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

thomblake comments on Beyond the Reach of God - Less Wrong

68 Post author: Eliezer_Yudkowsky 04 October 2008 03:42PM

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

Comments (271)

Sort By: Old

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

Comment author: thomblake 16 November 2011 06:59:00PM 0 points [-]

An argument is a series of statements ("propositions") that are intended to support a particular conclusion. For example, "Socrates is a man. All men are mortal. Therefore, Socrates is mortal." Just as one sentence is not a paragraph, one proposition is not an argument.

There is no question of whether "trying to make reality a safe and cozy haven will only cause harm" is a valid argument because it's not an argument at all. This is an argument:

  • If we try to make reality a safe and cozy haven, then we will only cause harm.
  • We are trying to make reality a safe and cozy haven
  • Therefore, we will only cause harm.

Note that this is a valid argument; the truth of the conclusion follows necessarily from the truth of its premises. If you have any problems with it, it is with its soundness, the extent to which the propositions presented are true. It sounds like you think the first proposition is false, but you are claiming Caledonian made an invalid argument instead. If that is the case, you're making a category mistake.

Comment author: Torgamous 16 November 2011 07:45:33PM 0 points [-]

And now we're disputing definitions. I was using argument to mean what you've defined as propositions; it was a mistake in labeling, but the category is the same. Regardless, the falseness of his proposition is not an issue. The issue I have is that his initial proposition, though it may possibly be true, has a wide range of possible truenesses, no indication which trueness the poster was aiming for, and may very possibly have been made without a particular value of potential truth in mind. If that's soundness, then yeah, I took issue with the soundness of his proposition.

Comment author: thomblake 16 November 2011 08:11:24PM 0 points [-]

The issue I have is that his initial proposition, though it may possibly be true, has a wide range of possible truenesses, no indication which trueness the poster was aiming for, and may very possibly have been made without a particular value of potential truth in mind.

I don't see how that's the case. It seems very specific to me. In the statement "X will only cause Y" are you confused about the meaning of X, Y, "will only cause", or something else I'm missing? (X="trying to make ... reality a safe and cozy haven", Y="harm")

Comment author: Torgamous 16 November 2011 08:32:07PM 0 points [-]

I take issue with Y. "Harm", though it does have a definition, is a very, very broad term, encompassing every negative eventuality imaginable. Saying "X will cause stuff" only doubles the number of applicable outcomes. That does not meet my definition of "specific".

Comment author: thomblake 16 November 2011 08:35:00PM 0 points [-]

Aha. Again, a definitional problem. I would indeed regard the claim "dropping this rock will cause something to happen" as specific, and trivially true; it is not vague - there is no question of its truth value or meaning.

I think this is resolved.

Comment author: Torgamous 21 November 2011 04:10:38PM 0 points [-]

I'm sorry, I want this conversation to be over too, and I don't mean to be rude, but this has been bugging me all week: where did you get that definition from, and where do you live? Literally everyone I have interacted with or read stuff from before you, including published authors, used the same definitions of "specific" and "vague" that I do, and in ways obvious enough that your confusion confuses me.

Comment author: thomblake 21 November 2011 07:10:19PM 2 points [-]

I live in (and am from near) New Haven, CT, USA, and I have a background primarily in Philosophy.

A vague proposition is one which has an uncertain meaning - 'meaning' is of course tied up in relevance and context. So observing that a patient coughs is a 'vague' symptom in the sense that the relevant 'meaning' is an answer to the question "What disease does the patient have?" and the answer is unclear.

In the above, Caledonian is stating that "trying to make reality a safe and cozy haven will only cause harm". I do not see this as in any way vague, since it has a clear referent. If anyone were to try to make reality a safe an cozy haven, and caused anything other than harm in doing so, then the proposition would turn out to have been false. You can unambiguously sort worlds where the proposition is true from worlds where the proposition is false.

I'm not sure from previous comments on this thread what definition of 'vague' you were employing or how it differs from this.

Comment author: TheOtherDave 21 November 2011 07:39:48PM 1 point [-]

You can unambiguously sort worlds where the proposition is true from worlds where the proposition is false.

Doesn't this depend on having an unambiguous test for whether reality in a given world is a safe and cozy haven? If one were skeptical of the possibility of such a test, one might consider the quoted statement vague.

Comment author: thomblake 21 November 2011 08:51:14PM 0 points [-]

Doesn't this depend on having an unambiguous test for whether reality in a given world is a safe and cozy haven?

No, you need to have unambiguous tests for "consequences other than harm" and "trying to make reality a safe and cozy haven".

The proposition leaves open the possibility that one might accidentally make reality a safe and cozy haven without trying and thus cause non-harm.

Comment author: TheOtherDave 21 November 2011 09:23:06PM 0 points [-]

You are entirely correct! Teach me to be sloppy.

Comment author: Torgamous 21 November 2011 08:06:04PM 0 points [-]

Ah, I see. That makes sense now; your previous example had led me to believe that the difference was much greater than it is. I had been using "vague" to mean that it didn't sharply limit the number of anticipated experiences; there are lots of things that are harmful that cover a range of experiences, and so saying that something will "cause harm" is vague. For the disease question, "vague" would be saying "he has a virus"; while that term is very clearly defined, it doesn't tell you if the person has a month to live or just has this year's flu, so the worlds in which the statement is true can vary greatly and you can't plan a whole lot based on it. Ironically, my definition seems a lot vaguer than yours now that they've both been defined.

And now I can happily say the matter's resolved.