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

prase comments on The Optimizer's Curse and How to Beat It - Less Wrong

44 Post author: lukeprog 16 September 2011 02:46AM

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

Comments (81)

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

Comment author: prase 19 September 2011 07:51:10PM 0 points [-]

Only high-status people like philosophers get this kind of treatment!

Are you saying that always when a sentence is translated, its author must have high status or gains high status at the moment of translation, because the default attitude is to ignore anything originally uttered in foreign language?

If this is what you mean, I find it surprising. I have probably never been in a situation when someone was ignored because he spoke incomprehensible gibberish and that fact was more important than the content of his words. Of course, translation may be costly and people generally pay only for things they deem valuable, which is where the status comes into play. But it doesn't mean that with low-status people it is more important that they speak gibberish than what they say.

(A thought experiment: A Gujarati speaking beggar approaches a rich English gentleman, says something and goes away. The Englishman's wife, who is accompanying him at the moment, accidentally understands Gujarati. The man can recognise the language but doesn't understand a word. What is the probability that he asks his wife "what did he say"? As a control group, imagine the same with an English beggar, this time the gentleman didn't understand because when the beggar had spoken, a large truck had passed by. Is the probability of asking "what did he say" any different from the first group?)

Comment author: komponisto 19 September 2011 09:45:10PM 1 point [-]

Are you saying that always when a sentence is translated, its author must have high status or gains high status at the moment of translation, because the default attitude is to ignore anything originally uttered in foreign language?

Yes. More generally, the default attitude is to ignore anything uttered by a member of an outgroup. By calling attention to the fact that a sentence has been translated, one is calling attention to the fact that the author speaks a foreign language and thus to the author's outgroup status. Omitting mention of a person's outgroup status is a courtesy extended to those we wish to privilege above typical outgroup members.

(A thought experiment: A Gujarati speaking beggar approaches a rich English gentleman, says something and goes away. The Englishman's wife, who is accompanying him at the moment, accidentally understands Gujarati. The man can recognise the language but doesn't understand a word. What is the probability that he asks his wife "what did he say"? As a control group, imagine the same with an English beggar, this time the gentleman didn't understand because when the beggar had spoken, a large truck had passed by. Is the probability of asking "what did he say" any different from the first group?)

Curiosity about what a low-status person says does not imply that one thinks the content of their words is a more important fact about them than their low status. With high probability, the most salient aspect of the beggar from the perspective of the Englishman is that he is a beggar (and, in the first case, a foreign beggar at that). Whatever the beggar said, if the Englishman finds out and deems it worthy of recounting later, I would be willing to bet that he will not omit mention of the fact that he heard it from a beggar.