of War officially communicated to the press the disaster that, on the following day, would be printed in the newspapers of Brazil and around the world.

Subsequently, Cunha Matos would deny being the author of a report cited in the report “A CATASTROPHE”, informing that he would only have reached the camp of Cumbe on the night of the 5th, “crippled”, without a head or “indispensable office resources” to write an official document. He said he only scribbled a hasty note, warning Souza Menezes of the setback. His official report would have been written only on March 10 and “to avoid his loss”, delivered personally by him to General Cantuária, in Salvador (FONTES, 1996, p. 332), but for some reason, it was not made public.

The federal government created two military inquiries to clarify doubts and gaps about the failure of the third expedition. A key figure, Cunha Matos was not heard, although he requested a War Council. The Army preferred to send him back to Canudos. A note or official document, there is an intriguing phrase in the report that Cunha Matos sent to Souza Menezes on the 5th: “The colonel cannot imagine how I am, and we are all with the great disaster; but I am also sure that you will have no difficulty in finding the culprit. And nothing else” (ARARIPE, 1985, p. 66).

Regarding the reference to First Lieutenant Alfredo Teixeira Severo, in the encrypted telegram, it is known that, when he saw the troops disbanding, he was contaminated by panic, in yet another example that the scene of the sacrifice of the “stoic garrison” of Salomão da Rocha did not happen exactly as recorded by history. Like gunners Villela Jr. and Paula Freitas, he fled, leaving the wounded captain behind. The author of the telegram seems to find it strange that Severo had already left Monte Santo when he arrived there. “It was believed that all the artillery personnel had been slaughtered when they tried to save the condemned cannons, but instead of this useless sacrifice many had followed the example of their comrades in the other arms”, dismisses Dantas Barreto (DANTAS BARRETO, 1905, p. 163).

On more than one occasion Lieutenant Severo and Lieutenant Paula Freitas detained some soldiers, who passed by the flanks, to help the men in charge of the cannons and take them out with their wrists, as a last resort, but then these pale men left everything and began to run, to run, along the length of the road, hallucinated by growing fear. The wounded, still entrusted to the pity of their comrades, were also abandoned and killed with a gun or with machete and club blows, because the fanatics no longer spent their ammunition on this weakness of the government, as they said when referring to the armed forces personnel of any of the expeditions

(DANTAS BARRETO, 1905, p. 153-154).

A report by First Lieutenant Marcos Pradel de Azambuja seems to indicate that he may have been the author of the encrypted telegram in which Lieutenant Severo and Major Cunha Matos are mentioned. In Expedições Militares a Canudos e seu aspecto marcial Military Expeditions to Canudos and its Military Aspect], Tristão de Araripe Alencar recovers the information that, visibly upset, Cunha Matos passed on Pradel’s report to General Artur Oscar, exempting himself from any explanation. “Not only because my pride demands it, but also because my superiors could take my comments as an insinuation”, murmurs Cunha Matos in the note he sent to the commander of the fourth expedition, along with the subordinate’s testimony (apud ARARIPE, 1985, p. 67). In his report, Pradel says that he and other officials were not consulted about the withdrawal on the 4th and that, like Moreira César, they thought there was no reason to back down.

What I can, however, anticipate, and this not only from information provided by infantry officers but also from what I saw during my stay, when I withdrew, is that this weapon (infantry) still had fifty thousand cartridges at dawn that day and ready personnel that could easily reach the number of one thousand men, fit for combat, not counting the artillery battery, which had almost all its personnel, including ammunition for sixty rounds, more or less

(apud ARARIPE, 1985, p. 69).

In the report, apparently written at the request of Cunha Matos himself, Pradel in a certain way accuses the major of having evaded his responsibilities, by commanding the vanguard of the column that “always advanced, and as always, left in its rear an enormous number of wounded”, afraid of facing a “group of 200 men, more or less, who made a real hunt in this mass of soldiers” (apud ARARIPE. 1985, p. 70). In his opinion, a new attack would have resulted in far fewer losses than the disastrous retreat. “Most of the personnel we have lost, whether officers or soldiers, have not been killed in regular combat; it was cut with a machete by the fanatics of Antonio Conselheiro, because, without means of retreat, they were abandoned on the road” (apud ARARIPE, 1985, p. 70-71).

Pradel will give a new version of Tamarindo’s death in his testimony. He says that, after he fell from his horse, wounded, he begged the passing force to form a square, like the one originally agreed, to resist the gunmen. “The latter, however, did not answer and, without command, I can say, since you were already close to Rosário, it always advanced, in complete disorder”, he says, addressing Cunha Matos, indirectly accusing the major of having been one of the first to flee. Penalized, some officers would have taken Tamarindo still alive to a nearby house, in a few minutes invaded by a gang of gunmen, who cut up the colonel’s body with a machete, “as they did to the brave Colonel Moreira César, and to the unfortunate comrades who, alive or not, were left on the road” (apud ARARIPE, 1985, p. 71).

Conclusion: what happened to Moreira César?

The panic that in a few hours gripped the third expedition to Canudos is visible in the four encrypted telegrams: they confirm that, attacked by the gunmen, the soldiers left Moreira César’s body behind, on the side of the road, without having had time to bury it. Surprisingly, one of these telegrams also informs that some officers had been taken prisoner by Conselheiro’s troops. However, hostages or prisoners of the gunmen were never known. Is it possible that they did not die immediately, some of them taken to Belo Monte and only then sacrificed? Or even switched sides? It may be just another one of the myths of Canudos War, such as that of the explosive bullets of the Conselheiro’s gunmen, which set fire to the imagination of military personnel and journalists. However, in his memoirs, Lieutenant Henrique Duque-Estrada Macedo Soares, who served on the fourth expedition, refers to soldiers who would eventually have served as shooting and military strategy instructors for Conselheiro’s followers (MACEDO SOARES, 1903).

The deserters instructed the fanatics in the handling of modern carbines and taught them small notions of tactics in the dispersed order, moreover, instinctive in the backwoods, and as they took possession of the armament left by the soldiers of the routed expedition, they used it immediately, with all the knowledge and excellent aim, thus demonstrating that at that time there was already a good nucleus of individuals in Canudos. connoisseurs of modern carbines, the use of which requires some practice

(MACEDO SOARES, 1903, p. 39).

The most revealing of the encrypted telegrams suggests the indignation of some officers with the option of withdrawal, making specific criticisms of Major Cunha Matos. Had the commander of the 7th Battalion repeated Tamarindo and indicated that from then on it was “every man for himself”, running to save his own skin and leaving the artillery, in the rear, to be sacrificed? Lieutenant Pradel, a possible author of the telegram, recalled in his testimony an intriguing detail: most of the casualties would not have occurred in combat, but during the abandonment of the wounded by the fleeing troops.

Another eyewitness, then Sergeant Villela Jr., would have heard from Captain Salomão himself that the infantry ammunition was low and the artillery would not be enough for half a day of bombardment. Villela Jr. states in his memoirs that the troops’ morale was very low, due to the commander’s serious wounding, their hunger, thirst and exhaustion. The clamor for a withdrawal order was widespread. Panic was also widespread. “Don’t be surprised, Va. As. if tomorrow we are alone with these cannons”, he predicted to Solomon (VILELA JR., 1988, p. 24)tag. That’s exactly what happened.

It’s easy to judge those who can’t defend themselves. It is known that Cunha Matos himself did not hesitate to blame the failure on Moreira César, despite being one of his right-hand men. In a telegram to General Dionísio de Cerqueira, then Minister of Foreign Affairs, dated March 13, he stated that “he was the only one to whom Colonel Moreira César sometimes listened; that only who had he been able to make him modify inconvenient orders on more than one occasion” (apud MILTON, 2003, p. 84). Among the unheeded pieces of advice, it was to give the troops rest before undertaking the attack. Another strategic mistake would have been to bring the entire brigade into line, with no reserve for support. According to Cunha Matos, “the enemy, protected, within half an hour had put half of the assailants out of combat, and had made the other half retreat to the ravine of the Vaza-Barris” (ibidem). With the death of Moreira César and the apathy of Colonel Tamarindo, the retreat would end up taking place “amid confusion and disorder”, with total disobedience. Cunha Matos even threatened the fugitives with his revolver, according to his account, but nothing prevented them from abandoning the latecomers.

In Expedições militares a Canudos e seu aspecto marcial, Tristão de Alencar Araripe will accuse the major of making “venous” references to the commander: “The tacticians and strategists of the tables of the cafes and newspapers chose Moreira César as a ‘scapegoat’”. According to him, “the insinuations of Major Cunha Matos contributed a lot to this” (ARARIPE, 1985, p. 73). Araripe considers that the major’s “frivolous” accusations served the desires of those who sought to deny that the government and the people of Bahia were sympathizers of Conselheiro. “Hence the rage with which they attacked the memory of Moreira César”, he criticizes, without hiding a certain sympathy for the colonel. “They tried to see errors in all the measures adopted, and in the end, they gave him up as mentally ill” (ibid., p. 74).

The tragic death turned the controversial Moreira César into a national hero, but not a universal one. In the Northeast, popular poetry immortalized him under the nickname “Corta-cabeças”. In Rio, Moreira César was marked by the murder of journalist Apulcro de Castro, in 1884, stabbed in the back. In Santa Catarina, where he was appointed governor in 1894, his renown came from the brutality with which he quelled the Federalist Revolution. “Nowhere in our territory has the rule of the states of siege weighed so firmly and so stragglingly”, narrates Euclides da Cunha, who paints a very negative portrait of the martyr of the third expedition in Os Sertões. “The shootings that took place there, with a sad apparatus of unforgivable wickedness, say so in abundance” (CUNHA, 2018 1902], p. 427).

Those who applauded him and those who invective him are justified. In that singular individuality, monstrous tendencies and superior qualities clashed, one and the other in the highest degree of intensity. He was tenacious, patient, devoted, loyal, undaunted, cruel, vindictive, ambitious. A constrained protean soul in a fragile organization

(Cunha, 2018, p. 423)

The pages of Os Sertões highlighted the unhealthy side of Moreira César, with descriptions of epileptic seizures during the trip to Canudos that would have compromised his decision-making capacity. “He had the unequal and bizarre temperament of a proven epileptic, covering the nervous instability of a patient in deceptive placidity” (CUNHA, 2018 1902], p. 424). Still covered in prejudice and mistrust, Moreira César’s epilepsy was exaggerated by Euclides as being the root of a kind of psychopathy. Thus, Moreira César’s expedition would have failed because it was led by a madman. In Epilepsia e crime Epilepsy and Crime], published shortly after the war, physician Afrânio Peixoto saw in the violence that the colonel demonstrated in Santa Catarina an example of “moral insensitivity to the extreme”, with unstoppable desires, “which the obedience or terror of his subordinates knows in no way how to refuse”, relating these characteristics to the disease (PEIXOTO, 1898, p. 195).

Epilepsy is indeed nourished by passions; it increases in the very expansion of sudden and strong emotions; but when, still larval, or translating into a merely affective alienation, it deafly undermines consciences, it seems to have in the free manifestation of these a saving derivative attenuating their effects. Thus, without exaggeration of phrase, it can be said that there is often in a crime, or in a rare act of heroism, the mechanical equivalent of an attack. Once the homicidal or immobilized arm of the hero in the glorious throw is restrained, the sick man may appear, ex-abrupt, succumbing to access. Hence these unexpected, incomprehensible or brutal acts, in which the victim instinctively seeks to deceive his own evil, often seeking crime as a derivative of madness.

For a long time in a semi-consciousness of his state, in a series of brief and fleeting delusions, which no one notices, which even she sometimes does not, he feels the instability of life growing. And fight tenaciously. The lucid intervals become a point of support for the vacillating conscience in search of inhibitory motives in an increasingly painful consideration of normal environmental conditions. These, however, are gradually weakened. The shaken intellect in the end hardly subordinates itself to external conditions or relates the facts and, in continuous decline, confuses them, disturbs them, inverts them, deforms them. The patient then falls into the twilight state, according to a happy expression, and condenses in the brain, as if it were the sum of all previous delusions, unstable, ready to unleash violent actions, which may throw him into crime or, accidentally, into glory, the potential of madness.

It is up to society, on this occasion, to give him the straitjacket or the purple

(CUNHA, 2018 1902], p. 428-430).

But does epilepsy have this power? Neurologist Elza Márcia Targas Yubian analyzed the nature of the disease, based on the description of Moreira César’s biographer, Oleone Coelho Fontes, as characterized by “focal crises with elementary and complex visual hallucinations followed by language deficits and episodes of complex partial crises and crises with secondary generalizations” (YACUBIAN, 2003). Is epilepsy related to a personality disorder, such as paranoia? It’s hard to say, she argues, since epilepsy was strongly related to the concept of degeneration at the time.

Because he was feared and hated by so many, who killed Moreira César is one of the great enigmas of history. “It could be said that never before have so many killed a single one”, ironized José Calasans in the preface to O treme-terra: Moreira César, a República e Canudos (FONTES, 1996).

In the popular imagination, the bullet that liquidated Moreira César came from different points, from different weapons, hired by countless enemies. Many people ordered the end of the famous Paulistano, for very different reasons. As it always appears in the language of the countryside. It was a death of order

Calasans gathers references that the fatal shot would have been fired by a subordinate, as revenge for the actions of the military governor of Santa Catarina, during the Federalist Revolution, when Moreira César ordered the murders of hundreds of rebels who surrendered after a peace agreement (CALASANS, 1979). Biographer Oleone Coelho Fontes recalls that Moreira César also did not enjoy a good reputation among Bahian politicians, for having contributed to the overthrow of Governor José Gonçalves, when he commanded a battalion in Salvador in 1891, “surely creating enmity that would only end with the death of one or the other” (FONTES, 1996, p. 364). Several friends of the Baron of Jeremoabo, a local landowner and avowed Gonçalvist, refer to Moreira César in unflattering terms in their letters: “An evil man, bloodthirsty by nature” (SAMPAIO, 1999, p. 143), “the cruelest man who wears the uniform of the army” (ibid., p. 142). “The murderer of the unfortunate prisoners of Ilha do Governador and Santa Catarina is going to cause a ghastly carnage in the ragged and almost helpless fanatics of Canudos” (ibid., p. 64), Francisco Pires pities. “I can assure you that, if César besieges Canudos, he will even kill the children and women. I’m not fantasizing, that’s what happened in Santa Catarina”, predicts Ubaldo Soares da Silva (ibid., p. 143).

Antero Galo wrote a letter to the baron saying that he had information, “with evidences, that Colonels César and Tamarindo were betrayed” (apud SAMPAIO, 1999, p. 160). Among the boxes that keep the memory of the Canudos War, in the AHEx, there is an investigation into a soldier identified only as Manoel, of the 7th Infantry Battalion, accused of having killed Colonel Tamarindo, not Moreira César, on the orders of a captain. The document is classified as reserved⁵.

Even the supernatural is invoked to explain the colonel’s death. According to Calasans, the widow of a doctor who died in Santa Catarina would have personally gone to Bahia and cursed Moreira César, when he left Salvador for Canudos: “You will go, but you will not return” (apud CALASANS, 1979). Another plague would have been the work of the vicar of Cumbe, Father Vicente Sabino dos Santos, supposedly forced to march in front of the soldiers towards Canudos, under accusation of being sympathetic to Conselheiro. Calasans has put together several stories like this.

Released after so much humiliation, he swore vengeance. He rushed to his church summoning the faithful immediately, who were surprised to attend a divine service. The vicar celebrated mass with the body present for Antonio Moreira César. It was hit-and-miss. The omen stuck. No living person survives a mass with the body present. A story, like so many others, of the folkloric cycle of Canudos. I did not find the slightest hint of Moreira César’s violence against the vicar Sabino. However, the “novelty” still thrives in the hinterland

(CALASANS, 1979).

About the body itself, survivors are divided, trying to explain why the man who was considered by Conselheiro as the Antichrist was not among the decapitated exhibited on the way to Canudos. A popular poetry found in a trench of the gunmen and collected by a soldier said that Moreira César went to Canudos “to give meat to the vultures” (EXÉRCITO DO BRASIL, 1997, p. 141). Another, collected by Euclides da Cunha in his field notebook, also refers to this hypothesis.

José Morera Sezar
14 battles won,
At 15 you see Bello Monte
And vultures ate it
(CUNHA, 2009, p. 151).

However, Os Sertões will reserve a less dark end for the leader of the third expedition: “Moreira César’s body was abandoned during the retreat and taken by the fanatics who incinerated it” (CUNHA, 2018 [1902], p. 487). Some survivors corroborated this version, others say that Moreira César was buried. However, among the documents collected in the Army archive by Lieutenant José de Macedo Braga in 1941, there is the testimony of two imprisoned residents of Canudos, informing that Conselheiro expressly forbade them to collect the colonel’s body. The document is part of a dossier with the police-military investigation into Father Martinho Codêzo y Martinez, listed in a process about his involvement with Antonio Conselheiro and ownership of a clandestine gunpowder factory in the Saí Mission and a warehouse in Vila Nova, in the interior of Bahia (SAMPAIO NETO et al., 1986, p. 100).

“It’s sad but it’s true – Colonel Antonio Moreira César, after his death, was devoured by vultures!!”, marvels the person responsible for typing the documentation that would serve to tell, from the point of view of Military History, what actually happened in Canudos War (MACEDO BRAGA, 1941). True or not, this version was never propagated by the military.

Figure 10
Title page of the police-military investigation into Father Martinho Codêzo y Martinez

• 1
  Article not published on a preprint platform. All sources and bibliography used are referenced.
• 2
  PhD in Communication and Culture from the School of Communication of the Federal University of Rio de Janeiro and postdoctoral researcher at the Department of Spanish and Portuguese at the University of California-Davis, under the supervision of Professor Leopoldo Bernucci. Associate Professor of Journalism at the School of Communication, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil.
• 3
  PhD in Systems and Computer Engineering from COPPE of the Federal University of Rio de Janeiro and Professor emeritus of the Computer Engineering Section of the Military Institute of Engineering, Rio de Janeiro, Brazil.
• 4
  Telegrams from March 7 to 19, 1897, were sent by Colonel Siqueira de Menezes and from April 7 to 13 by General Arthur. Contents: coded telegrams; disposition of forces and their respective commands; list of doctors in Sergipe; request for reinforcements; ordering horses; lack of war materials; lack of officers; request for quantification of the workforce; request for provisions; questions about the English telegraph company in Bahia. In: BR, RJ, AHEx CCan T 05.
• 5
  Letter from the session room of the Council of Investigations in the village of Santo Antônio das Queimadas, addressed to brigadier general Arthur Oscar de Andrade Guimarães. Document on the investigation of Corporal Manoel, of the 7th Infantry Battalion, suspected of the murder of colonel Tamarindo, on the orders of a captain. It also reports the decision to move the investigative council to the capital to hear witnesses about the case. The letter was signed by the colonel president of the Investigations Council, Francisco de Abreu Lima, on May 25, 1897. Note: The document is classified as reserved. In: LIMA, Francisco de Abreu. Letter from the session room of the Council of Investigations in the village of Santo Antônio das Queimadas, addressed to Brigadier General Arthur Oscar de Andrade Guimarães, dated May 25, 1897. In: BR, RJ, AHEx RI CCan Of 14.

Referências

• AHEX. Arquivo Histórico do Exército. Telegramas de 6 a 16 de março de 1897 [a]. Campanha de Canudos. In: BR RJ AHEx CCan T 07.
• AHEX. Arquivo Histórico do Exército. Telegramas de 7 a 19 de março de 1897 [b]. Campanha de Canudos. In: BR RJ AHEx CCan T 05.
• COSTALLAT, Bibiano Sérgio Macedo de Fontoura. Telegrama n° 886 da estação de Quartel Rio n° 588, de 24 de março de 1897. In: BR RJ AHEx CCan T 05 de 17/03 a 13/04/1897.
• GUIMARÃES, Arthur Oscar de Andrada. Telegrama n° ٨٨٣ da estação de Quartel Rio, de ٢٤ de março de ١٨٩٧. In: BR RJ AHEx CCan T 05, 03/1897.
• LIMA, Francisco de Abreu. Ofício da sala das sessões do Conselho de investigações na Vila de Santo Antônio das Queimadas, destinado ao General de Brigada Arthur Oscar de Andrade Guimarães, de 25 de maio de 1897 In: BR RJ AHEx RI CCan Of 14.
• MACEDO BRAGA, José de. Inquérito policial-militar sobre o padre Martinho Codêzo y Martinez 1941. In: BR RJ AHEx CCan IPM 02.
• BOXENTRIQ. Cryptogram Solver. Disponível em: https://www.boxentriq.com/code-breaking/cryptogram Acesso em: 23 jun. 2023.
  » https://www.boxentriq.com/code-breaking/cryptogram
• DCODE.FR. Solveurs, Crypto, Maths, Codes, Calculs en Ligne. Disponível em: https://www.dcode.fr/ Acesso em: 23 jun. 2023.
  » https://www.dcode.fr/
• FREQUENCY ANALYSIS. Enigmator. Disponível em: https://merri.cx/enigmator/cryptanalysis/frequency_analysis.html Acesso em: 23 jun. 2023.
  » https://merri.cx/enigmator/cryptanalysis/frequency_analysis.html
• PLANETCALC. Index of Coincidence. Disponível em: https://planetcalc.com/7944/ Acesso em: 23 jun. 2023.
  » https://planetcalc.com/7944/
• ARARIPE, Tristão de Alencar. Expedições militares a Canudos e seu aspecto militar: seu aspecto marcial. 2. ed. Rio de Janeiro: Biblioteca do Exército – Bibliex, 1985 [1960].
• BOMBINHO, Manuel Pedro das Dores. Canudos, história em versos São Paulo: Editora Hedra, 2002.
• CALASANS, José. Moreira Cesar, quem foi que te matou? Revista do Instituto Histórico e Geográfico Brasileiro, Rio de Janeiro, n. 324, jul./set. 1979. Disponível em: http://josecalasans.com/downloads/artigos/27.pdf Acesso em: 23 jun. 2023.
  » http://josecalasans.com/downloads/artigos/27.pdf
• COSTA, Cristiane; BREY, Maria Luise; AZEVEDO, Luana Neves. Anatomia de uma teoria da conspiração: O papel de Euclides da Cunha na divulgação de que Canudos seria o epicentro de um movimento de restauração monarquista patrocinado por potências internacionais. Revista Eco-Pós, Rio de Janeiro, v. 25, n. 2, 2022, p. 213-237. Disponível em: https://revistaecopos.eco.ufrj.br/eco_pos/article/view/27904/15307 Acesso em: 23 jun. 2023. DOI: 10.29146/ecops.v25i2.27904.
  » https://revistaecopos.eco.ufrj.br/eco_pos/article/view/27904/15307
• CPDOC – Centro de Pesquisa e Documentação de História Contemporânea do Brasil. João Tomás de Cantuária Rio de Janeiro: Fundação Getúlio Vargas, 2009. Disponível em: https://cpdoc.fgv.br/sites/default/files/verbetes/primeira-republica/CANTU%C3%81RIA,%20Jo%C3%A3o%20Tom%C3%A1s%20de.pdf Acesso em: 23 jun. 2023.
  » https://cpdoc.fgv.br/sites/default/files/verbetes/primeira-republica/CANTU%C3%81RIA,%20Jo%C3%A3o%20Tom%C3%A1s%20de.pdf
• CPDOC – Centro de Pesquisa e Documentação de História Contemporânea do Brasil. Bibiano Costallat Rio de Janeiro: Fundação Getúlio Vargas, 2009. Disponível em: https://cpdoc.fgv.br/sites/default/files/verbetes/primeira-republica/COSTALLAT,%20Bibiano.pdf Acesso em: 23 jun. 2023.
  » https://cpdoc.fgv.br/sites/default/files/verbetes/primeira-republica/COSTALLAT,%20Bibiano.pdf
• CUNHA, Euclides da. Caderneta de campo Introdução, notas e comentário de Olímpio de Souza Andrade. Rio de Janeiro: Fundação Biblioteca Nacional, 2009. Disponível em: http://objdigital.bn.br/acervo_digital/div_obrasgerais/drg1283650.pdf Acesso em: 23 jun. 2023.
  » http://objdigital.bn.br/acervo_digital/div_obrasgerais/drg1283650.pdf
• CUNHA, Euclides da. Os Sertões (Campanha de Canudos). Edição, Prefácio, Cronologia, Notas e Índices: BERNUCCI, Leopoldo. São Paulo: Ateliê Editorial, SESI-SP editora (coedição), 2018 [1902].
• DANTAS BARRETO, Emídio. Acidentes de guerra (operações de Canudos) Rio Grande do Sul: Editora Livraria Rio Grandense, 1905. Disponível em: https://upload.wikimedia.org/wikipedia/commons/0/06/Accidentes_da_guerra.pdf Acesso em: 23 jun. 2023.
  » https://upload.wikimedia.org/wikipedia/commons/0/06/Accidentes_da_guerra.pdf
• DANTAS BARRETO, Emídio. Última expedição a Canudos. Porto Alegre: Editores Franco & Irmão, 1898.
• EXÉRCITO DO BRASIL. Informações. A Defesa Nacional - Revista de Assuntos Militares e Estudo de Problemas Brasileiros, Rio de Janeiro, ano LXXXIII, n. 778, out./nov./dez. 1997. Disponível em: http://www.ebrevistas.eb.mil.br/ADN/article/view/6846/5912 Acesso em: 23 jun. 2023.
  » http://www.ebrevistas.eb.mil.br/ADN/article/view/6846/5912
• FONTES, Oleone Coelho. O Treme-Terra: Moreira Cesar, a República e Canudos São Paulo: Vozes, 1996.
• GUIMARÃES, Carlos Eugenio. Arthur Oscar: um soldado do império e da república. Rio de Janeiro: Bbliex, 1965.
• HORCADES, Alvim Martins. Descrição de uma viagem a Canudos Natal: Sebo Vermelho, 2011.
• LAREW, K.; KAHN, D. The codebreakers: the story of secret writing. The American Historical Review New York: Macmillan, 1968, v. 74, n. 2.
• MACEDO SOARES, Henrique Duque-Estrada de. A Guerra de Canudos Rio de Janeiro: Compositora Gráfica, 1903. Disponível em: http://objdigital.bn.br/objdigital2/acervo_digital/div_obrasgerais/drg1313065/drg1313065.pdf Acesso em: 23 jun. 2023.
  » http://objdigital.bn.br/objdigital2/acervo_digital/div_obrasgerais/drg1313065/drg1313065.pdf
• MANGABEIRA, Francisco. Tragédia épica (Guerra de Canudos). Bahia: Imprensa Moderna de Prudencio de Carvalho, 1900.
• MATHEUS, Letícia C. A imprensa e o desenvolvimento do sistema telegráfico brasileiro. In: XII CONGRESSO ALAIC, 2014, Peru. Anais Lima: PUCP, 2014. Disponível em: https://congreso.pucp.edu.pe/alaic2014/wp-content/uploads/2013/09/vGT17-Leticia-Matheus.pdf Acesso em: 23 jun. 2023.
  » https://congreso.pucp.edu.pe/alaic2014/wp-content/uploads/2013/09/vGT17-Leticia-Matheus.pdf
• MILTON, Aristides Augusto. A campanha de Canudos Brasília: Senado Federal, 2003 [1901]. Disponível em: http://www2.senado.leg.br/bdsf/handle/id/1070 Acesso em: 23 jun. 2023.
  » http://www2.senado.leg.br/bdsf/handle/id/1070
• O PAIZ. A Catastrophe Rio de Janeiro, 8 de março de 1897, n. 4539, p. 1. In: HEMEROTECA DIGITAL BRASILEIRA. Disponível em: http://memoria.bn.br/pdf/178691/per178691_1897_04539.pdf Acesso em: 23 jun. 2023.
  » http://memoria.bn.br/pdf/178691/per178691_1897_04539.pdf
• PEIXOTO, Afrânio. Epilepsia e Crime Bahia: V. Oliveira & Comp., 1898. Disponível em: https://www.obrasraras.fiocruz.br/media.details.php?mediaID=483 Acesso em: 23 jun. 2023.
  » https://www.obrasraras.fiocruz.br/media.details.php?mediaID=483
• PERNAMBUCANO DE MELLO, Frederico. A Guerra total de Canudos 3. ed. São Paulo: Escrituras Editora, 2020 [1997].
• RIBEIRO, Rondinele Aparecido. De O Berço do Herói a Roque Santeiro: análise da transposição do herói para a teleficção Dissertação de mestrado, Letras, FCLAS, UNESP, 2019. Disponível em: https://repositorio.unesp.br/handle/11449/190956 Acesso em: 23 jun. 2023.
  » https://repositorio.unesp.br/handle/11449/190956
• SALVADORI, Fausto. Nomes de Guerra. Revista Apartes, São Paulo, 06/05/2019. Disponível em: https://www.saopaulo.sp.leg.br/apartes/nomes-de-guerra/ Acesso em: 23 jun. 2023.
  » https://www.saopaulo.sp.leg.br/apartes/nomes-de-guerra/
• SAMPAIO, Consuelo Novais (org.). Canudos: Cartas para o Barão São Paulo: EDUSP, 1999.
• SAMPAIO NETO, José Augusto Vaz et al. (org.). Canudos: subsídios para a sua reavaliação histórica. Rio de Janeiro: Fundação Casa de Rui Barbosa, 1986. Disponível em: http://antigo.casaruibarbosa.gov.br/dados/DOC/FCRBbases/FCRB_canudos_subsidios_para_a_sua_reavaliacao_historica.zip Acesso em: 23 jun. 2023.
  » http://antigo.casaruibarbosa.gov.br/dados/DOC/FCRBbases/FCRB_canudos_subsidios_para_a_sua_reavaliacao_historica.zip
• VILLELA JR., Marcos Evangelista da Costa. Canudos: memórias de um combatente São Paulo: Marco Zero, 1988.
• YACUBIAN, Elsa Márcia Targas. Quando a epilepsia pode ter mudado a história: Antônio Moreira César no comando da terceira expedição na Guerra de Canudos. Arquivos de Neuro-Psiquiatria, São Paulo, v. 61 (2B), jun. 2003. Disponível em: https://www.scielo.br/j/anp/a/PZ9vgB5QzgH6jhjSzzm38Nd/abstract/?stop=next&format=html⟨=pt#. Acesso em: 23 jun. 2023. DOI: https://doi.org/10.1590/S0004-282X2003000300035.
  » https://www.scielo.br/j/anp/a/PZ9vgB5QzgH6jhjSzzm38Nd/abstract/?stop=next&format=html⟨=pt#.

Edited by

Responsible Editors

Miguel Palmeira e Stella Maris Scatena Franco

Publication Dates

• Publication in this collection
  01 July 2024
• Date of issue
  2024

History

• Received
  30 July 2023
• Accepted
  19 Oct 2023