Have you ever noticed people who you regarded as smart/intelligent, show themselves not to be that smart when they talk about something you know a lot about?
Say, you come to Hackernews/Reddit and usually see people on these sites as knowledgeable people. But when they talk about a topic you know very well, you realize they aren't that smart.
And, if this happens to the topics you know a lot about, what about the other topics you don't know much about? Are they wrong about those too?
So my question is: is there a name for this "effect", that you "lose trust" on someone you regarded as knowledgeable when they talked about something you know well?
We all sometimes have a tendency to talk about the things we know jack about while posing as experts because we read an article in the Guardian about it once, or an abstract from a paper. Some more than others.
In the world of show business and television, I’ve seen celebrities — singers, stand up comedians, those kinds of people — often being asked about their stance on complicated political issues or life advice as such — and they spoke bullshit with a mild air of authority instead of running away, which a sensible person should have done. The thing, though, is, that the fact they are celebrities means there are crowds of fans who will listen to whatever they say as if it was God’s own truth.
Hey, you can treat what I just wrote as uninformed bullshit, likely you even have reasons to, like data to contradict my statements. It’s ok. I’m likely a fool.
I think often they're asked because they've expressed interest in a topic or cause, and they realize that any benefit their name and the exposure they have at the moment is wasted if they ignore questions about it.
That they misstep is probably no different than any fledgling PR person the first day on the job, and sometimes I don't doubt it's media baiting them with questions to expose their lack of knowledge of some areas of the topic, as it makes for good coverage.
A lot of the times I give celebrities a lot of leeway in their knowledge, because most the time it's obvious that they just want to help because they feel fortunate for their current situation and want to give back. Sometimes that seems to be misapplied because of their lack of deep understanding on the topic, but at least their hearts seem to be in a good place.
It’s likely worse with celebrities who are generally smart and well-read on a wide range of subjects, who think that because they think they know a thing or two, that they should bestow their pearls of knowledge upon their audience.
The reality, though, is that they are in fact an illustration of the popular understanding of the Dunning-Kruger effect, because they don’t know how much they don’t know, sometimes even after a demonstration of the contrary.
I know about at least one such person who gave very thoughtful interviews on politics and blogged opinions and how he knew how to put things right, ran for a seat in the parliament twice, actually got elected both times, and failed miserably at his job both times, ending his mandate prematurely.
> the popular understanding of the Dunning-Kruger effect
You had me (my upvote) at hello (the acknowledgement of the popular misunderstanding). ;)
We are all fools on many topics, most topics. One of my mentors in the past used to say there was no substitute for experience in an area. If you were joining a project and the members had a head start you should listen and learn because their experience was way ahead of your opinions on how it could be done.
Why not participate in the norm of stating the source of your claims. It's easy enough to start a sentence with "I read in the Guardian" or "If I remember correctly" or "I heard from a professor once".
It even helps just to share your credence in your claim, e.g. "I suspect this is true", or "I'm pretty sure this is how something works".
> “Briefly stated, the Gell-Mann Amnesia effect is as follows. You open the newspaper to an article on some subject you know well. In Murray's case, physics. In mine, show business. You read the article and see the journalist has absolutely no understanding of either the facts or the issues. Often, the article is so wrong it actually presents the story backward—reversing cause and effect. I call these the "wet streets cause rain" stories. Paper's full of them. In any case, you read with exasperation or amusement the multiple errors in a story, and then turn the page to national or international affairs, and read as if the rest of the newspaper was somehow more accurate about Palestine than the baloney you just read. You turn the page, and forget what you know.” (https://en.wikipedia.org/wiki/Michael_Crichton#Speeches)
To what extent you think this should affect our trust on news sources/websites like Hackernews?
Before even trying to assess the accuracy of a story, I have to take a step back and assess the competence of the source. Read some other stuff they've written. Is any of it on matters I'm well-acquainted with? Do they seem fuzzy on basic facts, or have clear and cogent explanations? Does any of it open my eyes to new perspectives even on things I thought I knew? Is the new story in a field they seem to have competence in?
Then, I look at the facts presented in the story in question. Assume I don't have my own sources for the facts, but I can look at how they're presented: Are they clearly explained and sourced? Do they have the logic wrapped in a lot of inverting layers (judge blocks appeal of injunction against measure that would've prohibited another activity, wait what? is that a green light or a red for the activity?) which they successfully unwrap for the reader, or are we left to fend for ourselves? Can I take some of the named sources and go look at their own writings and see if they're being quoted accurately?
Then I look at the emotional language surrounding the factual statements. Do they seem to have a horse in this race and incentive to push one side, or is their opinion merely a product of being informed about it? (I'm willing to accept that researching a story naturally produces an emotional narrative, and conveying that can be an important part of context, but the facts should shape the narrative, not the other way around.) Is there enough information that I can imagine the other perspectives that might surround these facts? Do I want to go dig some up?
Safer and easier way: don't trust anything.
When you see this situation, ask yourself "why is this person wrong? why are they so confident in their opinion?" Often its because we think we know more than we actually do, we believed the things we read without bothering to give them the basic smell test for rationality. Where someone involved in the area has had exposure to the basic realities the readable sources do no or cannot impart.
Computer folks are especially subject to this error, I think, because all of our work is essentially readable, all the knowledge is communicable via text, and almost all the "rules" were created by human design that intended rational ends, even when it missed.
Carpenters by contrast get early and unavoidable lessons in the perversity and stubbornness of the real world, the effects of entrapped tradition, and the occasional need for black magic and blood sacrifice to get the job done.
This doesn't make them any less predisposed to error than the computer scientist in areas outside their competence, but it informs the shape of the errors they make.
I do know a couple of very successful people who aren't that smart but are good at selling themselves/their ideas. Some people look down on this, but it is a good skill to have.
HN has a problem where the people who comment are often not that knowledgable but can sound like they know what they're talking about. This is particularly bad in posts asking for advice. Allowing downvoting (HN didn't always have it IIRC) actually makes this problem worse. Also for some emotionally charged topics people will downvote true things that make them feel bad.
It makes me wonder: should this make us disregard many more discussions? Because what if the comments on those are as “untrustworthy” as the comments shown in that thread I know about?
Did you?
Often, on HN and Reddit, the most confident or authoritative sounding statements are also the most wrong. It always baffles me just how confident people can be in themselves when they really have no good basis for that confidence.
I think part of the problem is that we collectively tend to reward the "confident sounding" comments because we wrongly assume that only an actually knowledgeable person would have authored them.
"The fundamental cause of the trouble is that in the modern world the stupid are cocksure while the intelligent are full of doubt" — from The Triumph of Stupidity, 1933.
However to your point, Michael Crichton describes the rather sad "Gell-Mann amnesia effect": One tends to still trust the other pages of the magazine.
Let’s take an example: If I want to know what’s going on in Yemen this week, I’m probably better of reading The Economist than borrowing a book at the library. But if I want to get a deeper understanding on the situation in Yemen, I’m probably better off reading a book rather than a news article.
In the historical academic and intellectual epicenter of the US (East Coast college towns), the rule about not talking about a subject in which you are not an expert is still somewhat followed.
It is completely shattered in SV and coastal cities + DC
Case in point:
"Zero cases by the end of April, bruh...you gotta stop panicking bruh!" [0] AKA, the reason why Boston can't stand Elon Musk.
[0] https://twitter.com/elonmusk/status/1240754657263144960?lang...
As in: in academia if you say something, people are trained to analyze the veracity of that, and to judge you based on your methodological/academic rigor. Meanwhile, in SV people judge you on past projects and financial conditions, which are not tied to knowledge about reality at all. I don’t know if any of what I am saying is factually correct, I’m just guessing and wondering. Don’t take it as fact, just my biased, un-founded opinion created around what I see online.
When the science is knocking on heavens door after years or decades of struggles they feel the opportunity of commercial success and come out of the woodwork to make the last effort to finally push the pile over the goal line by making adjustments in how human capital and financial capital are deployed.
This may be part of the rub: there are a bunch of bullshitters on the internet.
Unfortunately, this kind of humility/modesty is not rewarded. As we see in the world around us.
I think this is only shocking to people who believe in a general g that simultaneously makes people smart in everything. There isn't a person alive that doesn't know enough facts that I don't know to take the rest of my life to enumerate, even if it has to get down to the level of knowing that their grandmother likes raisins in their oatmeal.
The closest we have to g is reason, and people whose reason is rigorous have rigorous reason everywhere that they aren't being willfully ignorant (like declaring that normal reasoning doesn't apply when it comes to X.) I'm not ever disappointed in the intelligence of people I know with good reasoning skills, unless you count groaning when I realize how long it's going to take to get them to speed on something I have a lot of knowledge about.
Smart is only capacity for noticing and understanding new things. It does not mean the person has presently a lot of understanding of things (knowledge) or that it has good judgment to not say untrue things in area they have no idea about (wisdom).
A person that is smart but has little knowledge will not be able to keep up in discussion with you in your area of expertise even if they are actually more intelligent than you.
Give them some time and see if they can pick up things fast or are slow to understand new concepts. This is better way to judge if somebody is or is not smart.
**
Imagine following situation: you are talking about something you know a lot about with a person that has no knowledge in your field of expertise.
That person decides to learn your field.
Let's say, in 2 years they know more in your field than you will ever know and they even are able to get a lot of original insight that leads to some astounding accomplishments.
Ask yourself, was that person "not smart" when you first talked to him/her? Or maybe they were always smart but they still needed to invest at least some time and effort to learn the field?
I have a new coworker who isn't very knowledgable, but she's able to understand new topics quickly, once I've explained them. She's often nervous about her lack of knowledge—rightly so, in all honesty—but I also think she's exceptionally smart. Certainly more than she realizes.
I change it up by assuming everyone is a dumbass but just informed on their one subject matter. Rare to meet people who are good at multiple things or, more importantly, good at learning multiple things. Usually I take smarts as being your ability to learn - not your amount of deep knowledge on one subject.
Fortunately, it’s easier than ever to be smart. We have an amazing assortment of great teachers online that do it all for free in a variety of formats that are accessible. Sadly, feels like relatively few people access the information.
Coincidentally, I met him later and he talked about his experience with the journalists. He was wise to their laziness, and he made them all read their notes back so he could be sure they got it right.
I think in principle the reporters know they should be careful what they write, but in practice... deadlines, you know.
Or, to put it the other way, stupidity and ignorance are two separate things. All of us are ignorant of more things than we're knowledgeable about -- even the geniuses among us.
Get a really smart person to talk about something that they have little special knowledge about, and they aren't going to seem any smarter about it than most other people.
Briefly stated, the Gell-Mann Amnesia effect is as follows. You open the newspaper to an article on some subject you know well. In Murray's case, physics. In mine, show business. You read the article and see the journalist has absolutely no understanding of either the facts or the issues. Often, the article is so wrong it actually presents the story backward—reversing cause and effect. I call these the "wet streets cause rain" stories. Paper's full of them. In any case, you read with exasperation or amusement the multiple errors in a story, and then turn the page to national or international affairs, and read as if the rest of the newspaper was somehow more accurate about Palestine than the baloney you just read. You turn the page, and forget what you know.
2. People usually have a lot of shallow knowledge on many things but deep knowledge on very few (if at all)
3. Many things are complicated enough that shallow knowledge==wrong knowledge from an expect point of view (the same way high school physics usually don't have friction, you sacrifice correctness for simplicity)
You could see that happening on Twitter during covid. VCs, who specialise in tech, were suddenly experts on medicine as well. Because the feedback loop has convinced them they are smart so it's hard for them to not buy into that.
And just to be clear these people are wicked smart a lot of times. Just that expertise in one area doesn't often translate to all others.
It's not modesty of sorts, I am often involved in life or work situations where everyone around me seems to have understood something highly relevant to the situation, but me.
And the few times I mention it, people laugh.
"Hu hu, good one!" "Yup..."
In any case, you read with exasperation or amusement the multiple errors in a story, and then turn the page to national or international affairs, and read as if the rest of the newspaper was somehow more accurate about Palestine than the baloney you just read. You turn the page, and forget what you know.
To be clear, I think I'm as guilty of this as anyone. It is very hard to recognize the limits of your own knowledge. It isn't really a matter of being "smart" as I see it. It's a matter of discipline and intellectual humility. You need to very intentionally practice checking yourself, being less certain on purpose, seeking to find knowledge rather than spread it, and even knowing all of that, I still find it very hard to do.
I don't think this is exactly the same thing as Gellman Amnesia, either. I'm not even sure Crichton was really correct about that kind of thing or at least not saying what people seem to be taking away from it. It can easily be the case that a person or publication you're trusting to be correct about something in one arena of knowledge really is correct there, but still doesn't know much about something you're an expert in. I think it probably is fair to apply this to reporters, because no disrespect intended, but they're not experts in anything except reporting, not even whatever specific beat they're assigned to. They're relying on the expertise of others to be correct but they may not be all that qualified to really evaluate the claims they're furthering. But it's equally bad to universally trusting everything you read in a newspaper to just assume they're wrong about everything because they were ever wrong about one thing you happen to know a lot about. That may be the case, but you shouldn't just assume that. This gets back to probabilistic thinking. You should understand that the probability of a claim you're reading somewhere being correct is rarely if ever 0 or 1, and your updates on the trustworthiness of a source when they prove to be correct or incorrect one time should not only be multiply by infinity or multiply by 0. You have to learn to live with uncertainty.
Case and point: Richard Dawkins is a genius in biology but a novice in philosophy. However, that doesn't stop him writing a best selling book which is disregarded as comical by atheist philosophers.
If you qualify it, such as "as I understand it ...", you will not.
I think the latter is seen in our age as a sign of weakness, since you are not expected to have a sense of shame if you are proved wrong. Sadly.
It’s when it comes to other topics that we wildly diverge in conclusions.
The above doesn't stop me from making guesses about other fields. Does that make me an idiot? No. I do try to be upfront about my naivety, but, I suffer Dunning-Kruger like everybody else, and sometimes I grossly overestimate my confidence. It happens. And that's how we learn.
If you want to talk to somebody smart, look for their expertise. If you want to feel smarter than everybody, only pay attention to your expertise, but remember, you're "not smart" too.
The funny thing about "Gell-Mann Amnesia" is that people make a career out of doing their best to understand 10 years of research in 10 hours, and repackage that for people to understand in 10 minutes. Recalibrate your expectations.
My theory is there are some fundamental cognition defects in our sapiens' brain which was evolved from "small data" environment (compare to today's bid data environment) so the biochemical reactions inside our brain on the conclusions from Bayesian model are not much different from the reactions to the conclusions from witness. That results in many people don't differentiate beliefs(very strong but have never seen the truth) from "solid facts" which is equivalent to the topics that you know very well.
The most common places to observe intelligent idiot are the areas related to ideologies. For example there are tons of politicians showing very strong intelligent signals: articulate and organize their thoughts very well, with solid logical reasoning, while actually talking bullshit which only can be noticed by the people knowing the ground truth.
In technology areas there are less but still some. People are easily confused with strong intelligence signals with the real intelligence.
It's not black and white. Some time real smart people also make the same mistake. Give a couple of examples: Sam Harris, Bret Weinstein , I consider both very smart, rational, evidence based,scientific thinking guys according to the depth of their thinking. However I noticed they have some serious mistakes on their judgements about some matters that they don't know the ground truth. I still highly respect their intelligence but not the same as before. They are not intelligent idiots but just made small mistakes. Meanwhile I think I might have the same problems that I don't know since they are smarter than me and still make mistakes.
Here in HN with a high density of smart people, there are tons of chances to observe intelligent idiots. Just check those non-technology related hot topics.
Gell-Mann Amnesia is rather strictly defined by its creator in the journalism domain, but perhaps it could be stretched to include the commentary here at Hacker News.
Still, I think that tagging your answer as "amnesia" does not work - unless you, yourself, manage to forget to be skeptical... on topics where you are not expert and commenters are still show-casing Dunning-Kruger.
And it's hilarious how many of them there are here!
