Criteria for the application of sortition in a political system

Preamble:

Representation by sortition is defined as “democratic” while representation by election is defined as “aristocratic”. Sortition is a democratic instrument because this way people are represented by “their peers” while in an election-based system people are choosing “the best” as “leaders” (= electoral aristocracy).

To illustrate the aristocratic nature of the electoral system, we can take the example of (what may be a somewhat romanticized description of) pirate ships. A hundred years before the French Revolution, pirate ships were run on lines in which liberty, equality and fraternity were the rule. On a pirate ship, the captain was elected and could be deposed by the votes of the crew. The crew, and not the captain, decided whether to attack a particular ship, or a fleet of ships.

The ancient Greeks (circa 400 BC) used the electoral system for the designation of “the best” as military generals. The legislative institutions, however, were based on democratic instruments: representation by sortition and the people’s assembly. Using an electoral system for legislative institutions mainly finds its origins in the Roman Republic [1].

The ECHR (European Convention for the Protection of Human Rights and Fundamental Freedoms) only mentions the “right to free elections”. Democracy as a right is not mentioned throughout the convention. The term “democratic elections” belongs to the misleading propaganda language of politicians and mass media.

Sortition is a democratic instrument which, together with other democratic instruments and elements (freedom of speech, freedom of assembly, …), can be deployed in a democracy. However, sortition is not a silver bullet. Not every application of allotment results in a democratic decision-making system. Here are some criteria that should be kept in mind when assessing whether a sortition-based institution promotes democratic outcomes.

CRITERIA:

1) Basic Criteria

With sortition, the basic criteria, which are applicable in the referendum system remain valid:

– Setting the political agenda by the citizens / initiative
– Information and debate – transparency of public data
– Decision-making power

2) The sampling system

Sortition systems can be complex, a fact which increases the possibilities of manipulation and errors. Even in simple, transparent systems such as the mechanical lottery drum a bailiff is present. In the Netherlands (2014), there was a suspicion of manipulation in the digital lottery.

The choice of the selection system, and the details of implementation, have to be justified by their function.

– The sampling system must be simple (e.g., simple random sampling) and conducted in a professional manner.
– The sampling procedures must be public.

3) The number of citizens appointed by sortition

A further criterion is the number of citizens appointed by sortition. This number will determine the margin of error, the reliability and representativeness of the panel appointed by sortition.

It may be that we pursue no representativeness for a particular application (e.g., G1000 Belgium: maximum diversity instead of representativity) or limited representativeness (e.g., Oregon Citizen Initiative Review: geographically and demographically representative).

Of interest is also the “rotation” which is a must in a  democratic system. Governing and being governed in turns is the essence of a democracy.

The number of citizens to be allotted should:
– Depend on the desired representativeness.
– Be a function of the allowed margin of error on the results.
– Be a function of the desired reliability of the results.
– Depend on whether mass participation is a goal.

A statistical calculation yields of required size as a function of the allowed margin of error and desired reliability.

– A large number of participants will be less susceptible to coercion and corruption and will also increase the likelihood that efforts to manipulation be reported more quickly.
– The panel of citizens appointed by sortition has to be “a true image of society” as a whole.

A properly implemented system with between 400 and 600 citizens is usually sufficient to ensure reliable operation. Having fewer people has to be justified in detail.

4) Period of service of the allotted citizens

A short period of service has several advantages. Longer times will therefore have to be weighed by their advantages and disadvantages. (Here: short = a day or several days, long = years.)

– A short service period increases the total number of participants over time.
– A long service period gives rise to greater opportunities for manipulation and slows down rotation.
– Longer service discriminates against participants who can not afford a long absence, or do not wish to do so.
– A long period of time may increase expertise. This may be considered both as an advantage and a disadvantage. More expertise means that the panel differs from an “image of society”.

5) Decision-making power

With panels who have only advisory jurisdiction there is on average a participation rate of 2%. This means that if you ask 100 people to participate in such a panel, only 2 people will show up. This could not possibly be representative.

It has been proven that when citizens have effective decision making-power they are more likely to take part in such initiatives.

In principle participation in a legislative panel appointed by sortition is obligatory (a civic duty). Such an obligation will of course be difficult to enforce but various measures that encourage participation may be introduced (fee, assistance, motivation, …). With voluntary participation, there is the risk of “professional participants” and paid participants (funded by civil society, businesses, interest groups, …).

This way, the “image of society”, which is the basis of the jury system, is no longer valid.

– It is of interest to increase the turnout and have it really representative in order to speak of a democratic instrument.

6) Manipulation

It would be unrealistic to expect that the strong financial and economic powers that now influence political decisions would quietly relinquish that influence when sortition is introduced into the political system. We have to anticipate that these powers will mobilize in order to maintain their power.

We can look at a system which we know and which has long demonstrated its reliability (Belgium: at least until the recent changes): the jury system in the juridical system [2]. A jury of 12 citizens designated by sortition (from the electoral registers) makes decisions of guilt or innocence. In comparison with countries (such as the Netherlands) where the court decisions are made by professional judges only, it appears that the jury system is certainly not inferior in the quality of justice it provides. We also see that both parties (accuser and accused) can call witnesses and experts, completely independent of each other.

Juries with mixed systems composed of professionals and citizens appointed by sortition give disappointing results. This is not surprising, such a panel is not at all “an image of society”.

– Avoiding manipulation by “independent neutral staff” who assist deliberation
– Determining which experts and interested parties have to be invited
– No composite panels of professionals (i.e. politicians) and people appointed by sortition

7) Considerations regarding so-called “deliberative” panels

A system that is often used is the division of a big allotted panel, which has heard the experts and interested parties, into small groups which discuss among themselves, under the supervision of “neutral facilitators”, the matters at hand. Then the whole group decides.

Although this system facilitates deliberation, it carries major risks of manipulation, especially if it is not about non-binding “recommendations” but about decisions where large sums of money may be involved (some examples of corruption in Belgium: army shells, Agusta army helicopters, special waste containers, windmills in the North Sea, …). Thus what can and what can’t work will depend on the application.

– We note in this regard the appearance on the scene of the “participation industry” that professionally accompanies participative events. This industry involves academics as well. Overcoming the potential of professionals to manipulate the outcomes can be potentially overcome through the use of bodies such as “supervision juries” which are part of the design proposed Terrill Bouricius.

– One must also distinguish between the “satisfaction of the participants” and the “results” that depend on other parameters.

– Participants of panels appointed by sortition can, in principal, not discuss among themselves but just listen to the experts and stakeholders and vote afterwards.

8) Comparison to referendums

One can also judge proposals (with or without sortition) on the basis of the following criteria: cost, the sampling system used and type of representation, and outcome. One will soon find out that even gathering the data for making such judgments is not easy.

One of the great advantages of referendums (citizens’ initiative) is that all discussions and events take place in public and are accessible to everyone. This freedom to participate and full openness is not found, at least not to the same extent, in the sortition system.


[1] https://www.amazon.co.uk/Beasts-God-Democracy-Changed-Meaning/dp/1783605421

[2] For what reason the French Revolution has established the jury system in criminal cases on September 3, 1791?

“Ce que caractérise Ia cour d’assises, c’est l’indépendance de cette jurisdiction. Elle offre la garantie que les jurés, en raison de Ia durée momentanée de leurs fonctions, n’abuseront pas de leur autorité.”

Translation:

“What this juridical system features is offering the independence of this form of justice. A guarantee that jurors, because of the short duration of their duties, will not abuse their authority.”

Advertisements

47 Responses

  1. Hi Paul,

    Thanks for this: a very interesting and important topic and a very stimulating article. There is lots that I agree with and lots that I disagree with and it would be fruitful to go over the ideas here in detail.

    Before I start, I’d like to link to a few posts of mine that I think are relevant for this topic:

    1. Institutional design power parameters
    2. Internal-dynamics design parameters
    3. A theory of sortition, part 1 of 2
    4. A theory of sortition, part 2 of 2

    Now, to your points. I’ll start with a few short comments on the first points. The later points are more complex and would require more detailed discussion so I’ll postpone those to separate comments.

    1. The preamble: (a) I agree to the general thrust. (b) I am not sure what the purpose of bringing up the example of the pirate ships is. (c) I think there is good reason to doubt that the Athenian assembly (and it should be “the ancient Athenians”, not “the ancient Greeks”) was a particularly democratic institution. After all, many non-democratic poleis had assemblies as well. Also, and perhaps for this reason, the ancient texts do not tend to see the assembly as a primarily democratic institution. (They do assert that having the assembly as sovereign is democratic, but that is a different matter from having the assembly make day-to-day decisions.)

    2. Basic criteria: setting the agenda by “popular initiative” is not a democratic process. Just like elections it grants powerful people and organizations a hugely disproportional influence over the agenda.

    3. The sampling system: (a) Indeed, unnecessary complexity should be avoided – in sampling as in other aspects of the system. (b) It may be worth mentioning that the elections mechanism suffers from the potential for manipulation and fraud as well.

    Liked by 1 person

  2. The example of the Pirate ships shows a clear difference between the appointment of a leader (the captain in this case) and the “democratic” decision (at ships level) of the crew to attack or not a target. For many people the aristocratic nature of elections is not clear at all.
    It is not because powerful people can influence or start an initiative that common people should be excluded. It is far better to improve regulations, demand transparency and put limitation of funding campaigns in place, then taking away for everyone the possibility to put a subject on the political agenda. Therefore I kept the possibility to place a subject on the political agenda in our sortition proposal (evaluation jury, a Multi body system as proposed by Terrill Bouricius)). The right to put a subject on the political agenda is an important democratic right.
    And yes, also powerful or rich people can put a subject on the agenda. But also some poor citizens. Everyone has an equal right to put a subject on the political agenda. It is possible that you need to prove some level of public support but it can be a far more modest requirement then you need now for a referendum.
    And for me, but this is a personal choice, the referendum system stays the ultimate expression of sovereignty of the people. When all else fails, people must have the right to decides themselves. And yes, powerful people might have more influence. But at least, at the end, it is my own decision.

    Like

  3. For those who don’t know our (democratie.nu) proposal it can be downloaded here http://blogimages.seniorennet.be/democratie/attach/137395.pdf The history of the word “democracy” by Francis Dupui Déri can be downloaded here https://www.researchgate.net/publication/259648867_History_of_the_Word_Democracy_in_Canada_and_Quebec_A_Political_Analysis_of_Rhetorical_Strategies : An examination of the speeches of modern Canada’s “founding fathers” lays bare their openly anti-democratic outlook. I-low did a regime founded on anti-democratic ideas come to be positively identified with democracy? Drawing on the examples of similar studies carried out in the United States and France, this analysis of the history of the term “democracy” in Canada shows that the country’s association with “democracy” was not due to constitutional or institutional changes that might have justified re-labelling the regime. Instead, it was the result of the political elite’s discursive strategies, whose purpose was to strengthen the elite’s ability to mobilize the masses during the world wars.

    Liked by 2 people

  4. Regarding this part: “Sortition is a democratic instrument because this way people are represented by “their peers” while in an election-based system people are choosing “the best” as “leaders” (= electoral aristocracy),” I would highlight that sortition is democratic because anyone can be chosen simply due to the fact that they are a human being — that sortition is the absence of any qualification to rule.

    I would say something like: “Sortition is a democratic instrument because anyone can be chosen to take part in decision-making simply based on the fact that they are a human being. It requires no qualification to rule. In sortition, all human beings have an equal right to take part in decision-making and are equally able to be selected to a governing position.

    Election-based systems are based on the idea that some human beings are more qualified than others to rule over their fellow human beings; in an election-based system all human beings are not equally able to be selected to a governing position and not all human beings have a right to an equal say in decision-making. This is an undemocratic system that by design privileges certain human beings over others in decision-making that affects everyone.”

    Rancière (drawing on Plato) emphasizes the key part of the definition of sortition: “What thus characterizes a democracy is pure chance or the complete absence of qualifications for governing. … Democracy is the specific situation in which there is an absence of qualification that, in turn, becomes the qualification for the exercise of democratic arche.”

    (Although I think it’s important to keep in mind that Rancière’s “no qualification” is masking a qualification, that of being human, that is rendered invisible due to the anthropocentrism here.)

    Like

  5. Regarding this: “The ECHR (European Convention for the Protection of Human Rights and Fundamental Freedoms) only mentions the “right to free elections”. Democracy as a right is not mentioned throughout the convention. The term “democratic elections” belongs to the misleading propaganda language of politicians and mass media,”

    I would not concede the point that the ECHR lacks a right to democracy. I think there are very strong arguments that the ECHR contains a robust right to democracy that requires systemic reforms (such as sortition) to fulfill. I’m working on this in detail and will post a paper on this at some point in the future.

    I would reword this to something like: Leading interpretations of the ECHR have often advanced the problematic viewpoint that elections are the foundational component of a “democratic society” [the term “democratic society” is mentioned throughout the ECHR and is a requirement and human right]. However, read together with the larger body of international human rights law, the ECHR contains a right of everyone to take part directly in decision-making that affects one’s life. This is the essence of a democratic society. Electoral systems systemically fail to respect, protect, and fulfill this right. Sortition is a mechanism that can help secure this right for everyone. …

    Liked by 2 people

  6. Hello Jonathan, very good, but “no qualification” is also not completely true. Even in the juridical system (here in Belgium) you have to prove that you can read and write (to some extend) and understand the language of the trial. Nevertheless sortition stays also a system of representation because you, yourself, have no say. Anyway, I don’t see any problem to change the text. Maybe keep both arguments.

    Liked by 1 person

  7. About the ECHR indeed they mention ” the ECHR contains a right of everyone to take part directly in decision-making that affects one’s life” but we may be sure that every word in that convention is very well considered. The right for “free elections” (and regular) is described in detail and installs this way an obligatory “electoral aristocracy”. On the other hand the “right of everyone to take part directly in decision-making” is left open to interpretation. We, here in Belgium have indeed the right to “petition” government. And thats it. They are in compliance of the ECHR. But suppose that the Swiss referendum about the introduction of sortition is a success, it is against the ECHR to abolish the electoral system completely.

    Liked by 1 person

  8. It’s great you’re doing this. These comments are just offered for consideration in the event you think they may strengthen the work.

    Regarding the sortition criteria [and regardless of one’s stance on sortition], it may be helpful to make even more explicit three criteria against which to analyze sortition on a sliding scale from the absence of sortition to hybrid sortition decision-making systems to pure sortition decision-making systems. Below are three key criteria necessary for a pure sortition system. [Maybe one thinks that pure sortition is a bad thing, but these criteria below help make more explicit where decision-making control lies so that the justifications for the relevant control can be examined more carefully and explicitly.]

    (1) A sortition body controls the sortition methodology.

    A sortition body has control over the methodology of the sortition selection process. Randomly selected participants have control over the methodology for random selection (e.g., what methodology is used for random sampling, how many people are randomly selected, how the boundary of the population from which randomly selected persons are drawn is defined, etc.).

    This does not necessarily mean that one sortition body has control over decision-making regarding both the sortition methodology and policy-making. It is possible that the methodology decision-making could occur in one special sortition body and the policy decision-making could occur in a different sortition body. It’s also possible that one sortition body could delegate the methodology decision-making to experts (but with the sortition body always maintaining ultimate power). The key is that a sortition body of some kind has ultimate authority over the methodology of sortition itself.

    (2) A sortition body controls policy-making procedures (including agenda setting).

    A sortition body has complete control over the policy-making process, can hire and fire moderators, can set questions (agenda setting), can change course and ask and answer different questions or pursue a completely different course than that originally intended. (The same issues as mentioned above apply here; it’s possible that one sortition body sets policy-making procedures for another sortition body that does policy-making; it’s possible that this function is delegated to a non-sortition body (e.g., experts), but with a sortition body retaining ultimate control; the key is that ultimate authority is maintained by a sortition body of some kind.)

    (3) A sortition body controls policy-making.

    The final, authoritative decision-making power is located in a sortition body. The sortition body does not act as an advisory body or consultative body for a different body that has final, authoritative decision-making power. Of course, the sortition body can call on and receive input from experts and others, etc., could be informed by elected institutions, administrative institutions, etc. — but the sortition body has final, authoritative decision-making power.

    Liked by 1 person

  9. not on topic but it might interest you: our proposal about the ECHR : Proposal of modification of article 3 of the “additional Protocol of the “ Convention for the Protection of Human Rights and Fundamental Freedoms ”
    Motivation:
    Article 6 of the Declaration of Human and Civic Rights of 1789 : “The Law is the expression of the general will. All citizens have the right to take part, personally or through their representatives, in its making”…

    ARTICLE 3 :

    Right to free elections
    The High Contracting Parties undertake to hold free elections at reasonable intervals by secret ballot, under conditions which will ensure the free expression of the opinion of the people in the choice of the legislature.

    ARTICLE 3 (proposition) :

    Right to free elections and referendum
    The High Contracting Parties undertake to hold free elections at reasonable intervals by secret ballot, under conditions which will ensure the free expression of the opinion of the people in the
    choice of the legislature.
    They also guarantee the right to launch a binding referendum on popular initiative at all legislative levels and to appoint representatives by sortition at all legislative levels

    and yes, I still am in favour of the “multi body” idea of Terrill Bouricius. I think your proposals fit well in our idea.

    Liked by 1 person

  10. >Of interest is also the “rotation” which is a must in a democratic system. Governing and being governed in turns is the essence of a democracy.

    How is that possible in a large modern state?

    Liked by 1 person

  11. Paul,

    > And yes, also powerful or rich people can put a subject on the agenda. But also some poor citizens.

    You could say the same thing about elections: “powerful people or rich people could get elected (or select their candidates) but so could some poor people.”

    The point is that both elections and the “popular initiative” process are both ways for the powerful to exert disproportional power. In both cases there are exceptions, but it is the rule that matters.

    It seems that if referenda are to be used, and they probably have their place as a specialized constitutional procedure, the agenda should be set up by an allotted body.

    Liked by 1 person

  12. Hello Yoram,

    in our sortition proposal the political agenda is set by an allotted body that examins the proposals of the citizens. In order to make some pre selection there can be a signatures threshold.
    This way I think that a lot of referendums can be avoided.

    quote: The Evaluation Jury (7.) will evaluate legislative proposals, who are submitted by citizen petition (4.), on their societal importance. In case of acceptance they are submitted to the “elected representatives” (5.) and the “Legislative Citizens Jury” to vote on (3.).
    . unquote

    Hello Keith,

    >Of interest is also the “rotation” which is a must in a democratic system. Governing and being governed in turns is the essence of a democracy.

    How is that possible in a large modern state?

    It is even not necessary to meet in person, unless for the more “important” juries (For the moment I don’t trust electronic voting and especially internet). People can work on their own or in small groups in their own community.

    quote:The Agenda Jury is made up of a, appropriate for this purpose, sample of 600 citizens (for example, through preliminary survey). The appointed citizens stay available to be selected for selection in the other jury groups, with the exception for the time that they are serving in the Agenda Jury. The appointed Agenda Jury is available for a month (not during the holiday period) to evaluate the societal importance of submitted bill/proposals within a certain time frame (for example 14 days). This means that they will, at the most, be active for a month and 14 days. During exceptionally busy times multiple Agenda Jury’s can be called up.
    Agenda Jury members don’t need to have meetings (the member list is not public.) unquote

    And even with meetings, the large numbers of the populations today make it possible to have vast rotations even without being called up more than a couple of times in a lifetime.
    Even in a small country like Belgium we have 8.000.000 people on the electoral list. This means, if all juries have 500 members we can call up 16.000 juries for serving one day, every day. Even if we all can be politically active one day in 50 years this means 320 juries a day. Even with a participation level of 20% this is 64 different juries a day. Serving on different levels of government, this seams to be enough. Be careful with my math, especially with the number of zero’s :-)

    Like

  13. Paul,

    In a small polis like Athens, randomly selected citizens had significant power so rotation was a substantive concept (hence Aristotle’s remark on its essential role for democratic freedom). Rotation cannot deliver significant power to citizens in large states with tens or even hundreds of millions of citizens, hence the need for the modern versions of isonomia and isegoria to involve representative mechanisms. You don’t need to be a mathematician to work this out, its just plain obvious. Modern discussions of sortition should drop any reference to rotation.

    Like

  14. Hello Keith,

    in this I don’t agree. Rotation is very important. In local discussions I repeatedly had the remark (and it also counts for me personally) that there are a lot of “poll’s” taken from a “representative” number of people but, even I am in my 70 ties nobody ever asked my opinion about anything. Participation (or at least having the possibility) in political decision making is for me essential in what we want to call a “democracy”. I think with a multi body system at all levels of government we don’t have a problem to involve everybody (or at least a “representative” number) op people. In our proposal we even had to be careful not to exeed the number of people available. Of course participation level stays unknown but we can use the number of people that involve themselves in referendum activities (about 40% , for us Switserland is a reference http://www.swissinfo.ch/directdemocracy/the-swiss-vote-more-than-any-other-country/8483932 ).

    Liked by 1 person

  15. Paul,

    Sortition does enable a descriptively representative sample, but Aristotle’s argument about rotation was a moral one to do with the nature of equal freedom. Either all citizens get a realistic opportunity to rule and be ruled in turn (most Athenian citizens could have ended up “king” for one day if they wanted to) or they don’t. We really do need to keep participation and representation distinct. My publishing company was planning to launch a new journal of Deliberative and Participatory Democracy and I asked Carole Pateman (author of the seminal text on the latter) to join the ed. board — her response was that the two forms of democracy have little in common.

    Like

  16. Hello Keith,

    yes indeed, representation and rotation are very different items. But unavoidably, the chosen system of representation (many participants and a multi body system and short duration, versus the opposite ) has an effect on rotation and has thus to be taken in consideration.

    Liked by 1 person

  17. Hi Paul,

    > in our sortition proposal the political agenda is set by an allotted body that examins the proposals of the citizens.

    Ok – but as the “basic criteria” section reads now it seems like an endorsement of the “popular initiative” process, and no reference is made to having allotted bodies determining the agenda for referenda.

    Also, are you implying that allotted bodies cannot come up with their own proposals and must accept or reject as is proposals that are handed to them by someone else?

    How about the following?

    1) Basic Criteria

    With sortition, the basic criteria, which are applicable in any democratic mechanism remain valid:

    – Setting the political agenda – the agenda should be set by the citizens rather than being imposed by an elite
    – Information and debate – open discussion and transparency of public data
    – Decision-making power – the citizens should have effective power to set policy from the full range of feasible options, not merely to make recommendations or select from a restricted set of options defined by an elite.

    Liked by 2 people

  18. I would oppose Yoram’s suggested amendments for reasons that I have provided at length elsewhere.

    Like

  19. Regarding the number of members in an allotted body:

    It seems that you keep pushing for bigger numbers: more representativity, lower margin of error, a truer “image of the society”. But that must be missing an important – indeed crucial – consideration. If there was no strong reason for reducing the number of members, we would be dropping the whole sampling mechanism and going for a mass participation mechanism (“direct democracy”).

    What are the considerations for reducing the number of the members?

    Liked by 2 people

  20. >What are the considerations for reducing the number of the members?

    In order to overcome rational ignorance.

    Liked by 1 person

  21. Hi Yoram and Keith,

    of course you can reduce the members. We know that very small juries can work in specific circumstances (juridical jury of 12 citizens). We also know that the results of the Oregon CIR (and alike) with 40 citizens (stratified) presents reasonable work (a citizens perspective of an initiative). The results of the Washington Jury (wages of politicians) is to review (10 citizens stratified, mixed with 7 professionals ) and so on. But this are not legislative juries. In that case I would stand for a non disputable number to start with ( around 500). And in special cases with a good explanation these numbers can be reduced if wanted. But I think this is not only a matter of knowledge about statistical systems but also about “perception” of reliability and representation. And of course some risks are greater with small numbers (coercion, corruption, influencing,..)

    Like

  22. Hello Yoram,

    “” Ok – but as the “basic criteria” section reads now it seems like an endorsement of the “popular initiative” process, and no reference is made to having allotted bodies determining the agenda for referendums.

    Also, are you implying that allotted bodies cannot come up with their own proposals and must accept or reject as is proposals that are handed to them by someone else?””

    Yes, we kept the “referendum” system apart from our sortition proposal. I don’t think it is wise to merge them. It is difficult enough when they are discussed apart from each other. We know the example of BC where a proposal of the jury was “defeated” in a referendum but politicians had imposed a “special majority” which is in our view against the equality rule. (proposal rejected by a minority)

    The use of the “evaluation Jury” as we proposed is in the example limited to evaluate citizens proposals. But there is no “theoretical” limit in what it can do. These possibilities have yet to be developed and only depends on your imagination or knowledge of examples.

    Like

  23. Hi Paul,

    In fact I agree that for a high powered body a membership of hundreds is desirable (although I would put the number somewhat lower than 500). But that was not my point. The point is that the considerations or criteria that you mentioned all point in the direction of increasing the number of members. The point is that that cannot be the entire picture, or else we would not rely on sampling at all. There need to reasons for reducing the number, and the document should discuss them.

    Referring to the posts I linked to above (“A theory of sortition”), the ability of an allotted body to generate representative policy relies on two planks: (1) self-representation of the decision making body, and (2) representation of the interests of the population in the allotted body. Increasing the number of members in the allotted body strengthens the second plank but weakens the first plank. A balance has to be struck.

    Sutherland, bowing as always to conventional wisdom in elite circles, offers the standard “rational ignorance” argument as the reason for reducing the number of members in a decision making body. And it is indeed one such consideration. But it is in fact only a secondary consideration, for even if it were possible to convince a substantial majority of the citizens of a country to educate themselves on a particular topic (and this is not a fantastic scenario), this would not enable mass-political democratic decision making on that topic.

    Like

  24. Paul, agree 100% with your last two comments.

    Liked by 1 person

  25. Paul,

    >In [the case of legislative juries] I would stand for a non disputable number to start with ( around 500). And in special cases with a good explanation these numbers can be reduced if wanted.

    The confidence interval is the important factor, but some statisticians argue for as many at 1,000, although this is pushing nearer to the rational ignorance threshold.

    >But I think this is not only a matter of knowledge about statistical systems but also about “perception” of reliability and representation.

    Yes, and that may lead to a larger jury than would be strictly necessary. The numbers could be reduced once people have got used to the stochation principle.

    >And of course some risks are greater with small numbers (coercion, corruption, influencing,..)

    One way of getting round this would be to limit the mandate of the jury to listening to the arguments and then voting in secret.

    Yoram: >Also, are you implying that allotted bodies cannot come up with their own proposals and must accept or reject as is proposals that are handed to them by someone else?

    Paul: >Yes

    Agree. The 4th century Athenian legislative system maintained a strict separation between proposing and disposing, with sortition only having a role to play in the latter. Restricting the proposal role to a tiny group of citizens would be a form of oligarchy. How to ensure that the proposals properly reflect the preferences of the target population is a non-trivial problem.

    Liked by 1 person

  26. Hello Yoram,

    the discussion is proceeding a little to fast. I don’t understand your comment: “” Referring to the posts I linked to above (“A theory of sortition”), the ability of an allotted body to generate representative policy relies on two planks: (1) self-representation of the decision making body, and (2) representation of the interests of the population in the allotted body. Increasing the number of members in the allotted body strengthens the second plank but weakens the first plank. A balance has to be struck.””
    I have to read first the articles you refer to. I don’t think I know them. (next week I have some more time)

    As far as I can see now the criteria list I propose still stand as a “check list” that leaves open all posibillities.
    The only things I want to add for the moment is the possibility of a “multiple body jury system” with his advantages. Also the possible use of additional involvements of citizens by all means available (Iceland constitution) has to be mentioned. But I have to study that case a bit more in detail.

    Liked by 1 person

  27. We face several dilemmas in designing a sortition body. To name four:

    1. Size (bigger is more representative, but increases rational ignorance or intellectual free-riding).
    2. Term (shorter preserves representativeness – higher participation rate and resistance to corruption, while longer increases skills and knowledge)
    3. Active deliberation vs. just listening to pro/con and voting (diverse groups deliberating together can uncover better win-win proposals, but also suffer from group-think and unrepresentative distortion).
    4. a full charge mini-public deciding its own agenda and rules is maximally free of elite domination, but can also thereby become corrupted and self-serving.

    The natural instinct is to compromise, balance conflicting benefits and costs and find the sweet spot (the optimal size, term of office and deliberation style, etc.) in a single perfect mini-public. But the better approach is to take the best of EACH design option and use it in a separate sortition body so we get All of the positive benefits of each design option and protect against the negatives. For example, one mini-publc establishes an agenda, but does not draft nor vote on policy proposals, numerous small active deliberating groups can develop proposals which a mute large jury hears and votes on.

    Liked by 1 person

  28. Hello Terrill,

    also a concern of mine is that the more complex the task of the alloted body, the more influence (manipulation) is possible from the organizing organization.
    Something that was a surprise for me was that most people who participate in such events are happy or even very happy (enthusiastic) that they did it, even there is no real result (as far as they and we know).
    This let me think that they are well received in a “professional” way. A good welcome, a listening ear, you are important, what you say is important and we are here to listen (at last there is someone listening to you) and so on… Some wining and dining (something most or many of us are not used to) , maybe stay over night in a hotel, expenses payed (this is normal but “unusual” for most of us). And I know from personal experience that corruption and manipulation start this way, at “very low” level.

    Liked by 1 person

  29. Terry,

    Your analysis of the four areas of trade-offs is spot on but why do you insist that sortition has to do all the heavy lifting and not consider the possibility of a hybrid system involving sortition, election and direct initiative? I can’t think of a single historical example of a well-functioning system of governance that was “pure” in the sense of relying on a single selection principle (classical Athenian democracy was certainly not a “pure” system). In the past you have acknowledged that your approach is utopian but Paul’s thread seems to be concerned with practical possibilities rather than abstract ideal types.

    Like

  30. Ben Margulies’ review of David van Reybrouck’s book on the Democratic Audit website outlines some of the problems of governance by multi-body sortition:

    Finally, the book’s model for a system run on these ‘aleatory-democratic’ (‘aleatory’ meaning ‘based on chance’) lines, though very thoughtful, is also hugely complex. It involves no fewer than six types of assembly: An assembly to set the overall agenda; ‘interest panels’ to propose specific legislative topics; a review panel to prepare detailed bills; a ‘policy jury’ that would vote on the bills (but not debate them); an oversight council to handle complaints and judicial review; and a rules council to set rules for the other bodies. It is unlikely even the most educated, wonkish voter would find such a Rube Goldberg machine an attractive model.

    http://www.democraticaudit.com/2016/10/30/book-review-against-elections-the-case-for-democracy-by-david-van-reybrouck/

    The adjective that I generally use to describe such a complex system is “byzantine”. Much better to use allotted juries for functions that we know they do well (listening to opposing arguments and deciding the outcome) and leave the other functions to electoral and direct-democratic procedures.

    Liked by 2 people

  31. While I like Terry’s list of dilemmata, it also smuggles some unsubstantiated – in fact demonstrably wrong – assumptions through the back door.

    “Size (bigger is more representative)” – This “more representative” implies a higher decision quality with higher numbers, which sounds right for frequentist statisticians but is demonstrably incorrect in consideration of a real life situation with subjective probabilities. We know that errors do not improve beyond a certain sample size. We tested this in many ways. The simplest one is by asking people to decide what a coin toss will result. Their prediction is 68% and the error does not change if you make the sample bigger. So I argue for strictly limited sample sizes at the level of efficiency (we like 99%), and no bigger.

    “Term (longer increases skills and knowledge)” – This implies a rule that foresight in a topic is not considered which contradicts et least our model. We envisage rapid appointment on single and connected decisions. Higher topical foresight increases likelihood of sortition (through more lots). Clearly, retaining a huge pool of good candidates, facilitates fast rotation. Key is to strive for a good 50% of the population participating in the virtual chambers of “higher” democracy actively in any given year. And hope that 100% vote for the party committed to the true, participative democratic process.

    “Active deliberation vs. just listening to pro/con and voting” – A restriction to “just listening” is grotesque, given freedom of speech and the importance of the wisdom of the crowd for good decisions. The deciders may and should deliberate on a proposal as much as they like before voting. However, if they want to improve/change something of the proposal, it is the sole discretion of the proposer to incorporate the change (or have his proposal potentially rejected). Interactive deliberation at all three virtual chambers is not only a good thing, but extremely important.

    “Full charge mini-public can become corrupted and self-serving” – This is only true when lacking fastest rotation, which is an absolutely indispensable requirement given that specialisation is progressing faster and faster. See “Term” above.

    Like

  32. HJH:

    >A restriction to “just listening” is grotesque, given freedom of speech and the importance of the wisdom of the crowd for good decisions.

    The 4th-century Athenians (who actually tried this stuff out) certainly would not agree with you, as this was the procedural rule for the legislative courts.* And Surowiecki (the author of The Wisdom of Crowds) argued vehemently against deliberation. Have you read his book? I think you are conflating Surowiecki’s argument with the epistemic case for cognitive diversity (Page, Landemore etc). Landemore is adamant that her book has nothing to do with the wisdom of crowds.

    * Freedom of speech for the allotted few actually harms the (virtual) freedom of speech of the disenfranchised many as the law of large numbers doesn’t apply to the speech acts of small groups.

    Like

  33. Keith asks
    >why have sortition “do all the heavy lifting” [be used to select the various bodies]?

    I propose that one of the bodies be instead self-selected… all who wish may join an Interest Panel to try and craft a proposal that can pass muster by sortition bodies…though I also reserve the option to create some Interest Panels by sortition as well. There are three reasons sortition is used for the other bodies.
    1)Blind break anti-corruption protection applies to all of the bodies,
    2) diversity applies to the Review Council that engages in active deliberation rejecting, accepting and melding the different proposals from numerous Interest Panels, and
    3) statistical sample size representativeness applies to the final Policy Jury.

    Some bodies are small to facilitate active deliberation (but thereby forfeit statistical representativeness), but always, the final decision is made by a larger mini-public.

    HJH notes that there is no need to over-do the size of a body… and this is correct for any sub-level body that is not making the final policy decision. It is also true that a sample of 500 is nearly as accurate for a population of 10,000 as for a population of 10,000,000,000. The question is simply how often we are willing to have a statistically inaccurate body. But likewise, a sample of 45 has a fairly high chance of being an inaccurate representation, regardless of population, UNLESS there is some sort of supermajority voting requirement (which unfairly benefits the status quo).

    Like

  34. Terry,

    >I propose that one of the bodies be instead self-selected

    However you reject the idea that people might prefer to select others to represent them or that persons should be selected on the basis of proven merit and expertise. In short you appear to reject the aristocratic principle tout court although one would be hard pressed to find historical examples of effective governance that did not rely on selection by merit to some degree. That’s why your proposal is utopian/revolutionary, rather than evolutionary.

    Like

  35. In my view there are only two criteria for a political system.
    – Trust in the system
    – General acceptance of the decisions
    We do know that at this moment elections and elected politicians are loosing on both terms for many years now.
    But this are long term criteria.
    When I look at the WCCSEO http://www.salaries.wa.gov/ My first reaction is that I don’t trust this system (composition of sortition and professionals). But on the other hand it is working now for about 30 years and I think Washington is a referendum and initiative state, so I presume that in that time people would have started an initiative to ameliorate the system if they think it doesn’t work well. It would be interesting to convene a jury of 500, hear the experts and then vote on the salaries and compare both results.

    Like

  36. Paul,

    >– Trust in the system
    >– General acceptance of the decisions

    I think this is exactly right. Some epistemic democrats argue that general acceptance will depend on the “rightness” of the decisions, but I’m sceptical as I believe democratic norms* require that the decision should reflect the informed preferences of the majority/plurality, irrespective of epistemic considerations. I also agree that the votes of a randomly-selected and silent jury of 500 would be the best way to uncover the informed preferences of the target population. That’s certainly how the 4th-century Athenians did it.

    *These two criteria would lead to very different outcomes in aristocratic, meritocratic, monarchical or theocratic societies, but that doesn’t undermine their validity.

    Like

  37. Hello Keith,

    quote:
    I also agree that the votes of a randomly-selected and silent jury of 500 would be the best way to uncover the informed preferences of the target population. unquote

    I am in doubt a bit about “the best way”. I would rather say “the safest way, all things considered”. Deliberation has his merits but is also very dangerous as far that manipulation is concerned. Also a definition of a “good result” is very difficult I think. We must take into account that vested interest will try to influence politics, no matter what political system is used.
    What can be done with “unlimited” money to influence a Jury? We can estimated the amounts available by looking at referendums. That money will still be used. The only question is “how”.
    If my (imaninary) “participation company” is allowed to organize the event I can deliver the desired outcome you pay for. Don’t you think? Those companies are performing very well these days (stakeholder engagement) but there is nothing at stake compared with political power. (Example : Process design & project management of a project in participative democracy engaging citizens and stakeholders around specific challenges in Zaventem, Nossegem, Sterrebeek and St-Stevens-Woluwe.)

    Like

  38. Paul,

    >Deliberation has his merits but is also very dangerous as far that manipulation is concerned. . . What can be done with “unlimited” money to influence a Jury?

    The Athenian demokratia made a categorical distinction between deliberation (isegoria) and judgment (isonomia). I think we agree that the 500-strong jury is the “safest way, all things considered” for the latter, the remaining problem being the former. This is why deliberative advocacy needs to be provided exogenously and in a balanced manner. In the case of the Brexit referendum the two camps were nominated by the Electoral Commission and there is no reason to believe that the result was “bought” by either camp — if anything the Remainers were better funded yet they lost, despite the additional benefit of support from the majority of the mainstream media. The only problem with the Brexit referendum was rational ignorance and this would be overcome by the use of a 500-strong jury.

    Like

  39. Hi Keith,

    with our organisation we make a distiction between plebiscites (referendum at the initiative of government) and the referendum. The plebisite is often an instrument of party politics (Brexit) or dictators. In Switserland it is not allowed. And we oppose it strongly. The Brexit was also in a country where people are not used at all to a referendum, in that case a vote out of frustration with the government in general is (and not about the brexit) is to expected. Most people didn’t believe that the result of the referendum would be followed by the government (it was informative). It is a reference but in the right perspective ;-).

    Like

  40. Paul,

    I only used the Brexit referendum as an example of how not to do it. I published a piece on Open Democracy arguing that a sortition-based alternative would be better: https://www.opendemocracy.net/can-europe-make-it/keith-sutherland/brexit-lottery

    Like

  41. I’m looking forward to Peter Stone’s paper “Why I Am Not a Sortinista” at the sortition panel at the PSA conference next month. This is from the abstract:

    “It is no longer enough to make the democratic case for sortition; it is now necessary to consider the respective democratic contributions of sortition and other institutional devices, notably election. Some sortition advocates have shied away from this challenge, out of a conviction that sortition is the uniquely democratic selection mechanism. I call such advocates sortinistas, adapting (in an admittedly revisionist manner) a term employed in dialogues over sortition. In this paper, I shall critique the sortinista approach to democracy, focusing upon David van Reybrouck’s recent book Against Elections.”

    I agree with Peter’s argument but would prefer the term “Kleristocrat” over “Sortinista” as full-mandate sortition is a form of oligarchy.

    Like

  42. Keith, your reference to the Athenians keeps confusing deliberation with the two responsibilities for the designing of, and the deciding on, proposals. They are different things. The activity of deliberating designs or decisions is not “conflating” anything. Unless we can agree to distinguish these entirely different responsibilities, our debate will be moving in circles.

    “And Surowiecki (the author of The Wisdom of Crowds) argued vehemently against deliberation.”

    I am not aware of anything like this in the book. To the contrary, S. explicitly mentions supervisory boards as an example of crowd wisdom. These boards very much deliberate on management proposals before voting on them. If you believe otherwise I cannot comment without an exact reference.

    Like

  43. Addendum: Just dug out my copy of Surowiecki. Sure, despite making a general case for crowd wisdom, S. warns of mistakes with badly managed deliberation or deliberation in badly designed small groups.

    Fair enough as long as we do not throw out the babe with the bath water. Dangers of bad execution exception does not invalidate the principle.

    Like

  44. For convenience I have dug out the exact conclusion of Surowiecki on deliberation (p.190):

    “There are two important things about these studies. The first is that group decisions are not inherently inefficient. This suggests that deliberation can be valuable when done well, even if after a certain point its marginal benefits are outweighed by the costs.”

    Note: Surowiecki is mostly concerned about efficiency…

    “The second point is probably obvious, although a surprising number of groups ignore it, and that is that there is no point in making small groups part of a leadership structure if you do not give the group a method of aggregating the opinions of its members. If small groups are in-cluded in the decision-making process, then they should be al-lowed to make decisions.”

    Note: Surowiecki sees deliberation very much as preceeding decisions.

    QED.

    Like

  45. HJH,

    >Keith, your reference to the Athenians keeps confusing deliberation with the two responsibilities for the designing of, and the deciding on, proposals.

    The Athenian distinction was between proposers/advocates and disposers. This involves separate personnel and the latter group do not deliberate (other than silently). Sortition is only a legitimate form of representation for the latter.

    If you look at the last chapter of Surowiecki’s book the reason he argues against applying the wisdom of crowds directly to politics is on account of deliberative democracy, which he rejects as a “quixotic” project (p. 260). I’ve had a long private debate with Helene Landemore in which she explicitly distances her position form the wisdom of crowds.

    Like

  46. While it may not be a focus of Surowieki, he does mention, and there is a lot written by others (such as Cass Sunstein) about the distinctly different benefits and dangers of active deliberation in tapping the “wisdom of crowds.” For some sorts of decisions (for example, problem solving) it is essential that the diverse group members be able to share their unique private knowledge or ideas with the other group members, for the group to hit upon a better solution to some problem. To ban such give and take deliberation would forfeit some of the most important benefits of diversity in group decisions. However, in other circumstances, active deliberation can prevent the uncovering of the wisdom of crowds…such as when “information cascades,” or social deference to those of higher status cause members to abandon their own opinions for bad reasons and defer to the opinions of others whom they assume know better (even if they don’t). As in the Condorcet jury theorem, there are times when full independence of members judging what they have learned is the ONLY way to elicit the wisdom of crowds. In short, BOTH participatory and internal deliberation are useful depending on the specifics, but they need to be performed by different groups of individuals gathered into distinct bodies.

    Like

  47. Terry,

    Although it’s just terminology, Helene refers to this as “collective wisdom” rather than the wisdom of crowds. I guess the point is that crowds (unlike a small group) don’t interact or deliberate. The wisdom of crowds is nothing more than a statistical aggregation of individual judgments (as in the Condorcet theorem). Helene scolded me for conflating the two things, and I’m just passing on the message. It always helps to have distinct concepts to stop misunderstanding, so I suggest we go with Landemore’s Collective Wisdom (small deliberative groups) and Surowiecki’s Wisdom of Crowds (statistical aggregate of individual judgments).

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: