Why random selection is necessary to create stable meritocratic institutions
131 comments
·July 14, 2025biomcgary
nashashmi
seems quite meritocratic with a pinch of (Lord's) randomness. The merit is "been with [Jesus] since baptism to taken"
retrac
The technical term is sortition. And it is my pet unorthodox political position. The legislature should be replaced with an assembly of citizens picked by lottery.
colmmacc
Ireland has a Citizens Assembly, which is selected by sortition. Ordinary citizens take time out of their lives to participate when assemblies are formed to examine issues of the day. The assembly receives expert and political testimony and evidence, and then votes and makes recommendations that often lead to country-wide referendums.
The process has been very successful at neutralizing contentious topics. The assembly on abortion showed that a healthy majority consensus could emerge, and led to abortion being legalized in Ireland after a constitutional amendment. The political parties generally support the process because it keeps socially divisive topics out of the main political sphere. Ireland also has relatively little money in politics, limits on donations, a standards in public office commission, independent constituency boundary commissions, a multi-seat proportional representation system, limits on media ownership, and the highest percentage of University educated citizens of any country. All in all it's helped Ireland come a long way from the 80s and 90s, when Ireland was much worse on corruption indexes.
eqvinox
My pet unorthodox position is also sortition, with an added (possibly transitionary) twist: hold elections and do it for non-voters, for a non-voter share of seats.
You can decide to vote, in which case you're removed from the sortition candidate pool. If you don't vote you're in the pool. A common representative body is formed at respective percentages.
This basically makes it so politicians have to race against "some random schmuck". If they can convince people they can do better, nice. Otherwise... too bad.
Of course some people will vote just to get out of the pool, but I think that's fine too.
gameman144
This may show that I'm biased, but the idea of a randomized group of citizens making the law of the land scares the heck out of me. There is a non-trivial amount of nuance and compromise that goes in lawmaking.
Now, the idea of electing a few thousand representatives and having sortition determine who is actually selected is something I could feasibly get behind.
woooooo
> There is a non-trivial amount of nuance and compromise that goes in lawmaking.
We just passed that "big beautiful bill" and it was quite clear nobody knew or cared what was in it, beyond it being "trump's bill he wants". I'm guessing staffers and lobbyists had a far more detailed understanding of their portions than any elected official did.
It's a reasonable guess that 100 randos would actually write a better bill.
shiandow
For what it is worth they probably wouldn't write the bill, just vote on it.
esafak
But the status quo is considered anomalous by most of the world, so I would not use it as a benchmark.
null
TimorousBestie
> Now, the idea of electing a few thousand representatives and having sortition determine who is actually selected is something I could feasibly get behind.
Since the linked article is to a substack called “Assemble America” I feel I should point out that if the apportionment House of Representatives had not been capped at 435 reps, the House would indeed be several thousand strong by now.
namlem
There are many proposed models for how to incorporate sortition into governance. Some examples:
- A randomly selected lower house with an elected upper house (or the reverse)
- policy juries which deliberate only on one specific piece of legislation, which then must be approved by a separate oversight jury before taking effect
- election by jury, where candidates are chosen by "elector juries" who interview and vet the candidates before selecting one
- multi-layer representative selection based on the Venetian model where randomly selected bodies elect representatives, of whom a random subset are chosen to then appoint officials
Right now the lottocratic/sortition-based bodies that exist are purely advisory, though in some places like Paris and Belgium they have gained a good amount of soft power.
It wouldn't be that hard to implement a conservative version of one of these in certain US states though. For example, add "elect by jury" to the ballot, where if it wins the plurality, a grand jury is convened to select the winner (counties in Georgia already use grand juries to appoint their boards of equalization, so there is precedent).
yesfitz
That's what bicameral legislatures[1] were meant to address.
Ideally, the lower house are representatives elected from the common people, and the upper house are the career politicians that understand how the government works.
In the U.S., the 17th amendment[2] changed that, for better or worse (probably both).
1: https://en.wikipedia.org/wiki/Bicameralism 2: https://en.wikipedia.org/wiki/Seventeenth_Amendment_to_the_U...
munificent
> the idea of a randomized group of citizens making the law of the land scares the heck out of me.
Here in the US, we use randomized groups of citizens to determine who gets locked away potentially for life or executed. Does a jury of peers also scare the heck out of you?
Supermancho
> Does a jury of peers also scare the heck out of you?
Those are screened.
Someone like https://youtu.be/00q5cax96yU?t=60 could be selected without some additional constraints than plain sortition. Ofc then those constraints are politicized.
gameman144
> Does a jury of peers also scare the heck out of you?
Honestly, yes. In the case of criminal culpability, it just happens to be the least scary of the available options of who gets to send someone to jail.
For lawmaking, this isn't the case: the work for lawmakers is much more detailed and gameable than a binary question of guilt.
connicpu
Regardless of how the average person may feel about it on a surface level, I think it's absolutely critical that congress has so many lawyers elected. These people write laws, we need people who actually understand the way law works doing that job.
namlem
Elected representatives do not write laws. Their legislative aides write the laws. While some state governments have highly professionalized legislative aides, in the federal government, such positions are typically poorly paid stepping stone jobs filled by people in their late 20s/early 30s who have little domain expertise.
f1shy
That is good for the form, OTOH the content (objet of the law, which is almost more important, one can argue) is more often that not, not related to the field where lawyers are experts (from sociology to engineering, through economics and medicine) that is typically handled by expert’s consultants, comities, etc.
So bottom line, I’m not so sure is so important that representatives are laywers. Maybe a good mix should be ok?
Ekaros
Maybe it is time to change how laws work if you need trained experts to understand them. Seems extremely harmful to everyone else who is not lawyer.
int_19h
In practice it creates a very strong incentive to write laws in a way that reinforces the "rule of lawyers", creating an exclusionary positive feedback loop.
pintxo
Given that legalese is still commonly prone to interpretation. I‘d rather have more Mathematics and Computer science people to ensure proper logic in the texts ;-)
almatabata
Aren't most of those lawyers the select few that can afford to go into politics?
vdqtp3
Our elected reps neither write nor even read the laws that are passed. Laws are written by lobbyists and aides, if we're lucky with direction from the representatives.
skrtskrt
Have you ever really paid attention to the members of the US House of Representatives?
There are some strong outliers but most are way below the bar of random selection. Do-nothing political nepo babies who are nothing but loud and in a gerrymandered district.
vkou
That's due to politics being a team sport, and everyone, including the voters, understanding that it's a team sport.
Getting your team control of a branch of government is way more important than having a 'good' rep in your district, because if you don't, they won't have any ability to do anything for it anyways.
If you couldn't get someone you wanted in the primaries, you just have to hold your nose, close your eyes, lie back, and vote for whomever made it through.
Whether this results in long term problems is a bit of an academic question, given that every election in the past decade is one where you either get to vote for the status quo, or an insane cult of personality.
keiferski
I think it could work well if you added two things:
1. A filtering mechanism after the selection process. E.g., basic civics questions like how many states are there, a background check, and so on. To make sure you don't pick anyone that's compromised or incapable of serving.
2. A training program that acclimates new members to the system. If terms are say, six years long, then the first year can be entirely devoted to training.
pintxo
I don’t see any need for that. There are enough weirdos in politics today that the weirdo rate might even go down when selecting people at random.
pintxo
This training thingy sounds sensible. But who controls the contents of the training? That body will have quite some power.
goda90
I saw someone on HN suggest the Supreme Court should just be randomly selected sets of federal judges on a case by case basis. Less opportunity for bribery and political games.
vannevar
It's an interesting idea, I've kicked something similar around with politically-minded friends for years. I don't know that a completely random group is the answer, but a hybrid approach might solve some critical problems:
- A largely unrecognized problem with our legislature here in the US is vote inflation: the number of representatives in Congress has fallen way behind the population growth, so that one rep is shared by a much larger group of constituents, devaluing the individual vote of each constituent and making it less likely that a given voter has a personal connection with their legislator.
- The increasing partisanship has reduced the number moderate and independent voices in the legislature.
We could increase the number of representatives in Congress by tripling the number of reps from each district, which would bring the rep-to-voter ratio back more in line with where it was when it was essentially frozen in 1929. Then one of those new reps would be chosen at random from a pool. Since the distribution of moderates in the general population is much higher than in Congress, this should have the effect of moderating partisanship.
hammock
In politics, sure. The way the headline is framed you can draw a similar parallel to genetic competition as well though. There are elements of both biodiversity and randomness required for successful genetic evolution
k__
Haha, mine too.
It would probably make sense to start with a new new "house" or something.
Might even make sense to have some quotas (at least 50% women etc.), so the whole things doesn't have to get to Chinese government size to reflect the populus.
That or pepple would have to be replaced with high frequency
pasquinelli
i've always thought a jury should be used instead of a supreme court. if the law people had their chance and couldn't settle an issue, kick it to the people.
programjames
> Directly select candidates at random for positions from an eligibility pool. Set and maintain the eligibility standard (such as an exam) by randomly selected oversight board to keep it updated and prevent the standard from being manipulated or gamed.
We don't want to discourage people from improving once they've met the bar. Learning a skill is often logarithmically distributed: it costs just as much to learn the first 50% as the next 25% and so on. At a minimum, to keep people cost-agnostic, we need
d/dx Pr(selected | didn't learn x%) ~ log(x%)
selection weight = [x log x - x + 1] * C
timdellinger
This perspective under-appreciates the role of a leader's charisma when it comes to attracting staff that will actually execute the ideas of that leader.
Anyone who has worked in a presidential administration (or a congressional office) can tell you that a leader is effective if and only if they have staff that believes in their message and agenda, and that is willing and able to execute on that agenda.
The practical reality here is that charisma isn't just a way of gaming the "getting elected" part of the job, it's also a requirement to be effective at the job.
braiamp
I think you didn't get to the part of how it would work in practice. It's not that the leader is selected randomly, it is that the people that select positions are randomly chosen. Also, your criticism only is valid if everyone through that being able to sell an idea is critical for the leader. The leader role is to manage the resources to accomplish the goal of the team, what the goal of the team is, is up to the team to decide.
Nicook
Article suffers a bit from the common hackernews intellectual bias.
ecshafer
Besides the curious absence of the word 'sortition'. Their historical examples are mostly totally wrong or missing key bits of nuance. Their example of picking the Doge of Venice misses that the convoluted process of "randomly" picking the doge isn't that random. They randomly choose electors only from the great families and randomly choose candidates from the great families and then choose. This is like if we chose the President my randomly choosing electors amongst the Senate, Governor, and the House, who would then choose candidates from amongst that same group, then randomly choose electors to decide amongst the candidates. Their example of hereditary monarchy assumes that murder and killing off competitors was common, however in European history that was pretty rare (instead putting them in the church was the way to thin the herd). If anything switch from gavelkind (all sons get a claim and split the lands between them) and going to a pure primogeniture succession greatly reduced said murdering and warring by reducing claimants.
My experience with KPIs also doesn't match the poster. KPIs are mostly ignored and it ends up going back to relationships and who has a better "deck" of accomplishments each year.
01HNNWZ0MV43FF
I think they're trying to keep it readable to people who don't know the terms.
I can say "sortition" and "ranked choice voting" and "LVT" and you'd understand what I mean, but to get a broad audience it pays to break it down into concrete ideas like "Random elections" and "More than two political parties" and "Why are we paying landlords to speculate on empty lots?"
wslh
> Besides the curious absence of the word 'sortition'.
Do you mean in common use? Wikipedia has a nice page on that [1]. There are also many papers on that [2].
[1] https://en.wikipedia.org/wiki/Sortition
[2] https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=sort...
ecshafer
The person just never says the word sortition. Its ends up feeling strange because it means either the person is trying to make this concept seem more their own, or they are that unaware.
torginus
* Campbell's Law (a variant of Goodhart's Law) states that the more a metric is used for social decision-making, the more it will be subject to corruption which distorts and corrupts not only the metric itself, but the very social processes it was meant to measure *
I just had a friend complain to me about LeetCode, saying that it's meaningless since everyone just mindlessly grinds the problem sets.
I pointed out to him that it's called studying for the test.
Nicook
that's true, but are you trying to measure people's ability to study for the test?
null
didibus
The obsession with meritocracy needs to be toned down a bit. In my opinion, the very idea of merit is fuzzy and lives right beside corruption and bias.
Merit is measured in imperfect ways, by other people, and fundamentally, we don't want a hierarchy of classes, even if we claim the higher rankings/elites have merited it.
Human dignity isn't contingent on outperforming others, and everyone would likely rather live somewhere that doesn't feel like constant competition is needed to enjoy leisure, food, shelter, pastimes, etc.
When it comes to who we should trust for critical work, taking decisions on our behalf, etc., we do want someone qualified. I find the idea of "qualification/qualified" much nicer than "merit". The latter seems to imply a deserved outsized reward, like it justifies not why you are given the responsibility of something important, but why you are allowed to be richer, higher ranking, etc., than others.
programjames
> and fundamentally, we don't want a hierarchy of classes, even if we claim the higher rankings/elites have merited it.
What do you mean by this? What creates a hierarchy of classes? Different social groups? Differing amounts of wealth? Different amounts of power to get stuff done? I think, in the end, it's got to come down to power, but I feel like it's good for society to distribute more power to people able to get better things done.
yesco
Meritocracy is simply a means of preventing elites from kicking the ladders down, nothing more, nothing less. Once the ladders are kicked down, which all elites will inevitablely try to do, society will start to stagnate, your country will start to fall behind the others, and your quality of life will start to rot.
The key here is that while meritocracy is championed as a means of finding the best, it in reality functions as a system to keep out the worst. You want harness the ambitions in people, even if not everyone's ambitions can actually be met, and you want to mitigate the harms of nepotism, even when eliminating it entirely is impossible.
So the difference between qualifications and merit evaluation are moot from my perspective, the question you need to ask is if whatever selection criteria you prefer is vulnerable to ladder kicking. If you preferred way is more vulnerable than the current system then you are putting the cart in front of the horse.
Also to make my position clear, I can't tell either way in regards to what you have suggested. As far as I was aware, we already select based on qualifications, so it's unclear to me what the exact change you are proposing is.
samdung
The Complicated Business of Electing a Doge
throw0101c
From the Wikipedia § Criticisms page:
> In his 2019 book The Meritocracy Trap, Daniel Markovits poses that meritocracy is responsible for the exacerbation of social stratification, to the detriment of much of the general population. He introduces the idea of "snowball inequality", a perpetually widening gap between elite workers and members of the middle class. While the elite obtain exclusive positions thanks to their wealth of demonstrated merit, they occupy jobs and oust middle class workers from the core of economic events. The elites use their high earnings to secure the best education for their own children, so that they may enter the world of work with a competitive advantage over those who did not have the same opportunities. Thus, the cycle continues with each generation.
* https://en.wikipedia.org/wiki/Meritocracy#Books
> In his book The Tyranny of Merit: What's Become of the Common Good?, the American political philosopher Michael Sandel argues that the meritocratic ideal has become a moral and political problem for contemporary Western societies. He contends that the meritocratic belief that personal success is solely based on individual merit and effort has led to a neglection of the common good, the erosion of solidarity, and the rise of inequality. Sandel's criticism concerns the widespread notion that those who achieve success deserve it because of their intelligence, talent and effort. Instead, he argues that this belief is flawed since it ignores the role of luck and external circumstances, such as social and external factors, which are beyond an individual's control.[91]
* Ibid
lapcat
I support the idea of sortition, which appears to be guiding idea behind "Assembling America". However, I'm not quite sure what this has to do with meritocracy.
From my perspective, the fundamental justification for sortition is that randomly selected citizens are more representative of the general public and, crucially, less corrupt and corruptible on average than elected representatives.
Why less corrupt? Because I think people who seek power are more corrupt and self-centered on average than those who have power thrust upon them. Why less corruptible? Because randomly selected citizens don't have to fundraise for political campaigns, and they are merely temporary occupants of their seats, not running for reelection and becoming career politicians. As far as I'm concerned, political campaign contributions are legalized bribery. It would be easier to police citizen legislator corruption, because we allow crap from elected officials—campaign contributions, gifted travel, post-legislator lobbying jobs—that we really should make totally illegally and jailable. A lot of "working class" politicians suddenly become super-wealthy after leaving office, and we all know it's quid pro quo. Just outright ban that crap and strictly audit former legislators.
hiAndrewQuinn
The underlying assumption here seems to be that there is no or even negative value in someone actively specializing their labor into politics, and I just don't think that's true. To the extent we have to "do politics" at all [1], it's probably best handled by the people who have dedicated their lives to becoming politicians, the same way that getting your house wired is probably best done by someone who spent their life becoming an electrician.
In fact, if anything, this system seems like it would be even easier to game compared to the status quo. If you select truly at random from the population you're going to pull a lot of people with not a lot in the way of resources, making for a very easy to bribe block, even if you have to repeat the bribes every few years as people shuffle through. If you don't - if you select randomly from, say, only the group of people who got perfect scores on the SATs, or from white land owning males - you're practically begging for tacit collusion as they realize they have essentially the same power that HOAs do when it comes to what we'll do next. Democratically elected politicians at least have enough sense to understand they have to balance their short run desires with their long run interests in continuing to be democratically elected politicians.
[1]: Which I don't admit we should in the first place, cf https://spot.colorado.edu/~huemer/papers/passivity.htm for one reason why.
LeifCarrotson
> best handled by the people who have dedicated their lives to becoming politicians, the same way that getting your house wired is probably best done by someone who spent their life becoming an electrician
Being an electrician makes you good at wiring houses in ways that work, that pass code inspections, and that don't burn down. The feedback loop isn't perfect (you're likely to succeed for a while if you produce flawed work fast that looks good enough to your boss), but it's at least feeding back in the right direction.
Being a politician makes you good at different things - fundraising, advertising, speeches, getting your name in the news - which are totally unrelated or even opposed to creating and executing legislation that is good for society. Sortition says that this relationship is so bad that the outcome under a lottery (the 50th percentile, eliminating the 49% of the population who would be better than average at the job) results in better outcomes than career politicians.
lapcat
> If you select truly at random from the population you're going to pull a lot of people with not a lot in the way of resources, making for a very easy to bribe block, even if you have to repeat the bribes every few years as people shuffle through.
This is incorrect: elected politicians are much easier to bribe, because bribery of them is totally legal via campaign contributions. It's both expected and indeed necessary for politicians to ask for and take large amounts of money from others for their job.
Policing corruption of randomly selected citizens would be much easier, because the expectation is that none of them would be asking for money or accepting money for their jobs. With strict auditing, anything out of the ordinary would be pretty easy to spot. The problem with the current system is that vast transfers of money to legislators are perfectly ordinary.
Also, with random selection, the odds are higher of finding one or more inherently honest and ethical people who will blow the whistle if there's some kind of mass bribery scheme. But our current pay-to-play election system is a mass bribery scheme. Ask any politician how much time they spend fundraising: it's just a crazy % of their time. You may think politicians are lazy because they take so many breaks from legislating, but they're actually taking breaks to go out and fundraise.
Anyway, I think it's a misconception that poorer people are easier to bribe than richer people. It's also a misconception that richer people are "more successful". In my experience, richer people tend to be more obsessed with money. Many average people just want to be happy, have a family, have friends, enjoy life. They are satisfied with what they have. The only purpose of their job is to make it possible for them to go home from their job. Whereas people at the top never seem to be satisfied with what they have and always want more, more, more.
int_19h
What does it mean to be "good at doing politics", though?
In a representative democracy, because of the very nature of the selection process at hand, it means "getting elected at all costs". Which is not all the same - and in many cases directly counter to - the desired goal of "governing well".
namlem
The French government and private interest groups alike attempted to manipulate the Citizens Convention for Climate back in 2019 and were not successful fwiw. When lobbyists tried to approach delegates outside the convention, they were quickly snitched on. Existing legal frameworks for preventing corruption among jurors and elected officials should suffice to protect assemblies from similar influence attempts.
hiAndrewQuinn
Would you necessarily know if they were successful? Can you actually prove that not a single person in that convention accepted some kind of kickback for e.g. changing their vote?
Mechanisms that effectively prevent this do exist in the literature, to be clear, but I rarely hear of those ones actually getting implemented.
namlem
Well we mostly know what positions these groups were pushing for. It's possible that some influence went unnoticed.
That said, the US used to have quite a lot of juror bribery in the late 1800s and managed to successfully crack down on it with harsh penalties, sting operations, and other strategies. Attempting to bribe a juror can get you 15 years in federal prison in the US, it's not taken lightly.
There is an interesting example of random selection of leadership from the Bible when the apostles replaced Judas. The criteria were agreed upon and then lots drawn.
Acts 1:21-26 Therefore it is necessary to choose one of the men who have been with us the whole time the Lord Jesus was living among us, beginning from John’s baptism to the time when Jesus was taken up from us. For one of these must become a witness with us of his resurrection.” So they nominated two men: Joseph called Barsabbas (also known as Justus) and Matthias. Then they prayed, “Lord, you know everyone’s heart. Show us which of these two you have chosen to take over this apostolic ministry, which Judas left to go where he belongs.” Then they cast lots, and the lot fell to Matthias; so he was added to the eleven apostles.
Can you imagine this practice replacing the Papal conclave? Or, pastor selection at your favorite Protestant group?