While Peisandros was in Asia Minor (likely Magnesia) meeting with Alcibiades, the members of the newly coordinated and revolutionized xynomosiai made important, preliminary moves of the coup d’état. As described by Thucydides, they implemented a two-pronged plan. The first part of the plan, carried out in secret, was to intimidate the population. That was accomplished, most notably, by the assassination of Androkles, “the foremost leader of the dēmos” and a man that rank-and-file democrats would have looked to for guidance in such uncertain times. He was not the conspirators’ only victim, however: Thucydides wrote that they killed anyone deemed to be “inconvenient.” The second part of the plan was to spread political propaganda. One of the goals of that propaganda, of course, was to remind citizens that, if a “different type” of democracy governed Athens, the Athenians would receive Persian financial support. But they also openly floated, perhaps in assembly speeches, a specific proposal: only those serving in the war should be paid by the state, and no more than five thousand men—those who could serve as hoplites and/or financial backers—should have control of the affairs of state (Thuc. 8.65).
According to Thucydides, the Athenians who were then in Athens and supported the democracy were unable to counter the campaign of intimidation and propaganda waged by the conspirators. His description of and explanation for their inability is very important for the present argument.
And no one of the others any longer spoke against them, through fear and because it was seen that the conspiracy was widespread; and if any one did oppose, at once in some convenient way he was a dead man. And no search was made for those who did the deed, nor if they were suspected was any legal prosecution held; on the contrary, the populace kept quiet and were in such consternation that he who did not suffer any violence, even though he never said a word, counted that a gain. Imagining the conspiracy to be much more widespread than it actually was, they were cowed in mind, and owing to the size of the city and their lack of knowledge of one another they were unable to find out the facts. For the same reason it was also impossible for any man that was offended to pour out his grievances to another and thus plot to avenge himself, for he would discover any person to whom he might speak to be either a stranger or, if an acquaintance, faithless. (Thuc. 8.66)
Thucydides describes the “official” overthrow of the Athenian democracy immediately after the passage just quoted. There were two important contributing events. The first event occurred during a meeting of the assembly—held, notoriously, not in the Pnyx, but a few kilometers away in the deme of Kolonos—wherein the dēmos ratified a motion, made by Peisandros: there were to be all new magistrates who would work without pay; four hundred men would be chosen to rule as they saw fit; those four hundred would convene a council of five thousand when it seemed advisable. Thucydides, using words that recall the dynamic of silence and intimidation described earlier, wrote that those measures were adopted “with no one objecting.” The second important event occurred soon after the meeting at Kolonos when the newly appointed Four Hundred, accompanied by 120 youths, burst into the Bouleuterion and ordered the bouleutai to leave their post. Thucydides, again stressing what is clearly a leitmotif of his narrative of the coup, wrote (8.70.1) that the bouleutai “quietly withdrew without making any objection” and that “the citizens at large raised no disturbance but kept quiet.” Thus the Athenian democracy was overthrown for the first time in its history.11
Thucydides, particularly in the passage quoted above, strongly suggests that the individuals in Athens who supported the democracy were unable to respond to the coup d’état because they had a “revolutionary coordination problem.” The term “coordination problem” refers to a situation in which individuals would like to participate in a particular group activity but do not do so because they are unsure whether or not other individuals will also participate: the cost of inaction or nonparticipation, that is, is lower than the cost of acting with an insufficient number of participants.12 Such problems are very common. Something as commonplace as meeting a group of friends for coffee, for example, can become a coordination problem: if an individual is not sure that the other individuals will attend, he or she might not make the effort to go to the coffee shop.
Coordination problems in revolutionary situations, like that confronting the Athenians in 411, have special characteristics and thus require separate analysis. A modern case study by Timur Kuran elucidates the general operative dynamic of the revolutionary coordination problem. Kuran examined the dramatic fall of communism in Eastern Europe in 1989 after four decades of stable rule. He was particularly interested in both accounting for the failure of experts to predict the catastrophic collapse and explaining why that collapse nevertheless seemed so explicable after the fact. But his theoretical insights are also of great use for an analysis of revolutionary activity in late-fifth-century Athens: they clarify the social dynamic described by Thucydides in the passage quoted above and account for the emergence, persistence, and ultimate collapse of that dynamic.13
Kuran’s theoretical analysis rests on the fundamental distinction between an individual’s “private preference” and his “public preference” about the regime in power. The former preference refers to how an individual actually feels about the regime and is, more or less, beyond his immediate control. The latter preference refers to how an individual appears to feel about the regime and is, to a considerable degree, under his immediate control. If an individual’s private preference about the current regime is not the same as his public preference—that is, if he appears to support the regime but actually does not support it—he is engaging in what Kuran calls “preference falsification.”14
Kuran suggests that every individual makes a simple cost-benefit analysis in order to determine the nature of his public preference concerning a regime that he does not actually support. Specifically, each person has two payoffs to consider. The first payoff is “external” and refers to what members of the ruling regime will do to him should he publicly oppose the political status quo. The value of that payoff is, under most circumstances, directly tied to the current extent of public opposition to that regime: the less opposition is publicly expressed, the more severely a regime will punish it, while the more opposition is expressed, the less severe the punishment will be.15 The second payoff is “internal” and refers to the psychological toll one inflicts on one’s self by preference falsification. The value of that payoff is directly tied to the extent to which one actually opposes the current regime: the more one dislikes it, the more “costly” the toll of acting like a supporter. If the actual internal cost of preference falsification is lower than the expected external cost of publicly opposing the current regime, an individual will falsify his preference and appear to support the regime. However, if the actual internal cost of preference falsification is greater than the expected external cost of public opposition, an individual will align his two preferences and publicly display his opposition. Kuran calls the tipping point at which that alignment occurs—that is, the point at which an individual will change his public preference to match his private preference—his “revolutionary threshold.”16
Kuran developed a simple model, called a threshold sequence, in order to explain the relationship between the public preferences of particular individuals concerning the ruling regime and the stability of the state.17 Here is an example of a threshold sequence.
{1, 1, 2, 3, 3, 4, 5, 6, 8, 10}
Each of the 10 numbers in that sequence represents the revolutionary threshold of one person.18 Thus the person represented on the far left with a revolutionary threshold of 1 will publicly display his opposition to the current regime only if one person has already done so; otherwise, he will continue to falsify his preference and appear to support the regime. Likewise, the person represented fourth from the left with a revolutionary threshold of 3 will publicly display his opposition to the current regime only if three people have already done so; and so on.19
The threshold sequence
presented above describes a stable political status quo (i.e., one in equilibrium) despite the fact that many individuals either are greatly dissatisfied with the regime or, if supporters, are nevertheless willing to participate in the early stages of a revolution. The person represented on the far left of the sequence is willing to publicly oppose the regime if only one person (or 10 percent of the population) does so first. According to the model, however, that will not happen. Likewise, the person represented in the position third from the left is willing to act if only two people “go first.” But, again, that will not happen. Thus the tragic irony of the revolutionary coordination problem is that many people want to depose the regime and, if they did act, likely would succeed. But they do not act, and the unpopular regime stays in power.
In the standard terminology of social scientists, the individuals represented in the threshold sequence who want to overthrow the current regime are handicapped by “pluralistic ignorance”: they do not know the relevant political preferences of their fellow citizens.20 Indeed, if the regime’s policies are repressive, it would be very difficult for such information to become widely known: individual A would either not take the risk to find out what individual B thinks or, if somehow he did find out, he would not risk telling individual C. Even if each citizen suspected, or knew with certainty, that the vast majority of his fellow citizens were falsifying their preferences, he still would not know those citizens’ revolutionary thresholds; that is, he would not know how many people would have to publicly oppose the regime before those citizens would also publicly display their opposition. It thus would be rational (and safer) for every individual to suppose that other individuals—despite the fact that they are adamantly opposed to the regime—will engage in public dissent only if a rather large percentage of the population has already done so. As a result, nobody publicly opposes the regime, because each person thinks that an insufficient number of people will follow him. In other words, the expected external cost incurred by action is too high.
The concepts developed by Kuran make it relatively easy to account for the inability of the Athenians to respond to the coup d’état in 411, in spite of the fact that a significant majority of the population supported the democracy. As noted above, the oligarchs implemented a two-pronged plan of intimidation (including assassination) and propaganda (promising Persian assistance against the Spartans in return for a change of regime). An Athenian citizen therefore knew that, if he should demonstrate political dissent without sufficient support, he would suffer a very high external or physical cost. He also had some reason to believe that other individuals might actually support the change of regime and thus would not support his public protest: they might actually believe that the Athenians would receive Persian assistance should Athens be governed by an oligarchy. In such circumstances, concluding that the expected external/physical cost of public opposition is greater than the internal/psychological cost of preference falsification, individual A falsifies his preference and appears to support the regime, although he does not actually support it. Then individual B, following the same thought process and noticing that individual A appears to support the coup, falsifies his preference too. Individual C follow suit, as does individual D, and so on. Thus an “ignorance cascade” sweeps through the Athenian population, resulting in higher and higher revolutionary thresholds for successive individuals and, consequently, a high average number in the population’s threshold sequence.
The supporters of Athens’s democracy, in other words, had a revolutionary coordination problem: despite the fact that most individuals do not support the coup, nobody does anything to oppose it. As Thucydides wrote in his description of the ramifications of pluralistic ignorance, they were “powerless because of the size of the city and ignorance of each other.”21 The fact that the Athenian fleet, and with it a significant percentage of the reliably pro-democratic thetes, was at Samos certainly contributed to the coup’s success.22 Nevertheless, the important point here is that those individuals who were in Athens and did not want the Four Hundred to dominate the city were unable to strike back. They wanted to act in defense of their democracy, but, given the high stakes involved in action and the uncertainty about their fellow citizens’ views, each person waited for others to act before acting himself.
I would like to suggest, if only for heuristic purposes, that the threshold sequence discussed above—{1, 1, 2, 3, 3, 4, 5, 6, 8, 10}—plausibly models the population in Athens at the time of the coup of the Four Hundred, when most of the thetes were with the fleet at Samos. This suggestion cannot be fully proven, of course, but three points argue in its favor. First, the sequence describes a population in which a slim majority of the population opposes the regime but does not take action: 60 percent of the population has a revolutionary threshold below 5 (the number at and above which generally indicates support for the status quo), while nobody has a revolutionary threshold of 0 (which indicates that an individual will oppose the status quo before anybody else does).23 Second, that threshold sequence describes a population in which a large minority supports the regime: 40 percent of the population has a revolutionary threshold of 5 or above. Third, the sequence describes a population susceptible to the type of revolutionary action analyzed in the following section.
MOBILIZATION
Thucydides strongly suggests (8.72.2) that fear felt by the leaders of the regime for the Athenian naval forces stationed at Samos was the ultimate cause for the collapse of the regime of the Four Hundred. They had reason to be afraid: when word of the coup came to Thrasyboulos—the soon-to-be general and future democratic hero—he had all the soldiers and sailors stationed at Samos swear an oath to (inter alia) support democracy and oppose the Four Hundred (Thuc. 8.75.2). They even threatened to kill ambassadors sent to them by the Four Hundred (8.86.2) because they “overthrew the democracy.” When the members of the Four Hundred learned of that determination, a split within their leadership emerged. Some—the so-called moderates—wanted to empower five thousand men to govern Athens in fact, not only in name, as had been the case up to that point, arguing that by doing so they would placate many of the Athenians stationed at Samos and thereby allow Athens to continue fighting Sparta and maintain the Athenian Empire. Others—“the most influential men” (hoi dunatōtatoi), as Thucydides refers to them—wanted to make peace with Sparta quickly. They clearly concluded that, should they ally with the Spartans, they could defeat the forces at Samos and maintain control of Athens, thereby saving their own lives. The members of this latter group thus sent ambassadors to Sparta to negotiate an end to the war and fortified Eetionia in order to facilitate the entry of Spartan ships into the Piraeus (Thuc. 8.90–91).
As described by Thucydides (8.92.2), the immediate cause for the collapse of the Four Hundred was the assassination of Phrynichos, a leading member of the Four Hundred and a participant in the embassy to Sparta. He wrote that, before the return of Phrynichos from the diplomatic mission, conversations critical of the regime were conducted in secret and between only a few individuals—and this despite the fact that the hard-liners were fortifying Eetionia more energetically and thereby increasing the probability that their city would be subjected to Spartan rule. Thucydides continues,
But finally, Phrynichos, after his return from his mission to Lacedaemon, was stabbed in the full agora as the result of a plot by a man of the frontier patrol, and before he had gone far from the council chamber suddenly died. The assassin escaped, while his accomplice, an Argive, was seized and tortured by the Four Hundred, but did not reveal the name of anyone who instigated the deed nor anything else, except that he knew that many used to assemble in the house of the commander of the frontier patrol and at other houses.24
Significantly, Thucydides causally connects the conspicuous assassination of Phrynichos with a large-scale uprising against the regime in the Piraeus. The uprising had three stages, each of which was defined by the status and number of its participants. Those who took part in the first stage were
“moderate” regime members. Thucydides (8.92.2) clearly connects the commencement of this stage with the assassination of Phrynichos: “then [i.e., after the assassination], when no single action had been taken in consequence of this, Theramenes and Aristokrates and all the rest who were of the same way of thinking went to work more boldly.” And in another marked phrase (8.92.3), which echoes and inverts an important leitmotif of his account of the coup, he asserts that Theramenes concluded that “it is not possible to keep quiet any longer.” The moderates then acted in earnest.
The second stage of the uprising involved the hoplites who had been fortifying Eetionia. After Theramenes and his allies had broken their silence, those forces—men who had been working to advance the interests of the hard-line conspirators—arrested Alexikles, a general of the oligarchic regime and “very favorably inclined toward the members of the political clubs” (Thuc. 8.92.4). Significantly, Thucydides adds (8.92.5) that “most important of all, the mass of the hoplites were in sympathy with all this.”
The final stage of the uprising drew in everybody else. After the arrest of Alexikles, chaos erupted both in the Piraeus and in the city. “Thereupon,” Thucydides reports, “the hoplites and many of the people of Piraeus at once mounted the fortification [at Eetionia] and began to tear it down.” The rioters shouted “whoever wants the Five Thousand to rule in place of the Four Hundred, let him set to work,” but Thucydides explicitly states that what they really meant was “whoever wants the dēmos to rule” (8.92.6–11). Clearly nobody was “keeping quiet” anymore.
Death to Tyrants! Page 4