Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Why Facts Don't Change Our Minds

newyorker.com

Post reply on HN
Print view View original

161–170 of 347 posts

Re: Why Facts Don't Change Our Minds

#161

Post by conistonwater »

dllthomas wrote: post #146
conistonwater wrote: post #114

Earlier quoted context omitted.

The way I parsed your comment: there are so many bad arguments possible even for good ideas that if you see people use a bad argument for their idea, it doesn't mean there isn't a good argument instead that can replace it. Then, you'd think good arguments would get accepted more universally than bad arguments, but probably not. Is it even in the same ballpark?

That sounds right. How is that in conflict with what you said? I saw it as expanding on the reasons to reject the "bad heuristic" under discussion.

I like the idea of seeking out good arguments in favour of other people's ideas, but I dislike the way it contradicts the principle that one has a responsibility to state good arguments for one's own position. Stating poor arguments for an idea should be grounds for not accepting that idea, which is why I thought what you said wasn't compatible with what I said. One reason I thought the bad heuristic was so bad is it didn't even reach the stage of assessing arguments properly, on actual merits, thus leaving the door wide open to myside bias, and to all the other biases too.

You know, it might be just my experience. Sometimes, it is like a game people play, where they present an incoherent argument for an idea they believe in, and then say "well, there could be a better argument, why don't you look for it?". That's something your comment reminded me of, even if it went in the direction of agreeing with consensus.

Re: Why Facts Don't Change Our Minds

#162

Post by Eridrus »

squozzer wrote: post #148
Eridrus wrote: post #9

Earlier quoted context omitted.

I think the discourse about "rights" is part of the problem in US politics, it makes the default mode of reasoning about ideology, rather than outcomes. The government should be in the healthcare business since the nature of the service precludes a functional market.

Sorry, I guess where I was going was when someone defines something as a right, they should be able to explain how they arrived at that conclusion. For instance, if you believe that we have a right to live, and healthcare has some relation to life, then haelthcare might arguably be a right. Another might be an analogy, such as with law enforcement. Some might say we have a right to a life free from crime. Most jurisd…

I think a rights based reasoning framework is problematic in general, it leads people to focus too much on what is considered their "right" and ignores both the harms associated with having that right and the reasoning on which that right was based.

You can fall into the trap of assuming your "right to private property" means that taxation is theft or you can start with a "right to privacy" and decide law enforcement should never be able to access the contents of your smart phone, rather than discussing the reasons for private property or privacy.

And in case case, rights tend to never be justified, just asserted and treated like true axioms.

Re: Why Facts Don't Change Our Minds

#163

Post by 77pt77 »

Florin_Andrei wrote: post #133
Mikeb85 wrote: post #49

Earlier quoted context omitted.

Of course. That would be an admission we were wrong which goes against our worldview where we're right.

Human beings are full of way-overcranked early optimizations. I guess that's what was required for the propagation of the selfish gene (and individuals be damned, Stalin-like), and you can't really blame a natural process - but sheesh, the whole thing just looks like a stinking pile of bad engineering. Maybe that's because I'm thinking from the side of the individual. The selfish genes, however, must be pretty happy…

> the whole thing just looks like a stinking pile of bad engineering

Well, there is no engineering at all involved, so not really unexpected.

Re: Why Facts Don't Change Our Minds

#164

Post by FabHK »

gregmac wrote: post #102
balabaster wrote: post #53

Earlier quoted context omitted.

and amazingly the amount of information that comes out with peer reviewed science that has been repeatable enough to be accepted as fact and advertised as such can be thrown away 15 years later as wrong... how can a fact be wrong? Either it was a fact back then and you were right or you were wrong back then and it wasn't a fact but information supporting a non-factual hypothesis... given how often we see this present…

> repeatable enough to be accepted as fact No. They are accepted as a scientific theory [1]. A scientific fact is "a repeatable careful observation or measurement (by experimentation or other means), also called empirical evidence" [2]. The media and general public doesn't really understand the distinction, and tends to call widely-accept scientific theories "facts" and treat them as if they are proven true -- when i…

> The media and general public [...] tends to call widely-accept scientific theories "facts" and treat them as if they are proven true -- when in reality the best science can do is just not yet prove something false.

That Popperian view (published 1934) is itself somewhat problematic, and I'd say obsolete. The distinction between pure unambiguous observation statements and the edifice of theory on top (concepts, operationalisation, hypothesis, test, theory, etc.), stemming from Positivism even before Popper, is also problematic. Even the distinction between analytic (logical, "by definition") and synthetic (empirical, "by observation") statements is problematic, see Quine's Two Dogmas of Empiricism (1951). Newer approaches see scientific theories as a "web of beliefs" - basically, "empirical facts" and "theory" and "auxiliary theories" are all intertwined.

The question, then, whether science can give us real facts about the real world (beyond what's immediately observable) is subject of an ongoing debate between instrumentalism and scientific realism.

Personally, I come down on the realist side - I think there are electrons and there were dinosaurs, even if I haven't seen them.

And thus, I think it's legitimate to call very well established scientific theories "facts"

(also, btw, because of the "oh, but it's just a theory" response of various anti-science trolls (not you, mind you!)).

Re: Why Facts Don't Change Our Minds

#165

Post by kpwagner »

This is a really tough problem. Like many in the HN community, I try to fight against confirmation bias, but sometimes it feels like a lost cause. Compare it to a reasonably complex math problem--I'm talking an Algebra II / Calc I type problem. You are relying on so much previous knowledge to complete the problem. That knowledge may have been "earned", meaning you really took the time to build up an intuition, or "unearned", meaning you just took the word of the instructor or book (essentially applying a procedure to get a result). This is a MATH problem that has defined parameters and a definitive right or wrong answer.

Moving to something in the social realm, it's like a laser hitting a geode. When I see a talking head like Ann Coulter on a "news" program, they are often quite adept at providing facts and statistics to support their argument. But ironically, this usually does nothing for me to strengthen their argument or change my mind. To start with, are these facts and statistics true and accurate? Is there really some subjectivity in these "facts"? Are they just outright rumors? Even if these facts are true and accurate, does it tell the whole story, or does it simply support their position? That's a pretty loaded question; no one really goes to the effort of presenting all relevant information and interpretation--nor would you want to listen. Denying facts and falling back on beliefs, ideally wisdom, is a defensive mechanism against information overload. To aggressively change your mind again and again when presented with new information is similar to the problem of overfitting a neural network.

Re: Why Facts Don't Change Our Minds

#166

Post by tn13 »

In Batman Zero Year, Bruce Wayne has made his first appearance as Batman and is almost 100% sure that people will easily find it out that Bruce is Batman.

Alfred then tells Bruce that during his performance as an actor he paid far too attention to minor details of the characters played including dialects, diction, consumes, makeup etc. just because he was terrified that people will see the person behind the mask. Only after some experience he learned that people pay to see the mask and not the face. They very willingly will refuse to see the face behind the mask because in their hear they want that mask to be real. People will not want to find out who is Batman.

Of course this is all fiction but it does ring a bell especially in today's super-heated political environment where we cant even know if a fact is a fact. Even when something becomes an obvious fact or an obvious lie it does not seem to matter to most people because in their heart they have an alternate view of the very facts.

Re: Why Facts Don't Change Our Minds

#167

Post by bambax »

Silhouette wrote: post #128
bambax wrote: post #18

> The fact that both we and it survive proves that it must have some adaptive function No, no, no, absolutely not, no, it doesn't. No. A feature can maintain itself in a population for a number of reasons: - it's neutral or not detrimental enough to drive affected individuals to extinction (ie, to get them killed before they have a chance to reproduce) - it's linked to some other trait that provides actual benefits -…

I suggest a more simple and straightforward explanation for the limitations of "reason": our brain is, in fact, write only. That seems to be inconsistent with the research described in the article where subjects were asked for an opinion, then in some way required to explain that opinion, and then asked for another opinion on the same issue. People apparently did weaken in their convictions, but only when forced to c…

If you're referring to this paragraph:

> A recent experiment performed by Mercier and some European colleagues neatly demonstrates this asymmetry. Participants were asked to answer a series of simple reasoning problems...

then no. While I don't know the exact content of the experiment or the wording of the questions, a reasoning problem isn't an opinion, it's a process.

A process (a function) can indeed yield a different result when some parameters are tweaked or a different input is used.

But my theory is that opinions, once formed, are difficult / impossible to change, not for some unfalsifiable evolutionary explanation, but for a very simple structural reason that the brain is a WORM system.

Also, my theory is just that: a hypothesis, a thought experiment. I don't pretend to know how the brain works; I'm just asking the question, if we think of the brain as WORM, can we explain more behaviors?

Re: Why Facts Don't Change Our Minds

#168

Post by AnthonyMouse »

conistonwater wrote: post #131
coldtea wrote: post #76

Earlier quoted context omitted.

> All you are really saying is its impossible to know anything for certain. Yeah, in the sense that all you are saying is its possible to have degrees of trust on difference sources of "facts". That said, you'd be surprised how often the fact (pun intended) that it's "impossible to know anything for certain" is lost on people. > Having a citation from a well regarded source is better than having no citation. Peer rev…

> That said, you'd be surprised how often the fact (pun intended) that it's "impossible to know anything for certain" is lost on people. I think David Hume was quite good on this subject. If chance of being right is all there is to knowledge, then maximizing your chance of being right is what knowledge is. So all the peer review, the numbers and statistics, for example, contribute to knowledge even if it might be wro…

Selection bias is a huge problem though. Statistics in politics basically have negative utility because people just omit the ones that contradict their favored narrative, and then you get this:

https://xkcd.com/882/

When all you have is that newspaper headline, your knowledge has gone down rather than up.

Re: Why Facts Don't Change Our Minds

#169

Post by leshow »

skybrian wrote: post #159

It's important to remember that most scientific disputes can't be resolved by two uninformed people talking on the Internet. If you can get people to share links to well-written scientific articles, you're doing well, but these disputes can only really be answered by actual scientific discussion between actual experts who take the time to dig into methodology (etc), and maybe not even then; the replication crisis is…

> can only really be answered by actual scientific discussion between actual experts

I fundamentally disagree with this. For one, who gets to decide who the 'actual' experts are compared to the experts that aren't actually experts?

The stuff you're saying is specious; on the surface it looks plausible but it actually doesn't mean anything.

Anyone can develop expertise in a topic of interest if they follow it ardently and have good critical thinking skills. One obviously needs the ability to read and comprehend more than just an abstract, but that's not what separates an 'expert' from an 'actual expert' anyway.

Re: Why Facts Don't Change Our Minds

#170

Post by FabHK »

philipov wrote: post #79
balabaster wrote: post #69

Earlier quoted context omitted.

So you're saying that my definition of the term "fact" is wrong... http://www.dictionary.com/browse/fact Fact: a truth known by actual experience or observation; something known to be true. So... are you saying that a fact isn't something that's known to be true? Because my understanding of the word "fact" is that it means something indisputable, an underlying truth upon which you can base other truths and know them…

Yes, correct. Humans are not capable of having access to indisputable truth. However, indisputable truth is not necessary for forming useful foundations for other knowledge. Humans exist in the space of the "probably good enough." What becomes a fact for you depends on how tolerant you are for uncertainty. You're probably going to die. That's almost 100%. A solipsistic universe is likewise very unlikely. Those are pr…

> If you want facts that won't change after 15 years, you should study Mathematics.

It's true, of course, in a sense, but a great book to get some perspective on the development and "historicity" of mathematical fact is Imre Lakatosh's great Proofs and Refutations.

It's a great read, as a fictional dialogue of students and teacher on Euler's "simple" formula

    V - E + F = 2
about the numbers of vertices (corners), edges and faces. They go and prove it, and oops, someone comes up with a counterexample, and they realise they have to refine things. Very instructive book.

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

Post reply on HN