MugaSofer comments on Wrong Questions - Less Wrong

34 Post author: Eliezer_Yudkowsky 08 March 2008 05:11PM

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

Comments (126)

Sort By: Old

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

Comment author: MugaSofer 13 April 2013 08:07:45PM *  -2 points [-]

Well, we're mostly discussing Maitzen's answer to the of the First Cause, the Infinitely Old Universe. Unless a First Cause is somehow (magic?) self-explanatory, it doesn't answer the question of "Why is there anything?" - but the same applies if you replace a First Cause with an infinite string of causes, or even a future cause + time travel.

Comment author: JohnH 13 April 2013 09:03:31PM 0 points [-]

Have you seen Gods as Topological Invarients? Note the date submitted as it is relevant.

Anyways the whole question seems a confusion: either the answer will be something that does exist or it will be something that does not exist, if it exists it would appear to be part of "anything" and therefore the question is not addressed, and if it does not exist then that appears to be contradictory.

Comment author: MugaSofer 13 April 2013 10:37:18PM -2 points [-]

That's not exactly a confusion, that's a paradox. And a faulty one; something might (somehow) "explain itself" or, more likely, we could discover a logical reason things had to exist. Or we might have some unknown insight into rationality and dissolve the question, I suppose, but that's not really helpful. The point is it's still an open question; the good Mr. Maitzen has not helped us.

Comment author: JohnH 13 April 2013 11:09:30PM 0 points [-]

Applying Greek thought to "Ehyeh asher ehyeh" is an attempt to get at something that "explain(s) itself", I am sure you are familiar with St. Thomas Aquinas and his five ways.

I suppose you are also familiar with Divine Sophia in Gnosticism? Saying we have a logical reason for things existing seems to be on that same level of reasoning and appears to just add another turtle to me.

Comment author: MugaSofer 14 April 2013 04:28:58PM -2 points [-]

Applying Greek thought to "Ehyeh asher ehyeh" is an attempt to get at something that "explain(s) itself", I am sure you are familiar with St. Thomas Aquinas and his five ways.

Yup. Being a theist, I suspect God is in some way the cause of everything, although I'm not really smart enough to understand how that could be. I leave the answer to some future genius (or, more likely, superintelligent AI.)

Saying we have a logical reason for things existing seems to be on that same level of reasoning and appears to just add another turtle to me.

Really? But logic, as a mathematical construct, "exists" (in the sense that it exists at all) independently of physical objects; a calculator on mars will get the same result as one on Earth, even if they have no causal connection. Logic seems like it can explain things in terms of platonic mathematical structure, not contingent physical causes.

Comment author: PrawnOfFate 14 April 2013 06:29:42PM 0 points [-]

Logic is independent of particular objects (multiply realisable), but there is is no evidence that it exists immaterially

Comment author: MugaSofer 15 April 2013 11:39:44AM -2 points [-]

Not entirely sure what "exists" even means in cases like this, but yeah, then I guess you're restricted to self-causing entities in that case, whatever that might mean.

Comment author: satt 13 April 2013 10:20:27PM 0 points [-]

I disagree. If I go looking for a First Cause and discover an infinite string of causes instead, that's reality's way of telling me that there just isn't a First Cause, and the premise of my investigation was simply wrong. Equivalently, then, discovering an infinite string of causes indicates that the question "What is the First Cause?" (and hence "Why is there anything?", since that question reduces to the First Cause question once one accepts Maitzen's argument) is a wrong question, since it hinges on a false premise.

Comment author: MugaSofer 14 April 2013 04:31:17PM -2 points [-]

"What is the First Cause?" is not the same question as "Why is there anything?". An infinite string would answer the former, not the latter.

Comment author: satt 14 April 2013 05:31:03PM *  2 points [-]

"What is the First Cause?" is not the same question as "Why is there anything?".

This is true in the absence of further assumptions. But once you or I assume Maitzen's argument is true (and I think we both do) the second becomes a mere instantiation of the first.

An infinite string would answer the former, not the latter.

This is false if Maitzen's argument is true. Conditional on Maitzen's argument, an infinite string answers the former and hence the latter. I could justify this by repeating what I've written in my two comments upthread, but it might be more productive if I give a different argument.

How do we usually answer "Why is there X?"? I think we usually pick out X's most salient cause. "Why is there an ambulance outside my neighbour's house?" "Because the neighbour had a heart attack." We're basically saying, "here's the most interesting antecedent node in the causal graph, and had we deleted or substantially altered it, there wouldn't have been X". If we'd deleted the "neighbour's heart attack" node, there wouldn't be an "ambulance outside neighbour's house" node.

This gives me a way to interpret "What's the First Cause?", or "Why is there anything?", or "Why is there everything?", or "Why isn't the universe in the counterfactual no-turtle state?" (to paraphrase you). These questions are asking for a node in the causal graph that's antecedent to everything. But how can I do that if the causal graph is an infinitely long string? There's no such node!

Comment author: PrawnOfFate 14 April 2013 06:02:48PM *  1 point [-]

There is still the question of why there is an infinitely long string.

Comment author: satt 14 April 2013 09:41:46PM 1 point [-]

That question fails in the same way as the others. It's asking for a causal node which, if altered or deleted, would turn the infinite string of causal nodes into a finite one. But no node in an infinitely long string has that power, so the question's implicit assumption is false.

Comment author: PrawnOfFate 14 April 2013 11:06:14PM 2 points [-]

A why question has more possible anwers than efficient causality.

Comment author: satt 14 April 2013 11:49:24PM 0 points [-]

This might be true of "Why" questions in general but I'm talking about the more specific class of questions that start "Why is there". Can you think of examples of the latter that have a sensible answer that isn't a salient cause?

Comment author: PrawnOfFate 15 April 2013 10:13:41AM 1 point [-]

Sure. "Why are there airbags in cars" is answered with "to protect the occupants". it would be inane to give a a causal answer, such as "because someone fitted airbags".

Comment author: ArisKatsaris 15 April 2013 11:26:48AM 1 point [-]

"Why are there airbags in cars" is answered wit "to protect the occupants". it would be inane to give a a causal answer, such as "because someone fitted airbags".

"to protect the occupants" is merely syntactically simpler than "because of the builder's desire to protect the occupants." -- the two statements equally well indicate causality.

Comment author: MugaSofer 15 April 2013 11:36:55AM -2 points [-]

To be fair, this could be phrased as "because someone decided they were the best way to protect the occupants, and fitted them." However, I would define an answer to a "why is there" question more broadly - what explains why the universe is not in the counterfactual situation of this not being there? If you count any causal antecedent as an answer, you can't explain causal loops, and you can only explain parts of infinite chains, not the whole.

Comment author: MugaSofer 15 April 2013 11:32:50AM *  -1 points [-]

But once you or I assume Maitzen's argument is true (and I think we both do) the second becomes a mere instantiation of the first.

Sorry, which argument is this? He makes several.

Comment author: satt 16 April 2013 08:07:09AM *  0 points [-]

The argument that once the elements in the string are individually explained, the string as a whole is explained.

Edit: maybe I should call it the Hume-Edwards principle instead.

Comment author: MugaSofer 19 April 2013 02:03:33PM -2 points [-]

Ah. Well, I agree with it in principle; I just think he misapplies it.