How would you govern Mars?

post by Bernhard · 2020-12-31T02:32:02.447Z · LW · GW · 14 comments

Summary: Good government is hard. Given the chance, how would you improve it? Below you will find one idea I like, to hopefully get a fruitful discussion started.

I started writing this in February, just as the Pandemic appeared on the horizon, and titled it "How would you govern Mars", hoping to get your attention (Of course Elon Musk beat me to it, but he is good at building hype after all). So now that I have your attention, here goes:

Government sucks. Demonstrably so, just watch the news. There may be some countries that are doing better than others, but most of them are either very rich (Switzerland, Norway, Singapore), and/or very small.

What I am interested in is how one would have to design a robust system that governs the day to day interactions between people, and that is able to evolve and adapt, and thus capable of serving the people now and in the future.

I'm not so much interested in any particular detail, and more in a method of how to identify and overcome newly arising issues (Not so much an updated reissue of the Ten Commandments, and more a system that generates useful commandments on demand). I'm, also not necessarily limiting this to "laws", and instead want to include "unwritten norms" and "culture" as well.

The system should be applicable to any (large, >>150) group of people, that wants to bootstrap a community. Possible examples would be the mentioned colony on Mars, or the Moon; an independent seastead floating somewhere in the Pacific; or maybe an existing country overcoming its former government (e.g. "The Capitalist Republic of North Korea", or "The Anarchic Free State of Portland").

By way of example, I'll start with listing a few high-level, long-standing issues that I think should be solved sooner rather than later. I'll also outline one idea that may address some of them. Then I would like to hear your opinions on what the important problems are, and how you would do it.

Problems:

Inequality: More unequal societies just do worse in general (I am not going to give an example/reference for this one. Instead I kindly suggest that you travel to new places once in a while)

Winner takes all: A special case of inequality. Exists on many levels (Country, state, corporation, personal)

Rules-as-written versus Rules-as-intended: Sometimes people do "bad" (e.g. negative-sum) things, without violating any laws (e.g. Martin Shkreli raising prices just because he can).

Responsibility: Powerful people commit crimes, and just walk away with no consequences (Neither legal, nor to their reputation) (one of a limitless number of current examples would be Kelly Loeffler). On a related note, I'd like to mention the concept behind the german word "Schlitzohr" (literally: "torn ear", meaning: "crook"). It refers to a visible form of punishment, that marks a dishonourable person for life, thought to have been inflicted by ripping out their guild earring.

Opaqueness: "Us down here" can neither see, nor influence what "Those up there" are doing. This perceived powerlessness erodes social capital, and trust in institutions. (see: conspiracy theories galore)

Polarization: People want to win and "destroy" their opponents, instead of finding a solution, or making a compromise. It's hard to see the "big picture" nowadays. Simple solutions presented by demagogues are often preferred. (e.g. immigration is a complicated issue, but a simple wall will solve it.)

Basic human needs: Their definition has shifted over the ages (as it should), but it is seldom kept up-to-date, or enforced. (Access to water is a human right? Good luck if you live in the Third World, or in Flint.)(How about access to air on Mars?)

Centralization: A few large, sluggish entities fare worse than many smaller ones, particularly in response to external stress. (This is true on many levels, e.g governments, corporations, creatures) (see also: Moloch [? · GW])

Coordination problems: Well this is a big one. For now I just mean the tragedy of the commons, and particularly externalities of everyday goods (Does the price of beef include the costs incurred by deforestation, pollution, etc.)

Legislation: When a new country is founded, people normally start writing a new version of "Our Constitution" (read: the best thing that humanity ever created). In practice, this means they just copy and paste an ancient text, and change a few things that they themselves are particularly interested in. Writing "normal" laws is barely better, even in the best case. (e.g. The Brexit deal mentions "modern" Netscape software. These people had years to come up with something, and in the end they invest as much time in finding a solution as a failing sixth-grader invests in homework). In the worst case, you end up with legislation written by lobbyists.

I think most people would agree on the "direction" of the required solution in each case. In my opinion, many of these problems are related to Law-thinking (as opposed to Toolbox-thinking [LW · GW]) that in practice just went wrong somehow.  To me the failure mode seems to be:

So instead of a perfect, top-down solution, here I want to present one possible just-good-enough solution to work on some of these problems.

Legislation-as-programming: A programmer invests as little time as possible when solving problems (or so I'm told). However, unlike a "lazy" politician, they are being "efficient", which means they neither write from scratch, nor do they copy just any old program. They instead import a library. A thematically grouped, prepackaged chunk of tested and mostly bug-free (or at least maintained) code.

My mental model right now, is that of laws being written and debugged just like code. Collaboratively, and using a version control system and wiki (let's say e.g. GitHub).

Why is this not a thing? While certainly not trivial to implement, this paradigm of legislation-as-programming certainly has advantages.

One particularly elegant idea (in my mind, at least) is a kind of community-curated news website but with a twist. Let's say a mashup of some parts of reddit, kickstarter, wikileaks and bellingcat.

I feel like this method takes humanities worst instincts and channels them into a (somewhat) better direction. It's certainly possible that it could degenerate into public show trials and vigilante justice, but at least it feeds the need for transparency and participation (both in the selection of important topics, and in their resolution), which will help build trust in the institutions.

One thing I like, is that this same procedure could be adopted for fictional events (e.g. which laws would be applicable if a president tries to preemptively pardon himself? There are none?! Well let's get to fixing that right away). People could be rewarded for finding (and disclosing) "bugs".

Similarly, it could be adopted for real events, but applying fictional, that is "forked", versions of laws. A campaigning politician would invite you to run your own test (e.g. the question of if you will have to pay more or less taxes) on his version of laws. And a crowd-sourced comparison of the differences between versions, should quickly find any hidden funny business.

Assuming that one could build a bare-bones version of the above, my hope is that it could be used to bootstrap the rest. If things function well enough, then a community might form organically (see: Wikipedia), and if not, then it wasn't a good enough solution anyway, no harm done.

Okay, so far the introduction to my attempt at de-Moloch-ification. I would be happy to hear your ideas, and comments. Is anybody aware of other attempts at creating a robust, more direct form of accountability? Or do you think it wouldn't work at all? What would you change, or how would you do it instead? Which tools are best suited? Which ones already exist (or can be repurposed), and which ones still have to be invented?

14 comments

Comments sorted by top scores.

comment by freyley · 2021-01-04T07:31:56.151Z · LW(p) · GW(p)

You need to differentiate the question of how law is managed from who has commit rights. Managing law as code, with patches and such, is an implementation detail. Current laws are actually written similar to git hashes - changes to the existing code that are then applied. That all of this is manual is not at all interesting, and automating it with git would not in any way change the fundamental power structures at play.

On the other hand, proposing that anyone can change the law would clearly be insane, just as large open source projects must have maintainers or go entirely off the rails. Currently you can call up your representative and propose a change to the law, they just will very very rarely bother to listen to you. Just like an open source project where the maintainer cares about their particular concerns, not yours. So the question is who has commit rights and how to manage them - in other words, it's fundamentally a question of political power and deciding who has it. 

Replies from: Bernhard
↑ comment by Bernhard · 2021-01-06T18:45:06.167Z · LW(p) · GW(p)

automating it with git would not in any way change the fundamental power structures at play

Well, currently, a lobbyist provides the desired changes, and your politician implements them. I am proposing to make it possible for everybody to propose changes easily.

proposing that anyone can change the law would clearly be insane

I agree. I was thinking more of something like wikipedia vs a classic encyclopedia. Many people determine what actually makes up wikipedia's content, but far fewer are in charge and oversee the final approval. As an example, in the past it would have been very hard to convince an editor of a classic (digital) encyclopedia to include a list of all Star Trek episodes. Nowadays, If you want one, you can write it yourself. And if it is factually correct, then it will probably be included.

So in short, the "masses" determine what is interesting, or needs fixing. They may also contribute ideas, or general directions as to how to fix it. The commit rights stay with the regular politicians.

Currently you can call up your representative and propose a change to the law, they just will very very rarely bother to listen to you.

Exactly. So by making it as easy as possible for everybody involved, things might get better. The most upvoted proposals, if implemented, make the most voters happy, which makes live easier for your local politician, if he ever runs out of interesting topics. 

Replies from: ChristianKl, Ericf
↑ comment by ChristianKl · 2021-01-06T21:39:04.075Z · LW(p) · GW(p)

Well, currently, a lobbyist provides the desired changes, and your politician implements them. I am proposing to make it possible for everybody to propose changes easily.

It's easy to propose changes in a representative democracy. That's what letters to politicians are for. They are not a tool that's limited to lobbyists. 

There are also other tools such as petitions or responding to requests for comments to engage with the process.

Many people determine what actually makes up wikipedia's content, but far fewer are in charge and oversee the final approval. 

Wikipedia has no process that's well described by the words "final approval".

As an example, in the past it would have been very hard to convince an editor of a classic (digital) encyclopedia to include a list of all Star Trek episodes. Nowadays, If you want one, you can write it yourself. And if it is factually correct, then it will probably be included.

That's misrepresents Wikipedia policy for most articles. "Factually correct" is not a category that Wikipedia cares about. It cares about whether the article is referenced with reliable sources and whether it's notable. 

Even if this would be how Wikipedia works, it's benefitial towards producing a quanity of articles. When it comes to laws you don't want quantity. You want as little laws as possible because with increased quantity of laws it gets more complex.

↑ comment by Ericf · 2021-01-06T21:38:42.646Z · LW(p) · GW(p)

A lobbyist is just a programmer for laws. "Everyone" already has the power to suggest changes, but only some people have taken the time to learn how to "code." The main issue is still "who gets the power?"

And Wikipedia suffers from the same coordination problem that plagues governments: if a small group values X highly, and 90% of everyone values "not X" just a little bit (or, more often,, would value "not X" if they were made aware of X), X gets implemented (because the people in charge don't know about the second group).

comment by ChristianKl · 2020-12-31T13:45:26.058Z · LW(p) · GW(p)

Hard cases make bad law is a general legal maxim. Your principles of constantly making up a bunch of new laws to solve possible problems raises the complexity.

Laws have complex interactions with the real world. They change people's incentives and get them to do different things. You can't reliably compute the effect of laws via unit tests. 

It can be "patched" more easily

This basically means it's easier for powerful people to change the law to their liking.

The Brexit deal mentions "modern" Netscape software. These people had years to come up with something, and in the end they invest as much time in finding a solution as a failing sixth-grader invests in homework

That seems like a bad interpretation of what happened. Writing a 1000 page document that deals with regulating everything and finding agreement is hard. The people who negotiated the deal likely focused their attention on the issues where they disagreed and thus didn't pay attention to the one page that talked about Netscape.

Replies from: Bernhard
↑ comment by Bernhard · 2021-01-06T19:20:01.693Z · LW(p) · GW(p)

Hard cases make bad law is a general legal maxim.

I agree.

I don't want to constantly create new laws, but instead constantly shine light on things that go wrong (This may or may not happen already, depending on which news you consume. Sadly however, mostly nothing comes of it, since change is hard). From there, if patterns emerge, then new laws should be proposed (For example, one pattern could be "unemployment in sector XY increases due to automation". For how many sectors should this have happened, before general action is warranted?)

You can't reliably compute the effect of laws via unit tests

I don't think there currently is an alternative that works. Politicians just claim that stuff will work, and either way, things just chug along. But for example, at the moment there are flesh and blood voters that think Trump lowered their taxes, when he did the opposite. A unit test like the one proposed above will work in that case.

I am also hoping that states in federally organized countries can copy each other. An example would be for example tax law or drug legalization in the US. Currently drugs are increasingly legalized. This could be a trivial change, just copy a state that does it well. Similarly, everybody seems to be fleeing California due to its high taxes.

In general, trying many different ideas, and letting the bad ones die, is quite an effective method of exploring a large number of options. Federalism in theory allows for this, if it weren't so sluggish.

and thus didn't pay attention to the one page that talked about Netscape

You may or may not suffer from a case of Gell-Mann Amnesia. The Brexit negotiations in general did not seem to me to be well-managed (or even well-intentioned). I don't think this is the exception to the rule. I still get your point though, but my answer again is, that if the process were easier, then it would probably work better.

Replies from: ChristianKl
↑ comment by ChristianKl · 2021-01-06T21:23:48.679Z · LW(p) · GW(p)

I don't want to constantly create new laws, but instead constantly shine light on things that go wrong 

The system you propose above is intended to fix things even when nothing went wrong (Trump didn't pardon himself).

I don't think there currently is an alternative that works. Politicians just claim that stuff will work, and either way, things just chug along.

Representative democracy allows accountability for the effects of laws that politicians pass. Evaluating politicians by their past actions instead of their promises for the future works much better. 

But for example, at the moment there are flesh and blood voters that think Trump lowered their taxes, when he did the opposite.

The idea that voters who don't get a good understanding of how the taxes they pay change from looking at their tax returns will get a good idea of that by them looking at unit tests on some government website sounds illusory to me.

I am also hoping that states in federally organized countries can copy each other. An example would be for example tax law or drug legalization in the US. [...] In general, trying many different ideas, and letting the bad ones die, is quite an effective method of exploring a large number of options. 

The whole point of unit tests is that it's a way not to try certain changes. 

You are essentially preventing that by getting people to focus on unit tests that can be made based on the text of the law.

If you evaluate laws based on an analysis of the law instead of an analysis of the effects on the law in empirical reality you reduce the amount that people learn from empirical reality. 

Currently drugs are increasingly legalized. This could be a trivial change, just copy a state that does it well. 

Do you seriously think that politicians passing new laws for drug legalization don't look at the existing laws to guide them?

The Brexit negotiations in general did not seem to me to be well-managed (or even well-intentioned).

The Brexit negotiations were dominated by power conflicts and as such the deal isn't good in a lot of respects. Both sides defects in the prisoner dilemma. 

I can completely understand bureaucrat from the UK to say to an EU bureaucrat: "I think the way the EU regulated IT security in 2008 was fine, let's just copy it over to our deal." Then the EU bureaucrat says "Fine, lets move on to one of the issue where we disagree"

Trade deals are not normal laws and the process that leads to them shouldn't be seen as normal laws. The process by which they are made is very different. 

You may or may not suffer from a case of Gell-Mann Amnesia. 

Gell-Mann Amnesia is about forming your opinions by reading the news. For me that's not where the foundation of my thinking about how law gets made comes from. My foundation rather comes from informal conversations which people who are involved in the process.

What kind of case do you have for not suffering from Gell-Mann Amnesia?

comment by YimbyGeorge (mardukofbabylon) · 2020-12-31T13:48:54.713Z · LW(p) · GW(p)

Use alignment of interests. The Captain must sink with the ship. 

Those in power and their immediate families  must be forced by law to use only the  services available to the lower classes while they are in office:

  1. Live in public housing.
  2. Use no private healthcare . Social healthcare is ok.
  3. Send their children only to state schools ( I think singapore enforces this).
  4. Mandatory conscription for their families if they send troops abroad.
  5. Use only public transport.

This will be a system of government where the fate of the elite is aligned with the lower half of the population. You cannot be allowed to take up a civil or political post if you do not agree. you can always resign and go back to your previous lifestyle at any point. I think that this way the standard of living will be raised for all. A lot of the elite  are not evil, they simply lack any undertanding  for the fate of those  at the bottom.  Being forced to live like common people  may help them develop some empathy.

To summarise - the design of incentives is just as important  as design of  the laws and structure of government.

Replies from: Viliam, Bernhard
↑ comment by Viliam · 2021-01-03T18:35:53.605Z · LW(p) · GW(p)

Possible objection is that this incentivizes people with power to hide from places you officially define as "people with power". Instead of becoming politicians, they will have their puppets elected as politicians... and the puppets will have to use mass transit and send their kids to state schools... and presumably get a reward that makes it worth all this suffering.

Counter-objection: this still adds another complication for the secret rulers, which may decrease their relative power compered to the non-secret rulers. That is probably good.

Another objection is that if the system works relatively okay, this motivates the rulers to keep it okay. But what if the system is broken in such way that fixing it would take a decade or more? For example, by the time you fix public education, it will be too late for your kids. Then the system would simply disincentivize people who care about their own kids (which doesn't necessarily imply they care about other kids).

↑ comment by Bernhard · 2021-01-06T19:46:31.701Z · LW(p) · GW(p)

I really like this, and agree that incentives are very important.

A related idea for example is to limit the pay of CEOs to X times that of the average/lowest salary of employees.

A trivial (as in too simple to actually work) toy example in the same vein is to continuously make the least powerful person the most powerful person.

That person than either helps themselves (consequently loosing office), or helps everybody at the bottom. Either way, inequality decreases (Yes, I know this won't work).

comment by Dagon · 2021-01-02T04:16:57.773Z · LW(p) · GW(p)

Why Mars for your thought experiment?  Currently, it's TRIVIAL to govern - there's nobody there.  For a long time yet, it'll be governed by military-like control structures, in order to keep people alive.  

By the time this question becomes interesting, the answers will start to look similar to Earth situations. 

On the object level, I'd probably go with a federated system - a world government to allocate resources to local groups, and let local groups govern however they like.  As long as resources are plentiful, guarantee right of exit for all people - they can emigrate anywhere that'll take them, paid for by the planetary tax fund.  

It's once resources get scarce that governance becomes interesting.  I don't know what to do about child labor, for instance, when that keeps the air breathable.

Replies from: ChristianKl, Bernhard
↑ comment by ChristianKl · 2021-01-02T12:11:34.074Z · LW(p) · GW(p)

A key question of Mars governance will be about what questions will be decided by people on earth and what will be decided by people on Mars. 

Replies from: Dagon
↑ comment by Dagon · 2021-01-02T17:46:37.553Z · LW(p) · GW(p)

Even that has earth parallels - colonies and low-resource areas that depend on outside economies cause interesting tensions.  Putting it on Mars makes it less interesting, actually, as we're fully guessing at the resource flows (and military threats) that drive the equilibrium.

↑ comment by Bernhard · 2021-01-06T19:59:15.252Z · LW(p) · GW(p)

Why Mars for your thought experiment?

I tried to think of a case where a "new" state is founded. On earth, every (dry) place is already owned, so that wouldn't have worked (without seceding). Seasteads might work nearly as well as space colonies. Assuming infinite monetary resources, in both cases it is conceivable to set up something completely disconnected from existing power structures.

In a way it's just a toy example, that disregards wars and other military actions.