The Presentation that Dr. Ober gave was based on the following paper published in the March/April 2008 issue of the Boston Review. Dr. Ober will also publish a book on the subject, more information can be found at http://press.princeton.edu/titles/8742.html.
Learning from Athens
Success by design
8 The great legacy of the 20th century may be the emergence of democracy as a universal value: the conviction that whenever people are subjected to power, their views about the exercise of that power must be taken into consideration. This democratic principle, it is now widely agreed, is a fundamental moral requirement on the governance of states, global institutions, and even nongovernmental organizations.
But if democracy is now generally regarded as morally superior to other forms of political organization, its effectiveness in delivering the goods remains a matter of sharp contest. How does democracy fare when it comes to assuring physical security, protecting health, and fostering economic growth? We know, for example, from the economist Amartya Sen that famines are all too common under authoritarian regimes but do not occur in democratic states with a free press. Yet Sen also acknowledges that we do not know the effects of democracy on economic growth: "If all the comparative studie s are viewed together, the hypothesis that there is no clear relation between economic growth and democracy in either direction remains extremely plausible."
Democracy may be right, then, but is it good?
What we know is that some democracies achieve sustained success. As democracy is universalized as an aspiration, it becomes increasingly urgent to understand what sets the successes apart. One answer—not very hopeful in its implications—is that successful democracies are the ones that got in on the ground floor of the developing world economy or benefited from such propitious circumstances as good weather, abundant natural resources, or the protection of oceans. An alternative argument is that democratic success is determined not by the unalterable facts of history and geography but by the kind of democratic institutions a country has.
This second theory is more hopeful but surprisingly hard to prove. We cannot for ex ample simply compare more and less successful democracies: the spread of democracy outside a small club of well-established "first world" states possessing substantial environmental and historical advantages is too recent to provide enough evidence to correlate the success and failure of states over time with specific institutional design. We could expand the universe by also considering nongovernmental organizations that employ democratic processes. But case histories of democratic nongovernmental organizations are sparse, and the fundamental differences between them and political institutions are too great.
Which brings us to classical Greece. With its premodern and illiberal characteristics—the exclusion of women from participatory citizenship, the prevalence of chattel slavery, low levels of technology—the world of the ancient Greek city-states is not where we would first be inclined to look for connections between democratic design and practical success. But on closer inspect ion, we have much to learn from ancient Greek democracies. Despite the slavery and exclusively male citizenship—which were present in both more and less successful cases—the Greek democracies provide just about as close as we can get to a laboratory of political experimentation.
Ancient Greek city-states (poleis; singular polis) existed in a highly competitive environment in which failure was severely punished, by loss of independence or even annihilation. Destruction, total or partial, of physical infrastructure (sacking) or population (mass expulsion, extermination, or enslavement) was quite common: between a quarter and a third of the better-documented Greek states are known to have suffered such destruction at some point in their history. Poleis responded to such internal and external threats by experimenting with a variety of constitutional forms, with more and less extensive participation by citizens. Women never held active political citizenship in classical Greece, although they played very prominent roles in religion and other central aspects of civic life. But the extent of adult male participation varied widely, from the periodic experience of autocracy in Syracuse, to the more or less narrowly framed oligarchy in Corinth, to Athens's very robust and highly participatory democracy.
These were essentially controlled political experiments, on a surprisingly large scale. There were, in the classical Greek world (ca. 600–325 B.C., from the shores of the Black Sea across much of the Mediterranean coastal zone), some 1,000 city-states, whose territory was usually made up of a central city and outlying villages. And while every polis had a somewhat different resource base and a unique history, the Greek poleis were relatively homogeneous in terms of resources and history. So history and geography probably accounted less for relative success and failure than the design of the political system. In addition, although Athens was more internally diverse than is usually recognized (more on this below), the population of Greek poleis lacked a "clash of civilizations" level of interstate cultural differences. So it is not likely that deep cultural differences between states can account for their differential success.
In short, many city-states, with relatively common levels of resources and a broadly shared history and culture, were responding to their practical problems by trying out different kinds of political systems. What do we learn from these experiments? First, democratic Athens did exceptionally well over time relative to its rivals: Athens beats all other poleis on all measures of practical success—for example, in notable public buildings, interstate acceptance of its money, overall cultural impact and fame. Second, the data for the ca. 200 poleis for which we have some constitutional information shows that although having a tyrannical form of government typically depresses productivity, being a democracy does not by itself guarantee success: some democracies do much better than others. Finally, by tracking the relationship between government form and overall state capacity—measured by military strength, building projects, and social programs—we find that, over 300 years of Athenian history, democratic participation is closely correlated with Athens's effectiveness at addressing economic, military, and social problems. Moreover, the Athenian democracy became more participatory over the course of its history, and we find that the increase in participation precedes the growth in effectiveness.
Athens, it seems, was successful at least in part because it was democratic. Was there something about the kind of democracy in Athens that distinguished it from the less successful democratic Greek city-states?
In fact, Athenian democracy had a distinctive design principle: it was designed for organizing the dispersed knowledge of citizens. Its central government al bodies, including the Assembly, enabled an active exchange of useful social and technical knowledge among diverse teams of citizens, promoted learning, and thus improved the chances for innovative and effective policies. And as a balance to innovation, Athenian institutions worked to codify rules, archive information, and standardize proven work routines, thus promoting organizational learning over time.
Although the relationship between democracy, knowledge, and practical success is not as widely recognized by modern scholarship as it should be, it did not go unnoticed in Athens. Historians and philosophers—Herodotus and Thucydides as well as Plato and Aristotle—all discussed the distinctive Athenian processes for the collection, coordination, and codification of useful knowledge and associated them with the polis's success.
* * *
The fifth-century B.C. historian Herodotus records the key meeting of the Athenian Assembly at which the Athenians decided to build a great fleet of warships:
The revenues from the mines at Laurion [a district in southern Attica] had brought great wealth into the Athenians' treasury, and when each man was to receive ten drachmas for his share, Themistocles persuaded the Athenians to make no such division but to use the money to build two hundred ships for the war, that is, for the war with Aegina [a small but prominent nearby island polis].
Herodotus drops us into the midst of a debate in 483 B.C. over state resources. The discovery of the Laurion silver mines had brought a windfall to the community, and a proposal had been made to divide the revenue among the citizenry on a per-capita basis. Ten drachmas was a considerable sum for a working man, perhaps roughly equivalent to a month's wages. Yet the Athenian citizens chose to forego individual payoffs in favor of Themistocles' proposal for building and manning a fleet of warships that would make Athens the greatest naval power of the Greek world.
What could have induced a citizen-voter in the Assembly to renounce his individual short-term advantage in favor of the grandiose project of turning Athens into a naval superpower? After all, as Herodotus notes, the only obvious naval threat to Athens at the time was posed by the relatively puny Aegina.
Themistocles was a prominent Athenian: a former magistrate (archon) who had successfully advocated a buildup of Athens's port facilities. But in 483 he spoke from the floor as a concerned citizen whose proposal would pass only if he could persuade the majority of the Assembly. Themistocles appealed to a body of social knowledge: it would have been meaningless to build hundreds of warships unless there were tens of thousands of men available to row them. Although in later decades Athenian war galleys were rowed by mixed crews of citizens, hired foreigners, and slaves, in 483 the state's limited finances manda ted that the vast majority of the rowers would be citizen-volunteers.
Themistocles' ship-building proposal came just after thousands of poor Athenian men were enfranchised by democratic reforms following a social revolution. Suddenly these men, who had no substantial property, were officially enrolled as participatory citizens. Yet they were denied the opportunity to serve in the military because military service was for the most part limited to those who could provide their own arms and armor. Themistocles' proposal offered a way to extend military service to all citizens. By drawing attention to the Athenians' public rejection of a policy of short-term individual gain in favor of one that deferred future advantages and appealed to social knowledge, Herodotus takes us beyond the familiar story (popular with some historians) about lower-class rowers gaining more political power as the navy becomes more important; Themistocles achieved good policy by making a persuasive argument in a public decision-making forum. And the argument was persuasive because ordinary Athenians knew enough to know that the proposal would work.
Three years later, in 480 B.C., the Persian Empire invaded Greece, and Athenians had to choose to fight the Persians with their ships, flee west, or surrender. Herodotus offers a detailed narrative of their decision-making. They began by seeking religious counsel. A delegation of citizens was dispatched to Delphi to ask the advice of Apollo's oracle. The response was, at first, extremely pessimistic: "Wretches, why do you linger here? Rather flee from your houses and city. Flee to the ends of the earth." The Athenians demanded clarification and were offered a cryptic message with a ray of hope. "All will be taken and lost . . . / Yet a wood-built wall will by Zeus all-seeing be granted / . . . a stronghold for you and your children . . . / Divine Salamis will bring death to women's sons." The Athenian envoys recorded both responses and presented them to the public Assembly of thousands of citizens.
Herodotus describes the debate in the Assembly as a deliberative process airing a variety of opinions. He shows how, in the course of the meeting, the Athenians integrated new information into their existing body of knowledge, considered various alternatives, settled on a policy of high risk and high payoff, and began to carry out its provisions. Herodotus notes that the "elder citizens" were sure that by "wooden wall," Apollo was referring literally to a palisade around the sacred Acropolis. Others believed that "wooden wall" referred metaphorically to the recently-constructed warships. So the debate was engaged: Flee to the ends of the earth, as the original oracle had advised? Or stay and face the enemy? If they stayed, should the Athenians depend on a wooden palisade or on their warships? Interpreting the ambiguous information offered by a culturally trusted source was not simple. Specialists weighed in: expert Athenian oracle-interpreters warned that the line about "divine Salamis bringing death" meant that Athens would certainly be defeated in a naval engagement.
In a hierarchical political order, there would never have been a public debate on the oracles. In a traditional republican Greek regime (e.g. Sparta), in which such issues were discussed in public, the authoritative opinion of elders, backed by religious experts, would prevail. But in democratic Athens the premise was that all citizens had the right to publicly express their views and that each knew something that might be important in deciding on the best policy. No plan could be adopted if it contradicted the knowledge and will of the majority of the Assembly.
Herodotus reports that some urged Athens to trust its warships. The Assembly included many men—including rowers, steersmen, and marines —who knew a good deal about the operational capacity of the Athenian navy. But they could not explain the oracle. At this impasse Themistocles, the same man who had proposed the great expansion of the navy three years before, stood up to challenge the elders' and experts' interpretation of the oracle. He pointed out that if the "women's sons" fated to die at Salamis were Athenians, the oracle would have called Salamis "cruel" rather than "divine." So, Themistolces reasoned, it was Persian women whose sons' deaths were foretold. Thus, he argued, the Athenians should regard their ships as the wooden wall and fight the invaders at sea.
Themistocles' novel interpretation provided the many Athenians who trusted the fleet with a way to align their personal knowledge of military preparedness with the religious message. Herodotus reports the outcome: "In their debate after the giving of the oracle they accordingly resolved that they would put their trust in the god and meet the foreign invader of Hellas with the whole power of their fleet, ships and men, and with all other Greeks who were so minded." The outcome was a dramatic victory at sea—the decisive Battle of Salamis in 480 B.C. The Persian invaders were thrown back, and the Athenians went on to win their own empire and enter a Golden Age.
Herodotus understood Athenian democracy. His story of the decision to fight the Persians at sea focuses on how Athenians sought out information and used it in their deliberations. Elders and experts in oracular interpretation had their say. But their interpretation did not carry the day because it did not adequately account for what many in the Assembly knew about Athens's new naval strength. Eventually, a speaker offered an interpretation that aligned a religious interpretation with the popular knowledge. Herodotus' account of this key Athenian decision—he had already informed his readers that the Athenian decision to fight at sea determined the course of Greek history—implies that the strategic plan adopted by the Athenian Assembly depended not only on oracles, experts, and leaders, nor simply on the bargaining power of newly enfranchised Athenian citizens. It rested on the conviction that that even the poorest Athenians, the ones who would be rowing the warships, knew something important about how to defend the community.
* * *
A century and a half after Salamis, in 325 B.C., the Athenian Assembly would again deliberate about the use of naval resources. Alexander the Great had extended his conquests to India, and Athenian foreign policy was constrained by the Macedonians. Athens would have to embark on a major overseas enterprise to keep a measure of control over its destiny. It would be another crucial decision.
The result was a decree authorizing a new Athenian naval station in the Adriatic. The military outpost would be expensive in terms of money, manpower, and the attention of government officials, yet the dangers they faced justified the expense. A supplementary decree arranging for the dispatch of warships to the station explained that the naval station and its fleet would guarantee Athenian commerce and the grain supply by providing security for Greek and barbarian ships alike against Etruscan pirates.
A copy of this "naval-station support decree" survives in stone and offers a snapshot of democratic knowledge in action in fourth-century Athens. Calling for some 14 institutional bodies to work cooperatively in carrying out a complex and potentially risky new venture, the naval-station decree gives a vivid sense of a system of public administration with many moving parts, encompassing inspection, supervision, funding, dispute resolution, reward, and punishment.
The decree reveals the many ways the state made use of public knowledge and the complicated interplay of legislative policymaking, established law, and executive responsibility. The primary decision-making and authorizing body of the decree is the demos: the Athenian citizenry in the form of the approximately 8,000 citizens who chose to attend the relevant Assembly. The decree assigns duties to a wide range of government agents: for example, to the curators of the shipyards at Piraeus—a board of magistrates selected for service in a public lottery—and the trierarchs—wealthy Athenians asked to take on the expensive public service of repairing warships for active service.
The curators and the trierarchs have primary responsibility for getting the right ships to the naval station at the right time. But the decree concerns many other public bodies as well, and it specifies a variety of incentives and sanctions. Incentives include golden crowns presented to the three trierarchs who get their ships to the dock first. Being knowledgeable enough to assemble the necessary experts‚—ship riggers and outfitters, competent rowers, a steersman, and so on—was the key to victory in the race to the docks.
The winning trierarchs' names were publicly proclaimed at a state festival. The proclamation had an important public purpose: to publicize "the competitive zeal [philotimia] of the trierarchs in respect to the demos." In other words, the elite officials appointed by the state were competing for the esteem of the citizens. The public proclamation made the identities of the elite Athenians who had proved themselves the most worthy of esteem known among the citizenry as a whole.
The decree also requires judicial magistrates to call people's courts into session to deal with legal challenges. Trierarchs who believed that their estates were being unfairly burdened were assured the opportunity of timely legal appeal. The funds for setting up these courts (mostly for the jurors' pay) are to be provided by yet another magisterial board: the treasurers of Athena. The decree then moves to punishments. Any individual, whether private citizen or public official, who shares some responsibility for the naval venture and who is discovered to have been derelict in his duty is to be fined 10,000 drachmas. This was a substantial penalty; anyone who could not pay the fine would become a state debtor and lose his civic rights. Public officials are charged with doing the investigation and are themselves subject to fines if they fail to assess penalties upon the guilty.
The decree assigns overall supervision of the operation to another governmental body, the Athenian Council of 500, and offers the councilors the chance to win a collective honor—but only if they are deemed by the demos to have acted well at the conclusion of the naval-station support project. The Council of 500 was to meet in plenary session on the dock until the dispatch of ships had been completed. But there were other demands upon the council's time, so the decree mandates the creation of a new magistracy: ten dispatchers who are to see to it that the council's orders are carried out. The decree concludes by authorizing the council to pass supplementary decrees related to the dispatch of the expedition, though specifically noting that the council lacks the authority to annul any decree of the citizen Assembly.
Like many other important enterprises undertaken by the Athenians, founding the naval colony required pooling a great deal of diverse knowledge. In the modern world we manage comparable operations with a professional executive, employing a command-and-control structure for coordinating a pool of experts. But very few of the many individuals and public bodies assigned responsible roles in the Athenian naval-station support decree could be regarded as professionals or experts in the ordinary sense of those terms. The Council of 500, whose membership was annually chosen by lot, did much of the work and oversight. And though magisterial boards were provided with quasi-professional "staff secretaries," the responsible officials were amateurs, chosen for a year's service by lot (the curators of the dockyards), by popular election (the generals), or by rotation (the trierarchs). In the Athenian model there was very little in the way of executive-level command and control, and nothing like a formal hierarchy.
* * *
Organizing collective action is famously difficult, but by considering the incentives and sanctions built into the naval-station support decree we begin to see how the participatory and knowledge-based Athenian system of self-governance could have worked so well. Yet this is not the whole answer to our question. In order to understand Athens's extraordinary, sustained effectiveness, we need to look closely at the design of the Council of 500, a linchpin democratic institution set up by Cleisthenes in the aftermath of the democratic Athenian revolution of 508 B.C. Today, the name of Cleisthenes is less well known than that of his predecessor, Solon, the famously wise Athenian lawgiver who revised the Athenian law code in the early sixth century B.C., or Themistocles, Pericles, and Demosthenes, the distinguished Athenian leaders of the fifth and fourth centuries. Yet it was Cleisthenes, a scion of an old aristocratic family, who saw that revolution had permanently changed the political landscape of Athens. Cleisthenes framed a series of new institutions that channeled the extraordinary energy of the revolution into coordinated collective action. The institutional design devised by Cleisthenes worked well over time because it was predicated on aligning the individual interests of diverse persons around policies that would benefit the community as a whole. Athens's success depended directly on the choices made by individual citizens, and the Council of 500 worked in surprising ways to shape those choices by ext ending and deepening social networks.
Consider a typical Athenian village in the late sixth century B.C.: Prasiai was a settlement on the east coast of Athenian territory, some 15 miles from the city as the crow flies, perhaps twice as far by the overland route. The free population was in the range of 700 persons. Of these, about 180 were adult native males. As a result of their long history of steady interaction—social, economic, and religious if not yet extensively political—the men of Prasiai probably knew a lot about each other. The ties between them were strong, in a formal sense: many of a man's friends and relations were also friends and relations of one another. As a result the general level of common social knowledge was very high. People knew who was technically skilled in various domains, who could be counted upon and in what circumstances, whose advice was valuable on what topics. The high level of common social knowledge meant that the community was transparent—commitments were credible and cooperation was unavoidable, because efforts to take a free ride would be noticed and would carry serious consequences for one's standing in the community.
Prasiai-like villages are celebrated by some historians of ancient Greece and communitarian political theorists, but liberal theorists point out that strong ties do not necessarily mean good values. Tight communities, they say, often show prejudice and intolerance toward outsiders. Perhaps so. But when it comes to making effective political decisions beyond the local level, the big problem with these communities is that they operate as cliques—efficiently processing knowledge within the small local network but resistant to the free flow of information outside it. Insularity defeats effectiveness.
Athens solved this problem through the Council of 500. Charged with deciding on the agenda of the full Assembly of Athenian citizens and partly responsible for the day-t o-day administration of state affairs, the council was composed of ten delegations, each with 50 members. Each delegation came from one of the ten Athenian tribes—these were entirely new creations, designed for political representation and not based on any ethnic or cultural connections. Each tribe in turn consisted of the residents of several demes (towns, villages, and neighborhoods), drawn from three regionally distinct parts of Athenian territory. So each tribe included men from the coast, from the agricultural inland, and from the city. The members of each tribal delegation were selected at the deme level. Prasiai was one of the constituent village-demes of the coastal zone of the tribe Pandionis. Each year Prasiai sent three men to Athens to sit as councilors as part of Pandionis's 50-man delegation.
Although we lack detailed first-person narratives from antiquity, a thought experiment can help us explore how the council's design shaped individual councilors' choices and ultimately improved the quality of political decisions. Imagine a man from Prasiai—call him Poseidippos—embarking upon a year's service on the council in the year of the council's foundation.
After being chosen—probably by a lottery—Poseidippos takes up temporary quarters in the city. Poseidippos is glad to have two friends from Prasiai on the Pandionis delegation, but he knows that he is going to have to work closely with 47 other men on his tribal team—all unknown to him—and then with the other 450 councilors from the nine other tribes. As Poseidippos looks around at the Pandionis delegation, he notices that it is a network riddled with structural holes. Councilors have few or weak ties outside their bonds to the members of their deme. This limits communication and coordination. But the council's institutional design made opportunities of these same structural holes; their p resence offers a strong incentive to an ambitious and entrepreneurial member of the council to build new bonds.
In using the language of contemporary network theory, I am of course speaking in deliberate anachronism. But network theory is based on social facts that are intuitively obvious: advantages accrue to those willing to build bridges between cliquish sub-networks within an extended network of persons undertaking some common enterprise. Suppose, then, that Poseidippos intuitively recognizes the advantages to being a bridge-builder. He leaps at the opportunities offered by the frequent meetings of Pandionis's tribal team of 50 to network with men from other villages. The personal interactions within the tribal delegation are intense as its members struggle to accomplish their duties. Poseidippos easily forms new friendships.
As the year goes on, Poseidippos becomes an increasingly well-respected and valuable member of his tribal team. He has a handle on more and more useful information; he has learned what people in other demes know. He learns something, for example, about pottery exports from his city contacts and about upland olive cultivation from his inland contacts. He also learns who among the members of his Pandionis delegation is trustworthy and on what topics, and who is a friend or enemy of whom. Soon he is in a position to bring together disparate types of knowledge for problem-solving. Meanwhile, he comes to see that sharing his own expertise as well as his new knowledge earns him trust and connections. As a source and conduit of useful knowledge, Poseidippos accrues advantages for himself and enables his tribal team to get its job done. Of course, Poseidippos is not the only one to see the advantages of building bridges across local networks; others on his team notice what he is up to and follow his example. The intimate conditions of service on the council make communication easy, and so knowledge flows quickly. The Pandionis team is soon working effectively as a knowledge-sharing network.
Pandionis is not anomolous; each tribal delegation has the same basic makeup. And so the same sort of bridge-building goes on both within each of the ten tribal teams and between them. Therefore, we can suppose that over the course of the year the membership of the council comes to operate as a single, extended social network.
The council, then, links small, preexisting, closely bonded local networks across regions, kinship groups, occupational groups, and social classes. Useful knowledge flows across the extended network with increasing ease. As communication generates trust, experts become more willing to share "proprietary" information. Some citizens realize that their tacit knowledge of people and processes, formerly simply taken for granted as common sense among members of their community, was valuable to a diverse group of people with very different sorts of tacit knowledge. Thus, both specialized technical knowledge and generalized tacit knowledge necessary to making good decisions are increasingly accessible to the deliberations of the group as a whole. As councilors learn more about who was good at what and who to go to for what sort of information, they become more discriminating about their recommendations and as a result the whole council is increasingly capable of doing its difficult job well. Moreover, because each councilor has a local network of contacts outside the council, each councilor is a bridge between the council and some subset of the larger population. The council as a body, then, is able to access at fairly low cost a good deal of the total knowledge available to the extended Athenian community. Athens as an organization comes to know a lot of what the Athenians know as individuals.
Because councilors ordinarily served for only a year and were judged on the basis of how well they served the polis, the council as an institution never developed a self-serving institutional culture. The rules of order remained sufficiently simple and transparent to be learned by each year's incoming class. As councilors built networks and worked together over the course of the year on problems of governance, they better grasped the larger governmental system of which they were a part. Government ceased to be opaque, and councilors quickly became expert at the work of politics. They became better able to judge the value of available knowle