kilobug comments on A Voting Puzzle, Some Political Science, and a Nerd Failure Mode - Less Wrong

88 Post author: ChrisHallquist 10 October 2013 02:10AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (180)

You are viewing a single comment's thread. Show more comments above.

Comment author: kilobug 10 October 2013 03:26:54PM 3 points [-]

Well, the whole point of instrumental rationality is that you need a correct map of reality (ie, to care about the truth) in order to be able to reach your goals whatever they are.

There is a strong signaling issue, appearing to be a conservative Christian can give a lot of political/social benefits in some circles, and it's easier to appear being one when you truly are one, but apart from that, having the belief of conservative Christian leads you to acts that are inefficient for reaching your goals, from rejecting your gay grandson to wasting time in prayer to not going to cryonics because you believe in afterlife.

Having a flawed map of a city means you'll not reach your goal efficiently (but either completely miss it, or use much more time/resources to finally reach it), and that's true whatever your goal is. The same is true with a flawed map of reality and navigating your life.

Even if you do not care about the truth for its own sake (if curiosity and preference for truth aren't in your terminal values), if you're intelligent, you should care about the truth as an instrumental value to reach whatever goal you truly have.

Comment author: Moss_Piglet 10 October 2013 04:35:36PM 7 points [-]

This is a good point, and holds in the majority of cases, although there are other considerations which should also be mentioned.

Since all maps are 'flawed' by definition, an important question is whether the flaws in your map actually interact with your goals, and if they do whether they are beneficial or harmful. It's usually not a good use of your energy to fine tune areas of your map which don't have any impact on your life and actively wasteful to "fix" them in ways which make it harder to achieve your goals.

Incorrect beliefs can be useful in the aggregate even if they fail in certain situations, as long as those situations are rare or inconsequential enough. I can be utterly wrong in my belief that there are no tigers in New York City (there are several in the Bronx Zoo, not to mention that more might well be kept illegally as pets) but it's completely orthogonal to my daily life and thus not important enough to spend effort investigating. And if I had a pathological fear of tigers, I would gain a pretty significant advantage from that same false belief; I would do well to maintain it even if presented with genuine counter-evidence.

I think that most religions are wrong to harmful degrees, but it's not an ironclad rule of rationality that beliefs must be maximally accurate. A pessimist is actually more accurate in their assessments of people, but optimists are happier and more successful; if your rationality insists you cannot be optimistic, then it is not useful and should be ignored.

Comment author: Lumifer 10 October 2013 03:59:36PM 7 points [-]

having the belief of conservative Christian leads you to acts that are inefficient for reaching your goals

I would submit that it rather depends on your goals.

Comment author: Desrtopa 10 October 2013 04:18:33PM 2 points [-]

That might be true, but you wouldn't be in much of a position to know whether it was true until you could conduct an unbiased analysis of your own motivations given a state where it was false versus a state where it was true.

Comment author: Lumifer 10 October 2013 04:40:54PM *  1 point [-]

But the same is true for everyone, isn't it? What's special about conservative Christians here?

Also, consider the goal "I want to be resurrected by Jesus and live forever".

Comment author: Desrtopa 10 October 2013 04:46:29PM 7 points [-]

Well, if it were really your goal to be resurrected by Jesus and live forever, and not just to be comforted by the belief that you were going to be resurrected by Jesus and live forever, then if Jesus didn't exist, it would be of prime importance for you to know that, since for there to be any chance of it happening at all, someone would have to make him.

Comment author: Lumifer 10 October 2013 05:10:09PM 0 points [-]

then if Jesus didn't exist, it would be of prime importance for you to know that

I am sure the fellow considered that possibility and rejected it :-) Or maybe he likes the Pascal's Wager.

In any case, getting back to the original issue, it was, to put it crudely, that Christians are necessarily stupid. That seems to be false on its face as there are a lot of people who believe in Jesus and are highly intelligent by all the usual measures of intelligence.

Comment author: Desrtopa 10 October 2013 05:25:03PM 3 points [-]

I am sure the fellow considered that possibility and rejected it :-)

Which is exactly the matter which, as Wedrifid pointed out, bears on the individual's intelligence and/or rationality.

In any case, getting back to the original issue, it was, to put it crudely, that Christians are necessarily stupid. That seems to be false on its face as there are a lot of people who believe in Jesus and are highly intelligent by all the usual measures of intelligence.

Nobody in this conversation made such a claim that I'm aware of. The point of contention originally raised in Wedrifid's comment was that religious conservatives may be intelligent and rational in spite of, rather than regardless of, their religious conservatism. That is, religious conservatism would be counterevidence to the overlap of intelligence and rationality.

Comment author: Lumifer 10 October 2013 05:42:40PM -1 points [-]

Nobody in this conversation made such a claim that I'm aware of.

I read wedrifid's post as stating that, in a bit more polite terms.

religious conservatism would be counterevidence to the overlap of intelligence and rationality.

So what does this actually mean? You see a girl, she looks intelligent and rational, you learn that she's a conservative Christian and you go "Oh, she isn't intelligent at all, my mistake"..?

Comment author: Desrtopa 10 October 2013 06:16:29PM 2 points [-]

How do you tell that she "looks intelligent and rational?"

If you have some other information that already screens off the evidence from knowing that she's a religious conservative, it doesn't adjust your probability, but if you don't, then you adjust your probability estimate that she falls into the overlap of "intelligent" and "rational" downwards.

If you know a particular human is three feet tall, but do not have access to other personal information about them, then it's possible they're an adult, but your best guess should be that they're probably not.

Comment author: Lumifer 10 October 2013 06:21:04PM -1 points [-]

How do you tell that she "looks intelligent and rational?"

By talking to her.

you adjust your probability estimate that she falls into the overlap of "intelligent" and "rational" downwards.

Would the downgrade from 99.999999% to 99.999998% be satisfactory? :-)

Comment author: wedrifid 10 October 2013 07:41:07PM *  3 points [-]

I read wedrifid's post as stating that, in a bit more polite terms.

I affirm Desrtopa's interpretation, as well as Eliezer's reminder about how conjunction works.

To reiterate: When you encounter "!(A AND B)" it does not mean "Let X equal whichever of !A and !B is most objectionable and claim that !(A AND B) is equivalent to X".

Comment author: pianoforte611 10 October 2013 08:09:16PM 0 points [-]

So what does this actually mean? You see a girl, she looks intelligent and rational, you learn that she's a conservative Christian and you go "Oh, she isn't intelligent at all, my mistake"..?

Not a very charitable interpretation. How about this instead: If someone is a conservative Christian then that fact makes it less likely that person is rational.

Similarly: If someone is deaf then it is less likely that they are a great pianist.

I can affirm that statement and still believe that Beethoven existed, without implying any insult to Beethoven.

Comment author: wedrifid 10 October 2013 07:12:54PM 3 points [-]

I would submit that it rather depends on your goals.

This is true in as much as the No Free Lunch theorem is true. As for the relevance to the beliefs and preferences of actual Christians, the testimony of the relevant religious texts, expressed beliefs of Christians and emphasis of Christian apologetics arguments do much to affirm that "long life of positive experience" is a goal that is in general shared by Christians. The "carrot" presented to reward belief is "eternal life". John 3:16 is the most famous quote from the Bible and the one used to express the core of Christian doctrine concisely:

For God so loved the world that he gave his one and only Son, that whoever believes in him shall not perish but have eternal life.

People, including Christians, tend to prefer long life---either in their physical body or after that body has been destroyed. If the beliefs of the Christian are false then the actions they choose when attempting to achieve this goal will fail.

Comment author: JQuinton 10 October 2013 05:35:27PM *  3 points [-]

I agree that having a correct map of reality is needed if you care about arriving at some (I hate this word) "materialist" goal, but not everyone can live in a more liberal area of the US/world. Areas where not fitting into the local community creates more burdens than necessary.

For example, when I was in the military, I identified as an atheist pretty openly, barely concealing my contempt for religion. One of my last Enlisted Performance Reports, my supervisor said I would do better if I were a Christian. I wasn't sure if that was a threat, or him implying that I was immoral and thus not quite a fit in the military community, but nevertheless I was penalized (in his eyes) for not being religious.

There's the map of reality, the map of the physical roads and buildings, but there's also the map of human interactions. One can signal consciously, but what I gather is that most people signal unconsciously and then generate a cached version of themselves that eventually becomes the real them (belief in belief and all that). While I might say I had an accurate map of reality when I was in the military, I didn't have, nor did I want to even accept, that there was another map of social situations that would also have helped me reach my goals more efficiently if I had an accurate map of it.

Comment author: WalterL 10 October 2013 06:11:43PM 1 point [-]

Hmm, its true that your map needs to be correct (in the sense that it corresponds to reality) in order to reach your destination, but it need not be wholly so.

Let's say I'm a firm believer that I must obey the laws written on a sheet of paper somewhere. I think they were written there by all powerful alien forces, obedience to whose dictates is the sole criteria for determining virtue. Any evidence to the contrary (say, the fact that no one has seen the sheet of paper) is part of the alien's test. Could you work with me?

It depends on what I think is on the paper.

If I believe that something approaching your, I dunno, call it ethics, are written on there, then it doesn't really matter to you too much why this is the case. The two of us ought to be able to cooperate (you holding your nose at my fundamentalism, me rolling my eyes at your relativism.).

My map of reality is wildly inaccurate at the edge of my neighborhood, but the two of us are only going to the drug store (voting on an issue we agree on).

I guess what I'm trying to say is that having a flawed map of a city only means you won't reach your goal if the flaws manifest themselves on the route between you and your goal. Flying Saucer Cult members still tie their shoes just fine (maybe, if you think they don't please substitute a task you believe that they routinely accomplish, perhaps donning their cult attire).

I believe that what you say about caring about truth as an instrumental value to reach my goal is true IFF my goal is essentially discovery-related. I need an accurate map when I venture beyond my neighborhood. But what if my goal is pretty much local?

I have a brother who is happily married, works to support his family, regular church goer, roots for his local sports teams, etc. He's pleasant and friendly to his many friends and associates, well respected. He learned what he needs to know for his job long ago, and is uninterested in improving at it (there's not much room to do so).

What does he need truth for? He is at his destination, his map was sufficient to reach it. How can truth improve his attending of church, or his time at the bar with his buds, or help him play with his kids? Its superfluous to his life.

Comment author: kilobug 11 October 2013 07:54:51AM *  2 points [-]

You may be fine with a flawed map, sure, if you're lucky. But then you're at a risk. Take your brother, what will happen if his kids end up gay or lesbian or polyamorist ? If he's really a conservative Christian, he'll believe his kids will burn in hell, you think that will make him "be at his destination" ? Or (and I really don't wish that to happen) if his wife ends up being pregnant again, but with a child having a severe disability that is detected at the early stages of pregnancy, will they abort ?

Yes, if your map is flawed in a location you never go, it doesn't matter much. But you're always at risk of potentially catastrophic failure if you do so.

Now, you may argue the risk of failure is low, and the cost of having an accurate isn't worth it... maybe, but there is no way to know that, to make that estimation, with a flawed map. You can't rationally chose to be biased, for to make that decision you've to know the truth already.

Comment author: Randy_M 08 November 2013 08:40:51PM 0 points [-]

"Or (and I really don't wish that to happen) if his wife ends up being pregnant again, but with a child having a severe disability that is detected at the early stages of pregnancy, will they abort ?"

Is it the "rationalist" position that it is advantageous or obligatory to do so?