proj-branchDemocracy-commentary

todo

The Councils: sortition; like a civil service; do-ocracy (like Wikipedia); encourage cross-factional deliberation

JAC: input from many branches

freedom of religion: the freedom to openly join and associate with members of any religion is also protected by 'freedom of association'. The ban on giving special treatment to religions applies both to religion in general and to any particular religions(s). However, some religions may consider some activities sacred and/or compulsory, and this does not prevent the organization from banning these same activities.

Overview

Q: what's everyone's job and composition:

A: Chairs: uphold procedures, safeguard rights, root out corruption, express the 'spirit' of the organization Board: factional ideologues who may not be good at mass-campaigning. Appoint individuals (esp the CEO), legislate Forum: popular voice, legislate; transitive proxy provides expertise Forum Speakers: popular ppl Councils: Popular voice. Less vulnerable than Forum to mass media campaigns. High Court: interpret Acts, safeguard rights JAC: diverse inputs to mix the biases of the other branches Seeds (if the Reputation Addendum is used): who is in the group

Q: Why the name 'branch democracy'? A: It's just a memorable, easy to say yet not too silly-sounding name. There are a lot of things in here that 'branch' could refer to (the emphasis on having multiple, differently-composed branches as a form of checks-and-balances (perhaps a form of mixed government, but with more emphasis on just having different mechanics of voting than on any fundamental difference of character, as if supposed in Plato's theory of types of government), branching proposal versions in the Forum, the Delegate Pyramid, transitive proxy).

Board

Q: There are indirect elections in Branch Democracy's 'hierarchical Delegate pyramid'. In the past, indirect elections have been associated with one-party systems, such as the USSR and China. Why won't indirect elections of Delegates lead to single-party dominance? A: Branch Democracy's indirect elections differs from both USSR and in China in two crucial ways: (1) in Branch Democracy, at each level, each Voter or Delegate is free to join any Constituency, or to create (with others) a new one. This aspect of these 'voluntary constituencies' creates a proportional voting system (in contrast to the majoritarian system of fixed constituencies employed by the USSR and China). (2) in Branch Democracy, there is a strict prohibition on any central control over nominations (in contrast to the both the USSR and China, where control over nominations was/is a key component of ensuring one-party rule).

Q: Instead of a complicated formula giving a limit on number of constituents based on total number of votes in the constituency, why not just say "you can use the votes of up to 5^L constituents, where L is the number of levels down from the board, with the boardmembers being L=1"? A: This is simpler but it also makes a delegate's limit dependent upon which constituency they join. The formula based on votes held is independent of that, and so is more encouraging of bottom-up organization.

Numerology

Q: Why do we often have at least 3 members being elected at once in a multiseat election? A: Single-seat elections are majoritarian. Two-seat elections strengthen two-party dominance, a degenerate case of a multi-party system. To avoid these outcomes, we choose a number of seats greater than two. Another explanation can be given by considering error-correction of fallible parts. Assume that officeholders vote the 'wrong' way with probability x. With one person making a decision, there is no protection against this; the probability of wrong outcomes will be x. With two people, there is the potential for deadlock (a tie vote, where neither officeholder is willing to back down). With three, there is no deadlock, and the wrong decision only occurs with probability x^2. For example, if each officeholder is wrong one times in ten, then a vote between three officeholders will only be wrong 1 time in 100 (assuming that the times when an officeholder is wrong are randomly (independently) distributed rather than synchronized).

Q: Why do we have 5 members for some bodies? A: Start with the number 3 as arrived at above. Now assume that we want some extra 'wiggle room' in case some officeholders don't quite represent their constituency very well; so go to the next smallest odd number after 3, which is 5.

Q: Why 9 High Court members? A: First, if this sounds like too much, remember that by default, there is no 'High Court' and the Chairs perform this function; the 'High Court' is an optional add-on to the system. We want 3 Judges elected in each election to allow for some degree of proportional representation (see other question), and we want staggered elections to provide averaging over short-term political trends, and we want an odd number of total members so that the High Court can always make a decision one way or another, and we want as few members as possible given these constraints. So we must have some number b in each election with b>=3, and the total number of members is b*s where b*s must be odd. The smallest such number is 9.

Rights

(Non)-conflicts with the bylaws

Q: I or my organization strongly objects to one or more rights in the Bylaws, what do we do? A: If you haven't yet adopted Branch Democracy, adopt a modified version of the Bylaws with the text of the objectionable right removed; if you have already adopted it, use the Amend the Bylaws procedure to remove that text

Q: What if the law of the country in which my organization resides requires us to violate some of the absolute freedoms? A: Follow the law. The bylaws explicitly say that the law of a government with jurisdiction over the organization takes precedence over the list of freedoms.

Q: Why do the bylaws allow the law to take precedence over human rights? Isn't that unethical? A: I don't want for-profit corporations to decline to use this system because they are worried that using it will get them in trouble.

Q: What if the government of the country in which my organization resides itself violates one of the absolute freedoms, but the law does not require my organization to do so? A: In this case, since the law does not compel the organization to violate the freedoms, the bylaws prevent the organization from doing so. What the government itself does is irrelevant. The organization may work for and with the government as long as it does not 'substantially and specifically assist in denying' the freedoms except as required to by law.

General wording

Q: What does 'which must be interpreted as absolute' in the list of human rights mean? A: For example, many organizations grant 'freedom of speech' but in a non-absolute way; this freedom is considered to be implicitly limited by saying you cannot shout fire in a crowded theatre, or you cannot publish certain types of things, or that the financing of speech may be subject to campaign finance rules. 'Absolute' means that, with respect to the freedoms in that list, there are no such implicit limitations. Aside from the requirement to obey the law, there are no exceptions or excuses whatsoever, however reasonable.

Q: What does 'without limitation' in the lists of rights mean? A: This means that there are other implicit rights that do not appear in the list; that is, the presence of the list cannot be interpreted to limit the set of actual rights to the set of rights in this list.

Q: Why don't you include right X, or strengthen the wording of right Y? A: I personally believe that people should have various rights not given in that list, however the very idea of embedding human rights into the bylaws of ordinary organizations (as opposed to governmental organizations or charitable organizations themselves focused on rights) is controversial, and i don't want this list to prevent many contemporary organizations from considering the use of these bylaws. Therefore i've mostly tried to stick to rights which are relatively non-controversial, with a few exceptions (see below). Furthermore, in the 'absolute rights' list, i've tried to keep the wording relatively narrow to minimize consequences that i didn't think of.

Q: Why are some human rights (eg privacy, self-determination) listed as non-absolute? A: Because i can't define them precisely.

Q: In the prohibition against assistance of denial of rights, why only forbid 'substantial and specific' assistance? A: We want to make it clear that the organization is not required to boycott any other organization which is somehow indirectly connected to bad actors, because that would prohibit from dealing with most large companies or governments, which is infeasible.

Thought rights

Q: Does the right 'to think privately' prohibit coercing a person to take a lie-detector test? A: Yes. A lie detector is a mechanism that attempts to measure some aspects of 'mental state' using modes not available to humans without technology.

Q: Does the right 'to think privately' prohibit all lie-detector tests without exception? A: It prohibits coercing a person to take a lie-dectector test. But a person could choose to take a lie-detector test at a party. To see the difference, think of boxing; if a boss were to order clerical workers to fight each other or be fired, that would violate their rights (because people have a right against involuntary or coercive violence), but someone could choose to box as a hobby or even as a professional boxer.

Communication rights

Q: "to assist another with communications, or communicate on behalf of another, regardless of the communication's content or purpose" is rather abstract, can you give an example? A: For example if you provide an email service or a web hosting service, you can't be held responsible for something illegal that your customers do with the service.

Q: "to receive any communication sent to the person or to the public" is rather abstract, what does it mean? A: This means that even if the organization chooses to punish people who disclose certain information, they can only punish the sender, not the recipient. For example, if someone emails you classified information, you can't be prevented from or punished for reading that email. For example, if there is a public website that you know contains classified information, you can't be prevented from or punished for reading that website.

Capital punishment

Q: Capital punishment is controversial, why do the bylaws take a stand on prohibiting it? A: My hypothesis is that the potential for capital punishment may be a big factor why some revolutions degenerate into witchhunts in which people falsely accuse others (eg the French revolution, eg the Cultural Revolution in China, eg the Great Purge in the USSR); my theory is that people are much more likely to falsely accuse potential political adversaries when they feel that their own lives are at risk if their faction should lose political power. This dynamic leads to democide. Therefore i feel that will be a great improvement in political systems if all participants can be credibly assured that they do not have to fear for their physical well-being, or that of their family, even if their political adversaries take power. Regardless of whether or not true criminals deserve to be put to death, i feel that the considerations of making democide less likely during times of political upheaval outweigh the value of giving criminals what they deserve. (However, i could be wrong; it's possible that allowing the leaders of losing factions to remain alive leads to more instability in the longer term)

Q: Torture is controversial, why do the bylaws take a stand on its prohibition? A: Similar considerations justify a blanket ban on torture, even for interrogation. Since during democide the victims are often portrayed as immediately dangerous violent enemies, the prohibition on capital punishment and torture must be absolute in order to have the desired effect.

Religion

Q: How can this system be used by a religious organization, when the Bylaws state that no religion can be given special treatment? A: A religious organization would have to Amend the Bylaws to remove that bylaw (or simply to initially adopt Branch Democracy, minus that bylaw).

misc


Old/to-redo

General

Q: Why two legislative bodies, that is, why both the Board and the Forum? A: Any group decision-making method is prone to its own sort of biases and gaming. The Forum is prone to mass advertising campaigns, to soundbites over substance, and to demagoguery, and the Board is prone to smoke filled backroom deals. By including multiple legislative bodies with very different types of biases, there is more chance of defeating each particular bias. The strength of the Forum is that it gives power directly to the people to decide issues they care about, and the strength of the Board is that it allows a small number of people, who do not have to do a mass campaign and so do not have to be good at soundbites, to deeply discuss the issues and to make unpopular decisions.

Q: Why does the Board appoint CEOs, Judges, etc, and the Forum has no say in these appointments? A: We don't want appointees to have to be good at mass campaigning.

Q: What are the "branches of government" in AC and what do they do? A: Legislative/executive (the Forum, the Board, and the CEOs), the Chairs, and Judicial. The Chairs protect the rights of the members and are keepers of the spirit of the organization; the legislative/executive creates rules, and sets and executes policy, and the Judicial branch interprets, sanitizes, and debugs the rules.

Q: What are the "organs of government" in AC and what do they do? A: The Chairs, the Forum, the Board, the Executive, the High Court, the Juries. The Chairs protect the rights of the members and are keepers of the spirit of the organization. The Forum allows voters to directly vote on rules. The Board is a small group representing all factions which appoints officials and votes on rules. The Executive is chosen by the Board and sets and executes policy. The High Court resolves disputes between organs of government or high officials, and suggests clarifying or sanitizing changes to the rules. The Juries, along with Judges, resolve other disputes and filter and improve proposed changes to the rules.

Q: Who are the High Officials in AC, and how many are there? A: There are up to 15: 3 Chairs, 2 Executive Officers, 3 High Court Judges, 7 Board members. However, the High Court is optional, and in a non-governmental organization one person may hold multiple offices, so in theory in a non-governmental organization one person could be all of these.



Branches

The Chairs

Q: Why so few Chairs and Judges? A: To enhance their prestige.

Q: Why not just a single Chair -- that would enhance hir prestige even more? A: So that the Chairs can investigate each other, and also so that anyone attempting to buy the Chairs' silence must buy three people instead of one.

Q: Why not two Chairs? A: We want an odd number.

Q: Why are all three Chairs elected at once, rather than having one elected each electoral cycle and having them serve three terms? A: If we elect them all at once, we can use proportional representation to ensure that large minority factions can be represented among the Chairs. If they are elected one at a time, the majority controls all of the Chairs.

Q: What's the idea behind the Chairs? A: The idea is to provide a check on the other parts of the organization. Each Chair has the ability to make sure that other parts of the organization aren't overreaching. Like the U.S. President, they are direcly elected for a set term so that they can exercise their individual conscience, and they have the power to prevent the government from overreaching, through supervision and through pardoning. Unlike the U.S. President, they only have negative powers -- the U.S. President tends to be in favor of increasing government scope, because they campaign on promises for the government to do more. The Chairs can only campaign on promises to keep the organization on a tight leash. There are three of them, elected by a proportionally representative method, so that even a minority can have a Chair to protect their interests. AC is a system that, like Westminster, separates the Head of Government from the Head of State; the CEOs represent the Head of Governments, and the Chairs represent the Head of State.

Q: Woah, those Chairs are pretty powerful. Isn't that too much power? A: Assuming that a High Court is elected, the Chairs have almost no powers that the U.S. President doesn't have, except the power to chair the Board's sessions (but without a vote on the Board; and the U.S. Vice President has a similar role in the Senate, with the addition of a vote in the case of tie). Also, the three Chairs can check each other. There are three Chairs elected by a proportionally representative method, so any faction with more than 33% can be represented by a Chair, checking the other Chairs. So, this is actually less centralization of power than the U.S. government.

Q: Why don't the Chairs also do X? A: Usually, the answer is one of (a) we want the Chairs to have only negative power, not positive power; e.g. they have the power to reduce others' power and to stop things from happening, but not the power to cause the organization to take action. and (b) we want to depoliticize the Chairs as much as possible, that is, we want to prevent the Chairs from having an impact on policy decisions. Both of these reduce the opportunities for the Chairs to become corrupt themselves, reduce the usefulness to individual factions of controlling the Chairs, and also reduce the ability of the Chairs to build a power base (because they have little to offer others who seek power).

Q: Won't the Chairs often just end up being obstructionist conspiracy theory crazies who somehow became popular? A: Yes.

Q: Won't a Chair sometimes investigate their partisan enemies, but not others in their own party? A: Yes, that's why there are three of them, elected proportionately.

Q: Isn't it possible that the Chairs, due to their direct election, will hold more of a popular mandate than the Executive Officer(s)? A: I hope so, this is by design.

Q: Why are the Chairs directly elected, rather than all of the other officials, who are either appointed by the Board or elected in the Forum with proxies? A: To give the Chairs the maximal popular mandate.


Forum

Q: Why doesn't the Forum have a mechanism to allow for rushed consideration of an urgent new proposal? A: This would be possible but it would make the Forum parliamentary procedure more complex, and we wish to keep them as simple as possible since everyone in the organization interacts with them. Instead, we say, if something has to be done urgently, let the Board do it, since either the Board or the Forum can initiate and pass Resolutions.

Q: Won't some demagogues amass power in the Forum and control a zillion proxy votes? A: Yes.

Q: Won't the Forum sometimes pass resolutions based on knee-jerk reactions? A: Yes, although the proxy system should alleviate this somewhat.


Board

Q: Why so few Board members? A: So that they can efficiently discuss with one another. Also, in a large organization, the fewer there are, the more chance they will all be considered celebrities, which would cause the public to pay more attention to political affairs.

Q: Why 7 and not some other number? A: First, we want an odd number so there is never a situation where the board is evenly balanced between 2 alternatives. Generalizing this, we want an odd prime so that there is never a situation where the board is evenly balanced between n alternatives. The first odd primes are 3,5,7,11. 3 is almost sufficient but if one of the three is not a very good representative then 33% of the voters are not adequately represented, which seems to be too much for me -- also, it means that a party must have at least 33% to be able to definitely get a seat, which seems like too much to me. Also, we want more Board seats than there are Chairs, to enhance the prestige of the Chairs. So 5 is the minimal palatable choice.

However, with 5 either a simple majority is sufficient to pass ordinary measures, or amending the bylaws is no more difficult than passing ordinary measures, or amending the bylaws requires consensus; so 7 may be better. In addition, many existing boards have 7 or more members, so people may be more comfortable with 7.

11 or greater is too unweildy for deep discussion, especially since, because of the board-coloring constraint, Courts and Juror Committees must also have as many jurors as there are Board members, plus some Judges.

So, the only good choices are 5 and 7, and 7 seems better.

Q: Seven Board members won't have the diversity of expertise necessary. A: In the U.S. Congress, the Speaker of the House and the House and Senate Majority and Minority Leaders and a few other people form a de facto small committee of the primary decisionmakers in Congress; but they are responsible only to their districts and opaque party procedures. In a large AC organization, if you took all the people on the Board, plus their constituencies, plus the constiutencies supporting those constituencies, that's on the same order as the number of people in the U.S. Congress. So we're not doing much different than what happens anyways. By formalizing the process of selecting the primary decisionmakers out of the larger body of representatives, we make these people more accountable to the voters.

The Board is a small body of people with the power to consider issues that cut across topical boundaries in a wholistic fashion. It is expected that most of the work on specialized issues will be done by Standing Committees elected in the Forum, rather than by the Board.

Q: Why do you think the Board won't just choose the same candidate the voters would directly choose, like the U.S. Electoral College? A: The Electoral College doesn't have any other powers; the Board does. This will keep them involved. Also, countries with a parliamentary system don't find their parliaments taking a passive role in the choice of a prime minister.

Q: Won't having the Board choose the Executive Officers weaken separation of powers, and let the majority faction do anything they want? A: No. Parliamentary systems seem to work okay. In addition, here, the Chairs provide a check on the Executive Officers and the Board.

Q: Why require a Delegate to be an indirect constituent of their constituency? A: We don't want a popular mass candidate (e.g. Bill Clinton) to just organize a ton of constituencies for the sole (if unofficial) purpose of voting for hir, making the delegates of those constituencies irrelevant like the electoral college. Since Bill Clinton can only be an indirect constituent of one board seat, this means they can at most do this for one board seat.

Q: Why not require a Delegate to be a direct constituent of their constituency? A: For higher level constituencies, this would mean that a Delegate would have to be delegated up from each lower level in turn. That means that each of those lower level constituencies has power over the higher level Delegate, more than other low level constituencies under the same Delegate which are peers with the ones that sent hir up. That's like when the U.S. Congressperson who sits on some committee that you care about won't answer your emails because you don't live in their state. That's not fair.

Q: Won't the Board members often just end up being party ideologues, deep if eccentric thinkers loved by their partisans and hated by everyone else? A: Yes.

Hierarchical delegation

Q: Why hierarchical delegation for the Board, instead of direct election? A: So that, in large organizations, the Board members can be people who are bad at mass campaigning, but good at discussing the issues in small groups. Also, it means that contituencies can be small enough each voter can actually sit down and speak with hir (immediate) delegate.

Q: Won't the constituencies organize around party lines? So isn't the system of hierarchical delegation for the Board kind of like just making political party nominations a formal part of government? A: Yes. By formalizing this process we make it more open, and make the delegates more accountable.

Q: Won't having hierarchical delegation result in top-down control as it did in the Soviet Union? A: No (at least, I hope not), in the Soviet Union, (a) Stalin controlled the nomination process in the constituencies, (b) the doctrine of "democratic centralism" suppressed dissent, (c) the Party Congress occasionally went into session, whereas the Central Committee and the Politburo were always active, so the Party Congress lost relevance. The constituencies' internal procedures cannot be controlled by any outside authority, intra-party dissent is encouraged by the default procedural rules for new constituencies, and the constituency delegates can continually propose and vote on changes to their super-constituency's platform.

Q: What if a constituency disregards its own parliamentary procedure, suppressing internal dissent? A: Anyone can leave their constituency. If enough people do this, it will fall below its minimum threshold.

Q: What if an operative from a rival faction signs up for a constituency to make trouble? A: A constituency can refuse or can expel its members. You are not considered a member of a constituency unless both you and the constituency says that you are.

Q: Why require a vote to initiate a recall election for Delegates, rather than just having a continuous score voting election for them? A: So that Delegates don't have to always be in campaign mode.


The executive

Q: Why does the Board choose the Executive Officers? A: So that the executive officers, along with the Board members who supported hir, can be held accountable for progress, rather than the Executive Officers blaming a recalitrant legislature and the Board blaming recalitrant Officers. Also, they will work together more efficiently this way. This sort of thing is called a "parliamentary system", in contrast to a "presidential system" where an executive officers is directly elected. The executives can be considered 'creatures of the Board' and hence part of the same 'branch of government' as the Board.

Q: What are the major differences between the method of Executive Officer election in this system compared to the Westminster system? A: In the Westminster system, I think that elections are called when a vote of confidence is lost or supply is blocked, after which there is an indefinite period during which a majority coalition must form to select a new Prime Minister. In this system, the score voting election for each Executive Officer is continuous, and the incumbent is replaced when someone else wins this election. The voters don't have to have an election when the Executive Officer is replaced. Crucially, there is no period in which the organization is without an Executive Officer, because there is no way to recall the current Executive Officer without replacing hir with a new one; but there is also no guarantee that the Executive Officer is backed by a majority coalition (for example, what if three out of five Board Delegates voted for only themeselves for Executive Officer, and the other two both voted for the same candidate; then the candidate favored by the coalition of two would win, but this candidate is backed by only 40% of the Board). This system may end up increasing the power of an incumbent Executive Officer because they might remain an incumbent even if a majority dislikes them, provided that the majority doesn't agree on another candidate.


Judicial

The High Court

Q: What if a High Court Judge goes insane? There is no impeachment mechanism. A: Yes there is. Another High Court Judge can resign, causing a reconstitution of the entire High Court.

Q: You let the Board and the Forum replace the High Court at will? Won't that compromise judicial independence? A: It has a >=2/3s voting threshold, so it's unlikely to occur often. If they have that they could almost amend the Bylaws, except that amendments also have a mandatory waiting period.

Q: Why give the High Court the powers of Found Rights and Clarification? Doesn't that politicize them? A: In the example of the U.S., we see that the way we actually run things has diverged from the text of the constitution, as the Supreme Court reinterprets the constitution with penumbras of privacy and the like. This means that you have to have a bunch of lawyers reading a bunch of precedents to know what's going on, instead of simply reading the constitution. In AC, there is no precedent, but the High Court fufills the same function in a more straightforward manner, via Found Rights and Clarifications. It just empowers the Court to do openly what it would have done anyway, in a way that makes the legal code easier to read rather than more difficult.

Q: Why is the High Court reconstituted when the Chairs say so, although its composition is determined by the Board; why not allow the Board, or the Board and the Forum, to reconstitute? A: To ensure their independence; if the Board could both choose and reconstitute the High Court, then it would be a creature of the Board; if the Forum was also required to reconstitute, that might be okay, but it might not because the High Court's chance of being reconstituted would be affected by popular opinion, unfiltered by representation.

Courts

Q: What's the idea behind board-coloring on juries? A: It makes the juries' decisions more representative, but the primary reason is actually that it provides a process by which partisans of all factions sit down with each other and try to come to consensus on issues which are occasionally politically charged. This should reduce political polarization.

Q: Why Juror Scores? A: To incentivize people to compromise, and to stack the appeals courts and juror committees and delegates with people who know how to compromise.

Q: Won't jurors just agree with the rest even when they would disagree, just to get points, and then become delegates or judges or appeals jurors? A: Yes. The same thing happens in traditional compulsory juries who can't get off the jury until they agree, though. Also, we have censorship courts who look over the juror transcripts for candidates to see if they did this; futhermore, the jurors on the losing side have the power to determine that another juror does not deserve juror points for non-participation.

Q: Since jurors are volunteers, doesn't this let one faction try to stack the juries by covertly paying their operatives to sign up as jurors? A: No, because of the board-coloring restriction. A faction could still instruct operatives to join other factions' constituencies, so as to stack the juries, but this causes the faction to give away representation on the Board to the same degree.

Q: Won't the jury committees often just end up being ordinary citizens who are busybodies who choose to spend time on politics? A: Yes.

Q: Why separate Juror Effort Scores from Juror Scores? A: todo So that everyone can afford to pay for initiating prosecution with their Juror Effort points, even if they are a stubborn person whose Juror Score points go deeply negative because they refuse to compromise.

Q: Why have judicial nominations vetted by two courts? A: The censorship court ensures that only people who appear to be 'of good character' become judges. The court in the topic ensures that only people who are experts in the jurisprudence of that topic become judges.

Q: What if the Judges in some area decide to interpret the rules in a certain nonsensical way, and refuse to change even if the rules are changes, and deny any judicial candidates who disagree with them? In this fashion, can't they gain absolute power within their topic? A: No, because they can be removed by the High Court and the Chairs. After all the crazy ones are removed, there is no one to block new judicial candidates anymore.

Q: Why elect judges three at a time? A: So that we can use our proportional voting procedure to ensure that minority factions are proportionally represented in the judiciary.

Q: Why confirm all three winning judges by 2/3s? The score voting procedure has already ensured proportionality. A: This procedure gives any faction with more than 1/3 power the ability to veto any judge (while retaining proportionality). This keeps out strongly ideological judges beloved by their faction and disliked by everyone else.

Q: Why do Juror committees, when reviewing legislation, still have to act by consensus minus two (80%, since there are 7 jurors and three judges)? A: This is by design; anyone on the committee can make suggestions to the legislation's originator, but in order to block the legislation or apply a Report, rough consensus is required. In case no agreement can be reached, the 'default' is for the legislation to go to the Forum as the originator wishes. Members of the Juror committee who have more controversial ideas for amending the proposal may offer these as an amendment in the Forum.

Q: Why no strict evidence rules? Why no voir dire? A: Apparently 'jury control', e.g. evidence rules, carefully choosing the jury, jury instructions etc, may be responsible for a good deal of the expense of jury trials in the U.S., because they make the trial longer ( http://digitalcommons.law.yale.edu/fss_papers/534/ and http://www.law.yale.edu/documents/pdf/Faculty/Langbein_Mixed_Court_and_Jury_Court.pdf ). I also feel that evidence rules can lead to miscarriages of justice in cases in which one believes the evidence rules may be 'wrong' and that evidence that would lead a reasonable person to switch from guilty to innocent is withheld, e.g. http://en.wikipedia.org/wiki/Ed_Rosenthal#Legal_battles , and furthermore that they demand jurors to do something irresponsible, that is, to make a serious decision without attempting to find impartial sources of evidence to ground one's thinking in.

Q: Why does the judge interrogate witnesses, rather than have cross-examination? A: http://digitalcommons.law.yale.edu/fss_papers/534/ seems to suggest that this may lead to shorter trials. I did a Google search to try and find scientific studies on the question of whether cross-examination leads to more correct trial outcomes, but i couldn't find anything. But i did find a zillion pages about how a lawyer should conduct cross examination, and the few i looked at suggested that it is commonly thought among lawyers that cross-examination is not about logic or truth but rather about about theatrics and tricking witnesses into looking stupid so that the jury will disbelieve them. Also i found some pages that claimed that civil law lawyers tend to think that cross-examination is a bad practice. So i'm inclined to think it is bad.

Q: Why is precedent not binding? A: To reduce the need for lawyers. If precedent is not binding, there is no need for people to hire lawyers to spend so much time pouring over past court decisions.

Q: Why ban plea bargaining? A: I'm not sure if it's a good idea or not, but this worried me: http://en.wikipedia.org/wiki/Plea_bargain#Controversy

Q: Where did you get the document discovery rule? A: Copied almost verbatim from International Bar Association (IBA) Rules on the Taking of Evidence in International Commercial Arbitration rule 3.3 as summarized in http://www.hugheshubbard.com/ArticleDocuments/648489.pdf .

Q: Why are opinions written for all cases? A: Langbein seems to think that this is very useful for decreasing errors in the German system which has both professional and lay judges deliberating privately; "Professional judges discuss the rules of law and caution against unfounded uses of evidence in camera when and if the need arises. Since these deliberations occur out of the presence of counsel, the requirement of written opinion is a central protection. Important findings of fact and rulings of law are meant to show up in the judgment and be subject to appeal." -- http://www.law.yale.edu/documents/pdf/Faculty/Langbein_Mixed_Court_and_Jury_Court.pdf

Q: If you don't bar tainted evidence how do you remove the incentive for the police to break the rules when gathering evidence? A: I'm not sure; still working on this. One paper i read said this: "In common law systems the rules of evidence are a means by which courts canexercise retrospective control over the way in which investigations have beenconducted. In inquisitorial systems there is contemporaneous judicial control of the investigation according to statute." -- http://www.academia.edu/933022/Comparative_method_comparing_legal_systems_or_legal_cultures . I'm not sure what that means but maybe it's a solution.

Q: Why do you allow one person to represent multiple seats on a jury when there are not enough jurors? A: Consider scaling down to small groups, or very unequal groups, where one person may be the only indirect constituent under multiple Board seats. Consider also the slightly more complex case in which two people jointly control more than two Board seats, and you see the necessity of this.

Q: Why can you ordinarily not appeal an innocent verdict from a three judge panel? Shouldn't the plaintiff get the right to be heard by a full court? A: This is to ameliorate harassment of clearly innocent parties by allowing them to quickly get the case dismissed.

Q: Couldn't this allow organization officials to break the law without fear of prosecution if the judiciary becomes corrupted? A: That is why either decision can be appealed if the defendent is the organization or a high official.



Topics

Voting

Q: Why (reweighted) score voting? A: It is proportional, it is simpler than STV, it is thought to give the Condorcet answer more than actual Condorcet methods in the presence of strategic voting, it has no spoilers, it allows centrist candidates who are favored by no one but liked by everyone to win. If you're curious, this guy made a website that gives arguments for it: http://rangevoting.org/. Note that in the presence of strategic voting, score voting may be 'more Condorcet than Condorcet" (e.g. http://www.rangevoting.org/ReasAssump.html http://www.rangevoting.org/StratHonMix.html).

Q: Who invented reweighted score voting? A: I think it was Warren Smith.

Q: Why only three integer choices in score voting? A: The strategy of score voting is complicated when you don't have a good idea how everyone else will choose scores given their preferences. 0, 1, 2 have obvious mappings to common human categories of preference that most voters will probably use: vote 0 for people you don't like or don't know about, 1 for people who you think are probably okay (these are likely to be centrists), and 2 for your favorite(s).

Q: Why not have the default vote be 0, and let voters leave it at zero for candidates they don't know, and choose -1 and +1 for ones they do? A: Because then there's a strong extra incentive for negative campaigning -- e.g. Republicans tell their voters, 'Be sure and remember to give Mr. Democrat a -1, s/he's so evil!'. That is, it's to your advantage to tell people who haven't even heard of your opponent about how evil your opponent is, for which purpose emotive, baseless slander will be effective. If the default is 0, it only helps if you can persuade former supporters of your opponent to drop their support, for which purpose you'll have to at least appear to them to be presenting rational arguments against a candidate they support (of course, baseless slander style negative campaigning will still be effective as a way of firing up your base and increasing voter turnout).

Q: Why multiply anonymous and open votes together and then take the square root? A: You have to have open voting so that voters known which delegates, and which people collecting proxy votes in the Forum, are actually representing them. You have to have anonymous voting so that the Mafia can't intimidate or pay you to vote a certain way. So we have both (except for non-Delegates with no proxies, who only have an anonymous/secret vote). They are multiplied together to give an incentive for people to vote their anonymous vote the same way as their open vote. See http://lists.electorama.com/pipermail/election-methods-electorama.com/2013-April/031706.html for the reasoning, and for criticism.

Q: Who invented taking anonymous and open votes and then multiplying them and taking the square root? A: I did.

Q: Why do you always have a runoff between "no action" and the winning proposal on legislative issues, and between the incumbent and the top challenger in single-seat elections? A: It's always good to have a runoff because when there are only two choices, there are no incentives for strategic voting. In cases of cyclic preferences, any option given in the last vote may have an advantage over options not given in the last vote. In a case of cyclic preferences in which there is no clear 'will of the group' and in which the status quo is in the cycle, we want to advantage the status quo, to discourage meaningless thrashing between alternatives over time.


Legislation

Q: Won't the supermajority requirement for passing rules prevent anything from being done? A: No, typically the Executive Officers will set and execute policy by themselves without everything having to go thru the Board. Rules are only needed if the Board disagrees with the Officers, or if they need to make a rule that constrains the members. The Budget does not need to be re-passed each year; it simply continues the way it was until amended. Also, in the Board, a majority is already 60%.

Q: What's the logic behind the specific voting threshold numbers? A: We require 50% < veto <= ordinary resolution <= extraordinary resolution <= amend bylaws. If possible, we want ordinary resolution < consensus,and then if possible, we want extraordinary resolution < amend bylaws, and then if possible, we want amend bylaws < consensus, and then if possible, we want veto < ordinary resolution, and then if possible, we want ordinary resolution < extraordinary resolution. Satisfying these constraints for a 2 person board leads to 50% < veto = ordinary resolution = extraordinary resolution = amend bylaws = consensus; for a 3 person board leads to 50% < veto = ordinary resolution = extraordinary resolution = 2/3 < amend bylaws = consensus; for a 4 person board leads to 50% < veto = ordinary resolution = extraordinary resolution = 75% < amend bylaws = consensus; for a 5 person board leads to 50% < veto = ordinary resolution = extraordinary resolution = 60% < amend bylaws = 80% < consensus; for a 7 person board leads to 50% < veto = 4/7 < ordinary resolution = extraordinary resolution = 5/7 < amend bylaws = 6/7 < consensus; combining these we have 50% < veto < 4/7 <= ordinary resolution = extraordinary resolution <= 2/3 < 5/7 <= amend bylaws <= 6/7 < consensus. A solution to this using simple fractions is veto = >50%, ordinary resolution = 60%, extraordinary resolution = 2/3, amend bylaws = 75%. However, 75% is difficult to achieve, so in the Forum the threshold for amending the bylaws is only 2/3s; it is still more difficult to amend bylaws than to pass extrordinary resolutions because of the requirement that they be confirmed after a waiting period.

Q: You let the Board and the Forum suspend the bylaws? Doesn't that make the system vulnerable to something evil like The Enabling Act of 1933? A: A Temporary Exception requires the same voting strength as is needed to amend the Bylaws, the only difference being that there is no waiting period. The purpose of a Temporary Exception is in case there is an emergency need for a sudden change, faster than the mandatory waiting period for amending the bylaws. Any single Chair can at any time veto or repeal a Temporary Exception. The Enabling Act allowed the German cabinet to further amend the Constitution; by contrast, Temporary Exceptions cannot make it easier to amend the Bylaws or to pass further Temporary Exception. I am uncomfortable about Temporary Exceptions, but I feel that in the event of a true emergency, if the Bylaws don't provide a way for rapid action, the Bylaws will simply be ignored; better to provide a mechanism to allow what is going to happen anyway to occur in a valid manner.

Q: Why do Board and Forum resolutions become Acts if the other house doesn't veto them, rather than affirmatively requiring the other house to pass the same resolution, as with the U.S Congress? A: This allows the Board to take action on matters too boring to interest the Forum, and allows the Forum to take action on matters too specialized to interest the Board. It is also more efficient.


The Budget

Q: Why does the CEOi have so much power over the budget? Why not give the legislature more power? A: In my opinion, corporations do a better job of budgeting than governments. It seems to me that this may be partly because corporations essentially let the CEO set the budget, although if the Board really disagrees they could threaten to replace hir. However, that system leaves no way to impose a bias against rising spending. Therefore, AC provides a hybrid between a system where the CEOi has total control and a system where the legislature has total control.

Q: Why can the legislature reduce but not increase budget items beyond the CEOi's proposal? A: Watching the U.S. Congress, it seems to me that allowing the legislature to increase budget items beyond the executive's wishes is mostly used for pork barrel spending.

Q: Why does the Board have the last chance to amend the budget, rather than the Forum? A: I expect that many voters in the Forum will choose to use their votes to express an ideological point, rather than choose to carefully consider the pros and cons and then to honestly offer a responsible budget. For example, in a national government, some people might choose to give 100% of funding to schools, and 0% to defense and prisons, even if the same person would be unhappy if all prisoners were immediately freed and the military were immediately disbanded. They will reason that their single vote does not decide the budget, that no matter what they do there will be too much funding for war and prisons and too little for schools, so they may as well push it in what they consider to be the right direction. This means that the weighted average of individual budget proposals need not actually yield a budget that a typical voter might like. Therefore, we allow the Board to fix up the results.

Q: Why is payment on interest always mandatory spending? A: Making it difficult to default improves the organization's credit rating, which reduces the interest rate that the organization will pay when it borrows.

Q: Why can the CEOi veto the budget if and only if they proposed it voluntarily? A: To remove any disincentive for proposing a new budget; otherwise, the CEOi would have to balance their desire to update the budget against the risk that they will like the new budget that the Forum and the Board gives them less than the current one.

Q: Why are the voting thresholds for increasing spending harder than those to reduce spending? A: In my observation, organizations have a built-in bias towards growth, leading to wasted money and harmful 'organizational bloat'. The voting threshold bias against growth is supposed to partially counteract this.

Q: Why do you make it so hard for the organization to decide to borrow money? A: In my observation, democratic organizations such as governments have a tendency to over-borrow. The voting threshold bias against borrowing is supposed to partially counteract this.

Q: Why do you think that organizations tend towards organizational bloat, and towards over-borrowing? A: My guess is that, in both these cases, the negative consequences are diffusely borne by many people over a long period of time, whereas some of the benefits are concentrated on individual legislators and concentrated special interest groups over a shorter period of time. In such cases, I think the imposition of a countervailing structural bias may be a good solution.

Q: Why does the current budget auto-renew if no proposal for a revised budget passes? A: In my observation, the U.S. government often fails to agree on a budget on time, and makes budget decisions by deadline brinkmanship, in a way that makes it difficult to make sweeping changes and leads to incremental patches and putting off problems, while also causing operational problems due to the lateness of the budget (not to mention credit rating problems due to the uncertaintly of whether interest payments will be made on time). I don't see any benefits to budget deadline brinkmanship to weigh against these disadvantages. Therefore, in AC we remove this deadline.


External Relations

Q: Why a separate External Executive Officer and External Relations Subcommittee? A: To allow the voters to separately express their preferences on external policy. My observation is that people tend to care more about internal policy, so if they are mixed together, candidates are selected based on their internal policy alone and the winners may not be very good at external policy. Most of the time I expect the External Relations Subcommittee to choose the same person for External Executive Officer as the Board choses for Internal Executive Officer (e.g. there will just be one Executive Officer), but the External Relations Subcommittee needs to be able to threaten to remove the EEO in order to stay relevant.

Q: Why is the External Relations Subcommittee a subcommittee of the Board, rather than a totally separate Board? A: I don't want the internal relations executive and the external relations executive to fight over policy. Having the Board groups responsible for internal and external policy be 60% the same should increase the probability of collegiality between them.

Q: Why does the Forum choose the External Relations subcommittee, rather than a vanilla score vote (without proxies) by all the voters? A: Actually, I couldn't think of a strong reason to decide between these two alternatives. I figured that in the absence of a good reason to do otherwise, it would be easier for groups to let the Forum do it, because then the candidates don't have to do quite as much mass advertising.

Addenda

Seeds

Q: Why have Seeds? A: In organizations that assign unequal and varying voting weight to members using a reputation system like PieTrust?, there is a need for a 'seed group' or 'inductive base case' in order to compute the reputation, and hence voting weight, of other members. Members of the seed group can have unequal 'seed voting weight', but this voting weight should not be able to be taken away by members outside the seed group unless permitted by a vote by the seed group itself. We call the seed group members Seeds and we call the seed voting weight 'Seedshare'.

Q: Why give the Seeds power to veto Amendments to the Bylaws which pertain to Seeds? A: Otherwise the other voters could amend the bylaws to allow them to control Seedshare, and hence the reputation system, and hence the voting weights assigned to voters.

Q: Why only a 1/3 voting threshold to block amendments which pertain to Seeds? A: If the Seed voting threshold to give away Seedshare to someone new is 2/3s, that means that reallocation of Seedshare should not be possible in the face of opposition from 1/3 of the Seedshare or greater. An opponent could get around this by Amending the Bylaws to reduce the Seed voting threshold for giving away Seedshare. So, we make it so that any faction of Seeds who has the strength to block reallocation also has the strength to block an amendment to take away their ability to block.

Q: Why give the Seeds power to veto Amendments to the Bylaws which pertain to the reputation system or to voting weight or to membership? A: These are all connected with the distribution of voting weight, which is the domain of the Seeds.

Fellows

Q: Why do Fellows have a vote on new Fellows? Why not just have the executive officers decide? A: I'm imagining something like IBM Technical Fellows, where the existing Fellows are better placed than executive officers to evaluate other candidates.



Misc

Q: Why should you not have indirect costs of voting? A: We want as many rational people to vote as much as possible. Rationally, if a lot of other people are voting, the chance of affecting the outcome is very small, therefore the cost should also be very small in order to make it worthwhile.

Q: Why not just make it a rule that everyone must vote? Then it is rational to do so. A: We could compel everyone to vote but we can't compel everyone to care. If you force someone who doesn't want to vote to vote, they may vote randomly or (perhaps worse) they may vote in a knee-jerk fashion without seriously considering the issue.

Q: In a situation in which a majority party has as much support as possible without having enough to unilaterally amend the bylaws or take all the seats in some of the branches (except possibly they take all the Executive Officers), how much support would they have and what would the factional composition of the High Officials? A: The majority party would have just short of 6/7rds voting strength, and they would have 2 Chairs, both Executive Officers, 2 High Court Judges, and 5 Board members including 2 External Affairs subcommittee members. So they'd have 11 of the 15 High Official seats, and the opposition would have the remaining 4 seats.

Q: There's only 15 High Officials! What if they all conspire together? A: They probably won't because the proportional representation ensures that all factions are represented, and because Chairs and Board members are elected in a different manner, and because it would only take one dissenting Chair and two Judges to find evidence of such a thing and put a stop to it. Also, the Forum provides a way for voters to express their will directly.

Q: It only takes 6 people to amend the bylaws! A: They can be vetoed by the Forum. Also, those 6 people represent diverse factions, representing 85% of the voters.

Q: No term limits? No prohibition against holding multiple offices? A: I think there things are a good idea for governments, but this system was also designed for small projects and companies, too. In such a case, a founder or founders might not be comfortable suddenly sharing control in so many ways with so many people. These are present in the optional addendum for governments.

Q: What fictional or historical or other roles are similar to some of these offices? A: The Chairs are like U.S. Presidents, or Heads of State in systems that separate this from Head of Government, or Roman tribunes or censors, or like Neil Gaiman's Kindly Ones. The Board is like a corporate Board of Directors, a U.S. Senate, an English Parliament, or a Roman Senate, or like a High Council in various fiction. The membership/censorship committees are like Roman censors, except that everyone can participate. The jury committees are somewhat like Fishkin and Luskin's deliberative polling, with representativeness enforced by the board-coloring constraint. The jury committees are also somewhat like Grecian government by lot (sortition).

Q: Which office would you run for, if you were forced to run, and could only run for one? A: I'd run for Chair, because obviously I'm into thinking about governance system rules and parliamentary procedures and making sure no one's rights are violated, and their job is to supervise that kind of stuff.



Links

Books

i havent read all these! this is a toread maybe list for me

common law vs civil law:


some design principles:

diversity of branches and sources of power (election methods). checks and balances

supermajority is best. when there is no supermajority, executive is free to execute mission eg to allocate funding and direct employees (but not to bind members, eg not to make rules/laws that are binding on members)

relatively strong executive; purpose of mechanisms to remove executive are meant to be used to prevent corruption or dictatorship and to remove incompetence, not to allow the board to participate in decision-making (except for the most major decisions) via (threat of) executive removal.

correct for certain biases of organizations (organizations are biased to empower organization and executives at the expense of members)

in electability, promises of action beat promises of restraint; so Chair position is separate from CEO

---

Q: Why does the Board need only a simple majority to replace the CEO, as opposed to 60% for most other things? A: Note that 4/7 < 60%, but 5/7 is > 70%. In a scenario in which the Board became factionally polarized into two 'parties', it would be relatively rare for anyone to vote to replace a CEO from their own 'party', and it would be rare for the 'party' on the other side of the current CEO to win 70% of the Board, therefore in such a scenario, once a CEO was chosen, it would be rare for them to be replaced.

---

The idea of having each person write down on their ballot one recommended person, and then having a lottery to choose out of these recommendations, has many good properties; it allows ordinary ppl to (sometimes) be elected, while also having the expected value be proportional to party loyalty. Why not extend this to the whole legistlature then, or at least more than 1 of them? The reason not to do this is that every now and then, by chance, a slate composed totally of some offensive extremist group will be elected; this slate would then be able to attempt to amend the rules at will, who would then most likely simply declare the entire system illegitimate and ignore the new 'rules'.

---