If you were accused of a crime, who would you want deciding your fate?
- A panel of randomly selected jurors, all of whom have spent multiple weeks sitting in a courtroom, listening to all the relevant facts and arguments put forward by both sides
- A popular vote open to hundreds of thousands of people in your county, the vast majority of whom only know a few sound bites about the case, which they heard from a biased and one-sided source
The premise behind “Election by Jury” is simple: we believe that our government, just like the criminal justice system, will function better if our representatives are elected after weeks of deliberation by a panel of randomly selected jurors. These jurors would hear from the candidates and their expert-witnesses, deliberate among themselves, and cast their votes in secret.
Here are a few of the most compelling benefits of our proposal:
- An “electorate” that is better informed
- Better ways of combating misinformation
- Breaking away from echo-chambers
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
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 
Hi Clay,
It seems to me that the elections method you are proposing starts too late. The question of who gets to be a candidate that is evaluated by the jury does not get addressed. If the candidates are the candidates we have in our existing system then it seems almost beside the point which one of them eventually gets elected – they are all largely the same, all representing some elite interests.
We had a similar discussion on this website over a decade ago (in the context of a different proposed electoral system) – I feel that the points I made then still stand and are not addressed by your current proposal.
LikeLike
> The question of who gets to be a candidate that is evaluated by the jury does not get addressed.
first, this general umbrella of “ballot access” is largely orthogonal to the subject of election by jury, which is about who gets to vote (or HAS to vote), and what the requirements are for them to vote.
additionally, i consider ballot access be an incredibly minor concern. we already have plentiful examples where third parties (e.g. green, libertarian) are perfectly capable of making it on the ballot—but they have no chance of winning, because the choose-one plurality voting method means you’re “throwing away your vote” if you select them.
this is trivially solved via improved voting methods like approval voting and star voting. i co-founded the center for election science, which helped to get approval voting adopted in fargo and st louis, and we’re working on taking it to state levels. the more sophisticated star voting, of which i’m the co-inventor, will be on the ballot this may 21st in eugene, oregon.
https://www.starvoting.org/
we have extensive theoretical and empirical evidence (e.g. from exit polls) that the exact same voters massively increase their support for minor party and independent candidates when using these alternative methods.
https://electowiki.org/wiki/2012_Occupy_Wall_Street_polls
https://www.rangevoting.org/Maine2014Exit
this means these methods would almost certainly cause the usa to escape duopoly. i co-authored this piece arguing as such, hosted on the personal blog of ralph nader’s former running mate, matt gonzalez.
and a non-duopoly legislature would tend to pass legislation to liberalize onerous ballot access regulations.
so this entire issue is, in my view, already a solved problem, and we just need to keep advancing the proven electoral reforms i’ve spoken of.
election by jury is about the next link in the chain:
1. voter ignorance — the fact that most voters know virtually nothing about the issues or candidates (e.g. most voters couldn’t define “deadweight loss” or “tax incidence” to save their lives, yet they vote for the people who decide economic issues like “should we have a land value tax?” or “should we have a universal income?”)
2. non-compulsory voting means turnout is heavily demographically biased toward older/whiter/wealther and more conservative voters.
election by jury also means it’s okay to have a plethora of candidates running, because jurors have days or weeks or even months to exhaustively assess them, and “grade” their answers to complex policy questions, even if you get 30 people running for mayor/senator/etc.
sure, sortition also solves the “ballot access” issue inherently. but it comes with some very awkward problems. first, the legislators are not policy-crafting experts, and cannot spend years working on an issue, refining it, etc. because you can’t reasonably sequester a jury for years on end. these sorted members also aren’t people who inherently _want_ to craft legislation. i spend my days working on complex mapping software for a self-driving car company. i’m sure i could teach a random jury the basics of coding, but it would take years to get to proficiency, to say nothing of inherent algorithmic aptitude. it would be far more plausible that a random group of jurors could _assess_ who are the best software engineers, by hearing testimony from several candidates, watching them walk the jury through code and discuss algorithms, hearing experts cross examine them, etc.
another huge issue i see is that sortition doesn’t pick centroid representatives but is “proportional representation”. so if someone becomes ill or if they are allowed to opt out, then you get demographic disparities. sortition advocates have porposed absurdly complex algorithms to try to demographically rebalance, but aside from the complexity of such a process, it can only measure demographics the designers thought of, and only on approximate largely binary lines. you can say we’re going to ensure we have a reasonable mixture of “democrats” and “republicans”, or “men” and “women”, but these terms are not really binary—they actually lie on a spectrum. we mention this in our faq:
“It may be tempting to enforce demographic quotas based on race, gender, age, sexuality, etc. But this becomes harder and harder to implement in a bias-free manner, especially with the rise in multi-racial families, trans individuals, and the various intersections between all of these demographics. The most foolproof solution will always be to select jurors completely randomly, and have a large enough jury pool to achieve statistical robustness.”
it’s far more optimal to simply have a sufficiently large jury that you automatically get a closely representative sample of the whole population. then, by using a centroid-finding voting method such as score voting, you can ultimately select candidates at the approximate overall consensus position of the electorate. then if they resign, you can always back-fill with either the next-in-line “runner up” candidates from the original election, or hold a new election. that will work out without any significant demographic/ideological disparity, because you were already picking centrists.
LikeLiked by 1 person
Hi Yoram & Clay Shentrup, Roger Knights here. I’ve clicked the link to the 2013 discussion of, among other things, three new voting methods—i.e., Ranked Choice Voting, Score Voting, and Condorcet Voting. Trying to follow its intricacies and references has badly strained the old “bean” and put me thirsting after one of Jeeves’ pick-me-ups. So …
I suggest that instead of agreeing in advance on which method will deliver which sort of results, we agree to institute ALL the contending methods across all the initial jury electorates (say 9, so 3 per method) and decide which method we and the publc like best afterward. Methods with a low level of approval would be assigned to fewer than 3 juries in the next go-round. Etc.
Some such compromise agreement among reformers to avoid this stumbling block (IOW, kicking the can down the road) is probably the only way ANY alternative new voting method will get a real-world jury-trial.
——————-
Yoram wrote: “It seems to me that the elections method you [Clay] are proposing starts too late. The question of who gets to be a candidate that is evaluated by the jury does not get addressed. If the candidates are the candidates we have in our existing system then it seems almost beside the point which one of them eventually gets elected – they are all largely the same, all representing some elite interests.”
But scale matters. Drastically cutting the size of the electorate has dramatic effects on who gets to be a candidate and how good their prospects are. Ballot access obstacles (e.g., signature gathering) and campaign-cost-barriers dissolve, making campaigning and victory likelier for poorer, non-elite candidates.
In addition, sample-sized jury-electorates are likely to become better informed about the issues, and thus better aware of their true interests, than mass electorates. Presumably this ought to count against elite-interest candidates, and in favor of egalitarian ones.
Continuing this line of thought, the deeper a jury-electorate gets into learning about the issues and the candidates, the more closely its electoral choices will approximate the public’s general will. Therefore, it is desirable for such electorates to oversee the behavior of their elected officeholders throughout their whole terms and into the next election, regularly hearing from each one and his critics. (This can be done conveniently over a private intranet.)
And, of course, it is desirable for such electorates to elect only ONE officeholder, so that their attention is more focused and deeper.
LikeLiked by 1 person
Hey Yoram. I’m Clay’s co-founder and the author of the above text. I agree that there are many open questions remaining, such as the process we would use to decide which candidates get to present their case to the jury. The above proposal isn’t intended to fix every single problem, but rather, one specific problem: improving the way we “elect” a politician from among a small set of candidates.
I think there is great value in fixing one problem at a time, using an entire toolkit of different solutions for different problems.
That said, I have some ideas around how Jury-Elections can also be used to determine the “slate of candidates”. I think all political parties will be well served by replacing their Caucuses and Primaries with Jury-Elections. There could also be a multi-level Jury-Election for independents and 3rd parties (think Superbowl playoffs), the winner(s) of whom are “on the ballot” for the ultimate election. All of these ideas will allow for good leaders to present their case and get a fair hearing, without having to relying on billionaire donors.
– Rajiv
LikeLike
Hi y’all, Roger Knights here. The following two quotations implicitly illustrate how badly the enormous size and wide dispersion of modern electorates penalizes parties whose positions are outside the mainstream‘s Overton Window:
“… we have the discriminating tolerance today already, and what I want is to redress the balance. … The chance of influencing, in any effective way, this [monster-sized, diffuse] majority is at a price, in dollars, totally out of reach of the radical opposition.” —Herbert Marcuse, in Prophetic Politics, 1970, 109
“At present city bosses have little trouble, for example, fending off insurgents…. Only by the dint of the most heartbreaking labors can they even make their names known to the urban mass and even then they cannot win the city dwellers trust, for they have no opportunity to do so.” —Walter Karp, Indispensable Enemies: The politics of misrule in America, 1973, 308
A subtler way that DeMockery penalizes “outsider” parties is by its bundling of all issues, some of them with emotional power, into a pair of package deals. This enables a party win an election with a bundle that is not in the public interest, but that attracts votes because of its few key positions. These few key issues in effect “set the agenda” and displace meaningful discussion of ten or twenty issues of equal or greater importance, at least collectively. For instance “waving the bloody shirt” was a big issue-displacing vote-getter for decades after the Civil War.
“There exist a great number of potential conflicts in the community which cannot be developed because they are blotted out by stronger systems of antagonisms.” —E.E. Schattschneider, The Semi-Sovereign People: A realist’s view of democracy in America, 1960, 68
“The substitution of conflicts is the most devastating kind of political strategy.” —E.E. Schattschneider, Th e Semi-Sovereign People: A realist’s view of democracy in America, 1960, 74
But this can’t happen under Demiocracy because of its “topical firewalls”—i.e., its topic-focused legislatures and electorates.
“If it is true that the result of politics contests is determined by the scope of public involvement in conflicts, much that has been written about politics becomes nonsense. … The scope factor overthrows the familiar simplistic calculus based on the model of a tug of war of measurable forces.” —E.E. Schattschneider, The Semi-Sovereign People: A realist’s view of democracy in America, 1960
“A change in scope makes possible a new pattern of competition, a new balance of forces and a new result ….” —E.E. Schattschneider, The Semi-Sovereign People: A realist’s view of democracy in America, 1960
“Nobody knows what American politics would be like if we had the institutions to facilitate the development of a wider span of political competition.” —E.E. Schattschneider, The Semi-Sovereign People: A realist’s view of democracy in America, 1960, 72
LikeLike
Clay,
> ballot access … minor concern
As I wrote, the fact that the “credible” candidates all represent elite interests is the most fundamental feature of the electoralist system and its effects determine the outcomes of the entire system. Perceiving this fundamental matter as one of “ballot access” is a grave mistake. Once you have cut down the list of candidates to a “reasonable” size (i.e., to a size that is short enough so that the alternatives can be evaluated on a rational, substantive basis – say less than 1,000 candidates) you have already narrowed down the options to include only poor, unrepresentative candidates. The rest of the process (i.e., how the winner is picked from the short list) matters little at this stage.
LikeLike
Rajiv,
> I think there is great value in fixing one problem at a time
Indeed. However, the problems need to be prioritized. Focusing our energies on fixing a minor problem (picking among a short list of candidates) while ignoring the fundamental systemic problem (that all the candidates on the list represent elite interests) is surely misguided.
LikeLike
> I suggest that instead of agreeing in advance on which method will deliver which sort of results, we agree to institute ALL the contending methods across all the initial jury electorates (say 9, so 3 per method) and decide which method we and the publc like best afterward.
this doesn’t make sense, because there’s no way to read their minds and know which result they “liked best”. to do that, you’d have to USE A VOTING METHOD in the first place.
experts have already run voter satisfaction efficiency calculations on a variety of voting methods to know that score voting is statistically superior. so while you can’t know with certainty that score voting (or any other method) got the best result in a specific situation, you can know that it gives you the highest expected utility.
https://www.rangevoting.org/BayRegsFig
LikeLike
> the fact that the “credible” candidates all represent elite interests is the most fundamental feature of the electoralist system and its effects determine the outcomes of the entire system.
we have lots of evidence that approval voting and related methods massively decrease the impact of “indicators of electability”, thus making money and other methods the elites use to influence elections inherently less impactful.
https://www.rangevoting.org/Cash3
https://www.rangevoting.org/PuzzNEP.html
giving all candidates effectively “equal air time” via a lengthy jury examination process virtually nullifies this issue of elite influence altogether. and without the complexity and other downsides i described with sortition.
> Perceiving this fundamental matter as one of “ballot access” is a grave mistake.
well, it’s that plus a limited set of other things, such as name recognition (which requires ads, which requires money, which allows for elite influence) and strategic voting (which makes the perception of electability more important, which makes money more important). there are only so many mechanisms by which the electoral system invites elite influence.
> Once you have cut down the list of candidates to a “reasonable” size (i.e., to a size that is short enough so that the alternatives can be evaluated on a rational, substantive basis – say less than 1,000 candidates) you have already narrowed down the options to include only poor, unrepresentative candidates.
the quality and representativeness of the candidates obviously depends on the filtering process. i just enumerated the chief mechanisms by which that filtering process is biased by elite influence, and how our proposal mitigates those. so i see no reason to expect “poor unrepresentative” candidates.
and you don’t need 1000, or even 100, candidates to get qualified highly representative candidates. we could have 5-10 mayoral candidates in portland for our next election, and i’m certain the vast majority of voters would find at least one of them highly impressive.
> The rest of the process (i.e., how the winner is picked from the short list) matters little at this stage.
the available evidence says otherwise. exit polls with advanced voting methods show the exact same voters giving like 20x the support to third party and independent candidates when you simply let them use score voting, approval voting, etc.
LikeLike
If I understand correctly, you think that the filtering process works well (or can be made to work well) so that the field of millions of potential candidates can be narrowed down to a field of, say, 10 candidates at least one of which is a “good” one.
But then, if this filtering process is so effective, why not use it to narrow the field down to a single good candidate? Why is the final narrowing from 10 to 1 more difficult than the much more dramatic narrowing from 100,000,000 to 10?
LikeLike