An argument against sortition that is fairly common among academics is that allotted bodies are not representative because the acceptance rates of offered seats are low. It is often claimed that “experience has shown” that “less than 10%” of people are willing to serve on allotted bodies. Such a finding, it is claimed, is grounds for not using sortition at all, for limiting the powers of allotted bodies, or for various forms of meddling in the way allotted bodies are made up.
Despite the fact that it is sometimes admitted that acceptance rates change depending on the circumstances, the “fact” of low acceptance rates is largely treated as being an immutable, if unfortunate, obstacle to the representativity of allotted bodies. In fact, however, it is obvious that acceptance rates can be easily increased, quite possibly reaching fairly close to 100%, if compensation for acceptance is high enough. How many people would refuse to commit a few weekends to participating in an allotted body if they are paid a few months’ worth of the median salary for their efforts? The answer to this question is obviously that (while we can probably make a good guess) we do not know for sure. But equally obviously it would be fairly easy to find out by running a few experiments.
It is a small miracle then that all those who busy themselves with attacking sortition by arguing that low acceptance rates make allotted bodies unrepresenative have not argued strongly for running such experiments. One may suspect that complaints about low acceptance rate is a tool for resisting the democratic power of sortition, rather than a real concern coming from people with a genuine interest in democratizing society.
A larger miracle it seems is that even promoters of sortition never seem to make an effort to refute the claim that the democratic nature of allotted bodies is irredeemably undermined by low acceptance rates by offering generous compensation for accepting allotted seats. Jaquart (2017), in an article devoted to analyzing non-acceptance of allotted seats and which seems generally sympathetic to the idea of the use of sortition, barely mentions the matter of monetary compensation. In a single sentence he acknowledges that “of course, all the incentives (e.g. money) and efforts made to facilitate participation (e.g. child care) can influence the rate of participation”, before moving on to analyzing various modes of justifications given by potential participants for rejecting allotted seats.
Yet, even a single or a couple of experiments showing that generous compensation does wonders for participation rates would in all likelihood change the entire tone of the conversation on the issue of acceptance rates. No doubt the detractors would then shift their position and explain that such high compensation is unfeasible, or unseemly. However that is far less comfortable rhetorical position than the high ground they are taking now. Could it be that even sortition promoters who have the resources for running such experiments, James Fishkin being a good example, are uncomfortable with bodies that are too representative (and/or with participants who are too well paid)? Could it be that the organizers of allotted bodies want it to remain clear who is in charge and who are mere foot-soldiers in the sortition mechanism?
Filed under: Academia, Experiments, Opinion polling, Press, Sortition |
Adam Cronkright
Ali
Ahmed R. Teleb
avanderven
avergne
Alex Kovner
azakaras
Brett Hennig
campbellwallace
Simon Threlkeld
Nick Coccoma
codyhipskind
Common Lot Sortitionist
Conall Boyle
constitutionalism
davidschecter
graytmatter
gtridimas
H.J. Hofkirchner
Hanspeter Rosenlechner
John Burnheim
jgarry20151
Jochen Krattenmacher
johnedwardtaylor
johnhuang865
Jonathan Crock
keithsutherland
Jacob Richter
kkoreilly
lukewhito
SimonPek
Nathan El
Nicholas Gruen
Owen Shaffer
Patrick Chalmers
Paul Gölz
peterstone
Pierre-Etienne Vandamme
pronfeld
Emil
rcaze
richwbrown
Roger Knights
roslynfuller
Pierre-Etienne Vandamme
shaharavin
tbouricius
Ted Wachtel
Joshua Laferriere
theodorestathis
towardsautonomyblog
Wayne Liebman
Yoram Gat
Oliver Milne
Equality by lot 
>grounds for limiting the powers of allotted bodies
Alex K and I merely argue that, if statistical representation is the legitimising (democratic) principle, then sortition is only applicable to situations where the law of large numbers applies. This has entailments for both the composition and mandate of the allotted body.
>the “fact” of low acceptance rates is largely treated as being an immutable, if unfortunate, obstacle to the representativity of allotted bodies.
Not so. We argue the case for quasi-mandatory participation and don’t much care whether it is implemented by the stick or the carrot.
>Could it be that even sortition promoters who have the resources for running such experiments, James Fishkin being a good example, are uncomfortable with bodies that are too representative
That verges on the libellous. Of all sortition advocates, Jim does everything he can to ensure that as many of those who receive the golden ticket turn up to participate. And, as you well know, I have long been an advocate of experimentation, albeit with the focus on ensuring invariant decision outcomes.
LikeLike
There is always a lot of argument not to change…. Yes low acceptance is a problem and that is why you run a second lottery with people who accepte to participate after you pool them in group to represent the population.
Is-it perfect? No. Would it be better if it was mandatory? Yes.
But the real question is, will it be better representative of the population that the actual system just at looking how many workers, middle class, homeless people, people who can not afford medications and so on are in congress and senate? The question is yes definitely, of course it is without any doubt.
So we should not look for perfection but for the best system we have to represent people and election prove not to be this one. People arguing that we should not do anything because it is not perfect without comparing to the actual system have a strange thinking to me.
This is the argument we should oppose to those who use this not to change everything and i would be curious to hear a counter argument on this.
LikeLike
The stratification criteria suggested by anonymous are primarily socioeconomic. Of course that’s important, but voluntarism could be taken to be an equally important population parameter, as there is a danger of a sample that is skewed towards activists and those with strong views on the issues under discussion, as opposed to the Average Joe. Given the degree to which the public sphere is already dominated by anoraks and headbangers on X (Twitter), institutionalised voluntarism could be disastrous. At least with the current arrangements, people are able to vote for the guy who (at minimum) performs speech acts that they find more agreeable than the other candidate. Note that my argument is not defeatist, it’s just if you don’t do it properly you are likely to discredit sortition as a viable tool of representation. I’ve been working in this field for around 20 years, and I don’t want it all to be in vain. I hope you feel that is a valid counter-argument. Note also that I’m fully in favour of Yoram’s proposal for bribery experiments — I don’t much care how we arrive at a representative sample (although I prefer the notion of civic virtue), that’s why we add “quasi” for our case for mandation.
LikeLiked by 1 person
I understand your view but right now it is how it works for most of the experiment (in Ireland or in France) and I have not seen the activist taken over…. One solution if sortition is used to answer only one question would be not to tell people on what subject they are going to work on.
If you replace our deputies, then an activist is activist on one subject most of the time but he will have to think to a lot of different subjects so at the end of the day I do not think it is a problem.
Speaking of voluntarism do you know who are volunteer to tell our society how it should be? Politicians… From day one to their election. They may be not activist but they serve the people who give them money. I do not know wich one is the worst. You may have read the Princeton study that show that the average Jo as no influence on political influence.
So again it is not telling sortition is perfect it is just comparaison between two system and choosing wich one has people interest first.
LikeLike
Yes, elected politicians are volunteers, but Average Joe/Jane chooses which ones get to serve.
>If you replace our deputies, then an activist is activist on one subject most of the time but he will have to think to a lot of different subjects.
That would suggest a longer period of service than the ad hoc panels (based on the classical Athenian template) that many of us are proposing. There is a danger of going native.
>You may have read the Princeton study that show that the average Jo as no influence on political influence.
EBL hosted an extended debate on that paper: https://equalitybylot.com/2014/04/17/commentary-on-gilens-and-page-average-citizens-have-no-political-influence/
>It is not telling sortition is perfect
But it could be near to optimal if a) the mandate was limited to activities legitimised by the LLN and b) participation was quasi-mandatory. Looks like we all agree on b) [the only disagreement being whether to prioritise the stick or the carrot], and a) has been debated ad nauseam on this forum.
LikeLike