The Challenges of Military Organizations

However, while no empire can live without military organizations, living with them is not easy either. Two central problems—how to pay for the military and how to control it—have come up again and again.

4.1. Funding Military Organizations

The simplest way to fund military organizations is through plunder, because in addition to killing people and breaking things, militaries are also good at stealing things. We find the same basic pattern, which I like to call “low-end empire,”^[354] in most of the earliest empires. A ruler surrounds himself^[355] with great warriors, whom he charges with the task of recruiting retinues. The warlords then bring these troops to fight for the ruler, who shares the plunder he wins with his lords. These lords then pass on smaller shares to their lieutenants, who in turn pass smaller shares to their foot soldiers, and so on. By the time the ruler has rewarded everyone adequately, he himself might keep only slightly more loot than his greatest magnates had received. That, however, is not necessarily a problem; even though the ruler has little wealth coming in, even less is going out, because he does not have to pay for much in the way of bureaucracy or standing armies.

Decentralized militaries funded on this basis can create great empires. In the ninth century bce, plunder fueled Assyria's conquest of much of the Middle East and the Zhou dynasty's conquest of much of China's Yellow River basin, and five centuries later, Rome conquered most of Italy using a cleverly modified version of low-end imperialism. The great weakness of this system, though, is that it requires a constant supply of plunder. One result is a powerful internal dynamic favoring end­less war, even when war is not necessarily in the empire's interests; another result is the military's disinclination to perform tasks that are not likely to generate much loot, even when those jobs are clearly essential.

In some cases, this weakness forces empires to move toward an alternative, “high-end” approach to military funding. This basically flips the low-end model on

its head. Instead of inviting aristocrats to join him in war and then sharing out the takings, rulers cut these old-fashioned elites and their decentralized forces out of the loop. In their place, rulers hire their own officers and troops, centralizing mil­itary institutions. Instead of funding the entire enterprise, plunder now becomes just an incentive for extra efforts (even in mid-nineteenth-century Western Europe, “prize money” from the sale of captured ships could make naval officers rich men), with financing instead resting on tax revenue.

The tax-and-spend model had many advantages over plunder. Most obviously, it gave the ruler much more control over the military, and freed him from the ad­diction to short-term profits to pay off his aristocrats. The steady income it pro­vided made it easier to maintain standing armies and navies, which in turn made it easier to raise military standards (either by hiring long-service professionals or by recruiting mercenaries). The scale of revenue involved—premodern rulers reg­ularly captured 2-3 percent of their empires' gross domestic product (GDP), and twentieth-century governments drove this up above 50 percent—also meant that militaries could become much bigger. It is hard to imagine an empire keeping 2 per­cent of its adult males under arms for centuries at a stretch, as Rome did, without a high-end, tax-based system, and plunder could not possibly pay for modern empires' air forces and nuclear deterrents. The historian Keith Hopkins suggested that mobilizing enough cash to pay for a standing army of300,000 men in fact acted as a major fiscal stimulus for the Roman Empire, which was forced to raise tax revenues in its rich interior provinces and then send the bulk of them to the army, which spent them on food in the frontier provinces where it was stationed.¹⁵

However, using taxes to pay for centralized military organizations has equally obvious drawbacks.

Empires had two options when, as often happened, they found themselves struggling to pay their militaries: they could lower their expenditures or increase their revenues. I will begin with cost-cutting strategies, of which there were several. One was simply to ignore the problem, and even the best-run empires regularly

allowed their troops' salaries to fall months or even years into arrears. Troops typ­ically tolerated a lot of financial abuse, because their own options were of course limited; but when rulers miscalculated and went too far, the results could be cata­strophic. When England stopped paying its sailors in 1667, for instance, they struck sail to prevent their ships from being put to sea (hence the expression “striking” over wages), and Dutch raiders promptly burned or towed away much of the Royal Navy. China's Han dynasty had an even worse experience when it stopped paying the troops guarding its western frontiers in the late second century ce. Their rebel­lion led not only to the loss of the western provinces, but also to the fall of the ruling dynasty and the breakup of the entire empire.

A second cost-cutting trick, used by almost all empires at one time or another, was to inflate the problem away by debasing the coinage or printing more paper money. The classic case here is the Roman Empire, which introduced a new coin called the antoninianus in 215 ce. In theory, this was worth two of the older denarii, but never contained more than 1.6 times as much metal; and between the 240s and the 270s, its silver content was reduced from about 40 percent (the rest being copper) to about 2 percent. The result was runaway inflation and a downward spiral into economic and military chaos.

A third way to cut costs to bring them into line with income was to reduce the military's bills by making troops self-sufficient. Armies were often full of men conscripted straight off the farm; so, their masters reasoned, why not put those who were not actually fighting back on the farm and have them grow their own food? Many empires tried this, particularly for garrisons that were likely to stay in one place across multiple agricultural seasons, such as those along China's thousand- mile frontier with the Inner Asian steppes. Modern empires have found their own variants, such as putting officers on half pay in peacetime, enrolling veterans in re­serves that can be called up quickly in emergencies, or allowing military men to run businesses of their own. However, the inevitable loss of government leverage when soldiers do not depend on it for their incomes and of professionalism when troops put fending for themselves ahead of fighting or training mean that this is only ever a partial solution.

Instead of reducing expenditure until it matches income, governments can of course try to increase income until it matches the demands of military organiza­tions. This, however, is difficult. There are usually strict limits on how high an em­pire can raise taxes. In most premodern empires, the peasants who made up the bulk of the population were exploited just about as much as was possible without making famine a constant problem, and the variations in different empires' tax rates came down not to humaneness but to the balance of power between governments and aristocrats.

The most important tool in increasing revenue, and arguably the most impor­tant tool in the entire history of imperial military organization, was debt. Ancient empires found it difficult to hoard vast amounts of bullion during peacetime to be ready for wartime demands, and confiscating elite fortunes, extracting forced loans, or abruptly increasing taxes to pay for wars usually created major political problems. Some rulers found a partial solution in taking loans (forced or not) from temples and churches, which were better placed to amass land and bullion in peace­time, but the bottom line was that through most of history, instruments of sover­eign debt were so poorly developed that all empires struggled to pay their bills. Alexander the Great was effectively bankrupt when he invaded Persia in 334 bce.

The outlines of a solution began taking shape in twelfth-century ce Europe, with the creation of Italian banks big enough to lend rulers the kinds of sums they needed to make war. However, the fact that bankers could not sue empires that defaulted made the system highly unstable, and the answer only emerged in the early seven­teenth century, when the Dutch Republic hit on the idea of creating a secondary market for government bonds.

England added a crucial refinement to this fiscal revolution in 1694 by opening a national bank to manage the public debt, backed up by allocating specific taxes to pay interest on each bond issue. The novelist Daniel Defoe was only half-joking when he wrote in 1725 that “credit makes the soldier fight without pay, the armies without provisions.”^[357] During World War II, after almost running out of money in 1941, Britain was able to borrow £3.6 billion, roughly three-quarters of the nation's entire prewar wealth. The bulk of it came from lenders within Britain's own empire, although the United States took over most of the debt in 1946.^[358] The last repayment was only made in December 2006. Without the fiscal revolution that made this kind of long-term borrowing possible, the military organizations that created modern empires would have been inconceivable.

4.2. Controlling Military Organizations

After finance, the biggest challenge in running an empire's military organizations was probably what social scientists call the principal-agent problem. People form organizations to accomplish tasks that are too big for any individual, but to make such an organization succeed, the person at the top—the principal—has to delegate powers to agents lower down the chain of command, who in turn delegate powers to subagents even further down the chain. Because all these agents are only human, they have agendas of their own, and these are not necessarily consistent with their principals' plans. Consequently, principals constantly have to deal with agents who will not do as they are told. Clausewitz saw this as a primary factor in what he called “friction”: “The military machine,” he explained, “is basically very simple and very easy to manage. But we should bear in mind that none of its components is of one piece: each part is composed of individuals, every one of whom retains his potential of friction.... A battalion is made up of individuals, the least important of whom may chance to delay things and sometimes make them go wrong.”^[359]

A principal walks a tightrope. If he keeps his agents weak by delegating few powers to them, they will find it harder to thwart his wishes—but also harder to carry them out; while if he makes them strong by delegating great powers to them, they will be better placed to carry out his instructions—but also better placed to resist them, and even, since the powers delegated to agents include control of armed men, to kill him. For thousands of years, principals have tried to limit these problems by imposing strict discipline in military organizations, with gruesome punishments for any hint of disobedience, but discipline only works if other members of the organization side with the principal rather than the rebel. “Orders” are always negotiations. “The machine itself begins to resist,” Clausewitz noted, “and the com­mander needs tremendous will-power to overcome this resistance... the inertia of the whole gradually comes to rest on the commander's will alone.”^[360]

Generally speaking, the less centralized the military, the more severe its principal­agent problems. Hence the crucial importance of legitimacy: unless a ruler can con­vince his agents that it would be wrong of them to push every negotiation to the limit and exploit every opportunity for self-aggrandizement, friction will soon par­alyze the organization. Kings consequently devoted enormous ingenuity to raising the psychological costs of disobedience. Ninth-century bce Assyrian kings, for in­stance, spent considerable amounts of time networking with the “Sons of Heaven” (mar banuti), the major landowners who ruled their own estates as virtual mini­kingdoms and provided the bulk of the empire's troops. Monarchs visited them, brought them together at the capital for hunts, feasts, and festivals, and showered them with gifts and honors—but not too many, because making the mar banuti over-mighty spelled disaster.

The second major strategy for principals in low-end empires, pursued in tandem with cultivating the major agents, was to promote the idea that the gods wanted them to rule, which meant that agents who resisted the principal were guilty of blas­phemy as well as treason. This required kings either to devote enormous energy to presiding over religious activities or to delegate such activities to religious agents whom they then had to cultivate just as assiduously as their military men—yet al­most every empire followed this path. Rebels pushed back with theories justifying resistance, such as the famous Chinese concept of a “Mandate of Heaven.” After the Zhou dynasty overthrew the Shang dynasty around 1046 bce, its rulers argued that the gods and ancestors had been so appalled by the Shang kings' drunkenness and license that they had transferred their mandate to the Zhou instead. This was a propaganda success, although it had a built-in time-bomb: if the gods could with­draw their mandate from the Shang, they could withdraw it from the Zhou too, and after 810 bce, as the Zhou began suffering military reverses, increasing numbers of lords insisted that the gods had done just that.

Shifting toward a high-end empire solved some of these problems, but created new ones, too. Scale was the greatest of these. While low-end kings had typically led their armies in person, high-end rulers often just had too much to do. The classic case here is the Roman Empire, which regularly faced multiple simultaneous threats everywhere from the Tyne to the Tigris. Emperors constantly had to delegate con­trol over enormous armies to agents, and their agents constantly repaid their con­fidence by turning these armies against them. Thus, Tacitus says, was “the secret of empire”^[361] revealed—whoever has the biggest army has sovereignty.

Roman and other emperors tried out every imaginable response, including fre­quent rotation of commanders, appointing only low-caste generals whom other important agents might be unwilling to follow, delegating power only to their own kin, recalling or killing agents who proved too successful, and breaking their empires down into smaller units so a principal was always available to lead large armies in person; but at some point or other, every empire would find itself with a commander-in-chief whose willpower (as Clausewitz put it) was inadequate to overcome the machine's resistance. Even China's Tang dynasty eventually fell victim to the iron law of agency, despite its astonishing seventh-century ce successes in keeping control over a million-man army as it conquered much of Central Asia (an officer who moved even 10 men without permission faced a year in prison; one who moved a regiment risked strangulation). According to legend, in the 740s ce its em­peror Xuanzong became more interested in pleasing his courtesan Yang Guifei than in managing his subordinates, and agreed to give her favorite general An Lushan command of much of the army on the northern frontier. In 755, An rebelled, and the only way for Xuanzong to defeat him was to give other generals even bigger armies (and to agree to their demand to execute his girlfriend Yang). The empire never recovered from the civil wars that followed.

On a day-to-day basis, imperial rulers have to cut deals with mighty agents and make constant cost-benefit decisions. Keeping the military paid, controlled, and busy either conquering or managing other problems is not easy. A few rulers have what Clausewitz called the “very highly developed mental aptitude for a particular occupation”^[362] that it demands. Most just muddle along, while a not insignificant number—including Xuanzong—are downright terrible at it.

5.