Category Archives: 1

I like Scott’s post on what LessWrong has learned in its lifetime. In general I approve of looking back at your past misunderstandings and errors, and trying to figure out what you did wrong. This is often very hard, because it’s hard to remember or imagine what absurd thoughts (or absences of thought) could have produced your past misunderstandings. I think this is especially because nonsensical confusions and oversights tend to be less well-formed, and thus less organizable or memorable than e.g. coherent statements are.

In the spirit of understanding past errors, here is a list of errors which I think spring from a common meta-error. Some are mentioned in Scott’s post, some were mine, some are others’ (especially those who are a combination of smart and naive I think), a few are hypothetical:

• Because I believe agent-like behavior is obviously better than randomish reactions, I assume I am an agent (debunked!).
• Because I think it is good to be sad about the third world, and not good to be sad about not having enough vitamin B, I assume the former is why I am sad.
• Because I explicitly feel that racism is bad, I am presumably not racist.
• Because my mind contains a line of reasoning that suggests I should not update much against my own capabilities because I am female, presumably I do no such thing.
• Because I have formulated this argument that it is optimal for me to think about X-risks, I assume I am motivated to (also debunked on LW).
• Because I follow and endorse arguments against moral realism, and infer that on reflection I prefer to be a consequentialist, I assume I don’t have any strong moral feelings about incest.
• Because I have received sufficient evidence that I should believe Y, I presumably believe Y now.
• I don’t believe Y, and the only reason I endorse to not believe things is that you haven’t got enough evidence for them, therefore I must not have enough evidence to believe Y.
• Because I don’t understand the social role of Christmas, I presume I don’t enjoy it (note that this is a terrible failing of the outside view: none of those people merrily opening their presents understands the social role either).
• Because I don’t endorse the social role of drinking, I assume I don’t enjoy it.
• Because signaling sounds bad to me, I assume I don’t do it, or at least not as much as others.
• Because I know the cost of standing up is small (it must be, it’s so brief and painless!), this cannot be a substantial obstacle to going for a run (debunked!).
• I know good motives are better than bad motives, so presumably I’m motivated by good motives (unlike the bad people, who are presumably confused over whether good things are the things you should choose)
• I have determined that polyamory is a good idea and babies are a bad idea, therefore I don’t expect to feel jealousy or any inclination to procreate, in my relationships.

In general, I think the meta problem is failing to distinguish between endorsing a mental characteristic and having that characteristic. Not erroneously believing that the two are closely related, but actually just failing to notice there are two things that might not be the same.

It seems harder to make the same kind of errors with non-mental characteristics. Somehow it’s more obvious to people that saying you shouldn’t smoke is not the same as not smoking.

With mental characteristics however, you don’t know how your brain works much at all, and it’s not obvious what your beliefs and feelings are exactly. And your brain does produce explicit endorsements, so perhaps it is easy to identify those with the mental characteristics that the endorsements are closely related to. Note that explicitly recognizing this meta-error is different from it being integrated into your understanding.

An interview with the people of Cool Earth, a charity I investigated and ultimately recommended as a relatively good one while visiting GWWC last summer.

Rob and Federico of GWWC have made a nice summary of metacharities, which might be useful to those of you interested in cause prioritization as a cause. For future reference, I’ve put in my online collection of useful things (just for useful things which need help with their web presence).

I’ve also put up a few ‘structured cases’ for various claims there. These are intended to be well organized collections of arguments and considerations regarding a particular question, such as ‘should I invest in opening US borders?’. The claims of interest are mostly about how resources can be used to make the world better. The arguments are in progress, and probably not in particularly good formats, nor complete. However I talk to people about these often, so it is good to have them online to point to.

I’ve also put up some puzzles and other things.

Information and ideas percolate through society in many forms: via research papers, media publications, schools, advocacy campaigns, and perhaps most ubiquitously, private conversations.

Private conversations fulfill purposes other than information processing and transfer, so they cannot be expected to perfectly fulfill such roles, or even to fulfil them particularly well. They convey implicit information about the speakers and their qualities, they manifest social maneuvering, they embody humor and other good feelings, and they make a good circumstance for enjoying company more generally.

The information related roles that conversation can play most obviously include straightforwardly communicating information, and – in a more argumentative fashion – collaboratively figuring out what is true. Even if conversations are usually for other things as well, given that they are a major part of the social information  and dispersal system, one might wonder if they could do these jobs better when those roles are important. For instance, if you partake in ‘work’ conversations with the intention of productively progressing toward specific goals, should you be doing anything other than following your natural conversation instincts?

If you wanted to fulfill these information roles with conversations, here are some ways they seem to fall short in general:

1. They are rarely recorded or shared usefully. Which is bad because it means they must be repeated anew by many different groups of people, not to mention exactly the same people.
2. Relatedly, conversations seem to hardly build on one another over time. If I think of a good counterargument to your point, this won’t be available to any of the gazillion others having the same argument in the near future, because neither of us will do anything that makes it so, and both of us will probably forget all but the gist of the discussion by tomorrow. I posit that it is very hard to have a counterargument to a counterargument to a counterargument to an argument widely known, even when the argument and the counterargument are often repeated. Except when participants have a history or perhaps a shared subculture, each discussion basically starts the topic anew. A given discussion rarely gets through many considerations, and doesn’t leave the considerations it gets through in a state to be built upon by other conversations.
3. Discussions are often structured poorly for analysis, though perhaps ok for information transfer. It is natural for a discussion to take a fairly linear form, because only one sentence can be said at a time. But topics being discussed often don’t fit well in this form. A given statement has many potential counterarguments, and lots of possible pieces of supporting evidence, or vantage points from which to analyze it. The same is true of each of those arguments or supports. So a person may make a statement, then another person may offer a  criticism, and so on for a few levels, at which point one of the people will ‘lose the argument’ because he has no retort. But that corner of the tree was not necessarily critically important to the truth of the initial claim. If at the first level a different argument had been pursued, or different evidence had been offered, a largely unrelated path would have been taken, and someone else may have had the final say. If the parties do not remember the rest of the structure of their discussion, it is hard to go back to to a sensible juncture and hash out a different supporting claim. Usually they will just move on to something else that the last bit reminded them of, and in the future vaguely remember who won the argument, without further value being created. 
4. Relatedly, it is often unclear to the participants in a discussion what the overall structure of a discussion is, or how the parts relate to one another on even a small scale. For instance, which parts are important to carrying the point, and which parts are watertight. I find that when I write out an argument at length, in a structured way, I notice gaps that weren’t salient informally. And structured arguments look as if they help students reason more clearly.
5. Disagreement interacts badly with the social signaling purposes of conversation, as it tends to be taken as an aggressive move except when done skillfully. It’s not clear to me whether this is a fundamental problem with collaboratively critiquing ideas or a accident of the social norms we have, both for critique and for attacking people.
6. Similarly, allocating time in conversations tends to interact badly with social signaling. The person with the best point to listen to next is unlikely to always be the one who should talk next according to fairness, kindness, status, or volume. There have been some attempts to improve this.

This is not near exhaustive. Feel free to suggest more.

I am told that people have often tried to improve conversational norms, but I only know of a few such efforts. These are innovations such as randomized alarms while talking, hand gestures during seminars, argument mapping, and anonymous text conversation while everyone is in the room. I’d like to see a better survey of such attempts, but so far have not. Hopefully I just haven’t guessed the right keywords. Pointers would be appreciated.

For figuring out what is true, it seems many of these problems are resolved by writing a more permanent, public, well structured, outline of a topic of debate, then adding to it as you find new arguments. I think various people are in favor of such a thing, but I haven’t seen it done much (again, pointers appreciated). I have tried this a bit, with Paul Christiano, with the hope that it will either help somewhat, or allow us to figure out the problems with it.

Here are a few examples in progress:

1. The case for Cool Earth (linked before from my climate research)
2. US open borders advocacy
3. Animal activism

So far we don’t seem to have found any irrecoverable problems. However I’m curious to hear your thoughts on the merits of such interventions. Especially if you think this sort of thing is a terrible aid to discussion for the purpose of figuring things out.

Ben Kuhn makes some reasonable criticisms of the Effective Altruism movement. His central claim is that in the dichotomy of ‘really trying’ vs. ‘pretending to try’, EAs ‘pretend to really try’.

To be explicit, I understand these terms as follows:

‘Really trying’: directing all of your effort toward actions that you believe have the highest expected value in terms of the relevant goals

‘Pretending to try‘: choosing actions with the intention of giving observers the impression that you are trying.

‘Pretending to really try‘: choosing actions with the intention of giving observers the impression that you are trying, where the observers’ standards for identifying ‘trying’ are geared toward a ‘really trying’ model. e.g. they ask whether you are really putting in effort, and whether you are doing what should have highest expected value from your point of view.

Note the normative connotations. ‘Really trying’ is good, ‘pretending to try’ is not, and ‘pretending to really try’ is hypocritical, so better than being straight out bad, but sullied by the inconsistency.

I claim Effective Altruism should not shy away from pretending to try. It should strive to pretend to really try more convincingly, rather than striving to really try.

Why is this? Because Effective Altruism is a community, and the thing communities do well is modulating individual behavior through interactions with others in the community. Most actions a person takes as a result of being part of a community are pretty much going to be ‘pretending to try’ by construction. And such actions are worth having. If they are discouraged, the alternative will not be really trying. And pretending to try well is almost as good as really trying anyway.

Actions taken as a result of being in a community will be selected for being visible, because visible actions are the ones you will be able to pick up from others in the community. This doesn’t necessarily mean you are only pretending to try – it will just happen to look like pretending to try. But actions will also probably be visible, because by assumption you are driven to act in part by your community membership, and most ways such motivation can happen involve a possibility of others in the community being aware of your actions.

Many actions actions in communities are chosen because others are doing them, or others will approve of them, or because that’s what it seemed like a good community member would do, not because they were calculated to be best. And these dynamics help communities work and achieve their goals.

Of course people who were really trying would do better than those  who were only pretending to try. As long as they could determine what a person who was pretending would do, a person who really tries could just copy them where it is useful to have pretending-to-try type behaviors. But there is no option to make everyone into such zealous tryers that they can do as well as the effective altruism movement without the social motivations. The available options are just those of what to encourage and what to criticize. And people who are pretending to really try are hugely valuable, and should not be shamed and discouraged away.

A community of people not motivated by others seeing and appreciating their behavior, not concerned for whether they look like a real community member, and not modeling their behavior on the visible aspects of others’ behavior in the community would generally not be much of a community, and I think would do less well at pursuing their shared goals.

I don’t mean to say that ‘really trying’ is bad, or not a good goal for an individual person. But it is a hard goal for a community to usefully and truthfully have for many of its members, when so much of its power relies on people watching their neighbors and working to fit in. ‘Really trying’ is also a very hard thing to encourage others to do. If people heed your call to ‘really try’ and do the ‘really trying’ things you suggest, this will have been motivated by your criticisms, so seems more like a better quality of pretending to really try, than really trying itself. Unless your social pressure somehow pressured them to stop being motivated by social pressure.

I also fear that pretending to try, to various extents, is underestimated because we like to judge other people for their motives. A dedicated pretender, who feels socially compelled to integrate cutting edge behavioral recommendations into their pretense can be consequentially very valuable, regardless of their virtue on ethical accounts.

So I have argued that pretending to try is both inevitable and useful for communities. I think also that pretending to really try can do almost as well as really trying, as long as someone puts enough effort into identifying chinks in the mask. In the past, people could pretend to try just by giving some money to charity; but after it has been pointed out enough times that this won’t achieve their purported goals,  they have to pick up their act if they want anyone (themselves included) to believe they are trying. I think a lot of progress can come from improving the requirements one must meet to look like one is trying. This means pointing out concrete things that ‘really trying’ people would be doing, and it hopefully leads to pretending to really try, and then pretending to really really try. Honing the pretense, and making sure everyone knows what the current standards are, are important goals in a community with goals that might be forgotten if we undervalue pretending to really try.