‘On Poachers and Gamekeepers’ – Response to Review Article

Ethan:> What distinguishes Bouricius’s proposal is the systemic integration of these diverse bodies. Each features design decisions (size, duration, selection, etc) based on its role in the whole. Yes indeed, you are right to view Terry’s proposal through the lens of deliberative systems theory (c.f. Parkinson and Mansbridge, 2012). My concern though is democratic systems theory, especially Dahl’s insistence that the demos should have exclusive control of both the agenda and decision stage. Terry acknowledges that only large quasi-mandatory decision juries can claim (descriptive) democratic legitimacy. A sortition-only system would not be able to choose agenda items for a political system that had a statutory (rather than advisory) role, as this would be an aleatory oligarchy. >But if Bouricius’ design were adopted at all levels of government, everyone in society would have many opportunities to take part. A new democratic legitimacy would stem from, in Aristotle’s terms, the chance to rule and be ruled in turn. Aristotle was writing about a tiny ancient polis, in which all citizens had a good probability of serving on the council and even being president for one day. Rotation is not scalable to huge modern poleis. >elections’ egregious agenda-setting shortcomings None of the proposers of ‘electoral lottocracy’ (including Jordan Rose, Kai O’Reilly, Alex Kovner and myself) seek to defend existing electoral institutions. Our proposals are for some variant of proportional representation (to better reflect the ideological diversity of the demos) coupled with a reduction in the parliamentary proposal threshold, to ensure this diversity is better reflected in the agenda for the policy jury to consider.

I'm only around a third of the way throgh Bouricius’ book, so I'm ill equipt to comment on his proposed sortition model. I have gone through a significant protion of his critiques of elections, and I agree with just about everything he says. However, of the various reforms he proposes and dismisses, one fairly new and radical proposal remains unexamined: removing majoritarianism. I argue that nearly all the issues he expresses with electoral systems can be sovled, either in large part or in entirety, through removing majoritarianism from the electoral process. As Keith mentioned, myself, O'Reilly Kovner and him all have variations of this . I think the significance of removing majoritarianism is far more profound that is currently understood. As a demonstration, I wish to make the following observation. In a majoritarian system, which ever party or coalition has power, they are expected to make decisions on all issues, as any problem that faces society that can be solved through law, can only be solved by that majority party or coalition. When voters participate in elections, they are therefore vasting a vote on all issues. If we remove majoritarianism and allow parties to act propose legislation independently of one another, we transform this dynamic entirely. As there is no singular party who is responsible for all law, there is no expectation for every party to have a plan for every issue. Any issue a party neglects can be addressed by a different party. Voters can vote for parties that explicitly represent only issues that directly matter to them. This has two distinct effects on voter behaviour. The first is that voters do not need to educate themselves on all issues (something we know they cannot be reasonably expected to do) but only the ones that are important for them. It also reduces the rational irrational effect. In a majoritarian election, it's only when their vote is the deciding factor between winning and losing can they expect their vote to matter. In this proposed alternative, so long as their party sits in parliament, they can be expected to make a difference, making the threshold of their vote being important significantly lower. Such arguments can be made to address many of the flaws presented in electoral systems. From partisanship, to campaigning, such a change would have an effect on every aspect of what it means to be a legislator and the consequencing incentives.

Just wanted to make clear that Jordan is referring to election as a mechanism for policy proposals. All (four!) of us electoral lottocrats agree with Terry that the final decision power has to be in the hands of large, quasi-mandatory policy juries.

Thanks Keith! Yes, that's correct. We are arguing that elections can be salvaged, but only if final say is given to some form a sorition body, and majoritarianism is removed as the underlying structure of the elected body.

Jordan:> of the various reforms [Bouricius] proposes and dismisses, one fairly new and radical proposal remains unexamined: removing majoritarianism. However we all (including Terry) agree that the Policy Jury decision is majoritarian, but the majorities will shift between different sections of the demos, depending on the specific issue, on account of the opening up of the parliamentary proposal threshold. Whether or not deliberation can move simple majorities in the direction of unanimity is an open question, but my argument for a three-party alternative to highly polarising decisions (illustrated by the Brexit Referendum) suggests that it might do. And this would count against Terry's argument that cleavages are the product of partisan shenanigans.

> Terry acknowledges that only large quasi-mandatory decision juries can claim (descriptive) democratic legitimacy. I don't believe Terry ever said that. In any case, it's not so. Neither legitimacy nor representativeness is binary. To the extent that bodies derive legitimacy from descriptive representativeness, different formulations offer legitimacy in different degrees. If we step away from the theoretical, the success of civic assemblies to date largely derives from the legitimacy elected officials opportunistically borrow from sortition-selected bodies that are neither quasi-mandatory nor particularly large. A few dozen everyday citizens chosen by stratified selection have repeatedly offered legitimacy when the vote-granted legitimacy didn't suffice. This is not to imply that massively empowering a single such civic assembly with statutory authority would be perceived as a legitimate, but that is obviously not Terry's proposal. In his system democratic legitimacy is derived from the fact that, while all assemblies are far more descriptively representative than anything any electoral system could muster, no single assembly has enormous power. That includes the Agenda Council, which cannot credibly be considered oligarchic given the limits of their authority and the pace of their rotation. Their primary function is to initiate Review Panels, whose policy proposals they have no authority to oversee.

Ethan, I don't have Terry's book to hand, but during our exchanges over 10 years or so on EbL he has acknowledged that "descriptive" legitimacy depends on the law of large numbers (and Surowiecki's independence criteria). None of his other panels meet these exacting criteria. >the success of civic assemblies to date largely derives from the legitimacy elected officials opportunistically borrow from sortition-selected bodies. Yes that's why I find the (faux-legitimate) representative claims of citizens' assemblies so alarming. It's disappointing that, despite his acknowledgement of Pitkin's multi-faceted perspective, at the end of the day all his emphasis is on descriptive representation. Citizens vote for those who, rightly or wrongly, claim to speak for their preferences and interests.

Sortition, for smallish bodies in which those called can decline to serve, combined with stratified sampling, would not provide the level of legitimacy needed for making final decisions about adopting laws. BUT they can still be tremendously useful, while also being FAR more representative than any political party or elected body. The DIVERSITY of sortition bodies allows for superior collective intelligence (compared to relatively homogeneous self-selected politicians). Such Review Panels also allow for the seeking of win-win possibilities through active deliberation. But win-win scenarios are anathema to political parties, which use policies as weapons of war in campaigns and SEEK to maximize disagreement, fear, and hatred. Sortition bodies also provide a buffer against corruption and excessive self-selection bias. Since these smaller Review and Agenda bodies, which would have longer terms of service could suffer from group-think, information cascades and undue charisma influence, and would not have the level of legitimacy of a very large quasi-mandatory jury none of their decisions can be decisive. The final decisive jury would use the wisdom of crowds to make individual assessments, after hearing pro and con arguments (like the Athenian Nomothetai, and Keith's design as well).

Terry:> Smallish sortition bodies . . . can still be tremendously useful, while also being FAR more representative than any political party or elected body. Only for one variant of representation. Using a different variant, a plurality of voters decided that Trump was more representative than Harris. >win-win scenarios are anathema to political parties In our various responses, both Jordan and myself have argued that the problem is not parties, it's majoritarianism at the proposal stage. In the revised version of my Brexit referendum alternative, I showed how a proposal threshold of 26% could generate a win-win solution. If we are going to deliberate in a constructive way we need to address the arguments of our interlocutors.

> Only for one variant of representation. Using a different variant, a plurality of voters decided that Trump was more representative than Harris. I mean, this is just a fancy way of saying "dems da rules." Formalistic representation currently grants election winners power, so they get to be called representative. If we adopted a constitution that formalized sortition-only power, the sortition body would have formalistic representation. The argument is about whether dem ought to be da rules. Ideally, our legislatures would offer multiple forms of representation. Sortition-oriented bodies could potentially address all four of Pitkin's forms. The most important, she felt, is substantive representation. Many chapters of Terry's book explore the sources of alignment problems with electoral systems. Lotteries alone do not ensure substantive representation, but taken as a whole Terry's system design engenders deeper deliberation, greater checks and balances, broader participation, and fewer conflicts of interests, all of which improve substantive representation. Symbolic representation is the fuzziest, but when I contrast the abysmal reputation of politicians with the high regard for civic assemblies, I don't find the preference alarming at all. The closer we look, the more we see superior substantive representation in civic assemblies, and so the symbolism grows. I personally feel far more represented symbolically by civic assemblies than by elected bodies. As for descriptive representation, there is no serious argument that electoral bodies are descriptively representative. Perfect descriptive representation is impossible, but the difference in descriptive representativeness between smaller civic assemblies and large juries is the difference between Denali and Everest. Elected bodies are just Capitol Hill. So, today 3 out of 4. Some day may it be a clean sweep.

>Formalistic representation currently grants election winners power, so they get to be called representative. No, it's not 'da rules', It is because they were chosen by electors. It's not perfect, but electoral lottocrats seek to ensure that the choices available better match the ideological diversity of the citizenry. >Sortition-oriented bodies could potentially address all four of Pitkin's forms. No - see my quote above from Pitkin on the limitations of the descriptive mandate. Manin is also very clear that sortition disappeared as a result of the natural-right theory of consent. >I personally feel far more represented symbolically by civic assemblies. It's worrying when sortitionists have to resort to symbolic representation, especially given it's fascist leanings (Pitkin, pp. 107-9). >there is no serious argument that electoral bodies are descriptively representative. Of course not. Their legitimising principle is 'acting for' (Pitkin, ch. 6)

Terry: > But win-win scenarios are anathema to political parties, which use policies as weapons of war in campaigns and SEEK to maximize disagreement, fear, and hatred. I'm currently reading your book right now and enjoying it immensely. I have a model that is quite similar to Keith's. What are your thoughts on how this might change if the legislative body, while elected, were not majoritarian? In this scenario, parties could submit legislation independently of one another and without the ability to block one another. It would then be decided through a trial format, with a sortition body deciding. In doing so, electoral dominance would no longer be the way in which success is found. Parties would not need to create broad platforms, as no party would be required to create all laws. This would no longer pit parties against one another electorally, as it isn't a zero-sum game. It would allow small parties to succeed, and even create law (if they can convince a jury), allowing for greater diversity in parliament. I feel that this has potential to solve a great number of the problems we see with electoralism, as outlined your book. Is this something you've put any consideration into? If so, what are your thoughts on it?

Jordan:> Parties would not need to create broad platforms, as no party would be required to create all laws. Exactly. But parties (or ad hoc coalitions of individual legislators) would need to put forward single-issue proposals that appeal to a majority of the Policy Jury. This might suggest convergence towards a win-win solution, as it's primarily the votes of the undecided that need to be won over. Both Terry and myself appeal to Harrington's model of two girls dividing a cake, in which (ultimately) the chooser rules.

Keith: > This might suggest convergence towards a win-win solution, as it's primarily the votes of the undecided that need to be won over. For sure. The requirement on legislators that they need legislation that can be convincing changes the incentive structure entirely. Further, since there is no de facto leader of parliament (through majoritarian rule), there is no "parliament" (in a singular sense) for the sortition body to defer to. Parliament is not unified through the majoritarian rulers, but a collection of largely equally powerful voices, none of which can be said to be *the* parliament for which one could defer. I think this might alleviate much of the concern Terry has with bicameral hybrid solutions.

I apologize for being slow to respond to the remarks on "removing majoritarianism." Let me first broaden the context to argue for broad democratic experimentation. I would like to see many reforms with well-argued rationales field tested. I primarily work in the US, where I advocate for a flourishing of state constitutional conventions that largely follow Terry's model but include a referendum component. I say this not to deliver us into a debate on ideal mechanisms for constitutional reform. But it's worth making this point: if held today, even constitutional conventions that approximated Terry's multi-body design would be extremely unlikely to arrive at constitutional designs that approximate his ideal. We can't know with any specificity what such conventions would decide, but the obvious guess is that sortition-based conventions would think, well, rather conventionally. But not like conventional politicians, like conventional *people.* My hope and expectation would be that they would produce a wide variety of solutions. Many would embrace proven reforms, and some would push the envelope a bit, but none – again, *today* – would scrap elections altogether. In this context, yes, I would very much like to see improvements along the lines of what you describe as "removing majoritarianism," though I would think a term like "fluid electoral majoritarianism" might better capture the idea. As I understand it, the argument is not against majorities per se but against a calcified majority. I imagine sortition-based constitutional conventions would be very receptive to arguments and designs for fluid majoritarianism and would like embrace such ideas on a much nearer timeline than they would any sortition-only design. I certainly agree that fluid majoritarianism would open the possibility of some substantively representative outcomes that current implementations disallow. But the more extreme claims seem implausible: > nearly all the issues [Terry] expresses with electoral systems can be solved, either in large part or in entirety through removing majoritarianism from the electoral process Almost none of the electoral concerns Terry outlines are deeply addressed by fluid majoritarianism. Not the subjugation of, in Kahneman's terms, system 2 thinking to system 1 thinking. Not the distortions of electoral ego inflation. Not even partisan tribalism, which is fueled by many forces beyond parliamentary vote calls. Any salve for rationale ignorance concerns is miniscule; the core dynamic remains that voter effort will yield little effect. Perhaps there would be a diminishment of the deliberation-corrupting influence of elections through the emergence of parties with incomplete platforms. If you haven't made promises about an issue and don't need to for your next election, you can theoretically engage in deliberation unencumbered. Of course, it flips the common argument for elections on its head to suggest that politicians are at least able to deliberate in earnest on matters about their constituents don't care. But of course that benefit already exists in the many arenas that escape public scrutiny, which brings us back to Terry's more core concerns about whether electoral bodies' interests genuinely align with the public's. I won't retrace the book here. He makes a very strong case that they are simply far inferior to sortition bodies in substantively representing the public. Almost all of Terry's arguments focus on electoral dynamics that corrupt that substantive alignment regardless of the voting system. Ultimately, I think of this more as an empirical question than a purely philosophical one. Let's have conventions and make the case for fluid majoritarianism; surely some places will give it a try. My hypothesis would be that such reforms will, like most other "successful" electoral reforms to date, solve a bug or two. But it will not address elections' core problems. It will not produce generative deliberation that deeply reflects the perspectival diversity of the populace. It will not effectively counter the growing disillusion with electoral oligarchy. And it will certainly not produce what Landemore calls "civic love." Terry explicitly frames his multi-body designs as a north star, not as a next step. He's not under any illusion that anyone will implement his designs tomorrow. He advocates for an adoption strategy he calls "peeling," which entails gradually empowering more sortition panels over time. For this to work, you wouldn't need broad shared faith in the north star so long as the incremental adoption of mini-publics provided incremental benefits. There is every reason to believe they will. So it may be a while before we get to treat questions around Terry's full design as empirical, we can at least hope for near-term incremental solutions where we can ask incremental questions. Does increased authority change deliberative dynamics? Are increasingly empowered civic assemblies be able to sustain the high-quality deliberative standards that today drive enthusiasm for the substance of their work? Does the public increasingly view systems of highly complementary sortition bodies as *accruing* legitimacy via checks and balances? These (and many many others) are questions best answers by welcoming an era of broad democratic experimentation. In such a world, fluid electoral majoritarianism should get a shot in the near term. When its middling nourishment has been digested, the north star will still be there.

Ethan: >though I would think a term like "fluid electoral majoritarianism" might better capture the idea. As I understand it, the argument is not against majorities per se but against a calcified majority I'm not sure what exactly what you have in mind for "fluid electoral majoritarianism", but I'm actually quite literal in my rejection of electoral majoritarianism. You can review my model here https://www.ingentaconnect.com/content/imp/jos/pre-prints/content-impjos9, but what I propose is allowing parliament to approve legislation with a minority vote (which I generally use 20% as my base threshold). There is no requirement for any majoritarian approval from the elected body at all. Parties can, straightforwardly, pursue legislation independently of one another. Elections are not winner-take-all, parties are not comepting electorally for dominance. >Not the subjugation of, in Kahneman's terms, system 2 thinking to system 1 thinking. As legislators cannot pass legislation based purely on electoral domination, which they can achieve without appealing to views outside their own, all legislation must be written with the perspective it must convince a jury of randomly selected citizens. This cannot be achieved without activating system 2 thinking. They must be capable of antisipating objections. > Not the distortions of electoral ego inflation. As elected officials are not "the winners", in that they have victory over other elected officials, the inflated ego that arrives from such a victory, as described in the book, doesn't really apply. Each legislator, on their own, has the same power as any other, and aren't afforded benefits by being in a majority party. > Not even partisan tribalism, which is fueled by many forces beyond parliamentary vote calls While tribalism, certainly, has it's roots in-groups and out-groups such that the mere existence of parties may engage it, the fact that elections require victory and dominance over other parties feeds into this dynamic. I'm uncertain to what degree this would go away if we remove that dynamic, but I find it doubtful the answer would be "none". > Any salve for rationale ignorance concerns is miniscule; the core dynamic remains that voter effort will yield little effect. Rational ignornace ensures they do not take the time to understand all the candidates, and the broad platforms they offer because of the small amount of significance their single vote contains. I do agree that this effect is unlikely to change. However, I think you underestimate the degree through which the harm from this is mitigated. As voters can vote for specialized platforms, they do not need to consider broad platforms in the firstplace, but discover parties that look to address the area of their life for which they are already familiar and motivated to improve upon intrinsically. A party that gets a single legislator can pursue their needs by finding a small 20% sympathetic coaltion (significantly easier than a 50% majority), and so the significance supporting such a party increases dramatically. While I do not expect citizens to suddenly become more well informed broadly, I think this system aligns with what citizens already do naturally. > But of course that benefit already exists in the many arenas that escape public scrutiny, which brings us back to Terry's more core concerns about whether electoral bodies' interests genuinely align with the public's. Citizen interests are obscured by the broad platforms demanded by majoritarian governments. Every party *must* have a full, complete, broad platform such that if they were elected and were in charge of all laws, how they would approach all laws is at least somewhat understood. This limits the types of parties we can have, as coalition parties become neessary as we squeeze issues together. Thus, when a party is elected, the specific issues individual citizens are facing are unknown. If we remove that dynamic, citizens are free to vote for specialized parties, particularly when they feel those interests are being neglected. If there is an issue that parliament has not pushed forward legislation, demand for party that will increases, and citizens can push to get to get seats for the party. These are dynamics simply not possible in majoritarianism. I think the freeing nature of removing majoritarianism is still new and being explored, and I do not think the dynamics bear any resemblance to majoritarian elections, for which I agree with you entirely that all of those problems laid out in the book are of great significance.

Ethan:> fluid majoritarianism . . . will not produce generative deliberation that deeply reflects the perspectival diversity of the populace. Although there are different varieties of electoral proposals I don't think any of us would advocate fluid (issue by issue?) majoritarianism. Depending on the threshold there will just be a broader range of (minority) proposals. Majoritarianism only comes in at the Policy Jury stage. I think it's better to deal with concrete examples rather than generalities. A 26% threshold would have generated three alternatives in the Brexit case. Alex has suggested a similar outcome in his alternative to the French CCC. If three is not diverse enough, then drop the proposal threshold, but this then makes it harder for the policy jury to decide between them. For a concise rendering of this principle see https://www.academia.edu/44790587/Some_Problems_of_Citizens_Assemblies > it will certainly not produce what Landemore calls "civic love." Yes that's true. Those of us who look to Dahl rather than Rawls or Habermas for inspiration are more modest in our goals -- i.e. that democratic systems should reflect the informed preferences of the citizen body.

From a purely pragmatic perspective we should acknowledge that advocating binding decision making by large policy juries is swimming against the tide of current developments in citizens' assemblies and deliberative democracy. So it might be wise for all of us to focus on what we agree on, rather than appealing to North Stars, civic love and other desiderata. There is unlikely to be incremental "peeling" if the project appears to be a utopian dream.

> I'm actually quite literal in my rejection of electoral majoritarianism OK, I clearly misunderstood the proposal! Thanks for setting me straight. I see the idea's appeal and will digest more, but what immediately strikes me is the contrast between civic assemblies' consistent capacity to reach supermajoritarian agreement and the notion that electoral bodies need to be relieved of even majority thresholds to generate healthy proposals. Do I understand correctly that there is an argument that removing the majority constraint will incentivize win-win solutions? How? Is it all in the feedback mechanism from the decision jury? My frame is that the best outcomes emerge from healthy deliberation that seriously engages with perspectives that deeply represent those population. While I can see this system reducing some barriers to achieving that end from current electoral systems, it's hard to see them approaching the quality of civic assemblies.

> rather than appealing to North Stars, civic love and other desiderata Despite the fluffy sounding name, civic love has the very near-term potential for measurement and operationalization and should absolutely be a system design consideration. The role that current political infrastructure plays in generating disillusionment and antipathy and polarization and other forms of civic disgust is immediately and concretely deleterious. We have to treat all of that, and not just the legislation, as a critical system output and thus a key design factor. For example, I understand that there are a mix of different decision jury designs on the table. Any system that involves no group deliberation by sortition bodies and therefore sheds some of the most civically rewarding dynamics available is missing a beat by treating a highly valuable outcome as an externality. You're not going to get the civic love if you don't have group interactions, and it's entirely appropriate to consider that outcome a design cost. Furthermore, whether by that name or others, civic love already does and will continue to play a role in the adoption of civic assemblies, the broader acceptance of which is critical to any of the plans we favor. Re the north star, it's certainly valid that some will perceive Terry's north star design as a utopian dream. While we're talking perception, many others are so disillusioned with electoral systems that they will disengage with reform unless the notion of a sortition-only system is on the table. Again, we don't need to share a north star, but it's important to understand that Terry's is a long-term vision that is already inspiring many to action: NOT burning it all down, but integrating sortition wherever possible. In other words, Terry's north star is helping open the door to lots of other incremental solutions. In the meantime, I'm happy to continue the argument about whether it's utopian; obviously I don't think it is. At any rate, the sortition movement's best card is that, while we probably won't ever agree completely on a north star, we have a great shot at coming together around sortition-based solutions that determine what we try next. I emphasize the public's right to choose its form of government more strongly than I advocate for any given form. In that context, I can agree that anti-majoritarian designs will likely merit experimentation without believing they will generate either the direct (policy) or indirect (social) outputs that Terry's system would.

>Do I understand correctly that there is an argument that removing the majority constraint will incentivize win-win solutions? If the proposal threshold is 26% there is a high likelihood that the median proposal will win out. >How? Is it all in the feedback mechanism from the decision jury? Yes -- see Harrington's parable of the two girls dividing a cake (cited above). As for the deliberative alternative that you prefer, the difficulty is ensuring that the deliberations of the CA reliably reflect the target population ("impartial" moderation is not possible if the assembly has binding decision power).

>You're not going to get the civic love if you don't have group interactions, and it's entirely appropriate to consider that outcome a design cost. Small-group deliberation is OK iff it doesn't affect the representativity of the minipublic vis-a-vis the target population. That's an extremely demanding constraint that Kai is currently trying to operatlonalise. And, speaking as a child of the '60s, I now think it's better to avoid fluffy terminology.

> Do I understand correctly that there is an argument that removing the majority constraint will incentivize win-win solutions? How? Is it all in the feedback mechanism from the decision jury? The way Keith and I address this differ greatly. Keith suggests that the lowering of the threshold allows for competing legislation, with the median one winning. I take a very different approach. In my situation, legislation is freely proposed, regardless of whether there are alternative legislation to pick from. In my system, the sortition body (for which I have not yet described the structure, stay tuned for my next paper) will deliberate and produce a decision on the legislation. My argument on why it would produce win-win legislation works on various feedback-loops inherent in the basic structure. In majoritarianism, legislation is decided by number of seats in parliament, and nothing else. If you have the seats, you write the legislation. Deliberation is not necessary. In my system, it is not decided by seats, but by trial. The jury deliberates. Regardless of the structure (it isn't a simple majority rules body), the jury will deliberate and contain multiple perspectives, which the legislators will need to antisipate. It is in their benefit to deliberate, in some fashion, amongst the various groups in parliament who have differing perspectives (they will be more specialized parties, as I explained above) in order to better understand the various perspectives they will be encountering in trial. If they wish to better there odds of winning a trial, they will have the ability to initiate other sorition tools, like a citizens assembly, to make reccomendations, so that they can see what sorts of decisions varied groups would produce first. So, there is an incentive to create more win-win proposals in the first place, to ensure they are able to present it to a deliberative body. But even after that, imagine legislation is approved that some group is unhappy with. They are free to propose new legislation to alter it in some way, just as freely as the first group were able to propose the legislation in the first place. This fulfills Philip Pettit's notion of contestatory democracy, in that even after passing, legislation may be contested. No matter what, the views of the many need to be taken into account and addressed. Parties cannot stop one another, and thus, must work with one another.

Elections are among the worst possible ways to pick people for setting an agenda for issues needing legislation (regardless of whether majoritarian, or proportional to votes). Keith argues that parties and elections ensure that "the choices available [will] better match the ideological diversity of the citizenry." Most citizens don't have a considered ideology. They may have a superficial ideological overlay from friends, family, a charismatic personality, or social media feeds. But if they are serving on mini-public, where their decisions can actually matter (unlike the rational ignorance that are the underpinning of mass elections), people can learn, assess, and develop a considered ideology or mixture of preferences. While my criticisms of elections and parties in the book deal with EXISTING parties that seek and win POWER (rather than merely the right to propose policies to juries), some of that criticism may be less relevant to Keith's proposed alternate reality. However, parties would still be dominated by special interests and big money, while crucial agenda items that lack the emotionality , fear and hatred, would still be ignored. Also, if politicians were no longer seeking and gaining POWER, I suspect voter participation would drop to single digit percentages (as it currently does in many US municipal elections already). Instead, I embrace the Greek democracy principle of ho boulomenos - (from any who wish). Throw the gates wide and let any and all citizens who wish (I propose working in small interest panels) make proposals to a lottery selected Review Panel that uses stratified sampling so as to be more representative of the population than any elected legislature or political party in history. Such Review Panels would craft a final draft bill to offer to a large fully representative Policy Jury for adoption or rejection.

As to the notion of aiming for an ultimate goal of democracy (my North Star), let me say just a bit about strategy. I refer to my step-by-step strategy as "peeling" -- peeling one policy domain at a time away from elected politicians and transferring it to a sortition-based law-making system. For example, following a bribery scandal with a developer and spot-zoning, a city council might agree (to save face) to no longer make zoning law changes, and transfer that authority to a jury process, that would vote to adopt or reject changes proposed by the planning department. The Planning department staff might then decide to let a mini-public deliberate about goals and values, before drafting a zoning amendment. Another mini-public might be empowered to decide what land-use policies need to be re-examined (a sort of agenda body). Nothing earth-shattering, but an evolution towards multi-body sortition that can then be replicated for other policy domains and other cities -- all headed towards the North Star. My strategy is to NOT incorporate any election process in the new design, but rather to evolve a parallel system that is free of election campaigning and politicians. A useful analogy is what happened to many of the monarchies of Europe. They steadily lost power with an elected oligarchy taking on those powers. It wasn't all at once, and the monarchs often persist. I imagine the day when, even if Congress still exists, they are reduced to naming Post Offices after famous people from their state, while meaningful law-making is done by a sortition process.

The most common reactions I receive from laypeople in response to sortition ideas are that people will not trust the legitimacy of randomly selected citizens, and that ordinary citizens lack the expertise for governance. I think that these objections are addressable with the right design, but Terry's approach exacerbates both of them. I view the legitimacy of a randomly selected body in a large jurisdiction as being primarily derived from accurate statistical representation of the preferences of the general citizenry. This legitimacy is not present for a small body, or one in which the dynamics of service distort representativeness. The fact that "Agenda Councils and Review Panels each gather over 100 deliberators – quite large by mini-public standards" is not very reassuring when the margin of error is nearly 10 points in each direction with n=100, assuming perfect sampling and participation rates. As I argue in my preprint (http://doi.org/10.2139/ssrn.5493247, which I am currently revising with a more robust design and justification), even a representative sample does not guarantee representative outcomes. Long terms of service, as in the agenda councils and review panels, are likely to facilitate factionalism, inequality, insularity, and corruption over time, jeopardizing the legitimacy of such standing bodies. Moreover, the generative tasks delegated to sortition bodies such as review panels in Terry's system invite more competence concerns and raise the risk of dependence on experts and staff. The backstop of policy juries is designed to ensure democratic legitimacy, but as I argue in my preprint, this yes-or-no body creates an oppositional dynamic not unlike that Terry himself critiques in "Why Hybrid Bicameralism Is Not Right for Sortition." A body that derives its powers entirely through vetoing is likely to create gridlock and delegitimization. A system in which most of the bodies face severe legitimacy deficits cannot easily be rescued by one legitimate yes-or-no body at the end. An alternative which Keith, Jordan, Alex, and I advocate is for the final decision-making body to choose among multiple competing options, giving it real volition and preventing gridlock while keeping its short-term and evaluative nature. Instead of giving the only (robustly) democratically legitimate body a single option, take it or leave it, it makes sense for the policy jury to be able to choose among a range of diverse options. In Terry's design, this could be accomplished by having each review panel produce several proposals, or having multiple review panels for each issue. In Keith's design and mine, this role is accomplished by having elected legislators produce a range of proposals through some thresholding or top-n system. I am not necessarily opposed in principle to Terry's system in the (far) future if the policy jury is multi-choice, although I still think elected legislators may accomplish the role of generating competing proposals more effectively, but this is a separate question, and a hybrid approach could have both elected legislators and randomly selected bodies produce competing options. Regardless, I think that using sortition bodies for standing, generative, and oppositional tasks unnecessarily exacerbates the legitimacy concerns facing random selection currently, and thus there is a much greater chance of widespread institutionalization if the more clearly legitimate use cases of sortition are prioritized, which I argue requires large, randomly selected, short-term, evaluative, multi-choice, conclusive, and informed bodies.

> Any system that involves no group deliberation by sortition bodies and therefore sheds some of the most civically rewarding dynamics available is missing a beat by treating a highly valuable outcome as an externality. You're not going to get the civic love if you don't have group interactions, and it's entirely appropriate to consider that outcome a design cost. These civically rewarding dynamics (perhaps including a reduction in partisan and affective polarization) are arguably positive, but if the side effects of deliberation undermine aspects of legitimacy, it is hard to see civic love carrying the day: representing the informed will of the people is more important than reducing polarization for a final decision. I have no normative predisposition for or against interpersonal deliberation in a decision-making body, but my data analysis that Keith mentioned, which I hope to finish this summer, may end up casting some doubt on its legitimacy. > Do I understand correctly that there is an argument that removing the majority constraint will incentivize win-win solutions? Potentially, but I wouldn't necessarily frame this as the main point of removing the majority constraint in the legislature. By allowing a range of proposals to make it out of the legislature, decision juries are given real choice between multiple competing options. Given that large, randomly selected, short-term juries are the most congruent to the people (other than perhaps referenda), this gives the most representative body more real choices, increasing the democratic legitimacy of the whole system. A consequence of this is that if the jury votes via approval voting, the most broadly popular proposals will generally prevail, incentivizing legislators to craft win-win solutions to succeed with the jury, but the main point of a multi-choice jury is more representative decisions and less gridlock. > The putatively “obvious” notion, for example, that elections should be used for agenda setting would need to account for elections’ egregious agenda-setting shortcomings to date. I don't necessarily think that elections must or obviously should be used for agenda-setting, but I think they certainly could serve as an adequate and pragmatic source of agenda-setting. Although there are many shortcomings of current electoral systems, a hybrid model of proportional agenda items and proposals from legislators and a randomly selected jury choosing among them is unlikely to suffer the same flaws. As long as some substantial proportion of legislators supports effective proposals, the jury can select them.

Thanks to Kai for his clarification of our broad agreement on a number of items. It's interesting that Terry's up/down Policy Jury model rules out win-win solutions. I do believe that if my proposed Brexit jury had three options (Leave, Remain, Renegotiate), the latter would have won by a large margin. And Kai is right to focus on perceived legitimacy. Rather than peeling, I would prefer baby steps -- reducing the parliamentary proposal threshold and introducing Policy Juries. The lay reactions to sortition proposals that Kai has experienced match my own soundings (although I would add the scepticism over the representativity of volunteers). I would also like to hear Terry's response to Urbinati's claim that the political party is the reincarnation of ho boulomenos for large multicultural states.

I should mention that I've never heard Terry use the term "civic love" or highlight the idea. Landemore centers the phrase in the year's other "Without Politicians" book. It's not fluff, even if it sounds it. It's a deliberately non-academic term, the likes of "chunking" in cognitive psychology; that's doesn't invalidate the concept. It emerges from extensive observation of civic assemblies and their impacts on participants. She defines it, explains it via examples, and provides indicators that show the durability of the effect. And it clearly echos findings described in Gastil's Hope for Democracy and other empirical research. None of Terry's designs rest on the concept, and I'm happy to use other terminology if proposed. But the more core claim is that non-policy outcomes can be valid system design metrics. Terry's focus, of course, is on the policy outcomes, and despite three attempts, a single rhetorical Pitkin question does a poor job of wiping away his considerable explications (including copious empirical research) for how deliberative mini-publics far more deeply align with public interest than elected bodies. I do respect the system dynamics in the proposed hybrid system – quite a bit more than in most hybrid systems I've seen put forward. I've enjoyed pondering it and recognizing the depth of its improvement upon current electoral systems, which was understated by my comment about fixing a few bugs. But I do see continued overstatement of the mitigation of electoral problems. Election, not partisan team victory, is the source of the ego inflation. The Kahneman system critique was primarily about citizens' engagement as voters, not about the elected. And the notion that citizens don't merit substantive representation in the issues that aren't at the front of their minds seems deeply problematic. The ideal approximated by deliberative mini-publics is approximating the way we would collectively think about things if we had the time and incentive to do so. (That Pitkin quotation is about constituents; it's atomic and makes no sense applied to the whole body). I do like that the decision jury pushes the elected body towards system two thinking. In effect the substantive representation of the ratification process improves the substantive representation of the proposers. Perhaps that's a point of agreement? Elected bodies are so poor at substantive representation that they can only be salvaged if a sortition body reigns them in? As I say, worth a try. And more likely to get a try than sortition-only systems any time soon. Mostly because in the general populace sustains an emotional attachment to elections and a story of empowerment that they do not actually offer. Formalistic legacy and symbolic fumes.

Ethan, I'm glad we are moving towards convergence -- it might even restore my faith in deliberative democracy! >[civic love] emerges from extensive observation of civic assemblies and their impacts on participants. I'm sure that's true. But this only serves to increase my concern that the citizens' assembly movement is overly focused on the internal dynamics of the deliberative exchanges (as Habermas specified). We should never lose sight of the fact that the transition to sortitional democracy would effectively disenfranchise the overwhelming majority of citizens, who don't get to experience the civic love. This is why Fishkin, Gastil etc are putting their faith in the deliberative transformation of the whole of society, which strikes me as a quixotic project. Yoram Gat has opened an Interesting debate on this on EbL: https://equalitybylot.com/2026/06/04/the-ideology-behind-the-notion-of-the-deliberative-transformation/

> The fact that "Agenda Councils and Review Panels each gather over 100 deliberators – quite large by mini-public standards" is not very reassuring when the margin of error is nearly 10 points in each direction with n=100, assuming perfect sampling and participation rates. Though it's not explicit in Terry's system, it's worth noting that in practice most assemblies that size (which have far from perfect sampling) require supermajoritarian agreement of 70-80%. For agenda councils specifically, it's nice to emphasize what Terry often refers to as self-correction and I like to refer to as representativeness over time. Elections show many of the same biases cycle after cycle; they continuously misrepresent in the same way. In an arena like agenda-setting, where breadth of scope over time has value, ensuring many different areas of civic life are attended to, there is something to be said for the long view of representativeness. Assuming capacious room for deliberation, with agenda setting the cost of important issues being ignored far outweighs the cost of trivial issues getting attention. Electoral systems have tended to fail on both counts (and also have a lot less deliberative capacity than Terry's system). I'm not seeing how the proposed designs would address those concerns. Decision juries can't nudge elected bodies towards focusing on politically treacherous topics they'd rather avoid. > a hybrid approach could have both elected legislators and randomly selected bodies produce competing options Well, that would make for some interesting data, wouldn't it? If it were possible to make it such that the juries couldn't tell which kind of body sourced which proposal, even in the presentation of arguments for and against, then you'd have a nice experimental control in terms of who could produce proposals that most impressed juries. If not, public perception of those body types would be conflated, but even so, if juries consistently favored one type over another, that would really tell us something.

> the transition to sortitional democracy would effectively disenfranchise the overwhelming majority of citizens, who don't get to experience civic love (and earlier...) > Rotation is not scalable to huge modern poleis. Take the US. In any given year we might have many thousands serving on federal review panels and many tens of thousands serving on federal policy juries. A tiny fraction of the country. The other core legislative bodies barely move the needle. Terry's book focuses on legislation, but let's broaden his proposal to include all branches. Imagine judges, department and agency heads, ambassadors, etc appointed by deliberative juries. Imagine agency policy being subjected to deliberative review (currently no deliberative body attends to such things). Now we're into the hundreds of thousands, still a fraction of a percent of the populace. Now add in state governments, which takes us into the millions. Then local governments, which takes us into the, erm... many many millions. As we descend to lower levels of government (this is obviously not the order of implementation), we have fewer participants per polity, and eventually the design must compromise quite a bit. (I have more questions about Terry's system scaling down than up.). But we have far more participants overall. The point is, rotation could in practice scale with multi-level government. No, not everyone would serve on federal bodies, but I'd go so far as to argue that the system should be calibrated such that everyone does get a healthy dosage of deliberation. A target rate might be something like: - every 10 years for reluctant participants who only take part in mandatory juries - every 3-4 years for willing participants who accept roles in optional assemblies like review panels - every year for eager participants who join interest panels. Today I am in a tiny minority in feeling that my inclusion in a democratic lottery would be a deeper enfranchisement than my registration to vote. That would likely have to change to see the adoption of Terry's system.

Ethan, It's problematic to take a couple of lines from Aristotle's observations on a small ancient polis and turn them into universal democratic slogans. Classical Athens was a direct democracy in which the assembly was sovereign (I remember a talk by Peter Stone at my university in which he ridiculed the notion that serving on a minor magistracy involved ruling). Dahl's criterion for democratic legitimacy is that the demos (a collective entity) should both set the agenda and decide the outcome. Election (under PR), combined with a reduction in the proposal threshold ensures the former, whereas large, mandatory policy juries ensure the latter. >the system should be calibrated such that everyone does get a healthy dosage of deliberation. Your focus on individuals ignores the demos as a collective entity. That's what we mean by "democracy". It's interesting that Terry's smaller bodies require 70-80% supermajorities, whereas Kai, Jordan, Alex and myself are arguing for superminorities. Terry's proposal will lead to a serious reduction in choice for the "take it or leave it" Policy Jury (presumably on account of the pursuit of unanimity at the proposal stage), whereas Electoral Lottocrats accept that politics is agonistic in nature.

I feel somewhat in the middle of all this, as I have strong agreements and disagreements in both camps. Terry: >However, parties would still be dominated by special interests and big money, while crucial agenda items that lack the emotionality , fear and hatred, would still be ignored. I wish to go a bit deeper in my proposed system. I propose that the overall dynamics in a non-majoritarian parliament are so vastly different, that a reflexive application of current electoral dynamics are going to fail to apply in the new system. As I read your book, I agree with just about everything you say about elections, and if I had not spent the last year and a half analyzing my own system from numerous angles, I probably would be skeptical of keeping elections moving forward as well. But the very nature of elections would fundamentally change. Without the gatekeeping of single majoritarian party of coalition, it frees up the ability for a party with even a single legislator to write a bill and form a mini-coalition of 20% to get it a trial. They would be able to "shop around" for different parties that have sympathetic or aligned interests. This empowers even the smallest of parties. Combine this with proportional represntation, and the electoral strategies of parties can change entirely. Instead of going on TV with an emotionally charged narrative that rile up the masses, one could successfully campaign to individual niche communities. LGBT rights advocates could successfully spread their campaign message to LGBT communities, and would be able to fight for their rights, both by proposing legislation and acting as a prosecutor in trials for legislation that might harm the community. Someone could run entirely on rural rights, healthcare advocacy, religious freedom, or climate change. While we all agree that citizens are not capable, for entirely rational reasons, of having broadly formed and considered opinions on a large number of issues, they are fully capable of recognizing the few needs that are important to them. Because of the significantly lowered bar of entry and success in parliament, it becomes easier for new citizens who recognize gaps in the agenda to form a new party and find niche groups of citizens who recognize the gap and will put them into parliament, allowing them to fill it. This sort of electoral strategies do not require the massive, national level campaigning that is required of current elections. Since eletoral domination is not how legislation is decided, massive, nationwide platforms that are created to trigger the emotions of the masses to garner broad support won't actually do much to help them succeed. In fact, such a party will find themselves struggling once they take office, as all their bills still need to survive the scrutiny of a trial by a jury of considered and deliberative citizens. Parties, instead, will narrow into particular needs. Campaigning becomes significantly de-emphasized and localized. > Also, if politicians were no longer seeking and gaining POWER, I suspect voter participation would drop to single digit percentages (as it currently does in many US municipal elections already). This is an interesting prediction, and one I cannot argue at the moment. This may, in fact, be the case. If so, this would also continue to lower the bar of entry into parliament and make it even easier for smaller parties of niche interests to make it in, which works to help, rather than hurt, my argument above. I will note that my system also encourages the usage of other sortition bodies which I haven't yet outlined publicly yet. Without going into detail I'll provide the motivation for it. Since parliament needs to produce legislation that will sucessfully get through a deliberative and considered diverse jury of citizens, it is their benefit to take reccomendations from a deliberative and considered diverse body of citizens. Legislative ideas proposed by the latter are more likely to succeed in the former. This creates a strong incentive for legislators to make use of such bodies. I'll stop here, but I hope that it's clear the incentive structures and dynamics from current systems do not translate terribly well to this new system. I believe it is worthy to consider it seprately.

> Your focus on individuals ignores the demos as a collective entity My focus is on the civic health of the demos, which includes but is not limited to its governance. I'm moved to engage in the matter by the experience of living in a demos whose civic health is in steep decline. It would broaden the conversation too much to enumerate and weigh sources of social cohesion and division and their changes in recent years, but while it's not the only driver, I think it's fair to say that electoral politics has been a deeply divisive force. The electoral lottocratic design could be considerably less divisive than the current one and should be credited for that, but I suspect it will be less cohesive than Terry's design. > Electoral Lottocrats accept that politics is agonistic in nature I can't speak for Terry, but that's be a fair distinction to draw with me. I don't think politics can or should escape conflict by any means, but I don't accept that politics is agonistic in nature, except in the colloquial sense in which "politics" means "electoral politics." And at a time when many of our cohering forces are diminished, we, as a demos, need the coherence that deliberative mini-publics can foster.

>My focus is on the civic health of the demos, which includes but is not limited to its governance. I'm moved to engage in the matter by the experience of living in a demos whose civic health is in steep decline. Political theorists (and sortition advocates) need to be scrupulously neutral in these matters. especially when it comes to disagreements between communitarians and liberals (with or without the prefix). As for agonism, I'm of the view that the origin of electoral politics is the observation that counting right arms is better than letting the swords determine the outcome.

I would like to note that while Keith and I have similar design structure, we do diverge, sometimes strongly, in some of our justifications and theories that lead to some different design choices. I mention this because I don't view politics as being inherently agonistic. While I use an agonistic trial, this doesn't reflect, in my eyes, the true nature of politics. Agonistic trials are a great tool for scrutinizing legislation, bringing to the surface potentially hidden issues and flaws, and allowing the facts to rise to the surface. It provides opportunity for testiomony from various citizens and experts, and provides a fair way to bring diverse arguments as each side can bring their own. However, my fundamental viewpoint in how the jury should process this information is through mutual cooperation, deliberation, and making spaces for one another. I view the purest form of pluralistic democracy as one of a shared space, in which people acknowledge the needs of one another and set aside their own preferences for the needs of others. My jury design reflects that, as I will be discussing in my next paper. I point this out to demonstrate that the minority legislative model can implement differing design choices, and the most fundamental philsophy I put into my model is a cohesive society which makes space for the many differing needs of the population.

Jordan:> Without the gatekeeping of single majoritarian party of coalition, it frees up the ability for a party with even a single legislator to write a bill and form a mini-coalition of 20% to get it a trial. Yes indeed, and the single legislator would have been chosen by her constituents, rather than just being self-selecting. Terry's volunteers would also fall victim to the Isaac Asimov exclusion principle.

> Political theorists (and sortition advocates) need to be scrupulously neutral in these matters I suspect/hope there's a misunderstanding. By poor civic health I was not casting aspersions at any political party or policy. I'm referring to deep polarization, tribalism, skepticism in institutions, political violence, receptivity to authoritarianism, etc. Surely you didn't intend to suggest such that it's out of bounds to for political theorists to see authoritarian vulnerabilities as design flaws, or for advocates to speak about concerns of tribalism as part of their pitch for system reform?

OK, but if you take a look at the comments on Fishkin's book, there's a sense that deliberative democracy leads to a move in a left-field direction. https://www.ingentaconnect.com/content/imp/jos/pre-prints/content-jjfishkin As to the issue of polarization, we would argue that this is a product of the 51% proposal threshold in the elected chamber. Win-win needs a lower threshold.

> there's a sense that deliberative democracy leads to a move in a left-field direction Ah. Yes, agree that's a critical risk. There was an article in the New York Sun this week entitled "The Newest Word From the Left — Sortition".

> Though it's not explicit in Terry's system, it's worth noting that in practice most assemblies that size (which have far from perfect sampling) require supermajoritarian agreement of 70-80%. Instead of asking one assembly of 150 people to come to a near-consensus on one proposal, I think it would be much more effective to ask five assemblies of 30 people to each come to a near-consensus on a proposal, and then have a large jury choose between the resulting five options. Because the outputs from a relatively small and skewed body are not going to be very representative, there is more value in using them to generate a diversity of options rather than giving only one option to the more democratically congruent jury. I would appreciate Terry addressing directly why it wouldn't make sense to give the policy jury multiple choices. > If not, public perception of those body types would be conflated, but even so, if juries consistently favored one type over another, that would really tell us something. Yes, and the two sources of proposals would create a greater range of options for the juries while encouraging both legislators and randomly selected bodies to up their game. > Now add in state governments, which takes us into the millions. Then local governments, which takes us into the, erm... many many millions. Having many millions of people participate in deliberative minipublics defeats the efficiency gains you get from random sampling, which would massively increase the financial cost of the overall approach (assuming participants are compensated well, which is arguably necessary for high participation rates). Moreover, having more people deliberate overall doesn't increase the legitimacy of any one decision, which depends on the representativeness of the specific body that made that decision. I find Fishkin's (2025) critique of rotation-based legitimacy quite compelling: > Whatever the advantages for the ideal of rotation (giving everyone a chance to participate on some policy at some level at some time in life), such a plan does not satisfy our criteria for popular control. The multiplication of small citizens’ assemblies or citizens’ juries has the effect of siloing the decision process. Suppose a citizens’ assembly decides a controversial issue such as access to abortion or eliminating social security or refusing to fund Medicare expansion, for residents at some level of government (national, state, or local). Perhaps it is a bad draw of the lottery that comes in very unbalanced on that issue. Or perhaps there is a sharply differential response rate from those who feel strongly. The fact that thousands of assemblies meet at different levels on different issues is not much solace to me, if I really care that rights to abortion must be protected, or that the social security system must not be drastically cut, or that the poor must have access to medical care. A simple bad draw of the lottery (or in differential responses to the invitation) could yield such an outcome and make me virtually powerless to do anything about it. It would be little recompense if I am drawn into a different lottery to decide some other issue at some other level, say electricity regulation for my local community. Different siloed “assemblies” for different topics at different levels can leave me with no say at all on whatever issues happen to really matter to me.

I agree with Kai's last comment and thank Ethan for confirming the danger that sortition advocacy will be perceived as left-wing. Terrry:> I embrace the Greek democracy principle of ho boulomenos - (from any who wish). Throw the gates wide and let any and all citizens who wish. The UK political parties (apart from Nigel Farage's Reform) are boycotting the Clacton by-election. As a result there are 34 (primarily) independent candidates. That strikes me as more akin to the Tower of Babel than ho boulomenos, and supports Urbinati's view that the political party is the reincarnation of ho boulomenos for the modern age. (It would be an interesting experiment to see what would happen if these volunteers sat down to deliberate together.) Alex Kovner is reading the posts and will be commenting next week.

Ethan is travelling this weekend, so would be good if Terry or some other defenders of multi-body sortition were to step up to the plate. At the moment sceptical voices are predominating.

Terry's system has an analogous defense against, say, substantively unrepresentative abortion policy to the electoral lottocrat design: the ratification jury. It also has higher deliberative capacity for self correction. (By the way, I think you guys you consider calling the model PEDAL, for Proposal by Electeds, Decision by Assembly by Lot. You could embrace the bike metaphor. Two tires: generation at the back, decision-making in the front) But I think it's a mistake to focus on the sexy, combative issues that have become combative issues largely because of electoral political dynamics. The PEDAL designs don't seem to address electoral systems' flaws in rewarding folks skilled at inflaming passions and directing disproportionate focus to hot-button issues. They may even inflame them further, if it is true as suggested that politicians will avoid weighing in on all issues and focus only on those with major resonance. Terry's book explores the incentives that drive politicians to direct our attention to such matters rather that focusing on the pragmatic, long-term, often boring details of responsible governance. The question isn't just what issues voters are currently thinking about (or are being induced to think about) but what citizens would think and care about if they had the time and resources and deliberative framework to do so. Most voters aren't spending tons of time thinking about housing permits or security trading safeguards or how to keep AI agents from supporting the development of pernicious viruses (algorithmic and biological), but that doesn't mean we should wait for such things to catch the public imagination to deliberate and govern them wisely. Consider the case of US national debt, which is deeply entangled with health care spending and social security, the so-called third rail of American politics. The political calculus with such challenging issues often concludes it's best to avoid them altogether, rather than risk the popular backlash from any given plan. Politicians wait for their colleagues to make proposals and then pounce at the opportunity to saddle them with electorally difficult positions to defend. And the can gets kicked. Terry's system would thrive at addressing such difficult issues, because deliberative mini-publics would recognize the critical importance of doing so, and no political calculus would prevent serious deliberation. PEDAL systems would be more likely to replay the current dynamics; there's will often be more political reward in watching proposals fail than in fighting for successful ones. Or, more to the point, focusing on flashier issues that catch the electorate's eye. There seems to be a hope that in the PEDAL system politicians would stop acting like politicians. But the system would still reward the same sorts: the attention-getters, the horsetraders, the machiavels. They would still build their parties and use their electoral weight to bind them, using the many tools still at their disposal to build tribal loyalty and discourage breaking of ranks. They would still attract power-seekers with an eye to post-political careers. They'd still be politicians. The many strengths of the PEDAL system over the current one all come from the use of sortition. Lowering the agreement threshold doesn't inherently encourage win-win seeking, quite the opposite. But the sortition jury improves the incentive structure enough to give win-win a chance. Win-win has much MORE of a chance if you get rid of the system contaminant of elections. And you dramatically increase your chances of tackling hard problems without clear wins head on. > Instead of asking one assembly of 150 people to come to a near-consensus on one proposal, I think it would be much more effective to ask five assemblies of 30 people to each come to a near-consensus on a proposal. There is a bit of that dynamic at play in most large civic assemblies' methodologies, which involve lots of small-group deliberations. They then hone and merge separate proposals with a group deliberative dynamic rather than submitting proposals independently to a jury to pick one, which would seem more explicitly aimed at win-win. I personally don't have any deep opposition to the production of multiple proposals, but I think there's a lot of reward from consensus seeking, which civic assemblies have proven very good at. I once asked a practitioner about lowering the consensus threshold, and his concern was that if you set the bar below 70%, the group would produce *too many proposals*. Another point is that Terry's design heavily emphasizes the trade-offs of different sortition designs. Just one of many is that pure sortition delivers improved representation but requires mandatory participation, which entails all the costs of unwilling participants. So his mandatory bodies are short-term, narrow in responsibility, and make only individual assessments. My understanding is that the various PEDAL designs on the table here vary in the design of the decision jury, but all have elements that increase the demands of mandatory service. Considering multiple proposals, group deliberation in some cases, returning for further service when proposals are sent back.... all these are high asks for mandatory juries that I believe Terry was seeking to avoid.

Thanks for the thoughtful post Ethan. Nice acronym! (I prefer a two-word solution (e.g. Jordan's Electoral Lottocracy or Kai's Reflective Democracy.) >Consider the case of US national debt, which is deeply entangled with health care spending and social security, the so-called third rail of American politics. The political calculus with such challenging issues often concludes it's best to avoid them altogether, rather than risk the popular backlash from any given plan. Yes, that's very true, but I think that's more likely the result of democracy per se, rather than just the electoral variant (Athens was certainly not immune). Challenging issues will still need to be addressed, and fiscal policy inevitably involves losers as well as winners. I believe it the case that some 80% of UK government borrowing goes on paying existing debt interest. My own preference is for a mixed constitution that would include statutory requirements for balanced budgets over the medium term. I can't imagine this being popular under any variant of democracy. Other constitutional constraints would be required to ensure minority rights and stable governance.

> I think that's more likely the result of democracy per se, rather than just the electoral variant Here's why I'm willing to invest some hope in the idea that multi-body sortition might succeed in addressing something like Social Security reform: First, it separates agenda setting from proposal drafting. "Third-rail" issues illustrate the value of this separation. Nobody in leadership today wants to put Social Security on the agenda (beyond vague electoral promises), because they know the issue is fraught with electoral danger. Even in a pure sortition system, a single legislative chamber might prefer to deliberate about easier topics. If the agenda setting body were a separate sortition body, as in Terry's system, their only responsibility would be to ensure that important issues were attended to. (Note that this separation would not work in a fully electoral system, because if the agenda bodies would conspire or sabotage depending upon the partisan make-up of the proposal body). So, these bodies would be very likely put thorny-but-important issues like Social Security on the agenda. Second, the proposal body is not burdened by electoral dynamics. Nobody is fulfilling promises or eyeing opportunities for advantage in the next election as they do today and would in the EL system. The less sullied emphasis on problem-solving increases the chances of finding solutions. It won't always be possible to find solutions without losers (though win-win will be sought) – Social Security embodies this dynamic. But civic assemblies are far better than the public square at carefully weighing alternatives, maximizing benefits, and minimizing pain. (I would love to see this claim about Social Security solutions tested with an advisory civic assembly. Even in the current unhealthy political climate, I would not bet against supermajority proposals). Finally, to celebrate the common strength of the MBS and EL models, the ratification process would similarly ensure that the final decision rests on informed public judgment. (In the EL system, proposers will still always have one eye on public opinion. In the MBS system, they are concerned almost exclusively with outcomes.)

Ethan: > Consider the case of US national debt, which is deeply entangled with health care spending and social security, the so-called third rail of American politics. The political calculus with such challenging issues often concludes it's best to avoid them altogether, rather than risk the popular backlash from any given plan. Politicians wait for their colleagues to make proposals and then pounce at the opportunity to saddle them with electorally difficult positions to defend. And the can gets kicked. So, let's examine this in the current system, and then in an Electoral Lottocracy, and you can see why I would believe in a different outcome. In current systems, parties compete in elections, and only one party (or coalition) can win. During their victory, they know they have to repeat this again in four years and will have to demonstrate "what they've done" to convince the population to vote for them again. Large issues, like healthcare spending and social security, are issues that affect people in the long term, would represent a major disruption to the economy (a potential short term problem) in order to, hopefully, in the long term, provide people with more stable healthcare. To achieve this within four years, with the potential for failure likely to result in loss of power entirely, makes taking on this task highly undesriable. It's far easier to take on tasks that require little of you, but can be used to campaign on. Comparatively, in an electoral lottocracy government, citizens who want single-payer healthcare can vote for the single-payer healthcare party. This party's entire electoral mandate is basically limited to attempting to progress single-payer healthcare. So long as they have any seats at all, they have the power to pursue it and aren't in fear of losing power due to the loss of a few seats. Those who voted for them, the supporters are single-payer, are voters who are more likely to stick around, regardless of the outcome of other issues. They don't need to campaign to the swing voter or the general public, but simply explain their progress to those who voted form them. If they fail to make any headway, this is not a failure they will need to explain to the general public, but to their voting base, who may create the "Universal healthcare" party to replace them with. On top of all that, the single payer healthcare party doesn't need to negotiate the healthcare plan with other parties to get them on board, but they do need to explain the plan and defend it to a sortition jury. I'm not saying politicians will stop acting like politicians, I'm saying that politicians will continue to do so. They are going to follow incentive structures to maintain their own position and power. However, what I'm arguing is the fundamental incentive structure changes in this dynamic. So long as there exists a percentage of the population that cares about an issue, that issue can be addressed.

Some interesting ideas from Ethan and Jordan. I still think we need to refocus on the vast majority of people who don't participate directly, and who would be looking to the Policy Jury to represent them. I'm a little sceptical that win-win solutions will emerge on the big ticket items for the reasons that Ethan outlines, hence the need for safeguards derived from the (republican) model of the mixed constitution. Bluntly speaking, I guess what I'm saying is that there's a finite pot of money, so choices have to be made, and this is a serious challenge in a pure democracy (either elective, sortive or a combination).

Yes, I agree that a PEDAL system (I know you're unlikely to adopt that term, but consider it an affectionate nickname) would significantly improve on the current system. The core incentive shift is the jury ratification, but the proportional representation and minority proposals are also key to the system dynamic. If PEDAL were on the ballot, I'd peddle it. EL is also probably also implementable in a shorter time frame than MBS. Fwiw, I would consider its deployment an example of, rather than an alternative to, Terry's peeling strategy. "Peeling" refers removing one bit of authority from elected bodies at a time. The authority can be content-based (as in putting juries in charge of homelessness policy or the like) or process-based. This is the latter; a PEDAL approach would peel the extremely important ratification authority from elected leaders, and thus would be a very powerful alteration. A further peeling step, along the lines of what Kai mentioned, would be to see sortition and elected bodies share proposal authority. I find that solution a very elegant proving ground. If elected bodies produced more successful proposals, that would justifiably slow the progression towards a sortition-only system. For what it's worth, I suspect sortition bodies would produce more successful and impactful proposals than elected ones, thanks to their superior alignment with the public interest, further legitimizing them and supporting the argument for still more peeling towards something akin to Terry's north star.

> Most voters aren't spending tons of time thinking about housing permits or security trading safeguards or how to keep AI agents from supporting the development of pernicious viruses (algorithmic and biological), but that doesn't mean we should wait for such things to catch the public imagination to deliberate and govern them wisely. I agree that most voters aren't thinking about these detailed policies, but I would venture that they are in fact thinking about the underlying agenda issues. Voters do care about affordable housing and the risks of AI. In reflective democracy, if voters care about certain issues, they can select legislators who will put those issues on the agenda. Sure, these decisions about the agenda may be somewhat manipulated and not perfectly wise, but defining what normatively constitutes a "wiser" decision is quite difficult. Most important issues have at least some proponents in the legislature, and my system will give those topics a chance to be considered. > PEDAL systems would be more likely to replay the current dynamics; there's will often be more political reward in watching proposals fail than in fighting for successful ones. My proposed system is actually designed to prevent exactly this dynamic. Because legislators create multiple competing proposals, they cannot just try to make proposals fail, they need to provide compelling alternatives of their own. Reflective juries choose among competing proposals by approval voting, so the top option will become law, preventing gridlock and ensuring something gets done. In Terry's system, there could be gridlock because the only robustly congruent body (the policy jury) has power through a veto mechanism. > They would still build their parties and use their electoral weight to bind them, using the many tools still at their disposal to build tribal loyalty and discourage breaking of ranks. Although there are many flaws of electoral systems and partisanship, I think this framing ignores the real benefits they have. Electoral competition fosters public justification and debate, boosting transparency and giving the entire citizenry (as opposed to just random samples) a more legible presentation of politics and some opportunities for participation (see https://doi.org/10.1177/14748851231220555 for a discussion of these benefits). My system of reflective democracy retains and even strengthens these advantages by having elected legislators make competing proposals and debate them publicly. These attributes are key for avoiding the blind deference that Lafont objects to, making reflective democracy much more defensible from a legitimacy standpoint. > There is a bit of that dynamic at play in most large civic assemblies' methodologies, which involve lots of small-group deliberations. They then hone and merge separate proposals with a group deliberative dynamic rather than submitting proposals independently to a jury to pick one, which would seem more explicitly aimed at win-win. I personally don't have any deep opposition to the production of multiple proposals, but I think there's a lot of reward from consensus seeking, which civic assemblies have proven very good at. I think this may be an overly rosy portrayal of citizens' assemblies. It isn't necessarily possible for a hundred people to meaningfully all contribute to one shared product, so deliberations are often siloed into much smaller groups, with only minimal consideration by the others. The dynamics of consensus-based deliberation can create group polarization and favor certain demographics. Because small, non-mandatory citizens' assemblies are not very representative, it seems quite difficult to have confidence that their output is legitimate, and I don't think a yes-or-no body on top of that fixes much. Weakly representative bodies like elected legislatures and citizens' assemblies can most robustly be trusted to make a range of proposals, which gives a more congruent body real volition. > My understanding is that the various PEDAL designs on the table here vary in the design of the decision jury, but all have elements that increase the demands of mandatory service. Considering multiple proposals, group deliberation in some cases, returning for further service when proposals are sent back.... all these are high asks for mandatory juries that I believe Terry was seeking to avoid. I don't think that my mandatory juries have prohibitively higher demands than Terry's. Considering more proposals is the main difference (my juries don't need interpersonal deliberation per se and never return for further service), but the tasks are still entirely confined to evaluating proposals made by another body, which we agree ordinary citizens are capable of. I would argue that serving on a reflective jury is no harder than serving on a trial jury considering a complex case, and would be much better compensated and more politically influential. The tasks given to ordinary citizens in my system are much more limited than those of Terry's when considering all of the bodies at play.

> First, it separates agenda setting from proposal drafting. "Third-rail" issues illustrate the value of this separation. Nobody in leadership today wants to put Social Security on the agenda (beyond vague electoral promises), because they know the issue is fraught with electoral danger. Even in a pure sortition system, a single legislative chamber might prefer to deliberate about easier topics. If the agenda setting body were a separate sortition body, as in Terry's system, their only responsibility would be to ensure that important issues were attended to. (Note that this separation would not work in a fully electoral system, because if the agenda bodies would conspire or sabotage depending upon the partisan make-up of the proposal body). My system also separates agenda setting from proposal drafting as two different steps. Because both steps are designed to proportionally represent the views of legislators, there is no "partisan make-up of the proposal body" to worry about: a range of political positions will always be represented in the proposals that go to juries. > EL is also probably also implementable in a shorter time frame than MBS. Fwiw, I would consider its deployment an example of, rather than an alternative to, Terry's peeling strategy. "Peeling" refers removing one bit of authority from elected bodies at a time. The authority can be content-based (as in putting juries in charge of homelessness policy or the like) or process-based. I'm very glad we agree that "PEDAL" systems would be an improvement over the status quo. FWIW, in my reading of Terry's peeling strategy, he is focused on content-based peeling, and explicitly argues against process-based hybrid systems, but I certainly agree with you that "PEDAL" systems could provide a good transitional approach. A hybrid of elected and allotted proposers would also be very informative as you mentioned, and I think could even be a good system in the long-term.

Ethan: >Fwiw, I would consider its deployment an example of, rather than an alternative to, Terry's peeling strategy. I agree, and if Terry's vision is, indeed, the North Star, is provides a real path to get there. It would be far easier to insert a small number of representatives willing to put themselves out of work in a "anti-electoral party" and convince a jury, than it would be to convince a majoritarian government to do the same. If the merits of the system can be demosntrated in a trial, the lack of a single, monopolized party power makes such reforms far more plausible. > For what it's worth, I suspect sortition bodies would produce more successful and impactful proposals than elected ones, thanks to their superior alignment with the public interest, further legitimizing them and supporting the argument for still more peeling towards something akin to Terry's north star. Is this is true, and it very well could be, my system incorporates this as well. In my most recent substack article (https://trialsanddeliberations.substack.com/p/a-token-of-good-faith), I outline further details of the token system introduced in my first paper. These were ideas I had removed from the paper in an attempt to keep it more focused. Parties that are not yet ready to put forward bills can use their tokens to create citizens' assemblies (among other uses) in order to determine good paths forward in the creation of legislation. Since, ultimately, they legislators will need to convince a sortition jury of citizens, a citzen's assembly is a highly useful tool to determine what a sortition body would consider a good idea. Listening to those ideas increase their chance for success in a trial. If they choose to ignore them, other parties are free to take those ideas for their own bills. Ideas do not live or die based on the singular power in charge, making suggestions from citizen assemblies extremely likely to find their way through.

Kai:> Reflective juries choose among competing proposals by approval voting, so the top option will become law, preventing gridlock and ensuring something gets done. In Terry's system, there could be gridlock because the only robustly congruent body (the policy jury) has power through a veto mechanism. Would it be fair to characterise Terry's model as proposal unanimity followed by popular veto, whereas Superminority/EL/Reflective Democracy involves proposal multiplicity followed by popular choice? The irony is that the former closely follows the structural dynamics of electoral politics (with parliamentary proposal majorities replaced by sortitional majorities). The other irony is that it achieves Madison's goal of excluding the people in their collective capacity.

Keith: My model does not use popular choice in the jury. My jury model is likely to be controversial, so I'm working on creating my strongest case I can in favour of it for my next paper, but I have a fairly different design in mind than popular choice.

Jurgen Habermas's 1962 study of the historical origins of the public sphere (civil society) claims that it was the product of the deliberative conversations in the (bourgeois) coffee houses of early-modern Europe. However this was "transformed" (aka destroyed) by structural factors, including universal suffrage, welfarism, mass media and consumer culture. Following Habermas's lead, deliberative democrats (including Fishkin, Lafont and Urbinati) have called for the (effective) siloing of deliberation from sortitional bodies to (voluntary) civil society, where it can then 'influence' (and constrain) elected politicians -- provide them with cover to do 'the right thing'. Terry Bouricius's call for binding decision making by large sortitional policy juries is orthogonal to this approach, hence my claim that it is better understood as the plebeian element in the neo-republican model of a mixed constitution. Sortition and deliberative democracy need to be kept distinct, the latter being a primarily elite project. [cross-posted from Equality by Lot]

I have quite a bit of reading to catch up on, and I'm putting great effort into doing so. As such, I won't doubt your telling on the history of these arguments, or your interpretation of it, as I have insufficient confort with them to produce my own. That said, I don't think your last move falls out, necessarily, from that interpretation. If deliberation in the public sphere is an elitist project, that does not necessarily imply that deliberation amongst a sortition body is as well. Whetehr a sortition body that contains a representative sampling of the population is also elitist is not automatically true. "Elitism" doesn't have a transitive property, as far as I'm aware. Sortition bodies would not contain demographics that favour the wealthy, provided a proper random sampling, and especially not the elite by definition. Jury duty is not an elitist function. Talking and arguing (in a civil manner) is not inheriently elitist either. While you and I agree on a model in which elections and sortition occur together, I cannot deny the arguments Terry makes in his books that elections are elitist, and I don't even make the claim that this would stop being the case in my system. I doubt it would, honestly. And thus, if we are to have the elected officials act as advocates, and the jury only listen and then vote in silence, then we are guaranteeing that it is only the elite who get to speak and be heard. In my opinion, that would be a failure of our system and an appeal to elitism, not a refrain.

Jordan, Elites come in all shapes and sizes other than wealth. My concern has more to do with the culture wars and the hijacking of university and knowledge elites (I was watching a fascinating documentary on this last night: https://www.bbc.co.uk/iplayer/episode/m002v2w6/storyville-speechless-1-the-new-campus-revolution Bear in mind that the New Left abandoned the proletariat when they refused to perform their allotted historical role and shifted their focus to oppressed minorities. My claim is that a random sample has to be large and quasi-mandatory in order to be descriptively representative and this rules out deliberation (in the Habermasian sense of the word). Deliberative democrats, however are comfortable with small self-selecting groups acting in an advisory role. Such groups may, to use Bill Clinton's words, "look like America" but that doesn't mean they are representative in a substantive sense. >if we are to have the elected officials act as advocates, and the jury only listen and then vote in silence, then we are guaranteeing that it is only the elite who get to speak and be heard. Yes that's true, but a)the spokespersons are chosen by the people and b) if they are to prevail then they need to appeal to them on their own terms.

The video is not available for me. If it's also on YouTube, feel free to share in that format. Unfortunately, you may find me to be someone elistist, in the way in which you seem to view it, though I disagree with your perspective on the new left. But those are discussions for another day. University professors will not be the majority of sortition bodies, and thus will not dominate those bodies. I think there can be various mechansism to deal with the problem of large bodies and deliberation, namely, split them up into smaller bodies for the deliberative part. And there are various ways in which this can be done, as well. But there will likely be bises introduced if the only people who speak are part of the elite and the jufy are therefore only able to judge the arguments arguments produced by those elite. Deliberation from the sortition body allows for the arguments provided by the elite to be considered, reframed, and examined through a more representative group. Even a small group of 12 randomly selected deliberating will be more representative of the broader population than elected officials. If our goal is to remove elitist biases, I do not see how we can achieve that while only allowing the elite to provide the arguments.

>But there will likely be bises introduced if the only people who speak are part of the elite and the jufy are therefore only able to judge the arguments Take a look at the section of Terry's book where he describes the argument between lobbyists as one of the most informative exchanges he encountered as an elected official. And I don't know of any cases of unmoderated DD (they claim it isn't a problem as the role of DD is purely to advise in the 'public sphere' rather than make decisions. However you don't need to be a behavioural economist to understand the power of nudging.

Right, I'm not denying the value of an advocate, even one of the elite, can provide in terms of producing strong arguments. It can be quite informative. But being informed, and considered, are not the same thing. Being informed is definitely improtant to have a considered view, but it is not sufficient. Considering places weighting and context on information. The way in which this is done by the elite will have it's own biases, but those will be the only weighting and context provided to the jury. Different citizens may be able to relate to the information provided in different ways. Not in ways that dispute the information, but in ways that contextualize it or weigh it differently. A considered view that does not have an opportunity to hear the same information presented in various ways will be have considered it in a biased manner. It will still have an elitist slant.

>A considered view that does not have an opportunity to hear the same information presented in various ways will be have considered it in a biased manner. In the proposal by Alex and myself there would be (at minimum) three perspectives on offer before jurors make their judgment (more if the proposal threshold is lowered). There has been no empirical assessment of whether intra-group deliberation adds anything of significant value and (more importantly) whether the inevitable information cascades and perlocutionary imbalances will mean that the aggregate verdict no longer reliably reflects the considered judgment of the full citizen body (who don't benefit from the deliberative exchange). It should also be mentioned that Terry's Policy Juries (who don't engage in intra-group deliberation) will only benefit from a single perspective before coming to their take-it-or-leave-it decision (I don't believe his multi-body system involves any devil's advocacy).

"Number of competing arguments" is not the kinds of perspetives I'm getting at though. If all three competing arguments are all coming in from electoral representatives, there may still be some bias that comes in in the kinds of arguments they produce. Imagine legislation related to housing and rent, when all the legislators are home owners. They may be able to produce various pros and cons to each of their different proposals, but if none of them have real lived experiences in dealing with those pros and cons, there may be some emphasis on different points and an underemphasis on others that don't align with the seroiusness of each point. The jurors, who might have more experiences with *some* of the points (though not all, obvoiusly not every person who rents experiences every problem that may arise), and they can use their own experiences to evaluate the points, there is no ability for them to transfer the specific knowledge on those points to others. While the facts they are presented with may be varied, there still may be something missing in the nature of the arguments themselves. When the sortition group deliberates, they will take on the same facts, but discuss them in a new way, which will help mitigate the elitist bias initially presented. How to deal with relationship between the body and the population as a whole is important. As I'm working through my arguments, I realize I wont' be able to address that in my next paper, as the scope is too large so it'll likely be put off to the one after, but it is something that's on my mind. However, if the outcome favours the elite due to the bias in the arguments that are presented, I would question how aligned it would be with the general public, anyway (as we see great diverges from elected governments and the populations today). In regard's to Terry's policy jury, it's one of the few points of critique I would levy against his model. Like yours, I also disagree with the lack of deliberation in the final decision.

Coming into the discussion a bit late, but I'll try to catch up quickly. Which approach handles tough decisions better? The example of the national debt has come up, an issue that roils the UK at this point more than the US. I think we need to break this down a bit further, however. What part of such an issue really scares political actors? (in which I include anyone to the extent that they are involved in political decision making, whether they are professionals or not) Does introducing the issue as important and in need of attention cause a problem? No, politicians and activists do it all the time; as long as they can maintain sufficient generality they do not expose themselves to attack. Is it the final vote that causes an issue? Yes--but only if the final vote is public, and only if the political actor has a future stake in whatever change in their reputation may ensue. For this reason, Keith and I have final decisions made by single use juries on the basis of a secret ballot. Lastly, what about making concrete proposals? Here, alas, is where the real problem lies. Making a concrete, public proposal exposes one to the most vitriolic responses. Even worse, these responses are asymmetrical, because critics rarely field concrete proposals of their own. An enormous part of current politics revolves around getting one's opponents to make concrete statements while making none of one's own. EL sees this as a structural issue, whereas MBS sees it as an ethical and interpersonal one. The whole point of Superminority is that the legislature is forced to do the hard part. Proposing that something should be done to deal with the national debt is easy, the hard part making a concrete proposal. In the current system, it becomes a third rail because the governing party is deathly afraid of exposing themselves to the opposition; the opposition, by contrast, wants to pretend it has a real proposal in order to bash the governing party, but once the tables are turned, they run and hide like everyone else. Superminority works by forcing the parties to come up with concrete proposals at the same time, so that no party is structurally disadvantaged. MBS, by contrast, assumes that if we have the 'good people' meet in a room to find a policy, they will swim upstream in a way that the 'bad people' refused to do. This is an emotionally appealing view, but why are we asking anyone to swim upstream? The point of structural reform is that the entire process should flow downstream; that is, each step along the way should be designed so that it is advantageous for all parties to behave in a way that advances the process. My own view is that MBS just can't do that. Fundamentally it assumes that choosing people at random will make an assembly more likely to eat raw brussel sprouts than choosing them by election. I don't think that's the case. Any "noisy" assembly (i.e. one with public discussion and publicly recorded voting) puts people's reputation on the line. In the age of social media the population of people who will wish to brave the onslaught will be nothing like the general public. Are you going to make the deliberation secret instead? Then the process will be less transparent than Superminority, because while final votes are secret with superminority, the proposals are published well in advance. Either way, any claim to representation is highly questionable.

Great to hear from Alex Kovner again (for those who don’t know him, Alex is the original author behind the Superminority principle). Key EL: Electoral Lottocracy MBS: Multi Body Sortition >An enormous part of current politics revolves around getting one's opponents to make concrete statements while making none of one's own. Yes, in the UK all political parties know that the triple lock on pensions will bankrupt the country, but no-one will propose abolishing it. Jordan:> Imagine legislation related to housing and rent, when all the legislators are home owners. They may be able to produce various pros and cons to each of their different proposals, but if none of them have real lived experiences in dealing with those pros and cons, there may be some emphasis on different points and an underemphasis on others that don't align with the seriousness of each point. Then the proposal threshold needs to be further reduced in order to generate proposals from an (ad hoc) “renters’ party”. Given that a large proportion of the Policy Jury will be renters, then the successful proposal will be the one that is broad enough to include their interests. This doesn’t require any of the elected members to be renters themselves, as their arguments need to convince the jury, rather than changing their own minds. As Alex puts it, legislators don’t need to “swim upstream” as their self-interested behaviour is dictated by structural considerations. Or, in Madison-speak, government is necessary because men aren’t angels. Remember, also, that the etymology of deliberation is the “weighing” of arguments. People aren’t stupid, and they are quite capable of making their own judgment without the need to be buttressed by groupthink and other forms of consensus-seeking peer pressure. I would also really like to hear from Terry or Ethan (his representative on planet earth) who provides the devil’s advocacy for the take-it-or-leave-it Policy Jury. In the original Athenian model, the assembly elected spokesmen to defend the status quo.

Alex and Keith: I'm a little unclear how your formulation of the model operates. Who sets the agenda by which different parties submit the proposals?

> I would also really like to hear from Terry or Ethan (his representative on planet earth) who provides the devil’s advocacy for the take-it-or-leave-it Policy Jury Heh. I can't represent Terry well, but I do support his work and am grateful that the review may bring more attention to it. And, while I'll still be on planet Earth, I'm leaving for vacation this morning, so this will be my last post for a bit. Re take-it-or-leave-it, there is quite a bit of flexibility in Terry's model about what happens after a policy rejection. He offers a framework within which any given implementation may offer different solutions. The Review Panel may be afforded an opportunity to re-submit, the Agenda Council may call for a new Review Panel, etc. Ultimately the Rules Council would frame how this is addressed. In any event, the feedback from the Policy Jury would still be part of an iterative cycle, but the jury would typically not be empowered to govern the iteration. As to whether the Policy Jury should consider multiple proposals, I personally have no strict opposition (as mentioned), nor do I think including them would depart drastically from Terry's design, though I think he embraces group deliberation incentivized towards consensus as the source of win-win solutions and may be skeptical of diminishing that incentive. Multiple proposals would invite questions about voting systems, but again those questions would presumably fall to the Rules Council. But *neither* of these are really looking at the core of Terry's design or his argument and seem to be treating his book as if it's a response to your designs, which it very much isn't. His book focuses on the failure elections to generate substantively representative bodies and in favor of the capacity of sortition to do so. There is little doubt that elections are still *perceived* to offer substantive representation, which is the core source of their legitimacy. Terry systematically dismantles all rationale for that perception, showing how and why elected officials provide dismal substantive representation. He then offers a system design to show that elections are not necessary for democratic governance. Will one book immediately eliminate the perception of elections as substantively representative? Of course not. And, to be clear, that perception matters. If you implemented a sortition-only system today, many people would not perceive it as substantively representative and would thus likely not consider it legitimate. Not good. That is the relevance of the "north star" framing; he's not arguing for immediate implementation. Other uses of sortition, like the PEDAL models espoused here, could be seen as legitimate in a shorter timeframe. But as an idealized design, Terry's work effectively lays the burden of proof on the inclusion of popular elections (I say "popular" to allow for election-by-jury for executive roles). Can popular elections actually offer superior substantive representation to sortition, or merely the perception of it? I suspect the latter, and that over time with broad adoption of sortition bodies, this perception may change. Participation rights, rotation, deliberative experience, descriptive representation, self correction, etc... these will all contribute to the legitimation of sortition bodies.

Thanks Ethan, but I'm still unclear as to who provides the advocates for the Policy Jury. Presumably the Review Panel would argue for the new proposal, so does that mean the arguments against it are left to the (losing) minority on the Panel? If so, then that doesn't count as balanced advocacy, especially as the advocates are (effectively) self selecting. >the jury would typically not be empowered to govern the iteration. Yes, that right would appear to be taken from the only representative body and arrogated to the aleatocracy.

> I'm a little unclear how your formulation of the model operates. Who sets the agenda by which different parties submit the proposals? The main way is simply to pass an agenda at the beginning of a legislative session. Think of it this way: each legislative session starts with a single agenda item, namely to set the agenda. The Superminority process then goes forward with, say, five complete agendas being sent to a jury, which decides in the normal way. In order to make the calendar work, the agenda for the next legislative session might be set by the previous legislative session so the jury can deliberate during the recess. The point is, the agenda is treated as just another proposal, with a diverse set of alternatives going to a jury, and the jury picking one without ammendment. In practice, there may be other ways to supplement the agenda setting process. Routine items like budgets will go on the agenda automatically. There may need to be a provision for emergency items, though of course emergency procedures are dangerous as they tend to become loopholes that undermine the system.

Ethan: > But *neither* of these are really looking at the core of Terry's design or his argument and seem to be treating his book as if it's a response to your designs, which it very much isn't. This is a very valid point. I think a problem with forums such as these is that there is a large agreement about many things, and I think the arguments Terry makes about elections are extremely well made, convincing, but, in these forums, are preaching to the choir. What remains to discuss is the differences, and that encourages people with similar goals to come in like wolves (myself included) to bring in their alternatives. So I do wish to step back from that, and simply say to Terry that I greatly enjoyed the book and feel he made a truly excellent case for the need to an alternative to the modern form of elective representation, and presents a wholly plausible concept that can act as a structure to be iterated upon.

>:What remains to discuss is the differences, and that encourages people with similar goals to come in like wolves (myself included) to bring in their alternatives. Ironically, I view it more as an exercise in deliberative democracy.

I see it more as a symptom of our own little echo chamber which would improve with greater variety of voices. If there were people who disagreed with the main thesis of the book, we would be arguing about that more.

It would appear that the role of the Policy Jury is to rubber stamp whatever the deliberative councils and panels send them. In Federalist No. 63, Madison proposed the "total exclusion of the people, in their collective capacity" from direct government operations and multi-body sortition would appear to achieve this goal.

The echo chamber is a real issue, but it's also nice to have a space to hash things out. I do think it's import to address the question of what exactly a "win-win" outcome is. Usually, the two "wins" refer to the participants in a negotiation. But the people are never direct participants in a negotiation. So any assessment of what is "win-win" depends on some theory of representation. The other possibility is that the organizers of CAs have some notion of an outcome they want, and are using the CA to provide rhetorical cover to call their preferred outcome "win-win". Sadly, I think that is the more common situation in CAs as they are used now, most notably in France.

Agree with Alex. What worries me is: Ethan: > Terry embraces group deliberation incentivized towards consensus as the source of win-win solutions and may be skeptical of diminishing that incentive. This cashes out as a single proposal for the Policy Jury to accept or reject. Nobody gets to choose the proposers and we still have to learn who provides the advocacy defending the status quo. It's this sort of approach that makes me think deliberative democracy is oxymoronic.

Keith and Alex: In your views, what is the main mechansim leading to the perspective of the organizers winning in current CA? Is it issues with the selection process, bias in the presenting materials, influence from the facilitators, some combination of these, or something else entirely?

>If our goal is democratic legitimacy, then all citizens should be capable of influencing the outcome Democratic legitimacy requires we focus on the 999999.99% of citizens who don't even get to participate in the deliberation, and this requires some very demanding constraints on the mandate of those who get the golden ticket. That's why Alex, Kai and I focus on a) representation of discourses via election and b) representation of popular judgment via aggregate voting in large quasi-mandatory juries. I'm afraid there is an unbridgeable gap between this and the concerns of deliberative democrats and this is why we are talking past each other. I don't think there's much point continuing this conversation if you aren't prepared to re-focus on the above. There are other journals dealing with deliberative democracy, but we are the Journal of Sortition.