Intelligence in War: The Value--And Limitations--Of What the Military Can Learn About the Enemy

by John Keegan

  By the fourteenth century, the conditions of warfare in post-Roman Europe had altered greatly to the local rulers’ advantage. The overriding need to suppress the aggression of nomadic despoilers—Vikings in the west, Saracens in the south, horse peoples in the east—had stimulated the building of fixed defences, including continuous barriers and chains of castles, which had solidified frontiers, pacified borderlands and restored the possibilities of trade, with beneficial effects on the general prosperity. Kings had money to pay soldiers; they also found the money to buy intelligence and pay agents, who moved with reasonable ease among travelling merchants and, or so at least was suspected by royal governments, under the cloak of international religious orders. It is a mark of how commonplace spying had become during the Hundred Years War between France and England that heralds, the non-partisan arbiters of propriety on the battlefield, went to great lengths to defend their reputation for impartiality; so too did ambassadors, though they were less often believed.

  By the middle of the fourteenth century there were extensive networks of English agents in northern France and the Netherlands, usually foreigners working for money, with French counterparts in England, often identified by the royal government as expatriate monks or travelling friars, how accurately is now difficult to establish. What their information was worth is equally mysterious. Even more than would be the case in later ages of improved communications, messages were difficult to transmit quickly in the Middle Ages. The roads were bad, the hire of horses unreliable, the sea a barrier, particularly to the transmission of messages from France to England. The English kings tried to smooth the path. The port of Wissant in northern France, the nearest to Dover, was a usual point of departure, where crossing fees were fixed by law. On the English side of the Channel, post horses were maintained at royal expense for official messages. One piece of evidence suggests that the money was well spent. On Sunday, 15 March 1360, news was brought to the royal council, sitting at Reading, that the French had attacked Winchester, a hundred miles distant, that very day. There is no suggestion, however, that intelligence had brought advance warning.7

  Real-time intelligence, except over very short distances, was inherently difficult to acquire in the medieval world. It simply could not be carried quickly enough ahead of the movement of enemy forces. That would remain so for centuries to come. Sometimes critical information did not travel even within the confined space of a battlefield. At Lützen, for example, on 16 November 1632, one of the most important engagements of the Thirty Years War, the Imperial (Austrian) and Swedish armies both made a tactical retreat at the end of the day. The Swedish king, Gustavus Adolphus, had been killed and if Wallenstein, the Imperial commander, had renewed the attack, the Swedes would probably have lost. Neither side, however, was aware of the other’s movements. Next day the Swedes returned, captured the Imperial artillery, which had been abandoned for want of horses to drag it off, and so turned what should have been an Imperial victory into a defeat.8

  The European armies of the eighteenth century had become much more professionalised than those of the Thirty Years War. Even so, they found real-time intelligence hard to acquire. Frederick the Great’s campaign of Hohenfriedberg in 1745 was exceptional. The Imperial (Austrian) army was concentrating against him to wrest back the province of Silesia, which the Prussian king had illegally seized in 1740. He got general word of its movement but needed to put himself in a favourable position to resist its attack, by tempting it down into the Silesian plain from the surrounding hills. His first move was to use a double agent he had, an Italian clerk, in Imperial headquarters, to spread the word that the Prussians were retreating. He then concealed his army in broken ground and waited for the Austrians to appear. They made no effort to disguise their movements, and so he was able to make use of rules of observation (indices) which were known to provide rough-and-ready real-time intelligence when the enemy was in view. Dust was an important indicator. “A generalised cloud of dust usually signified that the enemy foragers were about. The same kind of dust, without any sighting of the foraging parties, suggested that the sutlers and baggage were being sent to the rear and that the enemy was about to move. Dense and isolated towers of dust showed that the columns were already on the march.” There were other signs. The gleam of the sun, on a bright day, on swords and bayonets was open to interpretation at distances of up to a mile. Marshal de Saxe, Frederick’s great French contemporary, wrote that “if the rays are perpendicular, it means that the enemy is coming at you; if they are broken and infrequent, he is retreating.”9

  Frederick, on 3 June, had positioned himself at a lookout point which commanded the level ground in front of Hohenfriedberg. Towards four o’clock in the afternoon he saw a cloud of dust, through which gradually resolved eight huge Austrian columns advancing towards the Prussian positions, illuminated by bright sunshine. As darkness fell, Frederick ordered a night march. Next morning the Battle of Hohenfriedberg began.

  Despite his enjoyment of advantageous intelligence, Frederick did not win an easy victory. His army was outnumbered, and the Austrians and their allies had manoeuvred during the night to outflank him. As so often in war, it was superior fighting power that carried the day; Frederick’s preliminary intelligence success was soon negated. It was his own quick thinking in the heat of action and the fierce reaction of his soldiers which turned the tide of battle.10

  The same would most often prove to be true in wars yet to come. In their wars outside Europe, particularly in the North American forests, where Indian allies knew the ground intimately and were masters of the arts of scouting and surprise, European armies were to suffer shocking defeats in the depths of the woods. General Braddock’s disaster at the Monongahela, near modern Pittsburgh, where a large British force was wiped out in a few hours in 1755, was entirely the result of walking blind into an ambush prepared by the French, led by their Native American allies, in uncharted and unscouted woodland. In what both sides came to call “American warfare,” intelligence remained at a premium and usually provided the basis of victory or defeat. In the familiar campaigning grounds of Europe, during the great wars of the French Revolution and Napoleonic empire (1792–1815), intelligence rarely brought victory solely by its own account. That was true even during the British Peninsular War against the French in Spain and Portugal, 1808–14. Intelligence, however good, moved too slowly to bring a real-time advantage. Indeed, Wellington in the Peninsula depended upon exactly the same means of intelligence as Scipio in his campaign against Nova Carthago (New Carthage) in Spain in the third century B.C. Wellington, Caesar and Scipio all operated as intelligence-gatherers in exactly the same way. Their earliest concern was to discover the lie of the land (Wellington was a great collector of maps and almanacs) and the characteristics of the enemy. The collection of tactical intelligence—who was where when, what he intended and of what he was capable—was left to the month, the week, the day.11

  Wellington had the population on his side, in both Portugal and Spain. France, the invader, was resented, and after the excesses of 1808, hated. Wellington did not have to seek intelligence. It was brought to him by the bucketload. The difficulty was to sort wheat from chaff. Much more illuminating, as an example of intelligence-gathering in the pre-electric age, was the organisation of intelligence during his campaigning days in unconquered India. Wellington (Arthur Wellesley) was in active command of armies in India from 1799 to 1804. Britain, through the East India Company, controlled large enclaves in Bengal, Bombay and Madras but huge areas of the sub-continent were under the rule of local warlords or freebooting hordes. The French, by diplomacy, bribe and direct intervention, sought to bring a majority of anti-British elements to their side. Wellington, operating with small armies of mixed British–Indian composition, was mainly concerned with putting down such independents as Tipu Sultan and Hyder Ali, feudatories of the effete Moghul emperor, who were effectively running their own armies and states.

  In order to win, Wellington needed a stea
dy stream of up-to-date information, from both far and near, so as to anticipate the movements of his enemies and gain forewarning of shifts of alliances, the gathering of stores, the recruitment of soldiers and other signs of offensives in the making. The conventional means of securing such a supply of intelligence was to form a reconnaissance corps, of troops either already under command or recruited from the population. The British in India had recourse to another method. They took over a pre-existing intelligence system and made it their own.

  The harkara system seems to have been unique to India. Because of the sub-continent’s enormous size, difficult terrain and—until the building of the railways and the trunk roads of the British raj—lack of long-distance routes, power tended to be local. Even when centralised under the Moghul conquerors of the sixteenth century, it remained quite diffuse. The Moghuls in Delhi ruled by devolution, either to mighty provincial officials or by arrangement with local princes, particularly in western and southern India. The system could be made to work only if the court was supplied with regular reports of events at the lesser courts. It came to be supplied by two groups of news-providers: writers, often scholars of high status in the Indian caste system, and runners, who carried verbal or written messages and reports over long distances at high speed.

  Over time the system yielded a peculiarly Indian product: the newsletter, usually written in Persian, the language of the Moghul court, in a highly stylised form and on a regular, typically weekly, basis. The letters began as official documents but became, as writers and even runners acquired independence, a sort of private newspaper. Eventually not so private; to whom to distribute the newsletter became a decision of the harkara, who himself acquired a blurred identity, part intelligence-gatherer, part distributor. He also acquired odd rights, to be paid, of course, but also to be accepted as a sort of local correspondent at court, known to be working for other powers at a distant centre.

  The harkaras survived because, through their indispensability to those at both ends of the system, they established their independent status. It was an uneasy independence; flogging or even execution could follow the provision of dubious or misleading news. The punishment, however, was personal; it was not intended to undermine the system itself. The system, by the time the British embarked on their progressive supersession of the Moghuls at the end of the eighteenth century, was deeply entrenched in the processes of Indian political and military life. Indian government could not work without it. The British, who were committed to reestablishing Moghul power on an efficient basis, ruling themselves while leaving the Moghuls nominally in charge, simply took it over. They “reconstituted under their [own] control the classic Indian intelligence system which allied the writing skills and knowledge of learned Brahmins with the hard bodies and running skills of tribal and low-caste people.”12

  Wellington could not have established himself as the leading sepoy general without the harkaras, whom he both cultivated and tyrannised. His successors continued to do so. Not until the arrival of the telegraph and the establishment of printed newspapers in the middle of the nineteenth century did the harkara system decline; and even so, training in long-distance message-running persisted into the 1920s, sustained by the Indian appetite for news, uncontrolled by official interference, which is such a distinctive feature of sub-continental life. The reason, it has been suggested, why India has become and remains the largest and only real democracy in the Third World is because of its citizens’ insatiable thirst for information.

  REAL-TIME INTELLIGENCE:

  WHAT, HOW, WHERE, WHEN?

  Who knows what in sufficient time to make effective use of the news—that is as good a definition of “real-time” intelligence, the gold standard of modern information practice, as is possible—was not often a military consideration in the classical world or even the age of Wellington. Alexander, Caesar and Wellington all operated within the peculiar constraint, to the modern way of thinking, of very slow communication speed over any distance not to be covered by a running man or a galloping horse. The best harkaras were credited with a speed of a hundred miles in twenty-four hours, but sceptics thought fifty more realistic. The modern marathon, whose runners achieve twenty-six miles in about three hours, gives a better indication of the nature of real-time intelligence before the coming of electricity. The armies and navies of the pre-electricity age operated within an intelligence horizon of considerably less than a hundred miles. Hence the enormous importance attached by the commanders of the past to strategic intelligence: the character of the enemy, the size and capability of his force, its dispositions, the nature of the terrain in his operational area and, more generally, the human and natural resources on which his military organisation depended. It was from guesses based on such factors that generals of the pre-modern world made their plans. Real-time intelligence—where the enemy was yesterday, in which direction his columns were headed, where he might realistically be expected today—was arcane information, rarely to be collected on a real battlefield. As late as 1914, ten divisions of French cavalry, beating the Franco-German-Belgian border for nearly a fortnight, altogether failed to detect the advance of several million German troops. French reconnaissance forces failed again in the same area in 1940. Strategic intelligence is a desirable commodity. It rarely, however, brings advantage in actual time and space. For that, something else is necessary. What exactly is it? How is it possible to assure that the key questions, what, how, where and when, are answered to our advantage, not the enemy’s? That is the theme of this book.

  The acquisition of real-time intelligence requires, first of all, that the commander should have access to a means of communication that considerably outstrips in speed that of the enemy’s movement over the ground or water. Until the nineteenth century, the margin of superiority was very small. The marching speed of an army, reckoned at three miles per hour, was exceeded by that of a scout’s horse perhaps six times; but a scout had to make an outward as well as return journey, so the margin was halved. In the interval between scouts making contact with the enemy and returning, moreover, the enemy might advance, reducing the margin still further. Little wonder that surprise was so difficult to achieve in ancient campaigns. When it was, as spectacularly by the Seljuk Turks at Manzikert in 1071, the reason was often treachery or a total failure to reconnoitre, or both. At Manzikert the Byzantine army’s cavalry screen deserted, leaving the commander blind.

  Manzikert was an “encounter” battle, with both armies advancing simultaneously. More typical was the situation in which an advancing army ran into the outposts of an army standing on the defence. They automatically raised the alarm and, not having to go out and back, as in encounter operations, but back only, could give early warning. Wellington, for example, during the Waterloo campaign was, though strategically surprised, not so tactically. The French ran into his outposts, allowed him to fight a delaying battle at Quatre Bras on 16 June and to retire to a previously reconnoitred main position at Waterloo two days later.

  Surprise was equally as difficult to achieve at sea as it was on land until recent times. Indeed, the traditional problem in naval warfare was for opposing fleets to find each other at all. Hence the tendency for naval battles to occur in narrow waters, or “the shipping lanes,” often in areas where battles had occurred before. In theory, with the invention of the telegraphic flag code at the beginning of the nineteenth century, an admiral, by disposing his ships at the maximum limit of intervisibility—intervals of twelve miles—could, if his chain was long enough, create an early-warning screen which would cover several hundred miles of ocean. In practice admirals never had enough ships; and anyhow, they preferred to keep those they had concentrated, for the danger of being brought to battle dispersed outweighed that of being surprised. An admiral surprised with his ships within range of recall could form a line of battle; an admiral with his ships scattered on reconnaissance beyond quick recall had no such hope. Not until the invention of wireless telegraphy at the beginning of the
twentieth century, and its adoption by war fleets, could admirals truly begin to command the far seas. Even then, old habits died hard. Sailors like to see.

  Sight is, of course, the principal and most immediate medium of real-time intelligence. It was so in the pre-telegraphic age, and it has become so again in the age of electronic visual display. In the intervening period, which embraces the invention of the electric telegraph in the middle of the nineteenth century and its supersession by radio at the beginning of the twentieth, hearing acquired a superior status. It still enjoys something like parity. Radio in all its forms is an essential tool of military communication. Strategically, it has become less important than the written message electronically communicated, by fax or e-mail. Tactically, it predominates, by immediacy and urgency. In the heat of engagement, voice-to-voice communication between commander and the front line, and in the opposite direction, is what makes battles work, a reality not altered since Caesar took personal, front-line control of the Tenth Legion against the Nervii on the river Sambre in 57 B.C. “Calling to the centurions by name, and shouting encouragement to the rest,” Caesar transformed the tempo of the battle, shifting the psychological advantage to the Roman side and assuring the defeat of the Gauls.

  The golden age of heard communication—the dot-dash of Morse telegraphy, the human voice of radio transmission—was comparatively short. It lasted in the military arena from about 1850 until the end of the twentieth century. It was a period punctuated by grave and frustrating blanks, particularly during the First World War, when the intensity of bombardment on the fixed fronts of west and east broke cable communication as soon as mobile operations began, and the signal services had not yet succeeded in acquiring compact radios independent of cumbersome power supplies. Intelligence in real time became an impossibility. Commanders lost voice, indeed any other contact, with their forward troops over the shortest distances, and battles degenerated into directionless turmoil.