The Rise and Fall of Classical Greece

by Ober, Josiah

  In this kind of game, each player moves (makes a choice) in turn. When no player has a move to make that would better his or her situation, in light of moves he or she knows could be made by other players, the game ends. The (Nash subgame perfect) equilibrium outcome of this game is a negotiated settlement in which the King declines to attack the fortified city, the city pays to the king a reasonable rate of taxes (substantially lower than would be demanded if the city submitted), and the elite citizen chooses to support democracy, despite a relatively high internal taxation rate.

  Of course, simple games do not capture the complexity of historical reality. In many real-world situations, the outcome would differ from that of the game sketched in appendix II. In the real Hellenistic world, kings did sometimes choose to besiege even very well fortified cities. Sometimes cities were destroyed or forced to submit and pay high taxes. Sometimes elites chose revolution rather than cooperation with democratic regimes. The game offers only a general model, not a reliable prediction of what would happen in a specific instance. But the model’s historical realism is affirmed both by general trends in Hellenistic military, political, and economic history, and by consideration of an often-overlooked passage in Aristotle’s Politics.

  Writing in the later fourth century, at the cusp of the classical and Hellenistic eras, Aristotle, in books 7 and 8 of the Politics, developed his own model for a “best practically achievable polis.” Most of his discussion regarding the “polis of our prayers” concerns social, political, and educational institutions. Notably, his best polis, although in many ways aristocratic, is also democratic insofar as all native free males turn out to be citizens. The best polis’ citizens possess both civil rights to property and legal redress, and participation rights, in the sense of “ruling and being ruled over in turns.”37

  A central part of the duty of these citizens is to defend the state as soldiers. Like Plato, in the Republic and Laws, Aristotle was very concerned to ensure that the citizen-warriors of his best polis had the right motivation (to this end, Aristotle specified that each citizen would own property in militarily sensitive border zones) and the right training. The goal was to ensure full mobilization of a military that was effective in time of war. Unlike Plato, Aristotle also specifically advocates for city walls. Moreover, he insists that fortifications be militarily advanced and defended by the best available technology. His reason is the necessity of deterring aggressors:

  As regards walls, those [i.e., Plato] who aver that cities which pretend to valor should not have them hold too old-fashioned a view—and that though they see that the cities that indulge in that form of vanity are refuted by experience…. [Because] the superior numbers of the attackers may be too much for the human valor of a small force, if the city is to survive and not to suffer disaster or insult, the securest fortification of walls must be deemed to be the most warlike, particularly in view of the inventions that have now been made in the direction of precision with missiles and artillery for sieges…. not only must walls be put round a city, but also attention must be paid to them in order that they may be suitable … in respect of military requirements, especially the new devices recently invented. For just as the attackers of a city are concerned to study the means by which they can gain the advantage, so also for the defenders some devices have already been invented and others they must discover and think out; for [potential aggressors] do not even start attempting to attack those who are well prepared.

  —Aristotle, Politics 7.1330b–1331a, trans. Rackham (Loeb) adapted

  The King, City, and Elite Game described in appendix II explains the relationship among four distinctive features of the Hellenistic Greek world—more democratic states, lower than expected levels of rent extraction by warlords, heavy investment in fortification by city-states, and elite cooperation with local democratic regimes. The key to their relationship is called out in the italicized phrase of Aristotle’s Politics: potential aggressors “do not even start attempting to attack those who are well prepared.” That disinclination of the potential aggressor to begin an attack is an important outcome of the game.

  Aristotle is sometimes criticized for being excessively “polis-centric”—for putatively failing to attend to the great changes that were already well advanced when he was writing the Politics. I have argued elsewhere that, quite to the contrary, Aristotle’s “best practical polis” was designed with the emerging world of Macedonian hegemony very much in mind.38 If this is right, then we may suppose that among the unnamed aggressors who will not “even start attempting to attack” the presumptively well-prepared (with Fortifications, artillery, and defenders) polis is, as in our game, a potentially predatory Macedonian king.

  Among Aristotle’s primary goals in the Politics is to show that the autonomous polis is the best, indeed the only possible, environment for the pursuit of an essential moral end: the fulfillment of human flourishing (eudaimonia). I would suggest that Aristotle was very well aware of the military developments of his age. He took those developments into account in designing his best polis and, most particularly, when writing the passage cited here. If the morally essential (as he supposed) autonomous polis were to continue into the era that had been set in train by Aristotle’s employer, Philip II, and by his pupil, Alexander, the polis would have to be able to defend itself against levels of coercion that could destroy its autonomy and thus ruin the environment in which human lives might be perfected.

  Aristotle knew from experience that Macedonian kings could often, although not always, expect to win when they chose to besiege major Greek cities. He knew that Philip, Alexander, and their warlord successors put a great deal of energy into developing technology (torsion catapults, siege towers) and techniques of siegecraft. Their sieges, when attempted, were frequently successful. Aristotle’s hometown of Stageira had, like its neighbor Olynthos in 348, been razed as a result of Philip’s effectiveness as a besieger of cities. Aristotle also knew, however, that there had been some spectacular failures: Philip failed to capture either Byzantion or Perinthos, despite major sieges, in 340. Moreover, even a successful siege was likely to be costly—tying up essential manpower and resources for months—Alexander’s famous siege of Tyre, a prominent Phoenician city-state, delayed him for seven months in 332. All of these facts were readily available to Aristotle when he was writing and revising the Politics.

  Aristotle imagines his best polis as a new colonial foundation but one that will exist in an ecology of other poleis—potential partners in exchange and allies in war, as well as potential rivals. Aristotle also knew that the major Greek cities were well fortified and well defended, and that there were many of them. While it goes beyond the text of the Politics, Aristotle might well have reasoned that if a king attacked cities within his realm without provocation in an obviously predatory manner, the rest of the fortified cities in his realm would lose their incentive to cooperate with him. They might stop paying taxes, coordinate with other cities in revolt, or invite in a rival king (as Perinthus and Byzantion had done in receiving aid from the Great King of Persia). Given these conditions, a potentially predatory king had good reason, on the face of it, not to “even start attempting to attack” a well-fortified, well-defended city—so long as he could gain the revenue he needed to pursue his ends by other, less risky means.

  Aristotle certainly knew all that. Importantly, the Hellenistic kings who dominated the Greek world in the era after Aristotle also knew it, as did the residents of the fortified Greek cities within their realms. Moreover, each participant knew that the others knew it, and so on. That is to say, the predatory king’s disincentive to attack, if a city were well fortified and well defended, was a matter of common knowledge.39

  Common knowledge regarding the probability that a king’s attack would succeed and concerning the costs he incurred in mounting an attack is a premise of the game between a predatory King, a democratic fortified City, and a potentially revolutionary Elite citizen of the City (formalized in appendix
II). The game predicts that the King will decline to attack the City; the City will agree to pay moderate taxes to the King; and the Elite will support the democracy. The King declines to attack because, although he has a good chance of taking the City if he does attack, he also faces a significant chance of failure. In light of that chance, the cost of attacking is prohibitively high. The City willingly pays some tax because to refuse to do so would change the King’s cost–benefit assessment in the direction of an attack that really does have a good chance of succeeding, leading to plunder and higher taxes. The Elite citizen supports democracy because an oligarchic city would be less able to defend against the King’s attack. Because the King also knows that, the taxes the King could demand from the City might be proportionately higher if the City were oligarchic.

  These outcomes generally track historical trends across much of the Hellenistic world in the third and early second centuries. Moreover, if we change the assumptions of the game in ways that violate Aristotle’s recommendations for the defense of his best city, the outcomes of the game change in ways that track historical reality less well. Appendix II offers a fuller defense of the claim that the equilibrium solution of the game is a reasonable approximation of historical relationships between cities, kings, and elites in the early Hellenistic era.

  There are various possible contributory causes of continued growth after the classical fall: Most importantly, with Alexander’s conquests the greater Greek world continued on the path, traced in chapter 9, of expanding in size while converging on a common polis culture—and this increased the potential payoff to Greek cities from specialization and exploitation of relative advantages in production and distribution of goods. Counterfactually, however, had predatory early Hellenistic warlords been free to tax the Greek cities at very high rates, the underlying conditions that had (so I have argued) produced the classical efflorescence—vigorous competition between relatively wealthy and relatively independent city-states in the context of social orders that encouraged individual and collective investments in human capital—would have come to an end.

  Most of the early Hellenistic kings had short time horizons and were ready to forego the uncertain prospect of long-term gains for short-term payoffs in the form of booty and high rents. The conjunction of fortifications and/or federalism with democracy forced upon the kings a certain level of restraint in coercive rent extraction. That restraint helped to create the conditions that sustained a strong economy and thereby enabled the Greek world to continue to make substantial cultural advances long after the fall of classical Greece.

  ENVOI

  The victories of imperial Macedon and the loss of full independence in the later fourth century were very painful for great city-states like Athens, and at least temporarily eliminated others, including Thebes. But by the end of the fourth century BCE, it was clear that the fall had not been fatal for the polis ecology overall. The Hellenistic equilibrium maintained both political exceptionalism and efflorescence through the third century and well into the second century. That equilibrium was the enabling condition for the survival of Greek culture in the vibrant form in which it was taken up by the Romans, when they conquered the Greek world. Because Greek culture flourished for long enough to be adopted by the imperial Romans, it was eventually, albeit in fragmentary form, passed along to us.

  While Greek city-states remained in existence long after the Roman takeover, the Hellenistic equilibrium did not endure indefinitely. Roman military capacity was of a different order from that of any Greek state, so a different game was to be played in the Greek world after Rome arrived as imperial hegemon. The Greek cities, like most of the cities of the Roman Empire, were eventually stripped of their fortifications.40 The federal leagues were disbanded. But by that time, they had done their work: Greek political culture, in the form of democratic exceptionalism, had survived the fall and had been deeply embedded into a developing literary canon. Greece, once great and though fallen, was on its way to immortality.

  Because of the immortality of classical Greece, democratic exceptionalism has been preserved as a real-world possibility ever since. Because that which once was could again be, domination could no longer be authoritatively presented as the only secure form of social order, as the only route to economic growth and cultural achievement, or as the inevitable fate of humanity. Although autocracy and domination have remained historically common, whenever and wherever domination is challenged, political theorists and legislators know that there is an alternative. They know that the alternative could be a brilliant age of citizen-centered politics and high culture. Readers of history also know that regression to the historical mean of elite domination is possible, that citizen-centered regimes can be oppressive in their own right and can be overthrown, and that political, economic, and cultural development can be reversed.

  The purpose of this book has been to present anew the inspiring story and cautionary tale of the rise and fall of classical Greece, using newly available evidence and the explanatory tools of twenty-first century scholarship. I have thereby sought to keep faith with my predecessors: the generations of scholars and writers—from Herodotus, Thucydides, and Aristotle onward—who made the political history of ancient Greece a living resource for all who aspire to end domination and to advance toward citizenship.

  APPENDIX I

  REGIONS OF THE GREEK WORLD: POPULATION, SIZE, FAME

  The 45 regions of the Greek world are those listed in the Inventory, although names of regions have in in some cases been anglicized: thus Sicily for Sikelia. Region numbers are assigned according to order of regions in the Inventory. Regions are located by number on map 1. Numbers of the poleis within each region are per the Inventory. For each region, I estimate a total population, list the number of known (i.e., listed in the Inventory) poleis in the region, and calculate the size and fame of an average polis of the region.

  Each region’s population is calculated as follows: Multiply poleis of known size by estimated population per size category (see table 2.1). Add known poleis of unknown size according to the formula used to generate table 2.1. Estimate the total number of unknown-size poleis of each size category (the standard size of a polis of unknown size calculated at (0.3 × 1,000) + (0.53 × 3,500) + (0.114 × 7,000) + (0.04 × 17,000) + (0.14 × 35,000) = 4,125). Add 6 unlocated poleis and 64 hypothesized poleis, each assumed to be of the standard size, to reach the hypothetical polis total of 1,100 (per table 2.1). Unlocated and hypothesized poleis are assigned to regions 5, 25, 27, 33, 35, 37, and 39. Total population thereby added = 288,750. Population added to some regions is based on the high average size of known poleis and the large number of unknown poleis (regions 2 and 3), on the basis of literary and/or documentary population figures cited in Hansen 2008 or on other evidence suggesting undercount (regions 4, 7, 10, 11, 18–20, 30, 36, 39, and 40). Total population thereby added = 439,375. Population of region 30 (Inland Thrace) is a guess, since there are no known-size poleis in that region. Population is reduced in region 44 (Cyprus), based on unusual distribution of size 5 poleis. Sum of regional totals = 8,248,500, per table 2.1. “Average Size” (based on 7 size categories in table 2.1) is of known-size poleis only. “Average Fame” (based on the raw data of Inventory text columns rather than the five categories of table 2.2) is of all Inventory poleis.

  TABLE I.1 45 Regions of the Greek World: Population, Average Polis Size, and Fame

  APPENDIX II

  KING, CITY, AND ELITE GAME

  Josiah Ober and Barry Weingast

  This game models the choices of a Hellenistic King, the democratic government of a well-fortified Hellenistic Greek City, and an Elite resident of that City. The King must decide whether to threaten to attack; if he chooses to threaten, the City must decide whether to defend or submit. If the City chooses to defend, the Elite individual must choose whether to support or subvert the democratic government. Depending on those choices, there will be subsequent choices made by the three players. The game assume
s rational (expected utility maximizing) decisions from the stylized players, who make their moves in the game under conditions of incomplete but symmetric information: That is, the outcome of the lottery that decides, in the case of an attack, whether the attack will succeed cannot be known with certainty in advance. But all players have the same level of knowledge about the lottery—that is, they calculate the likelihood of the attack succeeding in the same way, reaching the same conclusions about probabilities of outcomes. Other than the lottery, players are assumed to have complete information.

  The game is obviously an abstraction from the much messier real world of politics and decision-making, where there are many actors, decisions are not formally rational, and information is often assymetrical and incomplete at every step. However, as the application of game theory to the problem of why wars are ever fought has shown (Fearon 1995), formalization may be useful insofar as it offers a framework for explaining puzzling phenomena. In our case, the puzzles are the counterintuitive correlation between more democracy and more investment in military architecture, the relatively low rents demanded from Greek cities by the kings, and the acceptance of democracy by Greek elites.