I've thought a bit about sortition. My favourite way of explaining it: Suppose a billionaire is looking for someone to marry. They want a partner who is interested in them for more than just a paycheque. Say gold-diggers represent just a very small fraction of the population but are extremely good at feigning fascination in pursuit of their goals and finding themselves in the company of billionaires. In fact they are so good at ingratiating themselves with billionaires that any potential partner this billionaire picks from the people he or she knows is likely to be a gold digger. What is a good but non-optimal solution? The billionaire should pick 1000 random people of the gender he/she prefers and make their selection from this pack. This cuts out the influence of the adverse selection effects the billionaire experiences just from being a billionaire. If all they’re looking for in a partner is someone in at most in the ~99th percentile of people this approach will work with little cost, as candidates can then be selected from this pool using conventional means. This would not work if gold-digging is a reaction to circumstance rather than an in-built characteristic - the same is true of corruption.
For politics, one approach would be sortition applied to those in the 99th percentile on a widely used standardized test like the SATs -it's important that this test be used outside politics, too. Sortition would then be applied to narrow down the pool from ~160000 (about the number of people with sufficient scores over a 10 year time-span) to say a few hundred, a few dozen, or less and then a conventional election from there. Though, you may need very attractive salaries to ensure those elected choose to serve as most don't have an interest in politics - perhaps a ridiculous pension for life after a few years of service.
The critical parameters here are the probability of a malicious/non-malicious actor (or competent/non-competent, etc) accepting the job (let's call it Pma, Pna), and the probability of selecting a malicious/non-malicious actor (let's call em Pm, Pn). What you want is that the probability of selecting times the probability of acceptance be greater for non-malicious:
Pn Pan >> Pm Pam
If we assume we can't do better than random screening for malicious, we have a fixed Pn, Pm. So the crucial thing is to find incentives so that non-malicious actors are the least likely to take the job.
A good pre-screening (if possible) would have to raise the RHS or lower the RHS. For example, if the major incentive for malice is monetary, sampling from wealthy citizens could lower Pm, but it might also sightly lower Pan, so you may have to spend more resources to guarantee Pan=Pam=1.
I'd replace the SAT score with some small, fixed amount of people vouching for you so that you could enter the lottery.
(After all, there might be someone capable who because of some good reasons didn't manage to take SAT, and the system should be just and fair for them, too.)
The amount should not be so large that you'd need the gigantic ad campaigns you need today to be elected, but large enough that you need support outside your immediate family. Maybe ~20. Gathering their 'votes' demonstrates that you're at least interested in the job, and the element of trust is still there (someone trusts you would do a good job, and not doing that would be a betrayal).
One of the nice properties of sortition is it combines well with many other forms of selection. However, this seems to ruin the point:
>Gathering their 'votes' demonstrates that you're at least interested in the job, and the element of trust is still there (someone trusts you would do a good job, and not doing that would be a betrayal).
A big plus of sortition is it does not maximally select for the people most interested in political power. Having to be elected or self-selected in such a fashion would mar this.
I agree, tests have their uses. Like any kind of constraint, they can be used to diminish the search space to increase the desired yield. When devised and administered fairly, they can go a long way toward leveling a playing field (or excluding the crazies).
For example, if we want to exclude congress(wo)men who are powermad egomaniacs, we start by asking knowledgable sources to rank candidates by lust, greed, and vanity, then exclude from random selection all those who fall below a certain threshold. Voila, no more psychos (golddiggers) joining the the Washington Elite.
The lovely part of Sortition is that with a large number of fair & knowledgable 'rankers' (or 'standardized tests') it's basically impossible to game the system.
Taleb talks about this (though he doesn't use the term Sortition) in Antifragile: "instead of having the rulers randomize the jobs of citizens, we should have citizens randomize the jobs of rulers, naming them by raffles and removing them at random as well. That is similar to simulated annealing—and it happens to be no less effective. It turned out that the ancients—again, those ancients!—were aware of it: the members of the Athenian assemblies were chosen by lot, a method meant to protect the system from degeneracy. Luckily, this effect has been investigated with modern political systems. In a computer simulation, Alessandro Pluchino and his colleagues showed how adding a certain number of randomly selected politicians to the process can improve the functioning of the parliamentary system."
Although I very much like the idea of sortition I found the paper by Pluchino et al. very flawed, the simulation they made captures none of the effects of sortition, and the effects they measure have no equivalent in the real world.
Just to get an idea, in their model politicians make many laws that help the population a little bit, instead the randomly selected citizen make make laws that help the population a lot, but they make only a few laws. And things have been defined in such a way that the optimal solution happens when mixing the two. They do a pretty good job at analyzing this simulation, the problem is that the simulation has little to do with the real world.
(I read the paper a few years ago so I hope I'm remembering things correctly).
"It is accepted as democratic when public offices are allocated by lot; and as oligarchic when they are filled by election." (Aristotle, Politics 4.1294be)
It's a practice done by many Lutheran churches - each person writing who they'd like for leadership roles and writing it on a slip of paper, and then drawing one of those slips randomly.
From a theological point of view it allows for the Holy Spirit to act among us.
From a personal standpoint it works well as it allows candidates to put their names forward without the embarrassment of losing based on popularity.
I suspect this sort of belief is what legitimated sortition in the classical world: the choice is made by unknown or supernatural powers. Sortition and "casting of lots" is better known today for divination.
I've also wondered lately if we shouldn't add some random element to the way we fund scientific research. Right now, as I understand it, researchers write grant proposals and funding institutions evaluate them and pick the best ones to fund.
Presumably then, the researchers who get the most funding would be the ones who are the best at writing grant proposals, and they may or may not be doing the best research or researching the most important things. In theory, there could be brilliant researchers working on important things who don't get any money because they lack the ability to persuade people to give it to them.
One of my "if I had a billion dollars" dreams is to set up a foundation that would randomly award research grants to field of qualified applicants.
I think it's the same problem with elections. Governing and campaigning are two different skills. Elections elect people who are good at campaigning, but may or may not be good at governing.
The one missing part of your understanding: grants are generally reviewed by researcher-peers, not really "funding institutions". Your proposal is read by other researches, scores are given to the grant and then a cutoff score is determined. All grants above that score are funded, those below are not.
This doesn't really change your conclusion, that grant writing is an important skill for acquiring funding but it's not about impressing the funding institution, it's about impressing fellow scientists.
The only problem is the scoring system is highly biased towards track record of the grant applicant not the actual grant content. On top of this once you get the grant you can in practice spend the money doing whatever you want.
More fundamentally the peer review process can't actually accurately sort good proposals from outstanding proposals, but we only have the money to fund a fraction of the outstanding proposals. Under these conditions just being honest and moving to a system where all good proposals are put into a lottery and funded by chance would be better approach.
Then you would be flooded with a deluge of low quality proposals. This kind of thing only works with people because it's not trivial to instantly create a large number of new ones simply to game the system.
Obviously there will be some kind of filter and spammers will adjust their techniques to slip through the filter since you've now given them incentive to do so. This is hardly a new phenomenon.
Spam filters are pretty effective these days, I don't think this is an insurmountable problem.
Maybe something like sending a verification email with a link to a captcha. Not too much of a nuisance to do for one application, but unsustainable for submitting a million applications.
Edit: Alternately, an even lazier idea: Randomly award the grant from the pool of applicants rejected by other institutions.
>Random-sample voting can be used locally, nationally, regionally, or even globally, with results that are more irrefutable than with current elections but at less than one-thousandth of the cost.
Random-sample voting, as proposed in those links, is a very specific application of sortition, it is not identical to the broader concept of sortition in the same way that the system currently used for electing the President of the United States is an example, but not identical to the broader concept, of election.
The biggest risk of sortition is moving the balance of power from representatives to the appointed/hired bureaucracy. I suspect that non-politicians are much easier to "manage up" against - I fear having it wind up as political lottery winners getting "advised" into going a certain direction.
> The biggest risk of sortition is moving the balance of power from representatives to the appointed/hired bureaucracy.
That's a risk, but I don't think its the biggest even within the "balance of power between officeholders and X" category of risks; I think the biggest in that category is the shift in the balance of power between office holders and professional interest group influencers (lobbyists et al.) who aren't going to be randomly selected short-timers that get swapped out every few years, they are going to be long-term career professionals representing entities with lots of resources, durable interests, and a commitment to achieving them.
OTOH, sortition doesn't have to be a wholesale replacement for elections, either; it can complement them (as it does now, you just expand the scale.)
Say, have an additional house of the (state or federal, though some of these powers would be more familiar and perhaps appropriate at the state level) legislature (the "House of Citizens", say) selected by sortition, and given it some or all of the following powers:
(1) To take any bill passed by the other house(s) and delay it by putting a hold of up to 90-days on it, after which the other houses must pass it again before it can move forward for executive signature or veto (if there are any changes to the version passed later, its treated as a new
bill, and goes back to the House of Citizens again),
(2) To, instead of #1, invoke the power of referendum against any bill passed by the other house(s), such that if it is approved by the executive (or passed over executive veto), it is referred to a vote of the electorate at large who must approve it before it goes into effect,
(3) To directly initiate an initiative measure, that would go directly to a vote of the electorate at large, and become law with their support, without the need for approval of the other house(s) or the legislature,
(4) To (for a federal "House of Citizens"), on a supermajority vote, propose to the State legislatures amendments to the Constitution, without the other houses (the normal thresholds for ratification would still be required.)
Note that #2 and #3 are powers of the people in many states, but usually require an expensive process of gathering signatures on petitions to invoke, making them more accessible to well-funded groups; a sortition-selected body as an additional avenue could provide an alternate filter that might be less cost-driven.
Athenian democracy is the model people point to, but the worm in the apple is Athenian citizenship rules. The reason selection by lot could work is that a large number of people were excluded from the pool.
"Only adult male Athenian citizens who had completed their military training as ephebes had the right to vote in Athens. The percentage of the population that actually participated in the government was 10 to 20% of the total number of inhabitants, but this varied from the fifth to the fourth century BC. This excluded a majority of the population: slaves, freed slaves, children, women and metics (foreigners resident in Athens). The women had limited rights and privileges, had restricted movement in public, and were very segregated from the men. Also excluded from voting were citizens whose rights were under suspension (typically for failure to pay a debt to the city: see atimia); for some Athenians this amounted to permanent (and in fact inheritable) disqualification."
https://en.wikipedia.org/wiki/Athenian_democracy#Citizenship...
"In regards to Greek mythology, the ephebe was a young man or initiate, around the ages of 17-18, who was put through a period of isolation from his prior community, usually the world of his mother, where he was a child in the community. The ephebe would need to hunt, rely on his senses, on aggression, stealth, and trickery to survive. At the end of the initiation, the ephebe was reincorporated back into society as a man."
https://en.wikipedia.org/wiki/Ephebos
> The reason selection by lot could work is that a large number of people were excluded from the pool.
Its clear that Athens excluded large numbers of people from the pool of sortition (just as most democracies before the 20th Century excluded lots of people from the pool of electors), but its not clear to me how you get from that to the conclusion that exclusion is necessary for sortition any more than it is necessary for election.
I meant could work, as in the place where it actually did work. I wouldn't yet argue that mass exclusion is necessary so that it can work.
Whether sortition can work in a large, inclusive, and diverse society is an interesting question. Democracy can only work in a high-trust environment where people don't expect their political rivals to say, have them murdered or nationalize major industries. The expansion of enfranchisement in the U.S. (at least) has also coincided with the marginalization of very extreme political movements, like Communism.
Say you have an electorate with a very radical far-left or far-right minority that is anathema to the rest of the citizenry? In an electoral democracy, the minority would be excluded in elections (in a first-past-the-post system) or by inclusion as a clear minority (in proportional representation). In a sortitive democracy, it seems like they would have to be excluded from citizenship, or the government would have to have to be large and diffuse enough that the damage of insane or dangerous people could be constrained by the rest of society.
When people are introduced to the idea of sortition, mostly they are shocked by the idea. Because what if you randomly choose a schizophrenic or Hitler to put in charge of the nuclear weapons? I think you have to encourage this tactic at a more familiar level—rather than a group of friends voting on a restaurant, allowing one person to pick after drawing straws. Which, surely, is not at all like giving them a professional military.
> I meant could work, as in the place where it actually did work.
You are still leaping without justification from the way it was implemented and what was necessary for it to work; this is a form of post hoc ergo propter hoc.
Yes, it was implemented that way in Athens.
That doesn't establish that it had to be implemented that way to work, even in that environment.
Eh, it sounds like Athens had systems in place to deal with these problems. With cause you could remove people from office and there was some sort of test to determine if you were eligible for the office at all. Their tests for citizenship (which was the only test that mattered for putting yourself up for election) were bigoted, I don't see any reason to hold them up as useful.
It may also be necessary for election. But you can't disregard the fact that only people who were qualified in some way were allowed to take part in running the country, and in that society it was that democracy was born.
Sortition and election are both ways of scaling democracy to populations too large to effectively use direct democracy alone; neither is any closer to direct democracy, and it seems to me that each can be viewed as mitigating the distortions imposed by the other (as our system -- in the US, but the same is true in most modern democracies -- tends to use election much more than sortition, it may be that it could be improved by increased use of sortition,)
I feel like sortition is much closer to direct democracy because the distortion is limited by the law of large numbers and can be exactly calculated. Of course this excludes things like people feeling like participating, which happens in direct democracy but not in sortition.
> I feel like sortition is much closer to direct democracy because the distortion is limited by the law of large numbers and can be exactly calculated.
No, it can't. The likelihood of a particular degree of distortion (sampling error) can be exactly calculated; the actual degree of distortion cannot.
That is true, but you could argue that the same is true with elections. What if everybody that wants to vote in a certain way is sick on election day? Not very likely, and of course in case of elections the number is larger and therefore the probability of (this specific type) of distortion is much smaller. But then one is again talking about probabilities. And in case of elections there are many other arbitrary factors that can influence the results.
The problem of randomly select citizens for office is that they are as vulnerable to be corrupted as the current elected officials.
But, I would like an approach where we randomly select "electors". They are sequestered, listen to the arguments of the candidates, hold several election rounds and finally come out with he elected officials. Like a jury. They would be dismissed after that.
Right. I'd argue that most of the distortion in US governance is due not to incompetance but malice (like bias that serves moneyed interests), and this malice is deliberate. Random selection alone would cause the 'malice factor' to diminish to the mean. Thus future governors would be no more malevolent (or incompetent) than the average Joe.
Ordinary citizens will suddenly have this huge power. And then they have to go to their former lives. They would be very easy to buy. May be a job offer as consultants after their term is over.
An interesting mixture of the existing election systems with sortition might be the following:
Elections for seats in the parliament are done in the usual fashion. However the number of seats selected by vote in the parliament is only equal to the percentage of voters. The other seats, basically representing the non-voters, are selected by lot.
On can justify this by arguing that non-voters are on the one hand not satisfied with the given choices and on the other hand indifferent to who is ruling them. So selecting the seats by lot might actually represent the will of the non-voters best.
I can't find the link, but there was a good article about this recently (maybe in the last year) that spent quite a bit of time talking about the selection of the Doge of Venice. IIRC, they had a very complicated process that included sortition, nomination, election, and alternating large and small groups that allow them to select qualified officeholders while simultaneously limiting many of the problems with direct elections.
I'd like to restore the state appointment of Senators, preserve the House and add a third house, consisting of representatives chosen by sortition from the states, with representation allocated by (federal taxes paid - federal dollars spent), and with the odds of each citizen's selection proportional to the amount he pays in taxes. Legislation would need to pass all three houses.
The justification for sortition is equity. We all have an equal right to have our views taken into account on all issues. Sortition permits the formation of a truly representative chamber of parliament. Elections do not, for a raft of reasons.
Representation is necessary because direct democracy is unworkable in large modern states, and because no-one has the time to study and understand every issue, particularly as many are highly technical and complex.
It would be absurd to choose one person, or even a small group, by sortition (or elections!), and give them legislative authority. The group chosen must be reasonably large, since a small group cannot be representative.
Regarding sortition and tests: Any test is designed to exclude some people. Who has the right to say that these people should be excluded?
A common idea is that by excluding the stupid, crazy, corrupt, ambitious, malicious… one will get "better" decisions, "better" legislation. But for any decision to be "good" it must first be equitable. Any exclusion renders the chosen group unrepresentative, and hence is inequitable. Better far to include these "undesirables": they will be relatively few in number, just as they are in society, and they will not vote as a unified block when decisions are made. Further, the more diverse the group, the better the quality of decisions (when this can be measured). It is not necessary for all the members of a group to be intelligent or well-intentioned for the group to make intelligent choices.
As for corruption and the undue influence of multinationals, lobbies, religious groups, and bureaucrats, I believe it is possible to greatly reduce or eliminate this with good design of the institutions.
For a fuller discussion of these issues and a proposal for a government based on sortition, you might be interested to read "Down with Elections!", available as an e-book on Smashwords, Amazon and Apple's iTunes bookstore, for about $1. (A blatant plug for my own work).
I've long been interested in the use of sortition in political decision making, and it always surprises me how little it has been seriously studied and considered compared to the potential it seems to have.
Much of the information there is is of pretty low quality. It might of course just be such a bad idea that everybody smart enough to give high quality contributions on the topic does not want to waste their time with the idea. But if this is the case it is totally non obvious to me, and most criticism I've read seem to be from people that do not have a clear understanding of the potential advantages sortition might have.
Very briefly, for the uninitiated, the main potential advantage of sortition is that it would make political decision making a lot more democratic. People representative of the population at large would actually discuss to make the decision, instead of the citizens making their choice by casting one vote every few years among a set of very similar parties (I know, this simplifies the debate a lot, but it is the main idea). This is very interesting if you are of the opinion (as I am) that lack of democracy is a big problem of our political systems. I believe that most time politicians go agains the will of people they do so for the wrong reasons and with the wrong goals, and way too often.
The law of large number makes sure the randomness in sortition is limited and predictable. Whereas with elections there is a big number of arbitrary factors that can greatly influence the results.
Of course sortition in practice might have a number of problem often brought up, but none seems unsolvable to the point where it's not even worth exploring the idea further.
How do you separate expertise from decision power, while still being able to make proper use of the expertise? How to implement sortition in practice? Would they ever let us? Would people be able to handle the pressure? Would they accept the position? And all criticism to democracy in general applies even more to sortition.
I think however that if you talked about elections to somebody who never heard about it, you could come up with just as a big number of potential problems. I don't know if sortition really is a better idea, but maybe it's an idea worth thinking about.
"It might of course just be such a bad idea that everybody smart enough to give high quality contributions on the topic does not want to waste their time with the idea. But if this is the case it is totally non obvious to me, and most criticism I've read seem to be from people that do not have a clear understanding of the potential advantages sortition might have."
All major media and academic outlets are, by definition, powerful. To have a megaphone, to have reach, to be able to have thousands or millions of people listen to your message, is the essence of power.
These media and academic outlets, along with political parties, major activist groups, lobbyists, NGO's etc, have evolved to be powerful within the current political framework. The feedback loop between their ability to send messages to the population, and then get votes in return, is what makes them powerful.
Thus, no powerful outlet has any interest in sortition. Sortition would threaten their power. It would be a great unknown, who knows what the political system would look like under such a dramatic change? Thus nobody really talks about it, and if someone does talk about it, nobody retweets them, or cites them, or gives grants to study it, or otherwise gets excited about the idea.
But at any rate, even if you could convince the powers-that-be to implement sortition, I think the idea is fatally flawed. There is a difference between power and authority. A lottery could give people authority - but exercising actual power requires much, much more. It requires real talent at dealing with people, real influence, real knowledge of how things get done.
Democracy, as in real rule by the people, does not exist, cannot exist, should not exist. The Iron Law of oligarchy always wins, and attempts to ignore the Iron Law of Oligarchy and implement "true" democracy simply result in corrupt and degenerate subversion of the on-paper electoral system.
> The law of large number makes sure the randomness in sortition is limited and predictable. Whereas with elections there is a big number of arbitrary factors that greatly can influence the results.
With elections, there are constraints which make deviations from the preference of the electorate less likely the greater the salience of the issue is. Sortition distributes the issues on which the decisionmakers deviate from the represented population randomly, which means that salience of the issue does not affect the likelihood of deviation between the decisionmakers and the represented population.
This is a significant weakness in sortition compared to elections, that needs to be weighed against whatever advantages are identified for it.
For a particularly dramatic example of this, read G.K. Chesterton's _The Napoleon of Notting Hill_.
In the book, England, late in a long and uneventful history, is governed by lot; a randomly-chosen individual is crowned King and given dictatorial powers. Eventually, the lot falls on a bored nihilist with a fondness for 16th-century uniforms, and within twenty years London is a patchwork of feuding principalities with antiquated weapons, beautiful heraldry, and a whole lot of innocent people dead.
This book confirms what you're saying: it's not very likely that a barking madman will be randomly chosen as dictator, but it only has to happen once.
(The strangest thing about it is that Chesterton saw the most troublesome belligerents, who at one point threaten to use a water tower as a sort of late-Victorian atomic bomb, as the good guys; their willingness to use violence brought beauty and dignity back to a pacific, indolent world. This book was written before Chesterton became Catholic, and its bored nihilism reflects his own feelings at the time; I strongly suspect that it had an influence on the development of fascism, given that Mussolini was a fan of Chesterton's.)
Edit to say that your point is interesting and one I hadn't really thought about. This make elections a referendum on the most salient topics, which I guess makes sense. I still think that in elections there are a lot of other factors that influence decisions and given certain voting preferences there are a lot of arbitrary factors that influence the results, more so that in sortition, whose main drawback, as I see it is how little we know about how it would work in practice.
Original message:
But even for salient issues there are a lot of random factors in elections. Suppose there is a issue so important that everybody cares and votes based only on that issue. Suppose there are only two candidate, and they have a clear and opposing position on this issue so that things are very simple for voters.
Suppose candidate A gets 50,999,897 votes
Suppose candidate B gets 50,456,002 votes
B can still win, as happened with Bush, depending on the voting system. This is just an example and of course depends on the specific voting system. The real point being that for sortition you have simple statistical guarantees, always, independently from salience.
Why not use both? Select a random sample of candidates and then hold an election. By adjusting the number of candidates you can have the process act more like pure sortition (low number) or pure election (high number).
That's a grand idea, but because most of the random lot would be unknowns, I think it would quickly devolve into electing the outlier -- either the one person we do know, or the one who had the superficial traits that the electorate favors most. We'd choose on the basis of celebrity and beauty, since those skills needed for leadership (e.g. good ideas, fairness, capable at delegation, and courage) are too subtle to be detected by a popularity contest, 95% of which takes place only on TV.
To take a realistic example, let a panel of laypeople read 20 studies, half saying that smoking does cause cancer, and half saying that it doesn't. Then, let the experts of both sides speak to them (not needed to say that the ones defending it doesn't cause cancer are much more confident and charming than the others).
I don't think those people would have reached the correct conclusion.
For politics, one approach would be sortition applied to those in the 99th percentile on a widely used standardized test like the SATs -it's important that this test be used outside politics, too. Sortition would then be applied to narrow down the pool from ~160000 (about the number of people with sufficient scores over a 10 year time-span) to say a few hundred, a few dozen, or less and then a conventional election from there. Though, you may need very attractive salaries to ensure those elected choose to serve as most don't have an interest in politics - perhaps a ridiculous pension for life after a few years of service.