The Man Who Broke Napoleon's Codes (18 page)

If Hardinge and D'Urban were to profit personally from the carnage of Albuera, so too was Scovell. He may have been disappointed after the Battle of Fuentes de Oñoro, but on 24 May, his friend FitzRoy
Somerset set pen to paper with good tidings. Scovell was promoted at last to major. The slaughter of the 57th at Albuera had left most of the regiment's commissions vacant. After years of trying to impress generals with his enthusiasm, plans and extra labors, Scovell had gotten his step the traditional way in war: into a dead man's shoes.

“The Commander in Chief having been pleased to direct Lord Wellington to recommend such officers for brevet rank as may have particularly merited his notice and approbation,” Somerset's letter to Scovell began, “I have much pleasure in being the channel of communicating to you that his Lordship has taken the opportunity of recommending you for the rank of Major and I beg you to accept my congratulations on an occasion naturally gratifying to your feelings.”

If the losses of the 57th had created vacancies, those who worked most closely with Wellington—Murray, De Lancey and Somerset—must also have impressed him of the candidate's virtues. The staff had come to know Scovell's restless energy well, for they had observed time and again that while he might lack the laconic wit of a Warre or Somerset, he was utterly tireless in his dedication to the task of improving the army. However keenly they had started their campaigning, most of the officers around Wellington could not sustain their professional zeal after months of riding about, getting shot at and sleeping badly. A good number of them, including FitzRoy Somerset, had already enjoyed leave in London. Scovell, on the other hand, had stayed put, always jotting observations about the campaign in his journal, darting about on his beloved horses, comparing notes with others before sitting down again at some rough-hewn peasants' table to make further scribblings.

During the winter inside the lines of Torres Vedras, finding that his occupation with communications was taking up too little of his time, Scovell set his mind to inventing something. He had seen that every cavalry regiment in the army had trouble with the carts used by its blacksmiths. The animals used to draw these wagons had frequently been worked to death on the Portuguese roads that were little better than tracks. Scovell therefore designed a portable forge that could be broken down and packed on two mules who might walk anywhere a regiment's horses could. In time, this invention would be credited with saving dozens of horses.

The importance of the Corps of Guides had increased greatly once the army had emerged from the lines of Torres Vedras, early in 1811,
and gone out to the open border country again. The general's scheme of operations for taking Ciudad Rodrigo and Badajoz involved dividing the army, and this meant successful operations were vitally dependent on timely communications between these two wings.

Wellington had seen Scovell's Corps of Guides at work and had been impressed, believing that they were better at their job than British soldiers ever could be. “Our English non-commissioned officers and soldiers are not very fit to be trusted alone, and out of view of their officers, in detached stations at a distance from the Army, as the soldiers of the Guides are for months together, without occasioning any complaint,” he subsequently explained to London.

Later that summer, Scovell and the Guides were given the task of establishing a daily post between the two halves of the Anglo-Portuguese force. It soon became a service that Wellington almost literally set his watch by. The journey between the two headquarters was completed in stages of ten to twelve miles, with the progress of each packet recorded on its accompanying ticket. The messengers “rode post,” picking up a fresh horse at each of the wayside halts and leaving their tired mount to recover. Scovell later said, with evident pride, “There was no instance of any of these orderlies betraying his trust.”

By August 1811, Scovell was fulfilling his duties so effectively that Wellington considered putting the entire civilian and military post of Portugal under his control. If the job of postmaster seemed a rather long way from Scovell's early ambition to lead a regiment of cavalry, the newly made major understood very well that dreams of glory do not satisfy one's creditors. The myriad tasks he took on in Wellington's HQ each carried its own pay or allowances; by accepting them he was making himself a most prosperous officer. While in the end the reform of the entire Portuguese postal service did not fall to Scovell personally, Wellington nevertheless announced by General Order:

These many tasks also included supervision of the telegraph system in the country. Prior to Masséna ‘s invasion, the Portuguese had maintained links of signaling stations up to Elvas and Almeida on the frontier. These had been destroyed by the Allies as they fell back to Torres Vedras. By the summer of 1811, Scovell was trying to reconstitute some of these links, although they never attained the same strategic importance as his daily post between Wellington and the officer in command of the Allies' southern wing, Lieutenant General Rowland Hill.

With these duties, Scovell more than doubled his earlier pay. The promotion from captain to major was worth almost five shillings more each day. He had also moved from deputy assistant QMG to assistant, which also carried a raise in pay of several shillings each day. Once he became superintendent of military communications he was given extra pay of £50 per annum, which was soon increased to £80.

To Scovell's satisfaction, the extra money also came attached to the responsibility for, among other things, codes and ciphers. Major Scovell was handed a small notebook containing a handwritten copy of a most unusual text:
Cryptographia or The Art of Decyphering
by David Arnold Conradus. The origins of this tract are obscure. It is possible that Conradus may have been a monk, for men of the cloth were preeminent among the secret servants who made or deciphered codes for the princes and great captains of Europe during the seventeenth and eighteenth centuries.

Anyone in possession of Conradus, a detailed knowledge of French and a good brain could attack the kind of simple ciphers that had appeared that summer in Spain and do so with a good chance of obtaining results, which was exactly what Hardinge and D'Urban had done with Latour-Maubourg's letter before the Battle of Albuera. Conradus had evidently spent many years studying the principal languages of Europe and his little book consisted of propositions and rules. The first section, headed “General Theory,” began:

Conradus then explained the peculiarities of the main European languages that might show themselves in ciphers. Chapter V concerned French and contained many vital pointers: that
e
was the most commonly used letter; that words ending in double letters most likely ended in
ee;
that
et
meaning
and
was the most common word; that there were only thirty-nine two-letter words in French (which he helpfully listed); and that a single letter on its own was an a, y or a consonant with an apostrophe.

If one counted up the code numbers in any given message, the most frequently occurring was likely to be
e,
the second most frequent was
i
and so on.

The approach contained in
The Art of Decyphering
was, long before 1811, clearly understood by those who made codes. Among diplomats and royal princes ciphers had been growing in size and complexity: the Spanish had introduced a five-hundred-character cipher in the late sixteenth century; King Charles I had used one of eight hundred characters during the English Civil War; Louis XIV had a
Grand Chiffre,
or Great Cipher, of six hundred, but distributed several different sheets to his ministers overseas, allowing choices as to which particular coding table had been used. All of these steps were designed to defeat the basic approach contained in Conradus, which was to draw conclusions from the frequency of different code numbers. In the summer of 1811, such Great Ciphers had not appeared in the French Peninsular army, which fortunately for Scovell gave him time to cut his teeth on simpler codes. Change, however, was afoot.

On 7 May, two days after Fuentes de Oñoro, Marshal Masséna had been superceded in command of the Army of Portugal. His replacement was Marshal Auguste Frederic Marmont, duke of Ragusa. Masséna had embodied the swashbuckling verve of the revolutionary armies; he was a man who believed in himself so completely and was so comfortable with risk that he had many a time clinched victory in an apparently hopeless situation. Marmont on the other hand was altogether more methodical, more scientific, an artillery officer who knew back to front the business of tangents and trajectories. Marmont had shown himself an able administrator, reorganizing the artillery throughout the army. Whereas Masséna cursed in his Piedmontese dialect, the duke of Ragusa was a cultivated man,
un savant,
someone who knew about science, art and philosophy. And while Masséna had kept his mistress at headquarters during the Portuguese campaign, Marmont, although reputedly one of the most handsome men in Paris, brought no Venus to the field of Mars.

Marmont was a man of energy too; thirty-six when he took over the Army of Portugal, he was fifteen years younger than the marshal he replaced. It was the kind of radical change the emperor wanted, and Marmont reveled in his reputation as one of Bonaparte's closest confidants during his epic campaigns in Egypt and Italy. It was only natural, though, that arriving with such a reputation, Marmont should excite jealousy in some quarters.

Although Masséna had coaxed it into battle on 5 May, the Army of Portugal was still utterly exhausted by its expedition across scorched earth to Lisbon and back. Upon assuming control, Marmont began a thorough reorganization.

Napoleon had grown sick of the constant bickering between Masséna and the commander of his three
corps d'armée;
Marmont disposed of this layer of command and consolidated his troops in six strong divisions, all under his own hand. Napoleon thought it absurd that 100,000 French troops in western Spain could be kept in check by 40,000 British, and Marmont understood that the solution lay in close cooperation with his neighbors, the Army of the North and Soult's Army of the South. In order to do this, prompt and effective communication would be needed. The marshal knew enough about codes to be sure that the kind of simple ciphers appearing in his own and Soult's armies were inadequate, and Marmont had one advantage over the others. He had
used ciphers in the Balkans while in command of Napoleon's Army of Dalmatia in 1807. That code had been a hieroglyphic form of
petit chiffre,
or small cipher. It was not very strong, however, and its complex symbols made it hard to use in the heat of battle.

If any further incentive was needed finally to make the move toward communicating in code, it may well have come from an incident involving General Maximilien Foy, who, while traveling back to Spain in March 1811, after briefing the emperor, had narrowly escaped capture during a guerrilla ambush. One of the first things Marmont did in his new post was to elevate Foy to the command of one of his divisions. The two men had been in the same class at artillery academy in 1792, campaigned together in 1805 and trusted one another completely. Foy had fought the British more times than almost any other commander in the French army. He had been at Vimeiro, Wellington's Peninsular debut in 1808, had placed himself at the head of troops trying to storm the seminary in Oporto following the crossing of the Douro in 1809, and had been seriously wounded leading his men into action at Busaco in 1810.

Before his own departure from Paris, Marmont had already discussed the possible introduction of a
Grand Chiffre
for use in high-level communication with Paris and Madrid with Marshal Berthier, Napoleon's chief of staff. While this question was being considered, he got his staff to prepare a new code table for use between himself and the Army of Portugal's six divisional commanders. This fell in complexity somewhere between the simple ciphers of thirty or fifty numbers and the 1750 pattern
grand chiffre
with its printed tables of twelve hundred. It would come in its own standard table of 150 characters. Marmont and his chief of staff had grasped two important points: a cipher for use by the Army of Portugal would be much easier to distribute in his own area of operations than one for use by the senior commanders right across Spain, and since he only had to get copies of the cipher to his own commanders, its security could be maintained by occasional changes of code table.