I wrote the following on my blog last night. I thought that I'd run it past an intelligent audience. Note that what I have referred to as an idea is what we here at lesswrong would call a 'belief'. I changed the name to remove any strange foggy baggage that might appear in the heads of potential readers who are not familier with belief vs belief-in-belief and other concepts like that.

What are your thoughts?

I recently got into a discussion on Facebook that started with an assertion that free-thought/atheism/humanism/etc was no different than the certainties of fundamentalism. But that discussion moved into many topics, one of which is why it should not be controversial to assert that one idea can be more 'right' than another.


I asserted that the view that the universe was created 13.72 billion years ago was more 'right' than the view that the universe was sneezed out of a giant space cow. My interlocutor felt that the giant space cow could be 'right' for one person, even if it is not 'right' for others.

It is at this point that we ran into a problem, as it became apparent that her view of the meaning of 'right' and my own were different. As best I can tell, she felt that 'right' meant that it feels right or brings comfort. I, of course, use the word 'right' to mean that an idea contains explanatory and predictive power. The idea that the universe started in a big bang explains a lot about what we see in the cosmos and predicts what we will see as we keep looking - with a high degree of accuracy. That makes is 'right'. And it's more 'right' now than it was two decades ago because we have found places where it is 'wrong' (the increase in the rate of expansion of the universe) and revised the idea to explain it (dark energy), making it more 'right'.

So we had two versions of 'right'. It wasn't a given that she would accept my version, so I had to come up with a good reason why my version of 'right' was, well, 'right'.

What I came up with was to point out an ethical imperative to be as 'right' as possible - using my definition. Consider this: If there are two ideas, one of which is 'right' enough to predict certain unintended consequences of an action that the other idea fails to predict. If you consciously choose the less 'right' of the two ideas (perhaps because it is 'more right for you'), you have consciously chosen to risk harming others in ways that could have been prevented by choosing the more 'right' of the two options.

Perhaps it will be clearer with an example:  Sally is worried about vaccinating herself before travelling to another country. She knows that the doctor says that it is necessary and safer than not being vaccinated. But she's also heard some bad stories about the side-effects of vaccinations. She decides that not vaccinating is 'right for her'. After all, if she's wrong, what's the harm? She might get sick, but that's a fate she brought on herself. What she fails to realise is that the more 'right' idea (that vaccines are safer than not having them) also predicts that if she fails to vaccinate, she can bring those diseases back to Australia and infect others.

But this isn't just a problem on the left wing: Consider the case of Josephine, who concedes that there is little evidence for the existence of an afterlife. But she chooses to believe anyway because it is 'right for her'. Why not? If she's wrong, she'll never know it because she'll have ceased to have existed. But here comes those unintended consequences again. This time, they come in the form of predictions that the less 'right' of the two ideas makes - that death is not the end of all existence, but a transition to a greater existence. As it happens, Josephine is an Australian senator and is about to vote on an authorisation for the ADF to bomb some village  in Afghanistan. She briefly worries about the fate of any innocent bystanders but is comforted by the fact if the ADF's aim is off, any innocents will go to heaven. But she's so busy that she fails to remember that her assumption about heaven was an arbitrary one for her own comfort and shouldn't be used outside the confines of her own skull.

Now consider this case: The CEO of a company sees credible evidence that the government is about to change, leading to a major change in policies directly affecting the company's business environment. She should probably hedge her bets to prepare for the likely change. But what if she really liked the current government? What if the prospective change to the opposition caused her distress? Might she choose to believe that the government will almost certainly win the next election because that idea feels 'right for her'? I would suggest that the stock holders would feel that her due diligence required her to hedge the company's bets, whatever her feelings.

But change this CEO to a mother making a choice on matters of vaccines or faith healing, and now she hasn't made any kind of ethical lapse - she's has just exercised her faith. We owe it ourselves and to those who are affected by our actions (which is everyone really) to try to be as 'right' as possible as often as possible. Never chose an idea because it is 'right for you'. And always be on the lookout for ideas that are even more 'right' than the ones you already cling to.

16

New Comment
40 comments, sorted by Click to highlight new comments since: Today at 5:28 AM

This may be better received in Discussion rather than Main.

Note that the post is now in Discussion.

That was my immediate reaction as well

Nice read. That's exactly what I should do. Someone else made a similar comment about definition issues. In retrospect, the conversation on Facebook could have been wrapped up very fast.

See Clifford's "The Ethics of Belief" for a classic statement of this argument.

Nice. That seems to state my point far better.

the universe was created 13.72 billion years ago

This is non-scientific when interpreted literally. You may wish to revise your phrasing.

The blog post was aimed at normal people. It was trying to do the whole only-one-inferential-step-beyond-their-thoughts thing. I'm well aware that the big bang was simply what we call the transition of our region of the universe from an inflation phase into whatever it is in now.

Besides, the statement in the posting is still more accurate than the one that it was paired with, which was exactly my point.

What might have been a better phrasing that stayed away from words that would scare off laypeople?

Please note the two statements you paired together are not mutually exclusive, as "was created" is passive voice, allowing any particular creator, including giant sneezing space cows. And since we don't know how the known universe "was created", it's an interesting (inappropriate?) choice for showing variable certainty in knowledge, even though you're going for fallacy of gray. I'd suggest a simpler, less controversial metaphor to introduce the idea, then extend it to the desired topic.

I'd disagree that I was going for the fallacy of gray. The fallacy of gray is replacing a two coloured world (black & white) with a 1 coloured world (gray). The post you linked to goes on to say that it is quite appropriate to point out that there is such thing as 'less white' and 'more white' - in fact, a world with millions of shades. It's a great antidote to two-colour thinking.

I didn't mean you were committing the fallacy, I meant you were similarly trying to point it out ("going for this kind of article and explanation").

Ah. In that case, your comment makes a lot of sense. I apologise for the confusion.

The blog post was aimed at normal people.

So you chose to say something only partly right, because the complete truth would scare 'normal' people? I question the effectiveness of this strategy when arguing for an objective standard of right-ness.

What might have been a better phrasing that stayed away from words that would scare off laypeople?

'The universe as we know it came into existence X years ago". No need to say "was created" or bring in "what came before?" into the conversation at all, as it's not the main subject, just an example.

Yes. I chose to say something only partly right. If I was talking to a creationist, I would suggest that maybe god could have used natural processes to create the world - because if I told them that there was no evidence for a god's action on the universe, they'd assume that I was doing the devil's work and not listen.

If I ended a conversation with my interlocutor's beliefs now being one step closer to the truth, I would feel like I'd done a good job. I can always shift them again next time around.

I take your point about the alternative phrasing. I don't think that that would have undermined my point, so I should have used it.

I would recommend starting with the blatantly obvious. For example, it is "right" that stabbing yourself will cause you to die horribly. If someone decides that it is "right for them" that stabbing themselves would instead result in them winning the lottery, they will make a very bad decision.

Also, your examples for how belief in heaven could hurt you aren't the most common ones. I'd list things like abortion, fornication, drug use, and any other issues theists and atheists tend to disagree on. Even if your decision only effects you, it's still important, as with the stabbing yourself example.

She briefly worries about the fate of any innocent bystanders but is comforted by the fact if the ADF's aim is off, any innocents will go to heaven. But she's so busy that she fails to remember that her assumption about heaven was an arbitrary one for her own comfort and shouldn't be used outside the confines of her own skull.

What your evidence for the fact that believers actually reason that way? There are Christians who believe that everyone who isn't a Christian will go to hell after he dies. If you were right those Christians would have less of a problem killing fellow Christians than killing Muslims.

In the real world things don't work that way. The belief works differently than you propose.

Josephine won't tell you that she makes her decision to bomb the village because she considers the average Afghan life to be worth a lot less than the average Austrialian life. She might not even admit it to herself.

What if the prospective change to the opposition caused her distress? Might she choose to believe that the government will almost certainly win the next election because that idea feels 'right for her'?

But change this CEO to a mother making a choice on matters of vaccines or faith healing, and now she hasn't made any kind of ethical lapse

There are very good reasons why the behavior of a CEO is more constrained than the behavior of a mother. A CEO gets passed very specific duties via a contract. He has to act according to what the societal consensus consideres to be right. A mother on the other hand has a lot more freedom to deviate from the consensus.

I don't see how the example of the CEO adds anything to your argument.

I don't think that this post will convince anyone who doesn't already share your own notion of what 'right' happens to mean. It doesn't really address any concerns of someone who has a different concept of what 'right' means.

The word "right" may not have caused any confusion. It should be obvious to anyone that "right", applied to a factual assessment, is appreciating the correctness of the statement, instead of condoning the underlying facts. Replace "right" by "true", and you'll run in exactly the same problems (I did).

What do you mean by "should be" in the sentence "It should be obvious to anyone ..."?

Oh crap. I meant it as a factual statement: "It is obvious to nearly everyone". I guess it is less obvious than I thought.

Luckily, in French, we don't have a word as overloaded as "right". We have "true", and we have "good", but we don't have "right".

So let me update my statement a bit: Replacing "right" by "true" won't make the problem entirely go away.

So let me update my statement a bit: Replacing "right" by "true" won't make the problem entirely go away.

Sure. For instance, the Scientology folks teach "what is true, is what is true for you" — but they still go around making pretty strong claims that what is true for Tom Cruise et al. is relevant to the rest of the population, too.

It is at this point that we ran into a problem, as it became apparent that her view of the meaning of 'right' and my own were different. As best I can tell, she felt that 'right' meant that it feels right or brings comfort. I, of course, use the word 'right' to mean that an idea contains explanatory and predictive power.

These are two different meanings of the word "right". Yours is "correct fact about objective physical reality". Hers is "right for someone" meaning "appropriate [for someone]" meaning "serving well a certain purpose for that someone: e.g. making them happy".

So we had two versions of 'right'. It wasn't a given that she would accept my version, so I had to come up with a good reason why my version of 'right' was, well, 'right'.

Both are equally valid colloquial usages of the word 'right'! Your usage is not "more right" than hers. Above all avoid arguing over definitions. You're trying to get her to use a word ("right") to mean something else than she currently uses it to mean. Why? What do you care how she uses words as long as you can understand her correctly?

I suggest you describe to her that there are two different concepts, which really have nothing to do with one another, and which got confused because the two of you used "right" to mean these different things. Then you can give them two provisionally different names, and proceed to talk about the merits of each.

And remember that "what should I believe to make me feel well?" (or to achieve some other goal other than knowing the objective truth) is a valid question. The answer may sometimes be different than "always believe the objective truth". This doesn't make such a person irrational or "wrong". Believing the truth is a very common instrumental goal, but it is not a very common end-goal among humans.

These are two different meanings of the word "right". Yours is "correct fact about objective physical reality". Hers is "right for someone" meaning "appropriate [for someone]" meaning "serving well a certain purpose for that someone: e.g. making them happy".

Alternately, it implies a very muddy set of beliefs about the underlying nature of objective causality. Potentially, a belief that the universe is actually different for different people, or just some incoherent mess of sloppy reasoning and rationalization that doesn't compress down into a nice compact philosophy like reductionism.

It's easy to forget, when trying to understand the opposing side of an argument, that the bulk of humanity does not hold their ideas to the same (admittedly not very strenuous) standards that self-professed rationalists do.

It's problematic that the case of Josephine is the least realistic of all the examples. Religious people, in my experience, never believe that killing has no moral consequence. On the contrary, they often believe murder would get them sent to a very nasty end of the afterlife.

It's problematic that the case of Josephine is the least realistic of all the examples. Religious people, in my experience, never believe that killing has no moral consequence. On the contrary, they often believe murder would get them sent to a very nasty end of the afterlife.

Either that or if they kill people that those with power in their tribe consider enemies they will get an awesome after life!

Yes, though "killing people that powerful tribe members call enemies" doesn't seem to require any superstitions. I'm unsure about how belief in reward/punishment afterlives (that map onto social beliefs about morality) actually alters behavior. Presumably it makes people more fervent but I'm not so sure (witness Marxism).

I don't think you need to be superstitious to commit the fallacy of choosing a belief that is less accurate intentionally. All you need to do is buy into the meme that all belief is good. I know plenty of atheists-by-default who still think that there's nothing odd about intentionally choosing to believe something arbitrary.

Yes, though "killing people that powerful tribe members call enemies" doesn't seem to require any superstitions.

Of course not. In the same way not killing people doesn't require any superstitions.

I don't know how unrealistic it is.

Josephine's reasoning is exactly the reasoning Muslim suicide bombers offer for their willingness to inflict "collateral damage" by setting off explosions that kill both infidels and Muslims. Also, medieval justifications for torturing heretics (so they'll recant and therefore avoid the greater harm of going to Hell).

those who are effected by our actions

Minor nitpick: effected' should be 'affected'.

It's still the same in the original post. Maybe you forgot to change it?

I hadn't realised I could. I've just done so. I didn't write any kind of note to that effect. Is one needed for a spelling edit?

I generally don't mark edits that I do essentially immediately or that don't change the meaning.

To be fair, we do also owe it to those who are effected by our actions (e.g. our children).

The definition of "right" isn't "explanatory and predictive power". Something could have a lot of both of those whilst being completely incorrect. The definition of "right" is that that's what actually happened.

But these people I was debating refuse to accept that there is such thing as a 'right' by that definition. They say that 'what actually happened' is forever unknowable. I was trying to point out that while we may or may not find out exactly what happened, we can always tell if an explanation is 'more right' or 'less right' than another, based on how useful it is in explaining the evidence.

It's worth asking whether your interlocutors were physical anti-realists or moral anti-realists who got quite confused about the scope of their anti-realist position.

That's possible, but I got the feeling that they wouldn't know what those positions are. Their positions were so self-contradictory that it makes me think that they had simply absorbed some of the zeitgeist without any kind of formal study and then failed to propagate the change all the way across their belief networks.

New to LessWrong?