by John Keegan
The intelligence services of all states originated, nonetheless, in the efforts to avert an enemy’s achieving a military advantage but to achieve military advantage in return. In peacetime, intelligence services may merely tick over. In war, they are supposed to bring victory. How effective are they? How do they—or how do they fail—to do it?
The novelists of intelligence have disseminated an enormous amount of information about intelligence techniques. Some of it is accurate, some is misleading. Few of them, however, even such writers who are as personally experienced in intelligence work as John le Carré, have set out in full the essential components and sequence of effective intelligence operations. That is understandable. Much intelligence practice is mundane and bureaucratic, unamenable to treatment in readable form. Even the most mundane, however, is essential if intelligence is to be useful. There are five fundamental stages.
1. Acquisition. Intelligence has to be found. It may be readily available in published but overlooked form. A former director of the CIA warned his analysts against what he called the Encyclopaedia Britannica factor: do not waste effort in seeking information which may freely be found in newspapers, scholarly journals or academic monographs. Stalin’s Russia took precautions to make information as difficult to acquire as possible, by restricting the distribution of such everyday material as telephone directories and street maps. As a general principle, however, it may be taken that information useful to an opponent is what may be called “secret” and has to be collected by clandestine means. The most usual methods are spying, in all its forms, now technically known as “human intelligence” or “humint”; by the interception of an opponent’s communication, which will probably require decryption, “signal intelligence” or “signit”; by visual surveillance or imaging, through photographic or sensory reconnaissance by aircraft or satellite.
2. Delivery. Intelligence once collected has to be sent to its potential user. Delivery is often the most difficult stage, particularly for the transmitter of humint. The humint agent may be watched, or may rightly fear overhearing or interception, or may be vulnerable to arrest at points of meeting. Moreover, the sender is always under the pressure of urgency. Intelligence goes stale or is overtaken by events. Unless sent in timely fashion, preferably in “real time,” which allows it to be acted upon, it loses its value.
3. Acceptance. Intelligence has to be believed. Agents who volunteer their services have to establish their credentials; they may be a plant. One’s own operatives may have been turned or may have fallen under the control of an opponent’s counter-espionage service. Even what they honestly offer may be wrong or only half true. Intercepts appear more dependable, but they may be bogus. Even if not, they can tell only part of the truth. Henry Stimson, American Secretary of State, rightly warned of the difference between reading a man’s mail and reading his mind.
4. Interpretation. Most intelligence comes in scraps. For a complete canvas to be assembled, the scraps have to be pieced together into whole cloth. That often requires the effort of many experts, who will have difficulty in explaining to each other what they understand by individual clues and who will disagree over their relative importance. Ultimately the assembly of a complete picture may require a superior to make an inspired guess, which may or may not be correct.
5. Implementation. Intelligence officers work at a subordinate level; just as they have to be convinced of the reliability of their raw material, so also they have to convince the decision-makers, political chiefs and commanders in the field of the reliability of their submissions. There is no such thing as the golden secret, the piece of “pure intelligence,” which will resolve all doubt and guide a general or admiral to an infallible solution of his operational problem. Not only is all intelligence less than completely accurate; its value is altered by the unrolling of events. As Moltke the elder, architect of Prussia’s brilliant victories over Austria and France in the nineteenth century and perhaps the supreme military intellectual of all time, memorably observed, “No plan survives the first five minutes of encounter with the enemy.” He might as truthfully have said that no intelligence assessment, however solid its foundation, fully survives the test of action.
This book is a collection of case studies, beginning in the age of sail, when the supreme intelligence difficulty was to acquire information of value at any lapse of time which made it useful, and ending in the modern age, when intelligence of all sorts abounds but its volume threatens to overwhelm the power of the human mind to evaluate its worth. Its theme is that intelligence in war, however good, does not point out unerringly the path to victory. Victory is an elusive prize, bought with blood rather than brains. Intelligence is the handmaiden, not the mistress, of the warrior.
CHAPTER ONE
Knowledge of the Enemy
STRATEGIC INTELLIGENCE
NO WAR can be conducted successfully without early and good intelligence,” wrote the great Duke of Marlborough. George Washington agreed: “The necessity of procuring good intelligence is apparent and need not be further argued.” No sensible soldier or sailor or airman does argue. From the earliest times, military leaders have always sought information of the enemy, his strengths, his weaknesses, his intentions, his dispositions. Alexander the Great, presiding at the Macedonian court as a boy while his father, Philip, was absent on campaign, was remembered by visitors from the lands he would later conquer for his persistence in questioning them about the size of the population of their territory, the productiveness of the soil, the course of the routes and rivers that crossed it, the location of its towns, harbours and strong places, the identity of the important men. The young Alexander was assembling what today would be called economic, regional or strategic intelligence, and the knowledge he accumulated served him well when he began his invasion of the Persian empire, enormous in extent and widely diverse in composition. Alexander triumphed because he brought to his battlefields a ferocious fighting force of tribal warriors personally devoted to the Macedonian monarchy; but he also picked the Persian empire to pieces, attacking at its weak points and exploiting its internal divisions.
The strategy of divide and conquer, usually based on regional intelligence, underlay many of the greatest exploits of empire building. Not all; the Mongols preferred terror, counting on the word of their approach to dissolve resistance. If duplicity enhanced their terrible reputation, so much the better. In 1258, appearing out of the desert, Hulagu promised the Caliph, spiritual leader of Islam, ruler of the Muslim empire, his life if he would surrender Baghdad. As soon as he submitted, he was strangled and the horde moved on. The Mongols, however, as a wide-ranging nomad people, also knew a great deal and, like all nomads, when not on campaign, were always ready to trade. Markets are principal centres for the exchange of information as well as goods, and it was often a demand of marauders—by the Huns of the Romans, frequently by the Vikings—that they should be allowed to set up markets on the borders of settled lands. Commerce was commonly the prelude to predation. Trade may follow the flag, as the Victorians comfortably affirmed, but it was quite as often the other way about.
Empires in the ascendant, to whom nomads were an irritation rather than a threat, adopted a different attitude. They gave and withheld permission to trade and hold markets on their borders as a deliberate means of local control.1 They also pursued active “forward” policies. The pharaohs of the Twelfth Dynasty not only constructed a deep belt of forts on the border between settled Egypt and Nubia but also created a frontier force and issued it with standing orders. Its duty was to prevent Nubian incursions into the Nile Valley but also to patrol into the desert and report. One report, preserved on papyrus at Thebes, reads, “We have found the track of 32 men and 3 donkeys”; nearly 4,000 years old, it might have been written yesterday.
Ancient Egypt’s border problem was perfectly manageable. The narrowness of the Nile Valley, amid the surrounding desert, necessitated the minimum of protective measures. The Roman empire, by contrast, was encircl
ed on all sides by enemies, who might come by sea as well as land, and needed to be defended by elaborate fixed fortifications as well as mobile armies. At the height of their power, Rome’s rulers preferred active to passive defence and maintained strong striking forces at strategic points generally behind rather than on the frontiers. It was only as their power declined and that of the outsiders grew that the border defences were thickened.
Whether on the decline or in the ascendant, Rome devoted great care to the gathering of intelligence. Caesar’s conquest of Gaul was as much the result of his superior use of intelligence as the legions’ superior fighting power. He took great trouble to assemble economic and regional intelligence, just as Alexander had done, and he was a coldly cynical assessor of the Gauls’ ethnic defects, their boastfulness, volatility, unreliability, lack of resilience; he was equally cold in exploiting the advantage his knowledge of their weaknesses afforded. He accumulated a detailed ethnographic knowledge of their tribal characteristics and divisions, which he used ruthlessly to defeat them. Quite apart from this strategic intelligence, however, he also had a highly developed system of tactical intelligence, using short- and medium-range units of scouts to reconnoitre up to thirty kilometres in advance of his main body, to spy out the land and the enemy’s dispositions when he proceeded on campaign. It was an important principle that the leaders of these units had immediate and direct access to his person.
Caesar did not invent the Roman system of intelligence. It was the product of several hundred years of military experience. Evidence for that is already given, by the time of the Gallic wars (first century B.C.), by the existence of established terms for the different categories of reconnaissance troops: procursatores, who performed close reconnaissance immediately ahead of the army; exploratores, longer-range scouts; and speculatores, who spied deeper within enemy territory. The Roman army also made use of local informers (indices), prisoners of war, deserters and kidnapped civilians.2 If not the inventor of the system, Caesar may, nevertheless, be credited with professionalising it and institutionalising some of its most important features, notably the right of direct access by scouts to the commander in person. He also, when necessary, went to see for himself, a dangerous but sometimes essential intervention. Ultimately, the crisis of the empire in the fourth century required the almost continuous presence of one of the emperors (there were latterly two, sometimes more) with the army, a contingency that, at Adrianople in 378, led to his death on the field, progressive disaster and the empire’s collapse. The emperor Valens had been in close touch with his exploratores on the morning of the catastrophe, and they had correctly reported the enemy’s strength and dispositions. What ensued substantiates a profound and enduring truth, that “military and political survival does not depend solely on good intelligence.”3
Systems do not, however, much change, unless circumstances change, and there was little circumstantial change throughout the five centuries of the Roman empire’s greatness (first century B.C.–fourth century A.D.). Reconnaissance throughout the period was by hearing and sight, communication by word of mouth or written despatch, speed of transmission at fastest by that of a fleet-footed horse. What was true of Rome remained true of the world for another 1,500 years.
The collapse of imperial government in the West in the fifth century A.D. entailed also the collapse of organised intelligence services and such ancillary services as the publication of guidebooks and cartography (though Roman maps are strange to us, since they usually took the form of route-charts rather than two-dimensional displays of territorial features, their disappearance was a serious loss to campaigning commanders). Worse by far were the progressive degradation and eventual and complete decay of the road system. The Roman roads were built primarily for the purpose of rapid all-weather military movement and were maintained by the legions, which were as much engineering as fighting units. The dissolution of the Roman army led rapidly to the cessation of engineering work on such key elements of the Roman transport system as bridges and fords. The road network, of course, had not existed during the period of Roman conquest; Caesar had made his way through Gaul by interrogating merchants and locals and impressing guides. It was the roads, however, that had allowed Rome to defend its empire for five centuries and the break-up of their solid surfaces made long-range campaigning at speed impossible.
That was not important to the barbarian rulers who succeeded the Romans, since they sought no more than to maintain local authority. When, however, the attempt began again, under the Carolingian emperors, to reestablish wide imperial domains in the eighth and ninth centuries, the absence of roads was a serious impediment to reconquest. Things got even worse with the attempt to penetrate the Germanic regions which lay beyond the old Roman borders. In those wildernesses there were neither roads nor easily obtainable intelligence.
Some picture of the difficulties confronting medieval campaigners is conveyed by the experience of the Teutonic Knights in their effort to conquer and Christianise the Baltic shore in the fourteenth century. The Teutonic Knights, a crusading order dedicated to the conversion of the Prussians and Lithuanians, were wealthy and highly organised. They operated from a chain of strong castles built on the Baltic coast, in which they were secure from attack and could organise crusading expeditions into the hinterland. One of their principal campaigning grounds was a belt of unsettled land a hundred miles wide between East Prussia and Lithuania proper, a maze of marsh, lakes, small rivers, thickets and forest through which it was almost impossible to find a way. Local scouts were recruited by the Knights to blaze trails and report. Their intelligence was collected in a military guidebook, Die Lithauischen Wegeberichte (The Lithuanian Route Guide), compiled between 1384 and 1402. It explains, for example, that Knights wishing to get to Vandziogala from Samogitia, both near modern Kaunas in Lithuania, a distance of about thirty-five miles by today’s roads, had first to cross a patch of scrub, by a track, then a large wood through which they would have to clear their way, then a heath, then another heath, then a second wood, “the length of a crossbow shot and there you have to clear your way too,” then a third heath and a third wood. Beyond lay the true Wiltnisse (wilderness). A Prussian scout’s letter describing it was copied into the Wegeberichte. It reads: “Take notice in your wisdom that by God’s grace Gedutte and his company have got back in safety and have completed everything you sent us to carry out and have marked the way so far as 4H miles this side of the Niemen, along a route that crosses the Niemen and leads straight into the country.” The tone of the report recalls that of the Egyptian border patrol from Nubia 3,000 years earlier; the terrain described is that over which the German Army Group North advanced to Leningrad in 1941, encountering obstacles the Teutonic Knights would have found familiar.4
Curiously, the Holy Land Crusaders faced much less difficulty in getting to Jerusalem in the eleventh century. In 1394, the Grand Master of the Teutonic Knights had answered Duke Philip of Burgundy’s enquiry as to whether there would be a Baltic crusade the following year: “It is impossible to provide a forecast of future contingencies, especially because on our expeditions we are obliged to go across great waters and vast solitudes by dangerous ways . . . on account of which they frequently depend on God’s will and disposition, and also on the weather.” In different words, a modern intelligence officer might respond almost exactly similarly. The Holy Land Crusaders, by contrast, had found a much easier way forward, travelling either by sea or along the surviving Roman roads in Italy or inside the dominions of the Eastern Roman (Byzantine) emperor in southern Europe, where the imperial administration kept communications in repair and furnished supplies. Once arrived at Constantinople, they were provided with guides and escorts and were able to travel on the great Roman military roads that led towards the Taurus Mountains. In what is today eastern Turkey, however, already invaded by Seljuk Turkish migrants from central Asia, they found the roads in disrepair and likewise the other conveniences of travel—cisterns destroyed, wells dry, bridges
fallen, villages abandoned. It was a foretaste of how a nomadic, horse-riding people ruined a civilised countryside by rapine and neglect. The final stages of the march to Jerusalem were far harder than the departure from Europe.5
Campaigning inside Western Europe itself throughout the Middle Ages, the leaders of armies found conditions consistently inimical to conducting effective operations. The main problem was a chronic shortage of money in an effectively cashless society, which made the recruitment of armies difficult and their provision with food and supplies often almost impossible. Movement was laborious, because of the absence of an all-weather road system, but the lack of intelligence also impeded the efforts of rulers to deploy such forces as they could raise to the places where they were needed. That difficulty became particularly acute during the crisis of the Viking invasions in the ninth century. The Vikings, who had achieved a revolution in mobility by the development of their superbly fast and seaworthy longships, appeared without warning, overwhelmed local defenders by the ferocity of their assaults and, in the second stage of their terrorisation of the Christian lands, carried violence and pillage deep inland by learning to capture horses in large numbers at their points of debarkation. The antidote to Viking raiding would have been to create navies, but that was beyond medieval kings; another recourse would have been to maintain an intelligence system, to provide early warning, inside Scandinavia. Such sophistication lay even further outside the capabilities of ninth-century kingdoms; moreover, the Viking lands were no place for inquisitive strangers, even with money to loosen tongues. There was much more money to be made by raiding than by selling information, and the Vikings took pleasure in cutting throats.6