Live data from Hacker News

The Kolmogorov option

scottaaronson.com

161–170 of 476 posts

Re: The Kolmogorov option

#161
post #156

Earlier quoted context omitted.

A rational discussion of it does not mean defending it! As Aristotle apparently said "It is the mark of an educated mind to be able to entertain a thought without accepting it." If we can't even discuss certain topics, if we can't even think things through, then we've entered the realm of thought police.

But if discussion of abhorrent ideas is enough to spread them, is the benefit of that discussion worthwhile? I ask this in a bit more depth here: https://news.ycombinator.com/item?id=14967029

It's obviously difficult to prove any position on this matter, but I definitely think it's worthwhile.

I think the idea that we should suppress mere discussion of certain topics is very dangerous, and could lead society down a very dark path.

Let's not forget that things like interracial marriage, gay marriage and equality for women were once widely considered "abhorrent ideas".

Re: The Kolmogorov option

#162
post #85

Earlier quoted context omitted.

It should be obvious that anyone arguing against having a rational discussion is wrong about being in that camp.

Not all subjects are amenable to "rational discussion", since it's inherently a restricted format which has implicit rules about what does and does not count as "rational". It also tends to presume its own weightlessness. Men seem very unwilling to recognise that the memo and the discussion itself is harmful.

> Men seem very unwilling to recognise that the memo and the discussion itself is harmful.

...you realize where this goes, right?

Like -- suppose you're right; and suppose you're given the power to suppress all such harmful discussions. You apply it. No more such discussion. Great.

Now suppose this occurs but in fact you're wrong. We must then ask: How would you find out that you're wrong?

Well, in such a case, you probably wouldn't. I guess you might find out when the chickens, whatever they are, finally came home to roost. But ideally one wants to find out before then. Better hope the chickens are merely bad rather than catastrophic, seeing as you've been doing absolutely no planning for this case. And hopefully they come sooner rather than later.

(And that's assuming you're a reasonable person who would actually admit error at that point; see below.)

I mean, really... illiberalism, it always goes the same way. You think it's discussion that's harmful? Have you seen the alternative? Because, I mean, examples abound, and how it goes is pretty clear. You're talking about going down a path dominated by humanity's worst tribal instincts. I should hope that's not what you want -- but that's where that path leads. By the time the far-off disaster occurs, do you think it'll be people like you, who are capable of thinking clearly but just think certain discussions should be suppressed, who are going to be running the show? No, it'll the people who are the least reflective, the most tribal, the most doublethinky.

Liberalism, free speech, when working properly, is supposed to work as a negative feedback loop. If you're wrong, you find out. Someone contradicts you, supplies arguments, and then you can consider them and see whether they might be right. As a lot of people have noted, it... doesn't exactly always succeed at this. But suppression of speech... hoo boy, that fails so much harder. That's how you get positive feedback loops. As the professed beliefs of the group get further and further from reality, simultaneously the requirements that you agree, the punishments for disagreeing, get stricter and harsher. You sure as hell don't find the truth that way.

Truth, now... I notice that's something you didn't even mention at all. Because some of the points made in that memo, are, as best as people can currently tell, true. You haven't made any claims about to truth or falsity, only about harmfulness. But do you think the harmfulness of the claims in that memo exceed the harmfulness of shitty civilizational epistemic practices? (Nature can't be fooled, as they say!)

Like, OK, bad epistemic practices might not seem that bad, might seem like a worthwhile tradeoff, if you imagine suppression of specific facts or claims or discussions as an isolated thing. Maybe we don't need to know literally everything. But that's not how it goes. Free speech, liberalism, these are ideas that are unnatural to people, they had to be learned, and they are constantly seeking ways to slip them off or and go back to full-on tribalism (or pervert them in service of it). You may want suppression of particular claims... you will get the bad old days. The positive-feedback loop of doom.

Claims don't exist in isolation, after all; claims have relations between them. You can't just suppress one claim, because people will rederive it from other claims. And if the claim you suppress happens to be correct? Then people will definitely rederive it. So either surrounding claims have to go, or the process of inference itself has to go. Likely both. In fact definitely the latter; you can peel off surrounding claims all you want but eventually you'll have to attack inference itself. And hey, it happens already that people are constantly eager to do that anyway! They only need a little push... and then oops, there's your positive feedback loop. Once you encourage people to use bad methods, they'll use them to reach all sorts of bad conclusions... I expect many of them will surprise you!

(And what is the scope of this suppression? Shall the hidden truth be kept alive in the academy, say, with a strict cordon, so that the facts may be known by the chosen few but never applied outside where it might be necessary? Shall those who wish to learn a subject have to first learn only the public parts, and then apply to join, to learn the hidden secrets? Or shall it extend even to them? Is the pursuit of truth itself something that must simply be forbidden?)

It's a dark road you're suggesting here -- and not a new one; an old one, an ancient one, one whose failures we know very well. I'll take whatever harmfulness the truth might pose over that any day. I don't think it can really hold a candle to that.

Re: The Kolmogorov option

#163
post #57
post #47

Earlier quoted context omitted.

I think what he's saying that if Kolmogorov was cautious enough to keep his unpopular opinions to himself, so can you.

yeah, but he was very clearly casting the whole thing as an "emperor's new clothes" situation with the implication that the unpopular opinions were self-evidently true but unmentionable.

He's not saying that any unpopular opinion is self-evidently true if people oppose it with such zeal. He's saying that there exist self-evidently true opinions that will be treated that way, and if you harbor one of those you should choose the Kolmogorov option. How do you know you're harboring one of those? The evidence will support it.

Re: The Kolmogorov option

#164

What I find myself wanting to fight is not the prevailing views on a topic, it's the willingness to lie and mislead to promote those views. I saw that this past week with the Google memo situation. I also saw a shocking amount of it last year with the US election. Sloppy journalism feeds the people who seem to want to believe the worst in everything, who in turn feed entire social networks. Eventually sufficient peop…

Some of the worst offences are these new "fact-checkers" which all seem heavily biased towards the left. When so many people are committed to a motivated and shallow version of the truth, it's hard to find the people who care ONLY about the truth.

Links to support this claim?

Re: The Kolmogorov option

#165
post #107

Earlier quoted context omitted.

Why? Is it wrong to be against a rational discussion of the benefits of race-based slavery? Is it wrong to be against a rational discussion of the benefits of child-slavery? Is it wrong to be against a rational discussion of the benefits of childhood labor? Is it wrong to be against a rational discussion of the benefits of racial discrimination? Is it wrong to be against a rational discussion of the benefits of gende…

If you find a rational person willing to defend any of the points you made, then yes, it is wrong to be against that discussion.

But there are no points, just topics.

Re: The Kolmogorov option

#166

Earlier quoted context omitted.

> No, the suggestion is that you cannot have context-free universally true premises for every type discussion. This is not a prerequisite of all discussions. > Most of this discussion outrage is based on people on all sides not realising they start from different, inherently subjective premises, and then believe they can build a universally true objective claim on that card house. This is easy to discover and correct…

> This is not a prerequisite of all discussions. You're literally reversing your position here: > Are you suggesting that some topics can only be discussed if We were not talking about all discussions, we were talking about how some topics don't play nice with this method of having a discussion. > This is easy to discover and correct for if both parties approach the discussion without being disingenuous. You mean lik…

I'm not reversing my position at all. I am arguing that some topics can be discussed with context-free universally true premises and some cannot. Nothing I have said is inconsistent with that.

> This is not a prerequisite of all discussions.

is equivalent to:

Some discussions do not have this as a prerequisite.

Re: The Kolmogorov option

#167
post #86

Earlier quoted context omitted.

The famous "fire in a theatre" example was invented to justify censoring the distribution of flyers opposing the draft.

Just so: https://www.theatlantic.com/national/archive/2012/11/its-tim... I'll try and avoid "the most famous and pervasive lazy cheat in American dialogue about free speech" in the future. :0 I really think it's worth reading the employment lawyer piece I linked, it's good to get the actual legal perspective on this whole situation.

Is it? Why?

I think it's not a legal issue - clearly Google can fire him without cause. But it's a cultural issue; is it morally acceptable to still work at or with Google, even though they show this behavior?

Personally, this pisses me off enough that I've switched to DuckDuckGo for the moment.

Re: The Kolmogorov option

#168
post #30

Besides ideological self-righteousness, Aaronson left out one important component of oppressive regimes, which I think is actually far more important: it takes a relatively large amount of people to topple a regime, while small groups as well as individuals do nothing but expose themselves by rebelling. So no individual with any regard for his own personal safety has any incentive to rebel, and will instead go along…

I don't think you need something as serious as an oppressive regime to get into this state. It sounds like an instance of the Abilene paradox - https://en.m.wikipedia.org/wiki/Abilene_paradox

Non-mobile link:

https://en.wikipedia.org/wiki/Abilene_paradox

Re: The Kolmogorov option

#169
post #152
post #144

Earlier quoted context omitted.

This is an equivocation. There's such a thing as an is/ought distinction. The fact that there might be some rational description of morally abhorrent behavior doesn't lead to the conclusion that that behavior is acceptable . Understanding doesn't mean accepting. Darwinism, for instance, doesn't imply Social Darwinism. Neither do theories of criminal behavior attempt to legitimize crime. And if anything, anyone who's…

And if anything, anyone who's appalled by these things should actually be interested in studying the conditions under which these things might happen I agree with this very strongly. I'm not sure if (or why) this leads to the requirement to have a rational discussion with proponents of abhorrent ideas. It's interesting. I used to believe that abhorrent ideas should be confronted. But I'm not longer convinced - it see…

>I'm not sure if (or why) this leads to the requirement to have a rational discussion with proponents of abhorrent ideas

Besides the difficulty of working alone, how can anyone challenge something they aren't willing to talk about? If anything is to be done about it, then people will have to start talking eventually, even if only for the purpose of political coordination.

I get that there's the issue of trying to present anything to an uninterested or uneducated audience in a way that they won't get the wrong idea. But assuming beforehand that they will get the wrong idea regardless of what you might do is just assuming failure. If everyone thought this way, nothing would ever be done: slavery, serfdom and child-labor wouldn't ever have ended.

Besides, ideas don't spread like that. Specially morally abhorrent ideas, because being "morally abhorrent" usually implies being politically unfeasible. Even the most terrible regimes tried to whitewash themselves because they knew what they were doing wouldn't fly with the populace.

Re: The Kolmogorov option

#170
Yeah, but many of us, even if aspiring to be our best, will never reach the importance of a Kolmogorov, or a Terry Tao, etc. Our optimal mix of political iconoclasty and mathematical progress may be much much different.

There's a sense in which way the ideal mathematician (or ultramarathoner, or classical composer...) is like a monk. Many of us are not built like that.

Post reply on HN