Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Annoyance comments on Well-Kept Gardens Die By Pacifism - Less Wrong

105 Post author: Eliezer_Yudkowsky 21 April 2009 02:44AM

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

Comments (309)

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

Comment author: Annoyance 21 April 2009 07:35:52PM 0 points [-]

"I have seen few good ones in favor of being an atheist."

That misses the point. Atheism is the null hypothesis; it's the default. In the complete absence of evidence, non-commitment to any assertion is required.

Comment author: William 21 April 2009 08:10:54PM 3 points [-]

The idea of a null hypothesis is non-Bayesian.

Comment author: JGWeissman 21 April 2009 09:27:16PM 6 points [-]

A null hypothesis in Bayesian terms is a theory with a high prior probability due to minimal complexity.

Comment author: andrewc 23 April 2009 11:31:58PM 1 point [-]

I'm not sure it's so clear cut.

They key point is that when you do the p value test you are determining p(data | null_hyp). This is certainly useful to calculate, but doesn't tell you the whole story about whether your data support any particular non-null hypotheses.

Chapter 17 of E.T. Jaynes' book provides a lively discussion of the limitations of traditional hypothesis testing, and is accessible enough that you can dive into it without having worked through the rest of the book.

The Cohen article cited below is nice but it's important to note it doesn't completely reject the use of null hypotheses or p-values:

.. null hypothesis testing complete with power analysis can be useful if we abandon the rejection of point nil hypotheses and use instead "good-enough" range null hypotheses

Comment author: thomblake 22 April 2009 02:59:10PM 0 points [-]

I think it's funny that the observation that it's "non-Bayesian" is being treated here as a refutation, and got voted up. Not terribly surprising though.

Comment author: ciphergoth 23 April 2009 01:07:05PM 0 points [-]

Could you be more explicit here? I would also have considered that if the charge of non-Bayesianness were to stick, that would be tantamount to a refutation, so if I'm making a mistake then help me out?

Comment author: thomblake 23 April 2009 03:50:52PM 0 points [-]

The charge was not that the idea is not useful, nor that it is not true, either of which might be a mark against it. But "non-Bayesian"? I can't unpack that accusation in a way that makes it seem like a good thing to be concerned about. Even putting aside that I don't much care for Bayesian decision-making (for humans), it sounds like it's in the same family as a charge of "non-Christian".

Comment author: Vladimir_Nesov 23 April 2009 04:06:12PM *  3 points [-]

One analogy: non-mathematical, not formalized, not written in English, and attempts to translate generally fail.

See [*] for a critique of null hypothesis and related techniques from a Bayesian perspective. To cite:

My work in power analysis led be to realize that the nil hypothesis is always false. [...] If it is false, even to a tiny degree, it must be the case that a large enough sample will produce a significant result and lead to its rejection. So if the null hypothesis is always false, what's the big deal about rejecting it?

[*] J. Cohen (1994). `The Earth Is Round (p < .05)'. American Psychologist 49(12):997-1003. [pdf].

Comment author: Nominull 23 April 2009 04:06:20PM 0 points [-]

Being non-Bayesian is one particular type of being untrue.

Comment author: Vladimir_Nesov 23 April 2009 04:13:37PM 1 point [-]

Being non-Bayesian is one particular type of being untrue.

Now, what does this mean? Sounds horribly untrue.

Comment author: thomblake 21 April 2009 08:01:30PM *  2 points [-]

But atheism isn't actually the default. A person must begin study at some point in his life - you start from where you actually are. Most people I'm aware of begin their adult lives as theists. Without a compelling reason to change this belief, I wouldn't expect them to.

Comment author: Annoyance 21 April 2009 08:15:24PM -2 points [-]

"But atheism isn't actually the default."

Well... yes, it is. I do not know of any theistic infants. Actually, I'm not aware that infants have any beliefs as such.

Young children seem predisposed to attribute things to powerful but non-present entities, but I'm fairly certain there are logical fallacies involved.

The fact that many people accept certain concepts as given without questioning them thoroughly - or at all - does not constitute a justification for believing those things. I have often heard the claim that philosophy does not attempt to examine premises but only to project and study the consequences of the premises people bring to it; I consider that to be one of the reasons why 'philosophy' is without merit.

Comment author: JGWeissman 21 April 2009 10:49:19PM 4 points [-]

It seems that Annoyance and thomblake are using different definitions of "default".

Annoyance uses it the same as null hypothesis, the theory with the smallest complexity and therefore the best prior probability, that any other theory needs evidence to compete with. In this sense, atheism is the default position, supposing that the universe follows mindless laws of nature without the need for initial setup or continuous intervention by any sort of intelligent power is simpler than supposing the universe acts the same way because some unexplained deity wills it. This definition is useful to figure out what our beliefs ought to be.

Thomblake seems to mean by "default", the belief one had when achieving their current level of rationality, that they will keep until they find a reason to change it. For most people, who are introduced to a religion at young age before they get a chance to learn much about anything approaching rationality, some sort of theism would be this default. This definition is useful to figure out why people believe what they believe, and how to convince them to change their beliefs.

Now, I am not sure what we mean by "sanity", but I think someone who maintains a default position (in thomblake's sense) that they would not have adopted if first presented in their current level of rationality, while they may benifet from achieving an even higher level of rationality (or simply haven't reviewed all their default positions), they are not necessarily incapable of achieving the higher level.

Comment author: Annoyance 22 April 2009 04:19:51PM 0 points [-]

I'm not even entirely sure that we're all using the word 'atheism' to refer to the same things.

This highlights the problems that arise when people use the same terminology for different concepts.

Comment author: MrHen 21 April 2009 08:19:40PM 1 point [-]

Well... yes, it is.

You keep doing this. Simply stating the opposite of another statement is not helping. Even if you clarify a little later it seems to be indirectly and without a solid response to the original point.

Comment author: Annoyance 21 April 2009 08:20:56PM -2 points [-]

That's why you need to read the sentences following the one you quoted.

Comment author: MrHen 21 April 2009 08:26:52PM *  1 point [-]

Well... yes, it is. I do not know of any theistic infants. Actually, I'm not aware that infants have any beliefs as such.

Infants without beliefs do not last long. They get beliefs eventually. Trying to argue this point just pushes the relevant stuff up the tree and makes the argument about semantics that are not particularly useful for the topic at hand.

And... are you saying that the null hypothesis is whatever an infant believes? How is that useful? I think it degrades definitions of things like "atheism" by saying that if you make no choice it is the same as making the correct choice. Coming to the correct conclusion for the wrong reason is the wrong solution.

Young children seem predisposed to attribute things to powerful but non-present entities, but I'm fairly certain there are logical fallacies involved.

The null hypothesis could be wrong. Logical fallacies are irrelevant.

The fact that many people accept certain concepts as given without questioning them thoroughly - or at all - does not constitute a justification for believing those things. I have often heard the claim that philosophy does not attempt to examine premises but only to project and study the consequences of the premises people bring to it; I consider that to be one of the reasons why 'philosophy' is without merit.

This is irrelevant to the topic. So, at the end, I spent my time telling you your comment was mostly irrelevant. I should just downvote and bury it like I did the other one.