THE ROLE OF WAR TECHNOLOGY
Thus far we have said that mass warfare matters for taxation, but we have offered no explanation for when and why states engage in wars of mass mobilization as opposed to more limited engagements. Over the last two centuries countries like the United States have shifted from engaging in wars where only a small part of the population has been mobilized, to a period of mass mobilization, and then back to smaller scale mobilizations since that date. The scale of mobilization matters because when only a small fraction of a country’s population is mobilized, it is hard to suggest that the rich should make a special sacrifice if most others are not sacrificing either.
There are two key reasons why the material we present in this chapter is critical for this book. The first is to help better understand the deeper reasons why steeply progressive taxes happened when they did, and why it may be more difficult to build political support for them today. The second is to suggest that the process driving taxation of the rich has been far from accidental. It has instead depended on long-run trends in international rivalry and war fighting technology.
The era of the mass army, and therefore the conscription of wealth, was to a great extent technologically determined. It depended on a state of technology in which men and supplies could be moved en masse by rail yet where the remote delivery of explosive force was not yet advanced enough to avoid the need for mass infantry. The era of the mass army of course also depended on the fact that there were sizeable powers that saw themselves as rivals. These basic facts have clear implications for how we think about taxation of the rich and about social stratification more generally. When changes in technology or international rivalries produce a shift toward a war strategy of mass mobilization, this will press states to take actions that reduce social stratification. Taxation of the rich is one such policy. Current technological trends suggest a continued turn away from mass mobilization strategies, which will drive countries further and further from a context where “conscription of wealth” arguments can be used to justify heavy taxation of the rich.
THREE HISTORICAL EXAMPLES
Waging war means projecting armed force, and often over a considerable distance. For most of human history projecting force meant sending soldiers to the field of battle. It also meant mobilizing the resources that soldiers would use to fight: armor, weapons, ships, and food. In some cases these resources were substitutes for soldiers, so one heavily armed knight might be expected to have the same impact as multiple infantrymen. In other cases these resources were complements, so for example a ship in the classical era required many rowers to propel it. Over the course of history, the scale of military mobilization has depended heavily on the type of technology that states use to wage war. There is also good reason to believe that the scale of mobilization has then influenced inequality. In societies where the choice of technology leads to the great mass of citizens being mobilized, there have been pressures for social leveling.1 Technology compatible only with smaller armies has in contrast been associated with greater social stratification. It’s worth considering a few examples to show the generality of this phenomenon. In so doing we distinguish between two ways that technology has mattered. In some cases exogenous changes in technology, due to invention, have allowed states to fight wars in new ways. In other cases a change in the nature of a state’s opponents has led to a shift from employing one existing technology to another. Both of these changes are at work today.
Classical observers argued that when states employed a military technology requiring broad mobilization of the population, it was associated with a more democratic form of rule. Aristotle provided a succinct description of this phenomenon in section seven of book six of The Politics. In his words: “Where a territory is suitable for the use of cavalry, there is a favorable ground for the construction of a strong form of oligarchy: the inhabitants of such a territory need a cavalry force for security, and it is only men of large means who can afford to keep and breed horses.” In contrast Aristotle argued: “Light-armed troops, and the navy, are drawn from the mass of the people, and are thus wholly on the side of democracy.”2
But why should having a navy be associated with less social stratification? The short answer to this is that in the classical era, naval vessels were rowed as well as sailed, and in naval warfare it was useful to have as much massed manned power as possible for this purpose. Athens is the prime example where reliance on naval power is said to have led to a democratic form of rule, though the city made a deliberate choice to have a navy rather than having this be imposed by technology. A fascinating text by an author known to posterity only as “The Old Oligarch” describes why mass mobilization should go hand in hand with a more egalitarian society. Though the Oligarchic’s identity is unknown, he almost certainly hailed from the wealthier class of Athenians. The text provides a clear argument why mass mobilization should result in more equality—because this is the fair thing to do:
So, first of all, I will say this, that it seems fair enough that in Athens the poor and the common people should have more power than the noble and rich, because it is the common people who row the ships and so render the city powerful; indeed the steersmen, boatswains, pursers, look-out men, and shipwrights render the city powerful, far more than the hoplites, the noble and the good. Since this is so, it seems fair that they should all share in the offices of state by the processes of lot and election, and that anyone of the citizens who wishes should have the right to speak.3
Over the course of history changes in military technology have also pushed toward increased social stratification, an outcome opposite from that observed in Athens. One excellent example of this involved the way in which the exogenous introduction of a new technology, the iron stirrup, helped lead to feudalism in Europe. The historian Lynn White proposed this theory.4 In the two centuries after the fall of the Roman Empire in the West, when warfare occurred, it took place predominantly on foot. The reason for this was that for horsemen, with the primitive saddles of the day it was difficult and dangerous to expect to wield a sword or spear from a mount. Though the Dark Ages were certainly not the best of times, this style of warfare fit with a social structure in the Germanic kingdoms of Western Europe in which there were relatively few official class distinctions. In the Frankish kingdom all free men had both the right and the duty to bear arms.
The arrival of the iron stirrup, an import from Central Asia, changed the way European warfare was waged, and if White’s thesis is correct, it also changed society. The iron stirrup allowed horsemen to brace themselves effectively so that they could wield a sword, or especially a lance. As a natural extension of this, mounted warriors increasingly adopted armor to protect themselves. With these developments Western Europe was on the way to having armed forces in which heavily mounted warriors would play a critical role. Now, in the economic conditions of the day, maintaining a warhorse together with everything associated was a very costly venture. Military equipment for one man fighting in this manner was estimated to have cost at least ten oxen. We lack accounts from that time of the sort provided to us by the Old Oligarch for Athens. Nonetheless, some likely believed that if only a few were in a position to engage in this new style of warfare, then these few should enjoy special privileges. The eventual outcome was that the introduction of the iron stirrup helped give birth to a feudal system in Europe with rigid class distinctions. One should certainly not see this as a story with a single cause. There were also other factors pushing toward feudalism, and White’s thesis has been much debated.5 All the same, it is a fascinating example of the conscription of wealth in reverse.
Europe has not been the only world region in which technology has helped determine the ways states fight wars and therefore social structure. China during its initial period of unification provides us with a particularly stark example. In this case it was not technology that changed; it was instead a change in the type of opponents fought that prompted the Chinese state to shift between two existing technologies of war fighting. This in turn had consequences for social stratification. Prior to its unification under the Qin (221–206 BC) and Han (206 BC–AD 220) dynasties, China experienced intense warfare between several regional states. These states increasingly adopted a strategy of fighting with mass armies of peasant conscripts. The leaders of the state of Qin were the most advanced in this regard. Unlike in the case of Athens, mass mobilization in states like Qin was not accompanied by the extension of political rights to the broad populace. However, it is clear from numerous sources that benefits were offered to the peasantry in exchange for their service. States like Qin engaged in important agricultural improvements. They provided land grants to peasant soldiers, subsidized food and clothing, and helped to stabilize grain prices.6 This does not change the fact that the Qin was, to say the least, an extremely repressive state, but the provision of social benefits to those who fought is still noteworthy.
If the state of military technology under the Qin dynasty helped push in the direction of lower social stratification, changes under the Han dynasty would push in precisely the opposite direction. After the Qin dynasty’s quick collapse, the Han dynasty maintained many of the same institutions and policies including universal military conscription. But over time this situation changed. As a “universal” empire the Han dynasty faced no challenge from another large state. It instead found itself battling with nomadic tribes on its frontiers. The ideal military force for waging this type of conflict was an elite army of well-trained horsemen, rather than a mass army of peasant conscripts. Moreover, a mass peasant army could increase the risk of internal rebellion. In AD 31, the Emperor Guangwu issued an edict abolishing universal military conscription.7 This decision changed the way China fought wars; it also had direct social implications. No longer needing the peasantry to fight, the Han state retreated from the sort of social interventions that had been initiated by the Qin.8
The experiences of Han dynasty China, Classical Greece, and Europe during the Dark Ages are obviously very distant from that of the more contemporary period that we consider in this book. However, precisely because they are so distant, they suggest that the link between military technology and policies that reduce social stratification is a very general one.
THE RAILROAD AND THE MODERN MASS ARMY
With the broad historical context in mind, let us now return to the industrialized world over the last two centuries. Is it possible that changes in military technology, and thus mobilization, have led to similar social change?
To answer this question we need to consider how wars were fought in Europe at the outset of the nineteenth century, the starting point for this book. Warfare had evolved substantially since medieval times with well trained and drilled national armies composed mostly of infantry using firearms. However, in other ways things had not changed that much at all. Armies still marched to battle on their feet. Anything they ate had to be carried with them, foraged for on location, or supplied from the rear using horse-drawn wagons. The supply problem placed an upper bound on the size of an army that could be maintained in any one location at a given time. Though it is popular to think of the Napoleonic era as having been a watershed, for all of Napoleon’s tactical innovation, he was still bound by the same logistical challenges that had constrained armies since ancient times. If Napoleon knew that an army marches on its stomach, he also knew that this constraint could only be satisfied by effective foraging or by painfully slow provision of supplies from the rear.
The beginnings of a solution to the logistical problem arrived only four years after Napoleon’s death—a railway with cars propelled by a steam locomotive.9 The advent of the railroad, as well as the invention of the telegraph, fundamentally changed Western and other societies. Outside the circle of military historians it is less frequently appreciated that the arrival of the railroad also permitted a drastic increase in the scale of warfare.10 Railroads could move men rapidly, and they could also move the supplies to feed them. For the first time large armies could be fed without having to forage for food. The first passenger railways, dating from the 1820s, were still very primitive transport systems, and it was several decades before innovations in rails, locomotives, and car design led to railroads that could carry large numbers of troops over substantial distances. The first significant military use of railways was made by Napoleon’s nephew, Napoleon III, during his army’s Italian campaign of 1859.11 Railroads were, of course, heavily used in the American Civil War (1861–1865). This war in many ways presaged the destructiveness of European conflicts to come. As we noted previously, it also saw the beginnings of compensatory arguments for taxing the rich.
To assess the impact of the railroad on the scale of military mobilization, together with our co-author, Massimiliano Onorato, we analyzed military mobilization in thirteen great powers between 1600 and 2000.12 For each country for each year of war we estimated the total strength of a nation’s armed forces, not including forces from colonies that may have served under the flag of the colonizer, nor home defense units that are not part of the regular armed forces. We used information on the actual strength of the armed forces as opposed to what is referred to by military historians as an army’s “paper strength.” Governments have a natural incentive to exaggerate the size of their armed forces, and this needs to be taken into account.
We expect a country that engages in mass mobilization for war to have an army that is large in absolute terms while also representing a significant fraction of the total population. Figure 7.1 shows average military sizes over twenty-five-year periods across the thirteen great powers for the last four centuries. In order to consider maximal army capabilities, we consider army sizes during war years only. There was a general increase in army size from 1600 to 1900, followed by a dramatic increase in the first half of the twentieth century. Figure 7.1 also presents evidence on army sizes relative to a country’s total population. We call this measure “military mobilization.” This measure starts off somewhat high in the beginning of the seventeenth century. This is almost entirely due to the ability of a few small powers (Sweden in particular) to raise large armies with foreign soldiers. These were not mass mobilized citizen armies of the sort we would see in later centuries. More generally, the average scale of military mobilization remained relatively constant during the seventeenth, eighteenth, and nineteenth centuries. In other words, army sizes may have increased, but total population increased in the same proportion. Now consider what happened during the twentieth century—the scale of military mobilization rose drastically only to decline almost as dramatically toward century’s end.
Figure 7.1 Military Size and Mobilization, 1600–2000. This figure reports the twenty-five-year averages for the absolute size of the military and mobilization rates for great power states in war years from 1600 to 2000. See Onorato, Scheve, and Stasavage (2014) for sources.
How do we know it was the arrival of the railroad that led to this dramatic increase in military mobilization? There were, after all, many things that changed between the nineteenth and twentieth centuries. To consider this we collected data on the overall size of national rail networks and investigated whether this was correlated with the scale of military mobilization. The idea here is that the greater the size of a country’s rail network, the easier it would be keep a mass army supplied. To deal with the likelihood that a third factor might have influenced both army sizes and railroad networks, we also considered several other possibilities. One might think that richer countries could afford to mobilize a greater fraction of their population. One might also think that a spirit of nationalism would drive more people to serve in the armed forces. Since it has been argued that basic literacy is a necessary precondition for nationalist sentiment, we considered the effect of basic literacy.13 We finally considered whether democracy made a difference for the scale of mobilization. It is plausible to suggest that people will be more likely to fight if they enjoy the right to vote and thus to influence choices made by their government.
Even accounting for the other factors, there is a clear and unambiguous correlation between the extent of a country’s railway network and the scale of military mobilization. The same correlation is observed when using military size, as opposed to mobilization, as an outcome. Do these results reflect a causal effect of the railroad on the scale of military mobilization? We need to consider the possibility of reverse causality; governments may have developed railroads in order to be able to mobilize troops. It would be very difficult to argue that railroads were invented to allow for mass mobilization; they were instead a by-product of the Industrial Revolution. The invention of the steam engine made it possible to move goods by rail in a way previously thought impossible. Improvements in steel production allowed for rail networks of durable steel rails. Even so, one might still argue that once the steam locomotive and the steel rail were invented, governments improved and expanded their railroads in anticipation of the need to mobilize a mass army. There were certainly some instances where this did take place, as with Prussia during the second half of the nineteenth century. Yet the phenomenon does not appear to have been general. Daniel Bogart, an economic historian, has shown that when governments anticipated military threats, they often responded by nationalizing their railways, but they did not expand the scale of them.14 Therefore, as long as we focus on the size of a rail network in our analysis, and not ownership, we can suggest that the correlation between railroads and military mobilization reflects the impact of the former on the latter.
In emphasizing the role of the railroad for mass mobilization, we are also challenging an important piece of received wisdom. Many scholars see the mass army as a product of the French Revolution and the creation of the idea of “the nation in arms.”15 By calling up the entire nation through a levée en masse, the French revolutionaries led Western Europe into a new style of warfare on an unprecedented scale. The evidence contradicts this notion. There is no apparent break in average army size in 1789 or any other year associated with the French Revolution. Even for France itself, there is little support for the claim that the revolutionary or Napoleonic eras were a watershed. It is true that the Napoleonic era saw important innovation in tactics. It is also true that the French revolutionaries briefly experimented with a system of universal conscription. But the increase in army sizes during this period paled in comparison with the change ushered in during World War I.
At the close of the seventeenth century, Louis XIV mobilized an army of 362,000 men. This was equivalent to 1.9 percent of France’s total population at the time. During the year in which Napoleon launched his Russian campaign there were 800,000 men in service in France, or 2.7 percent of the population. Should we call this increase a watershed? It would make sense to do so if the Napoleonic increase was followed by a new plateau, but this is not what happened. Compare now the figures for mobilization under Louis XIV and Napoleon with mobilization during World War I. At its peak France had an army of 5.3 million men, or fully 16 percent of the total population at the time. If we are to speak of a watershed then this clearly took place in 1914, and it was arguably a product of the railroad. This French evidence is borne out in the data more generally. When we take the entire dataset and examine whether the French Revolution constituted a statistically significant break in army sizes, we fail to see that this is the case.16 It is certainly still possible that the French Revolution gave birth to the idea of “the nation in arms,” but it would not be until the advent of the railroad that this idea could be fully realized.
The railroad was not only important for the scale of military mobilization; it was also associated with a dramatic shift in the way that governments recruited armies. Apart from the brief French Revolutionary experience, European armies during the eighteenth and nineteenth centuries were generally recruited either by attracting volunteers, through periodic coercion of certain groups, or by a system of conscription that allowed for those with money to purchase a replacement. This is the type of system that existed during the U.S. Civil War. It was also the system used by France for much of the nineteenth century. The practice of purchasing a replacement seems contrary to modern sensibilities, and it certainly provoked protests at the time, but many advocates defended the French system by suggesting that the government should have no business in restricting a voluntary exchange between two individuals.17 Margaret Levi has argued convincingly that governments eventually removed such possibilities for replacement in order to improve compliance with conscription.18 If all were expected to be able to serve, then any resistance to conscription would be weaker. The shift to universal conscription was also spurred by the development of the railroad.
To see the link between the railroad and universal conscription, we compiled evidence for the thirteen great powers on the date at which they first adopted universal conscription as well as the date at which they abandoned it.19 We then repeated the analysis we performed for military size and military mobilization while considering universal conscription as the outcome to be explained. The results show that the arrival of conscription with replacement or opt-outs was not correlated with the arrival of the railroad. The emergence of this type of conscription in most cases pre-dated rail transport. The situation with universal conscription was much different. The more extensive a country’s railway network, the greater the likelihood that its government would shift to a regime of recruitment by universal conscription, a practice that would become the default strategy for raising a mass army. The arrival of universal conscription led also to demands for a conscription of wealth.
We have already made a number of caveats about our evidence and whether the correlation between the expansion of the railroad and mass armies can be given a causal interpretation. In addition, we make no claim that the railroad was the only technological innovation that led to the era of the mass army. There were prior or simultaneous developments, as well as subsequent developments with the internal combustion engine, that also pushed in this direction. We are simply using this as the most prominent example of an important technology that allowed governments to field armies on an unprecedented scale. Second, we also make no claim that the railroad was a sufficient condition for the mass army to develop. Transport by rail solved the supply problem, but the scale of mobilization also depended on the type of enemies that were being fought. If warfare during the first half of the twentieth century had been limited to large powers waging wars against smaller counterparts, or in colonial contexts, it is likely that the railroad would not have been associated with such a large increase in the scale of mobilization.
THE DEMISE OF THE MASS ARMY
Technological innovation in the nineteenth century helped make the mass army possible. Technological innovation in the twentieth century helped make mass armies less desirable to field. The era of the mass army depended on the existence of a particular technological state of affairs. In this state it was possible to transport men en masse and keep them adequately supplied, but the remote precision delivery of explosive force was not yet a reality. During the twentieth century this situation changed. It became feasible to deliver explosive force from a distance, and as time went on this delivery became more and more precise. Today those countries with advanced weapons systems can deliver explosives with precision measured in a few feet. This development has quite arguably spelled the end of the mass army and with it the potential for compensatory arguments involving the conscription of wealth. In the words of Major Leonard Litton of the U.S. Air Force, “It is no longer required to bring forces into the same geographical area to bring their effects to bear on the same target and, in fact, on the modern battlefield it may be dangerous as well.”20
When it comes to the technology that led to the rise of the mass army, it seems fair to say that the development of the railroad was an essentially exogenous development, a product of the Industrial Revolution. In strong contrast, the technologies underlying the move away from the mass army have been an endogenous development, heavily influenced by investments from national militaries, and in particular that of the United States. The radar, the laser, and the satellite have been among the technologies that have helped allow for more precise targeting of munitions. This has led to an increased ability to deliver explosive force from the air. There is every reason to believe that for great powers, the arrival of the nuclear age has also had an effect in shifting states away from a strategy of mass mobilization. However, the development of nuclear weaponry has itself been aided by the same developments in delivery and guidance systems.
One might object to the previous argument by saying that destructive air power has been in place since World War II, and that was certainly a war of mass mobilization. What is sometimes not recognized is how much more accurate aerial bombing has become over time and how inaccurate it was some seventy years ago. The most common metric for judging the accuracy of payloads delivered from the air is circular error probability. For a given device the circular error probability is the radius around a target within which payloads will fall 50 percent of the time. In 1944 while using the Norden bombsight, the most advanced technology of the time, American B-17 crews were able to deliver conventional bombs with a circular error probability of 1,000 feet.21 That may have been sufficient to wreak havoc on the civilian population, but it wasn’t very precise for hitting specific military or industrial targets.
Fast forward to the Vietnam War. While this was a war fought mostly with conventional (i.e., unguided) bombs, the United States did also make the first use of a laser-guided bomb in the conflict. In raids conducted in 1968 using M117 bombs equipped with laser-guided bombs, a circular error probability of seventy-five feet was achieved.22 One of the most fascinating elements of this story involving the United States introducing the laser-guided bomb is that its use had a rapid impact on the thinking of Soviet military planners. The Soviet military had access to North Vietnamese reports on how effective laser-guided bombs proved to be. For an army whose European military strategy relied upon the idea of a mass armored push across the continent, this was a serious challenge.23 Today, with the third generation of laser-guided bombs, the GBU-24 Paveway III series first introduced in 1983, under ideal conditions a circular error probability of 3.6 feet can be achieved. While Paveway III bombs are expensive, in 2008 in Iraq the United States deployed a less expensive weapon, the LJDAM GBU-54, that still has a low circular error probability. In sum, precision has increased remarkably, and it has also become less expensive. To complete the discussion of air power we would of course also need to include cruise missiles and various forms of rockets. Given these developments, it is not surprising to see that great powers, such as the United States, have mobilized successively smaller numbers of soldiers in recent conflicts.
We investigated whether the arrival of precision weapons spelled the end of the era of the mass army. To do this we constructed a measure that takes a value of one for each year in which a government had access to cruise missiles and zero otherwise.24 We then examined whether this was correlated with either army size or the scale of military mobilization. As we would expect if precision weapons led to smaller armies, we found a negative correlation between access to cruise missiles and both military size and the scale of mobilization. We should be very cautious, however, before suggesting that this result proves that the arrival of precision weapons caused the end of the era of the mass army. As we mentioned earlier, if the railroad was an essentially exogenous invention, precisions weapons certainly cannot be characterized in the same way. The underlying technologies that allowed for their development, such as gyroscopes, radar, and satellites, were developed in a context of intense military competition between the major powers. So, even if there is good reason to believe that precision weapons caused the end of the mass army, our statistical tests do not unambiguously demonstrate this fact.
Aside from the development of precision weapons, the other obvious reason why great powers no longer fight wars of mass mobilization is that since 1945 they have ceased to fight wars against each other. When they are deployed today, great power armies are much more likely to find themselves fighting insurgencies against which a mass army is less effective. The Han dynasty of two thousand years ago found that in such circumstances it makes sense to switch away from mass armies toward a more capital-intensive form of warfare. In recent decades the United States appears to have been arriving at the same conclusion.
IMPLICATIONS FOR THE FUTURE
The era of the mass army, one where countries have mobilized a substantial fraction of their citizens to fight, was dependent on a specific state of technological development. As the precision of weapons delivered from the air has increased, it has become unnecessary, and perhaps undesirable, to mobilize a mass army for conflict. It seems unlikely that technological trends will push in the opposite direction anytime soon. Given the nature of enemies that a country like the United States, or other large industrial powers, are likely to face going forward, it seems even more unlikely that mass mobilization will take place. What does this imply about taxing the rich? The twentieth-century conditions that created powerful compensatory arguments for taxing the rich are unlikely to be repeated anytime soon. These conditions were far from accidental; they were driven by long-term trends involving international rivalries and military technology.