In Defense of Politics

post by ChristianKl · 2020-04-10T19:26:50.747Z · LW · GW · 23 comments

I wrote most of the following post in 2013 and cleaned up the draft to publish it now. If the post feels a bit dated, that's why.

Politics is seen as the mindkiller. Alyssa Vance from http://rationalconspiracy.com argues:

Politics is suspicious as the best means to ends for other reasons. [...] It’s also zero-sum, and extremely competitive. So much human and social capital is going there already, yours probably won’t make a dent.

Is that true? Is it impossible for the very smart person without much resources to make a dent?

Years ago I was talking with someone from an East European country about how they demonstrate to get rid of their government. When asked for a detailed background on the reasons of why the particular government should go they said that the reasons aren't well described by any English source.

I suggested that instead of spending their time being the 10,001 person attending a demonstration they should write up the reasons why the government is bad in a detailed manner and publish it in the Guardian's Comment is free section.

In that case, knowledge of the politics of the country and good writing skills would have been enough to make a dent.

When it comes to whether their government steps down, obviously it matters how other countries see the situation in the country. For someone who thinks that the government should step down, it should be obvious that it's important that the best reasons why the government should step down should be available in English.

If you look at how Mubarak lost political power in Egypt, people in the West who thought their media informed them didn't really understand the situation.

They were unaware of the relevant politics. They didn't know that according to US embassy cables released by wikileaks:

Academics and civilian analysts painted a portrait of an Egyptian military in intellectual and social decline, whose officers have largely fallen out of society's elite ranks. They describe a disgruntled mid-level officer corps harshly critical of a defense minister they perceive as incompetent and valuing loyalty above skill in his subordinate

The Egyptian military runs part of the Egytian economy. The IMF pushed for typical Washington Consensus policies and as a result a lot of private business came into Egypt and made money. What happened to the military?

Lee Sustar wrote in February 2011 The roots of Egypt's uprising:

The top military brass, however, hasn't cashed in on Egypt's economic growth to the same extent as private businesses. Indeed, military officials see the privatization of state-owned enterprises as a threat to their economic standing and political clout.

If you look at that situations it much easier to understand why the Military didn't do something against the protesters that brought down Mubarak but now takes strong measures against the Muslim brotherhood. The West wanted to see the fall of Mubarak as a sign that democracy won instead of seeing it as a sign that the military got what they wanted.

Why does that matter to someone who wants to create political change? It shows that the general public is horribly informed about what goes on in the political sphere.

Before Bradley Manning submitted files to Wikileaks, Wikileaks dealt with information about corruption in Kenya. By leaking the Kroll report they contributed significantly to a different party winning the 2007 election.

Why do those examples from Egypt, Kenya and East Europe matter? You might say, of course people who spread information can change things in those countries but what about the USA?

If that sentiment resonates with you it should make you think about whether the USA is democratic in a meaningful way. Democracy is about citizens being able to affect the political process through free speech.

Why might you think such a thing? To be a politician in the US congress you need to gather a lot of money from donors. The Democratic party recommends that new US congressmen spend 4 hours of their day raising money. Fortunately, in practice the amount of time seems to be not that extreme. After all, fundraising isn't fun.

The problem is that it's not easy to reach politicians directly. A lot of people try it and it's going to be hard to get sufficient attention from a US congressman to explain to him in detail why a certain policy is wrong.

What options do we actually have to create political change? Julian Assange had arguably a large political impact given that he's a single person who's not that rich.

What's his self professed goal? Justice. Defining justice itself isn't an easy task. We can however observe that sometimes there are groups that do something that isn't in the interest of the broad public.

If one member of a group of 100 people thinks that the group is engaging in injustice that is a valuable data point. Groups where every member thinks the group is doing good should be able to outcompete groups where some members think the group engages in injustice.

That's where Assange comes into play. He wants to empower that single individual that thinks the group is injust. Assange also made the observation that if a group spends a large amount of resources on keeping certain information secret that corresponds to the harm that the group will suffer should the information become public.

He makes the further observation that if it's possible for a single individual who thinks that a group is unjust to bring down the whole group, members of the group won't share all information with each other anymore. When it becomes harder to share information inside the group the group has to effectively pay a secrecy tax. That means that the group is less effective in its battle against other groups in society that don't have to pay that tax because their members believe in their course.

Assange articulated that theory with using graph theory in a paper titled Conspiracy as Governance in 2006. From there he goes to accept every whistleblower that wants to contribute an internal document. He founded Wikileaks for that purpose. In some sense you could say that he took the idea that "everything that can be destroyed by truth, shall be" very seriously.

What is the problem with that political theory of change? It's not that it is ineffectual. It's that it is powerful enough that you get into problems. Assange made some powerful people really angry and as a result he spent years in the Ecudaorian embassy and is now imprisoned under conditions with are called torture in the Lancet.

I personally don't want to do that kind of politics. I don't want to attack other groups in a manner which endangers myself. That's my conscious choice. The founding fathers of the US payed the price of their victory in blood. But how do you do effective politics without harming yourself? To answer that question it's useful to look at the foundation of our current political orthodoxy.

Milton Friedman wrote in the 1982 preface of Capitalism and Freedom

Only a crisis—actual or perceived—produces real change. When that crisis occurs, the actions that are taken depend on the ideas that are lying around. That, I believe, is our basic function: to develop alternatives to existing policies, to keep them alive and available until the politically impossible becomes politically inevitable.

Milton Friedman developed a lot of political policy ideas that had a big impact. He knew that you can't just go and demand the government to change. Usually the government changes when there some form of crisis. When in crisis mode politicians want to be seen as doing something about the crisis.

As I wrote above, politicians don't really want to engage in hard work. That means that someone else often needs to have done the intellectual legwork beforehand. Politicans seek experts and then take the policy ideas from those experts.

Friedman and a lot of think tanks specialised into developing detailed policy proposals. These days there are think tanks that provide ready made bills. A congressman can just take the bill and copy paste it without having to do any work.

Thinks tanks have the advantage that they can spend money on feasible developing model bills.

We however have seen that Wikipedia managed to outcompete the Encyclopedia Britanica. There in principle nothing that stops a smart programmer from building a platform that provides for model bills that change society for the better.

The hard part will be around getting contributors to focus on practical effects of policies and move beyond idealism.

Some demands just aren't realistic and have no chance of being implemented. On the other hand I think there a lot of room for smart policies that are simply superior to existing policies.

Once you talk about whether different policy tools work, you also have a discussion that's a lot less mind-killed.

In building such a tool you have to make certain choices. You can build it as a hosted website. You can also build it as a distributed platform that runs on home computers.

Given that you don't want outside interference from NSA, you want a distributed architecture.

Why do these thoughts matter, even if you don't want to invest a significant amount of time into politics? Many readers of LessWrong are software engineers or move in the startup space. As such they are in a position to make important architecture choices. When deciding whether or not a software that you design relies on a central US server that stores all user data you are making a political choice.

I don't want to say that cloud architecture is inherently politically wrong but if the only reason you decide for a cloud architecture is because it's a cool buzzword, you are doing political evil. If you then complain that congress doesn't implement your pet political idea, you should look at yourself and judge the extent to which you are part of the problem.

Whenever you make big architecture choices in software that affect society you should think in detail about how that will change the power balance in society. Commenting on random reddit posts about the evil of software patents isn't being politically active.

To the extent that LessWrong is full of intelligent 20-somethings I would predict that in 10 years people of this group will be in positions to make influential architecture choices. The only way to make those responsible is to understand the political effects.

But what about those things that aren't in your direct sphere of effect? Think about policies. Even if you don't have a ready-made bill that a congressman can copy and paste, ideas about policy are important.

Zeitgeist Addendum is for example a political movie that opposes the status quo. It has some popularity according to some people who hate the status quo. Even if you disagree that the status quo is bad the policy ideas that the movie pushes are dangerous.

One of them is that all economic decisions should be made by a "scientifically" designed central computer who knows what's best for everyone. What about the problem of people not wanting to do what the central computer tells them? People only would do that if they are selfish, and people who get the right education won't be selfish and thus work for the common good and recognize that it's the common good to do what the "scientifically" designed central computer tells them to do.

Here we know about the problems of FAIs, however many people on the internet don't. Because of lack of well argued alternatives to the status quo they might follow insane ideologies like the one articulated in Zeitgeist Addendum. Unfortunately I have seen smart people advocate Zeitgeist Addendum as a means of being contrarian.

Mencius Moldbug effective at changing political opinions despite not writing in a way that optimized for attracting a large audience. Moldbug is right when he argues that most people don't understand politics. Moldbug solution of rolling back all social progress is still bad. You can meaningfully say that ending slavery was good. The case goes for legalising homosexuality.

I don't have a good answer of what our political system should look like. But I do think that's an important question. Discussing it is worthwhile. If we find a design that nicely fits together there are a lot of people who hate the status quo and who gladly take your political philosophy if they don't have to do the leg work of thinking up the fundamentals themselves.

23 comments

Comments sorted by top scores.

comment by beriukay · 2020-04-13T19:42:26.600Z · LW(p) · GW(p)

Just about a year ago, Assange was arrested, and is no longer at the embassy. https://www.bbc.com/news/uk-47891737

Replies from: ChristianKl
comment by ChristianKl · 2020-04-14T08:16:53.780Z · LW(p) · GW(p)

That part of the article was still written in 2013, I edited it now to make it up to date.

comment by Pattern · 2020-04-14T19:24:59.734Z · LW(p) · GW(p)
You can meaningfully say that ending slavery was good. The case goes for legalising homosexuality.

What do you mean by "meaningful"? In terms of values, I can see how such a statement can be accurate - but how is this different from ideology? Idealism?

We however have seen that Wikipedia managed to outcompete the Encyclopedia Britanica. There in principle nothing that stops a smart programmer from building a platform that provides for model bills that change society for the better.
The hard part will be around getting contributors to focus on practical effects of policies and move beyond idealism.
Replies from: ChristianKl
comment by ChristianKl · 2020-04-14T20:32:34.016Z · LW(p) · GW(p)

The policies behind legalizing homosexuality is to repeal the paragraph that criminalized it. That's a simple policy question. The amendment that ended slavery is also quite short. On the other hand a policy like Obamacare is >600 pages and much more complex.

It's one thing to say that you want to have a wealth tax. It's easy to hold that ideological position. It's quite hard to get down and write model legislation that would actually implement a wealth tax as you need to think hard through the mechanisms that you actually need.

I used both of the examples of slavery illegalization and homosexuality legalization because they are simple on the policy level.

Replies from: Pattern
comment by Pattern · 2020-04-15T04:43:01.658Z · LW(p) · GW(p)

I didn't say the policies were complex. Why do you think they are good?

Replies from: ChristianKl
comment by ChristianKl · 2020-04-15T07:16:00.800Z · LW(p) · GW(p)

Basically, you want to reduce a claim to good/bad instead of wanting to understand the point that's made. I don't see that as a productive discussion.

Replies from: Pattern
comment by Pattern · 2020-04-15T17:26:02.487Z · LW(p) · GW(p)

Your point has a foundation somewhere - perhaps it's "human flourishing is good" or "human freedom is good". I want to know what that foundation is. This allows not for "reducing a claim to bad/good" but seeing broader implications. For instance, if that foundation is "human flourishing is good" then that opens up the question "What policies that are in place now could be removed that would be good?"

Replies from: ChristianKl
comment by ChristianKl · 2020-04-15T20:27:32.525Z · LW(p) · GW(p)

My article is not about arguing for specific ends that should be achieved. It's about means of change.

Replies from: Pattern
comment by Pattern · 2020-04-16T01:02:14.265Z · LW(p) · GW(p)

I was not asking about specific ends, I was asking about motivations. What means should be used to accomplish an arbitrary goal is not clear in the abstract, though means of change may follow after having a direction.

Replies from: ChristianKl
comment by ChristianKl · 2020-04-16T06:51:18.491Z · LW(p) · GW(p)

It seems to me that you do a poor job to try to understand motivations if you focus on a single peripheral sentence near the end of the article instead of on my main arguments.

I don't believe that everyone should participate in the political process with the same motivations but I'm a believer in a pluralist society where different people have different motivations.

I have the impression that I want to have a political conversation on a level that's productive, while you want to have it more on the way most political conversations are traditionally done.

Replies from: Pattern
comment by Pattern · 2020-04-16T17:30:05.288Z · LW(p) · GW(p)

Your post does two things:

1. It tries to offer an abstract vision for the future.

2. It abstractly references what you want.

I don't believe that everyone should participate in the political process with the same motivations but I'm a believer in a pluralist society where different people have different motivations.

I think this is (more specifically) the answer to the question I have been asking throughout this thread - that this says what you want. (It's a little meta - it might be about what you want for everyone, as opposed to just yourself.)

I have the impression that I want to have a political conversation on a level that's productive,

And given what seems to be what you want, your lack of interest in how to get from is to ought (specifically) now makes sense - you don't know. What you want is general, not specific. (Maybe your vision is of a world where people who don't share the same vision work together, or where things get done even if people don't agree on everything.) You want to get from is to ought, not in one step (personally), but in two (with other people).


political conversation on a level that's productive

What was confusing about this, is that the post presents a negative vision.

Replies from: ChristianKl, ChristianKl
comment by ChristianKl · 2020-04-17T16:46:17.875Z · LW(p) · GW(p)

Thinking about this exchange a bit, it seems to me that the Kegan 4/5 split plays into this. I'm not at Kegan 4. I'm not operating out of a single system. If you model my as having a Kegan 4 motivation structure where a single system is driving me, your model will be off.

Replies from: Pattern
comment by Pattern · 2020-04-18T16:22:59.552Z · LW(p) · GW(p)

You said you wanted something. I asked why. It appears you didn't have an answer.* "I prefer means of change X" or "I don't prefer means of change X" seem like they probably aren't based directly on preferences about the means of change, but rather other things you value + ideas you have about consequences.

*Maybe you don't have an answer because:

  • "X is something I believe, but it is difficult to articulate." (because ethics/other things are complicated)
  • "I think people should focus on practical results, and I think the practical results of the policies I'm against would be bad."
  • "I am pointing out a current problem. I want solutions to the problem, not to extrapolate past 'progress' into the future. A process dedicated to creating value/making the world better should concern itself with creating value/making the world better, rather than concern itself with being an extrapolation of past trends."
  • "I want people to brainstorm solutions to this broad issue, rather than focusing on narrow political concerns and otherwise operating in (traditional) ways that aren't effective."


Moldbug solution of rolling back all social progress is still bad. You can meaningfully say that ending slavery was good. The case goes for legalising homosexuality.

I agree with this statement. I also believe that people who agree with it have other beliefs about how the future might be improved. Beliefs which might shed light on:

I don't have a good answer of what our political system should look like. But I do think that's an important question. Discussing it is worthwhile. If we find a design that nicely fits together there are a lot of people who hate the status quo and who gladly take your political philosophy if they don't have to do the leg work of thinking up the fundamentals themselves.

Perhaps I was jumping the gun, or you are interested in more people having shallower discussions, or focusing on what they want rather than why, because that is the best way to capture low-hanging fruit.

Replies from: ChristianKl
comment by ChristianKl · 2020-04-18T18:40:59.885Z · LW(p) · GW(p)
You said you wanted something. I asked why.

That's not true. For the point of illustrating a more central point that I care about I said that X is benefitial. The thing I wanted to achieve by mentioning the example was to stimulate useful thought in the reader.

Perhaps I was jumping the gun, or you are interested in more people having shallower discussions,

You were explicit about not being interested in trying to understand distinctions that are more central to my article then a particular point about a particular policy person of Molburg.

I'm interested in deep discussions of my central points but not in deep discussion of whatever a person who's not interested in the central points want to discuss.

If you want to have a deep discussion about how political action should be motivated, why not write a post that focuses on that?

Replies from: Pattern
comment by Pattern · 2020-04-18T21:48:01.709Z · LW(p) · GW(p)
If you want to have a deep discussion about how political action should be motivated, why not write a post that focuses on that?

In general, a fair point. Your post talks about two things. Let's call them 1. Discussing/calling for reasonable politics. 2. Ruling out Moldbug.

Comparatively, there was a post a ways back that said something like: 1. Freedom of speech, people can talk about whatever. 2.Unless they're nazis, then they should be banned.

Are there ways that can include 1 but exclude 2? Yes. I was asking which one of them it was. 1 reads like a call for ideas, idea generation, brainstorming. 2 is reducing things.

Both may be a part of finding/choosing solutions. But it isn't super clear what it is that is being ruled out in advance before brainstorming.

Loss of freedom? Pre-existing systems of thought (like communism, or the agendas of the left or right)? Things that are unethical? Politically untenable? Or things that don't emerge naturally as solutions to things that are considered problems?


I'm interested in deep discussions of my central points

Practical has to start somewhere. Is the post you made a call for political activism, or political thought (via blogging)?

Replies from: ChristianKl
comment by ChristianKl · 2020-04-23T19:40:50.195Z · LW(p) · GW(p)
Your post talks about two things. Let's call them 1. Discussing/calling for reasonable politics. 2. Ruling out Moldbug.

My post makes a lot more points then those two. The point about Moldbug is not one that I deeply developed. Besides proposing a specific way to do politics I spent effort to argue that the causes of the Egyptian revolution (a major political event) should be seen differently. I explained the philosophy behind Wikileaks and do believe it's valuable for people to understand it. Those are points that are developed over multiple paragraphes.

I don't think influencing people's opinion of Moldbug is important.

Unless they're nazis, then they should be banned.

I remember that case. I think there's a core difference between saying "nazis are bad" in an enviroment where most people believe "nazis are bad" and saying "nazis should be banned" at a venue which moderation policies that don't involve banning nazis.

Is the post you made a call for political activism, or political thought (via blogging)?

Those two don't feel like categories that I would use. It's the nature of political thought that isn't just play-acting to encourage action.

Replies from: Pattern
comment by Pattern · 2020-04-24T05:34:59.066Z · LW(p) · GW(p)
Those two don't feel like categories that I would use. It's the nature of political thought that isn't just play-acting to encourage action.

True. The action you wanted out of this post seemed to be coming up with figuring out action, rather than executing it:

If we find a design that nicely fits together there are a lot of people who hate the status quo and who gladly take your political philosophy if they don't have to do the leg work of thinking up the fundamentals themselves.

Something like political design/science/communication. Might have been a better choice of words, though I'm a bit fuzzy on which/what you're pointing at beyond "let's not rule out politics".

Replies from: ChristianKl
comment by ChristianKl · 2020-04-24T10:55:35.664Z · LW(p) · GW(p)

That's one example of things I pointed out as valuable in the post it's however not the only thing I argued to be valuable.

comment by ChristianKl · 2020-04-16T19:16:38.563Z · LW(p) · GW(p)
your lack of interest in how to get from is to ought

The people I know personally who do things that have political impact don't spend time discussing how to get from is to ought and the people I know who do discuss how to get from is to ought don't have any political impact.

I don't believe in action being grounded in abstract principles that you build up.

Replies from: Pattern
comment by Pattern · 2020-04-16T19:43:29.907Z · LW(p) · GW(p)

That makes sense. I'm curious about the positive aspects of your vision - though that does seem like the topic for another post.

comment by Chris_Leong · 2020-04-10T21:52:32.552Z · LW(p) · GW(p)
That's where Assange comes into play. He wants to empower that single individual that thinks the group is injust. Assange also made the observation that if a group spends a large amount of resources on keeping certain information secret that corresponds to the harm that the group will suffer should the information become public

The problem with it is unilaterists curse and the idealism involved in believing that everything should be public.

We however have seen that Wikipedia managed to outcompete the Encyclopedia Britanica. There in principle nothing that stops a smart programmer from building a platform that provides for model bills that change society for the better.

Hmm... I'd be surprised if this worked. In most cases there would be way too much disagreement.

Replies from: jessica.liu.taylor, ChristianKl
comment by jessicata (jessica.liu.taylor) · 2020-04-10T22:03:00.977Z · LW(p) · GW(p)

The presumption with conspiracies is that they are engaged in for some local benefit by the conspiracy at the detriment of the broader society. Hence, the "unilateralist's curse" is a blessing in this case, as the overestimation by one member of a conspiracy of their own utility in having the secret exposed, brings their estimation more in line with the estimation of the broader society, whose interests differ from those of the conspirators.

If differences between the interests of different groups were not a problem, then there would be no motive to form a conspiracy.

In general, I am quite annoyed at the idea of the unilateralist's curse being used as a general argument against the revelation of the truth, without careful checking of the correspondence between the decision theoretic model of the unilateralist's curse and the actual situation, which includes crime and conflict.

comment by ChristianKl · 2020-04-11T07:09:52.870Z · LW(p) · GW(p)

Disagreement doesn't prevent Wikipedia from having one website on every topic. You just need some process of finding consensus. Plenty of people don't like the consensus that Wikipedia finds, but that doesn't mean that there isn't useful output.

If we have an important topic like Supervulcano defense, few people have any idea about what to do about it and disagree. However if we had the draft legislation, it would be quite useful to pull it out in a time like this where we suddenly have the public more concerned with the risk of catastrophes.

There are plenty of different ways to create "consensus". You could have a process that looks like github where every project has a maintainer and if someone disagrees with the maintainer, they can fork.