Government Watch is a hypothetical NGO that popped into my head today.
Government Watch sponsors a large jury (100+ members), picked in a way as close to sortition from the whole population as possible. The jury discusses matters among itself and seeks testimony from experts and affected people on whatever it chooses to pursue. It monitors government business and the issues of the day. Ideally, the jury is paid an honorarium to make taking some time off work possible. Possibly multiple juries would run at once. Thus far, Government Watch is similar to several experimental “citizens juries” that have been set up. However, its function is different.
The jury’s role is to monitor government action and decisions and, where appropriate, for want of a better term “call bullshit”. “Calling bullshit” takes a two-thirds majority vote. The two-thirds majority is necessary because if the margin were slimmer, it might just reflect the random composition of that particular jury.
“Calling bullshit” is the jury’s way of saying that the government has taken a decision that there is no chance that the population as a whole would support if they understood the issue. Basically, it’s a way of saying “look here, the government has done something awful”. The jury (hopefully) won’t call bullshit on decisions they think are merely bad rather they will call bullshit only on decisions they think are indefensible. The kind of decisions that obviously only serve narrow sectional and partisan interest.
A good example of a decision that I hope, a Government Watch jury would call bullshit on is the decision to fold, time and again, to the tax complexity lobby. This is a real political grouping in America that lobbies to make income tax returns more painful.
The hope is that if you set this thing up, and did enough publicity for it, the fact that a Government Watch jury had called bullshit on something would, in and of itself, be news, and that would be a punishment, however slight, for the government pursuing policies on behalf of obvious special interests. It would be essential for the functioning of government watch that it be non-partisan, that its juries be picked as close to randomly from the population as possible, and that there not even be the appearance of anyone exercising undue influence over the juries.
The jury might choose to go a bit broader than just targeting special interests, and go after especially cruel, inconsistent, authoritarian or wasteful acts it considered indefensible. However the two thirds majority requirement, and the representation of the whole population, means that we can be pretty confident that anything the jury called bullshit on had serious problems, at the very least in how the government had explained its actions.
It’s not a cure-all. Most bad governance is better disguised than the tax-complexity lobby. However, I think there’s a sizeable and important niche here.
The counter against this charity: the "conservative silent majority" (or some massive latent outgroup) will be heard. This is akin to Curtis Yarvin's "Facebook Election" problem to differentiate civil democracy vs Aristotelian populism.
I love this idea! Sortition has always seemed like a potentially great system to fill exactly this kind of niche, because (like you said) it has such a potential to be non-partisan, common-sense, and generally good-faith. Actually getting a country's population and representatives to agree to such an initiative is usually the main issue, which this proposal elegantly sidesteps by making it an NGO.
A consideration: while sortition systems can reduce partisanship, there is no guarantee that it won't creep in and gridlock everything. There's a distinct possibility that Government Watch would simply split into two parties after a few years, which would make any two-thirds majority borderline impossible. Then again, something like 46% of Americans currently identify as Independents, so maybe that would keep things running.