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

Bound_up comments on Social Insight: When a Lie Is Not a Lie; When a Truth Is Not a Truth - Less Wrong Discussion

7 Post author: Bound_up 11 August 2017 06:28PM

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

Comments (58)

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

Comment author: lifelonglearner 12 August 2017 06:29:37AM 1 point [-]

[Quick comprehension check]: I think that you are saying that it is important to acknowledge when our notions of truths and lies break down because saying a thing that is apparently "true" can have connotations we didn't intend, thus making it "false". And you're flipping it around to say that the opposite is also valid-- that you can say a thing which is apparently "false", yet the way that it's interpreted could make it more "true".

I think you are saying that there are other factors when communicating, which is the context you convey with your words, i.e. meaning that is imparted which is distinct from the actual referents of the words in your utterances. And that this meaning is also important to keep in mind because we can't "just" communicate with only the words themselves, apart from connotation / context. It's just part of the package.

I think that you then took this to show that there are often times where knowledge about true things isn't easily transmissible due to a lack of prerequisite knowledge. And that this has problems when that information might be important.

[Actual response, if the above was accurate]: I think the part of this essay about how trying to get across points is often difficult is important. There are certain tradeoffs to be wary of, like when someone asks you a question, and you give an abridged / simplified question to optimize more for communication rather than accuracy. (EX: Giving someone a stripped-down description of your medical condition when they ask you why you're taking a pill.)

Thus, one of the questions we might want to take out of this is "How can we convey information many inferential steps away from the other party, especially when it's beneficial to them?" which seems like it could be resolved several ways:

  • 1) Take the time to build up their prerequisites.

  • 2) Convince them you're competent / trustworthy such that they can defer to your judgment.

  • 3) Tell them false things such that they do the thing the information would have convinced them to do.

(I don't really like these options. Feel free to take this as an open invitation to spend 3 minutes thinking of other things.)

Anyway, it's less clear to me that you can tell people false stuff to make them believe true stuff. It feels more like you can people false stuff to do either 2 or 3, but not 1.

Comment author: Bound_up 12 August 2017 06:19:40PM 0 points [-]

Suppose that a vast group of statements that sound (they really, REALLY sound) like propositions about economic cause and effect are ALL interpreted by a great many people always and only as either "Yay blues" or "Boo blues."

In that case, your ability to tell the truth is limited by their way of filtering your statements, and your ability to tell lie is equally hampered. All you can do is decide whether to say Yay or Boo or not say anything at all (which will also often be interpreted one way or the other if you're involved in politics). It is an illusion that you're saying something about the minimum wage, for example. All you're really saying is "Yay blues!" as far as a great many people are concerned.

And if you're aware of this and count on it, you can choose to use statements that way on purpose, such that that IS all you're really saying.

This is most of politics.

Comment author: lifelonglearner 12 August 2017 07:16:09PM 0 points [-]

<Nods> Thanks for the additional clarification.