Democracy Through Multi-Body Sortition: Athenian Lessons for the Modern Day

Mature Classical Athenian democracy is presented as a representative system, rather than the commonly described form of “direct democracy.” When viewed in this way, the commonly assumed problem of scale in applying Athenian democracy to modern nation states is solved, and principles and practices of the Athenian model of democracy continue to have relevance today. The key role of sortition (selection by lot) to form multiple deliberative bodies is explained. Five dilemmas faced by modern proposals for the use of sortition are examined. Finally, a new model of lawmaking using multiple allotted bodies is presented, which resolves these dilemmas and which can be implemented in many ways, from a small addition to an existing system to a more fundamental reform such as replacing one or both elected chambers of a legislature.


Introduction
A number of scholars have proposed democratic reforms that use random selection (sortition) to form deliberative bodies (minipublics). Sortition advocates often look back to Classical Athenian democracy for inspiration. However, most political theorists dismiss Athenian democracy as having no practical lessons for modern nation states, due to the issue of scale. I will argue that certain Athenian democratic practices and principles can overcome the challenge of scale, and that they can be used to design legislative systems that are superior to any legislative system used by modern representative democracies. My intention is not to idealize Athenian democracy -this was a society that held slaves, excluded women from citizenship, and created an empire by conquering other city-states. However, I will argue that certain aspects of Athenian democracy contain valuable lessons for reforming today's governments.
Three key Athenian practices were: 1) choosing law-makers and other deliberative bodies by lot rather than election, 2) dividing legislative tasks between multiple bodies, each with particular characteristics, and 3) utilizing both temporary bodies and ongoing fixed term bodies in the decision-making process. This structure allows for optimal performance by matching legislative tasks to the inherent characteristics of each type of body, while also minimizing the opportunity for power-hoarding and corruption.
The key Athenian democratic principles that underlie the system I will propose are the principle of political equality (isonomia), the right to speak and contribute (isegoria), and a belief in the ability of a cross-section of people to deliberate, weigh arguments, and make reasonable decisions.
1. Athenian democracy reinterpreted 1.1. Athenian model: non-electoral representative democracy We learn in school that the Athenian system was a form of "direct" democracy, where citizens made decisions in face-to-face assemblies, without representatives. We learn that although this kind of system could work on a small scale, such as a New England town meeting (Bryan 2003), it would be unworkable for a large city, let alone a nation. 1 We are led to the conclusion that there is essentially nothing of Athenian democracy that we can use today beyond its inspirational value. However, a careful review of the facts of Athenian democracy, and how the Athenians improved it over nearly 200 years, 2 reveals a very different story.
Athenian democracy -especially the mature democracy after 403 B.C.E., as described in Morgens H. Hansen's The Athenian Democracy in the Age of Demosthenes (Hansen 1999) was fundamentally representative rather than direct. At no time did more than a small fraction of the male citizenry of Athens gather to vote. The meeting place of the People's Assembly could hold only 6,000 and later perhaps 8,000, out of an estimated 30,000 to 60,000 eligible citizens. Thus, the People's Assembly was a sample of the demos. But the decisions made there were treated as if the entire demos had voted.
What is more, most governmental decisions were not made by the People's Assembly, but by smaller representative groups of citizens. These representatives were not elected. They were chosen by lot.
The People's Assembly did not generally debate a matter until it had been considered by the randomly selected Council of 500 (boule). Classics scholar Josiah Ober has noted evidence that a council selected by lot was the key institution in Greek democracy, and may even have been more central to the Greeks' concept of democracy than the People's Assembly (Ober 2007). Nondemocracies, such as Sparta, had assemblies, but the agenda was controlled by the aristocracy. In Athens, the allotted Council of 500 set the agenda, and prepared decrees and resolutions. Randomly selected Legislative Panels (nomothetai) of 1,001 citizens over age 30, had to approve new laws. The People's Courts (dikasteria), usually 501, 1,001 or 1,501 citizens chosen by lot, could over-rule the People's Assembly. 3 Nearly all of the magistrates who carried out governmental business were also chosen by lot, usually in panels of 10 citizens. 4 Only a few specialized executive positions, such as generals and financial officers, were filled by election at the People's Assembly. The Athenians regarded elections as inherently aristocratic, since only those with money and 2 It is common to imagine Athenian democracy as it existed at the time of Pericles and his famous funeral oration. However, the Athenians continued to make changes. Paul Woodruff writes in First Democracy: the Challenge of an Ancient Idea, "After civil war or a great military failure, the Athenians would adjust their system to conform better with the goals of democracy. A striking example of this was taking the full power of legislation away from the Assembly in the fourth century, and dividing it with a representative body" (Woodruff 2005). 3 The Greek biographer and historian Plutarch suggested the randomly selected People's Court was thus the ultimate sovereign authority, rather than the People's Assembly. 4 There were, of course exceptions. Taking the concept of "public servant" to the extreme, some government officials, such as the Approvers who ruled on the authenticity of silver coinage used in the public market (Agora) and at the port (Piraeus) were required to be actual slaves (Ober 2008). status could win. To the Athenians, selection by lot was an essential feature of democracy. In fact, this was the general view among political theorists from Aristotle to Montesquieu and Rousseau (Manin 1997).
Athenian democracy was not based on the principle that all citizens should participate in all decisions. That would have been as impractical in Classical Athens as it is today. However, there are important principles and practices of Athenian democracy that can be applied today.
The first principle is isonomia -the equal right of all citizens to exercise their political rights. Through sortition, all citizens who wished had an equal chance and high likelihood of serving in public office. This is fundamentally different from the extremely unequal chance of being elected to political office through election.
The second principle is isegoria -every citizen had the right to speak at the People's Assembly and make proposals. Few citizens ever actually spoke at the Assembly, but the right of any citizen to add new information or arguments was considered fundamental. This is not the same as an individual right to have one's vote counted. A single individual's vote in the People's Assembly in Athens, as in elections today, had little significance. For an individual's vote to make any difference there would need to be a tie that the individual's vote broke (or created). Indeed, votes on most matters before the People's Assembly were never actually counted. 5 Instead, nine randomly selected citizens simply estimated the show of hands.
Josiah Ober has argued that Athenian democracy's institutional ability to harness latent and diffuse knowledge spread throughout the population was a critical factor in allowing it to flourish (Ober 2008). The true significance of isegoria is the opportunity of any citizen to give information, rather than merely a vote. Unlike a single vote, a single piece of information has the serious potential to swing the ultimate decision. Isegoria was not only an individual right, but also a community benefit. The polis would be unlikely to suffer if one individual couldn't vote, but could lose a lot if a citizen with crucial information or argument was denied the right to contribute it, and the People's Assembly made a bad decision as a result. Isegoria protects such "speech acts" rather than voting rights.

The issue of scale
Understanding that Athenian democracy was representative -but in a very different form than we know today -leads us to another important insight. Most modern students of democracy dismiss the Athenian system as inapplicable to modern nation states (or even cities) due to the issue of scale. Some argue that democracy is simply not possible at a large scale, while others simply re-define democracy by substituting modern electoral systems for the original meaning. In fact, the Athenians solved the problem of scale --the core problem that has stymied democratic theorists and practitioners for the last several hundred years.
A population of 30,000 citizens may be small by modern standards, but it is far too large for face-to-face "participatory" democracy as we think of it today. The Athenians invented a system of government that worked at a larger than faceto-face scale, in which the citizens ruled through representative institutions. It was called "democracy." Even the People's Assembly, as noted above, had a representative character. With modern understanding of probability and scientific sampling, we know that a representative sample does not need to keep growing proportional to the growth of the population being sampled. A sample of 6,000 citizens (typical of the People's Assembly) could accurately represent a population of 300,000,000 as well as 30,000.
Some will dispute my contention that Athenian democracy was representative. Some have argued that sortition was simply an efficient means of achieving the principle of "rule and be ruled in turn" through rotation (Manin 1997) or perhaps leaving the choice to the gods. This argument asserts that office holders were not viewed as representatives of the communities, classes or tribes from which they came (Dowlen 2008). Some evidence to the contrary comes from the fact that each of the 139 geographic units of Attica (surrounding villages and neighborhoods of Athens, known as demes) were entitled to a number of seats on the Council of 500, in proportion to their population (Hansen 1999).
It can also be argued that the Athenians, despite their astonishing advances in mathematics, did not know probability, nor have a "theory of representation" (Pitkin 1967). But these bodies did effectively function as representatives of the citizenry as a whole. As sortition theorist Keith Sutherland has noted, any Athenian cook knew that by giving the soup a good stir and sampling a spoonful, one got a good sense of the soup as a whole (Sutherland 2008). We can also note that the Athenians had no "theory of gravity," yet went ahead and utilized gravity in daily tasks any way.
There were two design concepts that were central to Athenian democracy, and that can be profoundly useful today. Random selection (sortition) was one. The other was dividing political powers among multiple randomly selected bodies with different characteristics.

Multiple bodies
In Athenian democracy, most decision processes were divided between separate bodies. The Council of 500 set the agenda, and prepared preliminary decrees and resolutions for the Assembly to consider, but could not pass laws.
The passage of a decree by the People's Assembly could be over-ruled by a People's Court, but these Courts could not pass laws themselves.
Following the codification of 402 BCE, the People's Assembly could no longer pass laws either. Instead the Assembly could only initiate the process by calling for the creation of randomly selected single purpose Legislative Panels, which had to pass any new laws. As Hansen notes, this was a beneficial reform because "the double consideration of a proposal allowed the possibility of coming to a better decision." It also gave "breathing-space to overcome the effects of mass psychosis such as a skilled orator could whip up in a highly charged situation" (Hansen 1999).
The Athenian separation of powers between multiple randomly selected bodies and the self-selected attendees of the People's Assembly achieved three important goals that our modern elected legislatures do not: 1) the legislative bodies were relatively descriptively representative of the citizenry; 2) they were highly resistant to corruption and undue concentration of political power; and 3) the opportunity to participate -and make decisions -was spread broadly throughout the relevant population.
In the next section, I will discuss contemporary proposals to give sortition a broader role in government, especially in the legislative branch. I will argue that they don't meet the three goals above, primarily because they each propose only one all-purpose randomly selected body. Then I will present a design for a legislative process using multiple allotted bodies. Elements of this design could be applied at various levels of government (local, state, national, or international), and with various degrees of saturation. It might be used on a one-time basis for a single law, similar to the task of the Citizens' Assembly in British Columbia, or the randomly drawn Citizens' Panel charged with drafting the budget for the Sydney suburb of Canada Bay in Australia. It might be applied on an ongoing basis to one area of legislation. Such bodies might be particularly appealing to elected legislators who wish to avoid unpopular decisions, such as with the military Base Realignment and Closure Commission established by Congress. This design might be applied to replace one chamber in a bicameral system, while retaining one elected chamber. It could even be applied to the exclusion of elected legislators.

Modern sortition proposals and their problems 2.1. Modern sortition proposals
A number of modern advocates of deliberative democracy have proposed a variety of sortition schemes. Many of these proposals have been for one-off or advisory bodies (Dienel 1995;Dahl 1989;Crosby 1986;Fishkin 2009), while others have proposed schemes to institutionalize randomly selected deliberative bodies, often with genuine decision-making authority (Burnheim 1985;Callenbach and Phillips 1985;Gastil 2000;Carson 1999Carson , 2005Gollop 2007;Leib 2004;O'Leary 2006;Sutherland 2008;Zakaras 2011). All of these plans seek to increase genuine deliberation, increase descriptive representation, 6 and reduce corruption. They also seek to overcome rational voter ignorance, recognizing that in most elections, the chance that one's vote will actually change the outcome is so remote that it is irrational to spend time or effort learning about the candidates or issues. 7 Most of these authors have proposed systems that include only one randomly selected body. 8 There are some exceptions. John Gastil and Robert Richards have proposed a multi-body sortition model utilizing up to five distinct bodies, to add elements of democratic deliberation to the existing initiative referendum process in various U.S. states (Gastil 2012). Aubin Calvert and Mark E. Warren advocate multiple single-issue minipublics (Calvert 2012), and Lyn Carson and Janette Hartz-Karp discuss the concept of linking more than one deliberative method into a combined system, noting that the deliberative democrat "Luigi Bobbio, on one occasion, suggested the possibility of convening two [Citizen Juries] on the same topic with each jury having a different compositionone for activists and one for randomly-selected citizens to assess their respective outcomes" (Carson 2006). John Burnheim's landmark book, Is Democracy Possible?, describes a system he calls "demarchy," composed entirely of randomly selected bodies divided by functional area (Burnheim 1985). He also proposes separate "higherlevel bodies" that would oversee and provide a legal framework for the policymaking bodies to settle disputes. This concept of meta-legislative bodies, which do not initiate or make policy decisions, is one of the springboards for the model I propose.

Motivation to participate
Any proposed system that requires an increase in the amount of citizen participation, as this one does, must respond to the question of whether there would be sufficient motivation among broad swaths of the population to participate. After all, only a minority of citizens is willing to vote in most American elections, and that requires relatively minimal time and effort. New England communities that still have town meetings also see only a fraction of their citizens attending. The competing uses for personal time in modern society, and the "unattractiveness" of politics to most people, raise serious questions about the viability of any such democratic undertaking, especially one centered on deliberation and broad participation (Warren 1996). Even in Classical Athens, democracy was an activity only of that (relatively large) portion of the citizenry who chose to participate. The goal with this plan is to go Athens one better, and include the general population, rather than merely those who are eager to participate.
The (testable) assumption I make here, is that most citizens would readily participate for a set period of time, with appropriate compensation, in a process in which they believed their input really mattered (unlike mass elections). This democratic process would bear almost no relationship to "politics" as we know it today. While the high levels of satisfaction reported by participants in various deliberative processes, such as the BC Citizens' Assembly, or the Danish Consensus Conferences (Fischer 2009), may be misleading (since these were randomly selected from among those who already said they were interested), being one of those "selected" has the potential to overcome the "rational ignorance" problem of mass elections. Just as jurors in court systems may complain about the nuisance of serving, they almost universally take the job seriously. Indeed, many jurors go away with a heightened sense of citizenship (Matthews 2004). The system described below also seeks to accommodate varying levels of willingness to commit personal time to self-governance. The largest portion of participants would commit a very limited time -say, no more than a week.

Five dilemmas of sortition design
All of the single-body sortition proposals face five dilemmas -five pairs of opposing objectives -which can't be reconciled with only one type of body.
1. There is a conflict between maximizing descriptive representativeness, versus maximizing interest and commitment among members of a deliberative body. In Deliberative Democracy in America, Leib seeks to maximize descriptive representativeness and avoid the bias of "participatory distortion" by insisting on mandatory service as in a jury or draft (Leib 2004) Others put more priority on assuring interest and motivation. Their designs tend more towards volunteerism, or a lottery of the willing.
2. There is a conflict between increasing participation and resistance to corruption through short terms of office, versus maximizing participants' expertise or familiarity with the issues under consideration through longer or repeat terms.
3. There is the conflict between giving every citizen the right to speak (self-selection) -offering agenda items, information and arguments for the deliberative process (isegoria), versus the danger that the self-selection of those most motivated to speak will promote domination by special interests and steer outcomes away from the common good.
4. There is a conflict between wanting a diverse body that engages in problem solving through active deliberation, versus independent personal assessment ("private deliberation") that taps the "wisdom of crowds" and avoids information cascades, which can shut out private knowledge. There is persuasive research showing the value of cognitive diversity for problem solving, but also the value of independent, private assessment of information (Page 2007;Landemore 2012;Lorenz 2011;Surowiecki 2004). Group deliberations can also suffer from deference to high status members or group solidarity, leading either to groupthink or polarization. (Sunstein 2005) 5. Finally, there is a conflict between maximizing democratic power by allowing a deliberative body to set its own agenda, draft its own bills, and vote on them, versus avoiding the bundling of issues, with the resulting vote-swapping, as well as arbitrary decisions arising from the persuasive powers of a few unrepresentative charismatic members (Sutherland 2008). These five dilemmas (and proposed resolutions) are summarized in Table 1 in the conclusion section.

Overview
This proposal uses a variety of bodies, each with unique characteristics (such as selection method, and term of office) that are optimized for the task each body handles. The proposal below is comprehensive -a sort of "reference design" such as that used by architects. In real-world applications, it is likely that certain bodies described might be used, and others not. For example, a city government might utilize the Rules Council, Review Panel and Policy Juries within one specific policy area, but keep the Interest Panel, Agenda Council, and Oversight Council functions within the existing city council.

Setting the agenda (Agenda Council and petitions)
An allotted body called an Agenda Council would have responsibility for setting the agendas of policy-making bodies -but not for developing bills, voting on them, or anything else. I call this a meta-legislative body, because it legislates about legislation. The Athenian system did not completely isolate agenda setting from the drafting of proposals, since the Council of 500 could play a role in both. However, in cases where the People's Assembly initiated consideration of some new legislation, the task of drafting it fell to individual citizens, with the ultimate decision falling to an allotted Legislative Panel. Assigning meta-legislative tasks to a separate body from normal legislation follows the long-standing principle of "checks and balances," or separation of powers as advocated by the likes of Montesquieu and Madison.
This body might be selected using a two-tier lottery system of the willing, similar to that used in the British Columbia Citizens' Assembly in 2003-4 (Herath 2007). Such a two-tier lottery system was also used in Athens, where a group of 6,000 citizens over the age of 30 were selected for one year terms to serve on the people's Courts and Legislative Panels, with a subset being selected by lot for any given court case or law.
If several demographic factors are balanced to match the diversity of the general population, random selection will also tend to produce a body that closely resembles the general population in terms of other characteristics, such as political attitudes and cognitive styles.
The Agenda Council and their staff would seek out problems needing attention, rather than merely react to media or special interest group pressures. For example, the United States now faces a little discussed, but indisputable, infrastructure deficit (transportation, water systems, etc.) that arguably is ignored by elected representatives because raising the issue doesn't benefit re-election. The goal is to set an agenda rationally, rather than according to the dictates of electoral imperatives.
In the spirit of isegoria, it would also be desirable to have an alternate means for agenda setting, open to all citizens. Therefore, anybody would be allowed to initiate a petition drive to force a topic onto the agenda. Establishing rules that allow any citizen to promote an agenda item, but that do not encourage special interests to flood the agenda, is challenging The threshold and rules for such a petition effort should not be made by the Agenda Council, because it might be tempted to defend its prerogatives by establishing unreasonably steep barriers to petitions. Instead, a separate allotted body called a Rules Council (discussed below) would handle this task, adjusting the rules over time, seeking to optimize isegoria, while avoiding special interest domination.

Drafting bills (Interest Panels)
Once an agenda item is established, there would be a call for volunteers to serve on an Interest Panel, a group of around a dozen members (to facilitate active participation). The Interest Panels would generate legislative proposals, but have no power to adopt them. In Athens, self-selected citizens (ho boulomenos) could propose laws or decrees, but these generally had to pass through multiple self-selected and allotted bodies (the Assembly, the Council of 500 and Legislative Panel) before final enactment.
There would be as many Interest Panels on a given topic as the number of volunteers would fill. This is derived from, but also modifies, the principle of isegoria. Unlike Athens, in this case the individual is not speaking directly to the ultimate decision-making body. However, dividing into many smaller units means the amount of input would be far greater, with the potential for any individual participating on a Panel to affect the final legislation. Interest Panels at the local level might meet in person on weekends or evenings, but, as Internet access becomes more common throughout society, many would likely use Internet collaboration tools that allow members to communicate and work on proposals sequentially, rather than needing to coordinate meeting times.
The Interest Panels could be formulated in more than one way. Some Interest Panels might be self-organized by like-minded individuals. This might lead to the generation of extreme proposals. However, since these panels would know that they are not the final decision-makers, they would have an interest in tempering their proposals in order to win approval at the final stage. Alternatively, the volunteers could be randomly mixed on Interest Panels to promote diversity of perspectives and cognitive styles. Self-selection at the level of the Interest Panel allows experts who would be unelectable (due to their appearance, class, personality, or other traits) to contribute to governance. It also means that special interests and self-deluded incompetents could participate. While a panel of a dozen would be likely to identify and discount the ideas of "wackos," it is likely that some Interest Panels would put forth poor legislative proposals. This is one reason why it is desirable to have multiple Interest Panels, and why self-selected Interest Panels should not make ultimate decisions. Some of them might deadlock, or disintegrate due to quorum failures. In most cases, however, multiple Interest Panels would produce draft proposals for the next level allotted body -the Review Panel.

Reviewing bills (Review Panels)
There would be a single Review Panel for each policy area established by the Rules Council. The Council of 500 is the closest analog from Athens. The Review Panels fulfill most of the functions of a traditional legislative body, except the initiation and final passage of legislation. The process of the Review Panel would be significantly different than that of an elected legislature. The Rules Council might establish a process closer to that of the British Columbia Citizens' Assembly, with a "learning phase," a "deliberation phase," etc.
While the purview of each Review Panel is far wider than the Interest Panels, it is far narrower than existing legislatures, or the proposed second chamber sortition models, which deal with all issues. This allows the members to develop a deeper understanding within a defined area than is possible in an allencompassing legislature. In traditional legislatures, it is typically only a small portion of the members -those who serve on a particular committee -who have any likelihood of fully understanding any given bill. Most legislators never even read the bulk of the bills they vote on. They inevitably vote based on other considerations, such as vote swapping ("I'll vote for your highway earmark, if you vote for my education subsidy amendment"), or based on some heuristictypically following the lead of their fellow party members who serve on a bill's committee of reference. Thus the homogeneity of the legislature as a whole is compounded by having only a small fraction of the full body even attempting to understand each bill. The Review Panel concept promotes having every member seek to understand every bill, and eliminates, or at least reduces, vote-swapping and partisan gamesmanship.
The Review Panels would be selected in the same manner as the Agenda Council -a lottery of the willing. Unlike the Interest Panels, however, volunteers for the Review Panel lottery would not choose the subject they would be assigned to, in order to avoid special interest distortion.
The Review Panels would be larger than the Interest Panels (perhaps 150 members at a state level). They would be reasonably compensated, and provided with meals, childcare, and a pleasant working environment.
For a state or national design, these panels would be full-time, with overlapping terms of perhaps three years (to gain familiarity with the subject matter). For a municipal implementation, it might be appropriate to have meetings on weekends or evenings so as to not interfere with normal employment. This body would be far more descriptively representative than the Interest Panels. Screening of prospective members should be minimal (such as the ability to read and understand background materials) so as to not unduly distort the representativeness of the final body.
The Review Panel would perform traditional legislative activities: holding hearings, inviting and listening to expert witnesses, utilizing professional staff for research and drafting, and amending or combining elements from the proposals submitted by Interest Panels to produce a final bill. Review Panels might also set goals or criteria for final bills and refer drafts back to Interest Panels for revision.
The procedures used by the Review Panel require careful design so as to maximize problem solving potential and minimize both group-think and internal polarization (Sunstein 2006). These psychological tendencies are powerful, but can be addressed with good design (Manin 2005;Fishkin 2009). The micro-and macro-level elements necessary for genuinely "democratic deliberation" deserve intense study (Gastil 2012). These design elements could include developing a set of agreed upon facts as a basis for discussion. Too often members of groups simply talk past each other because each has their own separate understanding of "the facts." Alternating pro and con speakers for any particular amendment is preferable to a discussion style where the initial majority persuasion dominates. Psychologists have shown that people have a tendency to lean towards the apparent majority side simply due to a desire to fit in -a sort of social bonding instinct. This can also lead to polarization into subgroups that move further apart as members adhere to arguments that support their initial position, and dismiss arguments that don't fit their view (Sunstein 2005). Yet communication between group members that includes feedback about members' confidence levels has been shown to improve decisions in many situations (Koriat 2012). As the science of group decision-making advances, the procedures of democratic institutions should be adjusted accordingly.

Voting on bills (Policy Juries)
A key feature of this proposal is that we don't leave the final decisions on policy to Interest Panels or Review Panels. Because of the risk of group-think, or an extreme polarized majority supported proposal, final decisions are made by separate bodies, called Policy Juries. This separation would also reduce the likelihood of extreme polarized majority positions coming out of the Review Panel, since members of Interest Panels and Review Panels would understand that the final products must be able to pass muster in front of the Policy Juries. Finding middle ground and addressing the needs of minority perspectives might enhance their chances of ultimate success.
Each Policy Jury would vote on one piece of legislation, like the Legislative Panels of Athens. In a state or national implementation, jury service would be nominally mandatory, though with reasonable hardship excuses. It is unlikely that an early municipal implementation would have legal authority to mandate service, so means of encouraging participation would need to be examined. We have few, if any comparable opportunities for civic engagement in modern America, so we can only speculate about potential participation rates. Short service durations combined with compensation and some symbolic status honors might achieve adequate participation, and thus descriptive representativeness. A state or national Policy Jury should probably have at least 400 members to achieve a representative sample. A municipal Policy Jury would probably be significantly smaller, simply for financial reasons.
Since members of the Policy Jury, like the Legislative Panels and People's Courts of Athens, are simply listening to presentations, without discussion, it would be logistically feasible to create far larger juries -in the thousandsthrough the use of the Internet. However, a compelling reason for a more modest size is the goal of finding the "sweet spot" that motivates participants to overcome rational voter ignorance and put in the effort needed to make a considered judgment.
The task and procedures of the Policy Jury are fundamentally different than those of the Interest and Review Panels. Like the Legislative Panels and People's Courts of Athens, a Policy Jury would listen to pro and con presentations on the proposed legislation, and without further debate, vote by secret ballot. This procedure is intended to benefit from the wisdom of crowds described in James Surowiecki's book while avoiding the group-think and polarization dynamics that can arise when participants engage in discussion (Surowiecki 2004). The narrowing of the task to listening to presentations and voting, and the incentive of exercising actual power, will increase willingness to participate and thus descriptive representativeness.
The secret ballot also helps avoid the social pressures that can interfere with members' ability to vote as they think best. The secret vote also reduces the risk of vote buying. Jury tampering protections would still be needed, but potential corrupters are less tempted to attempt to buy a vote when there is no way of knowing if the vote was delivered.
Policy Jury members are not intended to "represent" geographic constituents (as in an electoral system), nor any particular constituents (e.g. based on demographics or political views). The traditional concept of accountability of representatives simply doesn't apply. This may seem strange at first, but once the jury analogy is fully understood, it is obvious. We expect jurors of a particular race, for example, to seek justice, rather than be "accountable" to citizens of their race. Each member of a Policy Jury is asked to vote for what they think is best, with the net result mirroring what the community as a whole would decide if they had the information and time to reflect. Experience with deliberative polls conducted by James Fishkin, in which randomly selected community members are asked to make decisions about public policy matters, suggest that such representative groups may be more able to favor long-term and community interests over egocentric self-interest than elected representatives (Fishkin 2009). Even if no elected official were corrupt or self-serving, re-election dynamics may pressure them to make "popular" decisions they believe their ill-informed, shortsighted or selfish constituency would favor, rather than decisions that might genuinely be in their constituents' long-term interest.
To enhance this promotion of the common good, research on priming effects (Aquino 2009) suggests it might also be beneficial to have members make a pledge, similar to the Athenian Heliastic Oath, that they will vote impartially as their conscience suggests is most just and best for the community. But even if we assume that most members of a randomly selected body will ignore the common good (or we assume that the very concept of a common good is a vain myth), and instead vote according to selfish interests, at worst we arrive at the ideal outcome envisioned by adversarial liberal democracy, which is finding the majority preference among competing interests.

Setting the rules of the process (Rules Council)
The rules committee of a traditionally elected legislature has a built-in conflict of interest, where the party in control tailors the rules or their interpretation, to favor the party in power and advance particular legislative goals.
To solve this problem, I propose the creation of an allotted Rules Council, similar in form to the Agenda council. The Rules Council would establish rules and procedures for all of the other Panels and Councils, such as the lottery process, quorum requirements, means for soliciting expert testimony, procedures to be used in deliberation, etc. Members would have limited terms and could not know how the rules might hurt or help any particular piece of future legislation. Their natural interest would be to assure the fairest and best functioning of all of the bodies. It might be appropriate to limit the lottery for this particular Council to those who have previously served on some other allotted body, so that they understand the dynamics involved.

Enforcing the rules (Oversight Council)
Who presents the pro and con arguments to the Policy Juries, and how do they decide exactly what content to include, in order to give a "balanced" presentation? Even the relative charisma, appearance or social status of different presenters may be significant. One possible approach is to have the same staff present both pro and con arguments crafted by members of the Interest and Review Panels who fall on either side of the pro con divide. However, this leaves the Policy Jury in danger of being steered by the bureaucracy promoting its own interests --a common problem with elected bodies. Many state and national legislatures try to solve this problem by allowing each legislator to hire their own staff. However, this individualized staff often ends up spending an inordinate amount of time on re-election concerns, such as public relations and constituent service, rather than policy.
Instead, I propose the creation of an Oversight Council, chosen by lot, which deals exclusively with staff performance and fairness, rather than the policy issues themselves. In addition to evaluating the general performance of the staff, they would rule on complaints about biased or unfair presentations given by staff. They should probably have the power to hire and fire staff serving the other sortition bodies. At a state or national level it might be appropriate to have three separate meta-legislative councils -the Agenda Council setting the agenda of issues to be tackled and establishing Interest Panels, the Rules Council establishing rules and procedures for good deliberative process, and the Oversight Council overseeing the staff of all of the sortition bodies, and perhaps one that oversees the performance of the executive in implementing the laws. For a single municipality, a single meta-legislative council seems appropriate; with the limited task of assuring that sortition-formed bodies have good process and staff that provide balanced presentations.

Conclusion
The principles and practices of Classical Athenian democracy provide a valuable starting point for designing a modern democracy, which can be scaled to any size. The model presented here fulfills the Athenian principles of political equality (isonomia) and the freedom to speak and contribute to the democratic decision-making process (isegoria). It capitalizes on the Athenian practice of using multiple representative bodies, selected by lot, providing a system of checks and balances. Finally, this model resolves the five dilemmas of conflicting objectives, discussed above (see Table 1.). Interest Panels and petitions allow any citizen to contribute, while Review Panels, and ultimately the Policy Juries protect against participation distortion maximizing problemsolving ability through internal discussion and debate avoid polarization and group-think Councils and Panels promote give and take problem solving, while Policy Juries protect against both polarization and groupthink maximizing democratic power by having a deliberative body with broad powers to set its own agenda, draft its own bills and vote them up or down avoid arbitrary decisions that are overly sensitive to the inclinations of a few unrepresentative charismatic members in an allotted body Councils allow democratic control of agenda and process, while Policy Juries protect against extreme influence from powerful individuals This sortition model is scalable from a municipal to state, national, or international level. The areas of jurisdiction are endlessly flexible, from relatively minor decisions up to and including all legislative tasks, to the exclusion of elections entirely. Obviously it is both practical and prudent to start at a small scale, and with a narrow mandate. Thus an initial focus on municipal adoptions makes sense, as happened with Canada Bay in New South Wales, Australia in 2012 (Thompson 2012). Situations where existing elected officials are reluctant to make unpopular (political no-win) policy decisions, and might be willing to transfer them to allotted citizen bodies, are especially plausible for early adoption.
This model can be implemented in a variety of ways, ranging from small incremental changes to fundamental reforms. For example, elements of the model could be used to: 1. Deal with one law -as with the British Columbia Citizens' Assembly.
2. Make all laws within one issue area -for example, an area where legislators have a conflict of interest, such as redistricting or ethics rules enforcement.
3. Enhance the deliberative quality of an initiative and referendum process (Gastil 2012) -for example, as in the Oregon Initiative Review process.
4. Replace one elected house of a bicameral legislature. 5. Carry out the entire legislative process in place of an elected legislature. It is my hope that this "reference design" can serve as a conceptual base for the development of concrete proposals, which may retain an elected legislative chamber and are thus more feasible to implement, but that benefit from many of the advantages of this model. One wonders what might have occurred if this competing vision of democracy had been widely known by Egyptians during the Arab Spring. The irony of leading democratic activist Dalia Ziada urging a boycott of an election, due to the lack of an acceptable candidate (Ziada 2012), drives home the limitations of the electoral model of democracy. Democracy might be better without elections.