Hacker News new | past | comments | ask | show | jobs | submit login
Intelligent disobedience (wikipedia.org)
212 points by hhs 41 days ago | hide | past | favorite | 84 comments

Intelligent Disobedience has also found its place in Children's Rights Education offering instruments that help to keep children safe from the rare, but traumatic instances when authority figures abuse their power.

I really don't think this is rare. People do crappy things all the time, especially when they imagine they know what is best for someone else. This is the default assumption of most adults in how they relate to any children in their lives.

In sense, the parent-child relationship is kind of weird to have compared to the flight-crew/pilot or the guide-dog/disable-person relationship. The later relationships are between trained individuals (or animals) acting in fairly specific circumstances. The parent-child relationship is something everyone winds-up with, with large portion also doing the parent part. And so whatever training a parent has is minimal yet the circumstance that they deal with their children in are very general.

So a lot of problem parent behavior is ... a problem. The main way something like this "Children's Rights Education" tries to resemble the other situation is teaching kids that there are few circumstances of abuse where they legally refuse/resist. Of course, there are lot of abusive situations that the child can't legally resist and of course being a child navigating such an overall situation presents many minefields. Which is to say this approach has to subscribe to the fiction (derived from the law) that there is a fixed line between abuse and non-abuse where in fact the line is fuzzy indeed.

For the most part, the real solution to child abuse or mistreatment is to deal with the parents. America does a poor job of providing adequate support to parents. We have a lot of family unfriendly policies.

Can you elaborate? Provide some examples of other communities that do better? (Honest question.)

The US is stingier with child care and maternity leave than the rest of the world


The US is on a short list of countries that don't provide maternity leave. All the others are dirt poor. We have no such excuses.

We have terrible healthcare policies. At one time, it was somewhat common for dad to be the primary breadwinner and have a decent job with benefits, including healthcare for the wife and kids. Mom worked part time or was a homemaker.

Now, even if both parents work, the kids may not be automatically covered on either parent's healthcare policy from work. They may have to pay extra for that. (Though I'm not clear how much Obamacere changed that.)

In much of Europe, it is still somewhat common for the extended family to help raise the kids, daycare is generally more readily available, maternity leave is the norm, etc.

You don't have to try hard at all to find countries with better family-friendly policies than the US, even without having sophisticated, well-thought-out ideas about what that should mean.

As I'm from Northern Europe and have lived in Western Europe (In a country where I speak the language) for some time I'm wondering what kind of basis you have for your claim.

It's especially the following part that I'm curious about as that has not been my experience but if you have statistics then I would be happy too look at them.

"In much of Europe, it is still somewhat common for the extended family to help raise the kids...."

Here is something from 2015:

In the United States, 24% of children under five have been cared for by grandparents in the previous month (Laughlin, 2013), and a study of 11 European countries showed that 58% of grandmothers and 49% of grandfathers looked after at least one of their grandchildren aged under 16 in the preceding year in the absence of parents


I've read extensively on such topics over the years. Sometimes, my stats are a little out of date because I did a lot more such reading in my twenties and I'm now 54.

But, overall, my understanding is that what I read about differences between women's rights and family-friendly policies between the US and Europe when I was a homemaker in my twenties trying to figure out why the hell I didn't magically end up with the two career couple family I had imagined would happen has not really changed all that much. The gap in some things is less wide, but it's mostly a matter of degree, not kind.

Are those numbers directly comparable? The US figure is “in the past month” for “under five years old” and the European figure is “in the past year” and “under 16 years old.”

Having lived in several countries including the US, the UK, and Switzerland, there doesn’t seem to be much difference. It’s pretty common for older kids to visit grandparents everywhere, and somewhat less common for grandparents to look out for younger kids while the parents go do something.

Among people I know, grandparent living in the same city taking care of child when parents cant is quite normal. It is not thought as "raising the child", but more of as helping to parents.

No, the US does have maternity leave. It just doesn't have paid maternity leave.

Also known as: it doesn't take money out of my pocket to pay for your kids.

Every dollar spent on helping to care for the nation's children that are below a certain age saves several dollars on down the line on things like incarceration.

This is a penny wise, pound foolish attitude. If you want to be a "stingy bastard," the "stingy bastard" option is to take care of the kids so they don't become bigger problems down the line.

(Not intended to name call. Just intended to characterize a certain position.)

It feels like there's some kind of weird amnesia that a lot of people experience, where for part of their lives they forget they were ever children.

They read about a child friendly policy and think about "other people". Somehow their mind doesn't click in such a way that they can generalise and understand that such measures might be universal and benefit every future person[1]. And that they would have helped them too.

[1] At least, those lucky enough to have families.

I know it's difficult to disagree, but please don't strawman.

Obviously I know everyone was a kid. But we agree as a society that parents are responsible for the financial burdens of their offspring. It isn't fair to make people who choose not to have kids to pay for those who do.

And maternity leave isn't really a benefit for the child--it's a benefit to the parent. No one will ever remember their first 8 weeks of life.

Just like it isn't fair to make people who don't take vacations to pay for those who do.

I don't believe one has a right to have children in the same sense that one has a right to fair trial or healthcare. If you cannot afford to skip work for 8 weeks, then save money or don't have a kid.

I'm sorry that I unfairly characterised your position.

My perspective is that universal worker's rights (like the 40hr week, paid maternity leave, minimum yearly vacation weeks) help prevent the "trap of moloch" (context: https://slatestarcodex.com/2014/07/30/meditations-on-moloch/). The kind of system traps that ends up screwing human beings. The purpose of work, money, capitalism, government is to benefit human beings. When the pursuit of system goals ends up getting in the way of human needs then the system is broken and needs to be fixed. When we define good baselines there is no competitive penalty for organisations that provide sane benefits.

Taking vacations as an example, we need laws to define some vacation minimum because it seems self evident that all human beings need some level of downtime in their life. If we don't define a minimum then competitive pressure will squeeze this out such that only the well off ever get time off.

Our systems need human-shaped holes carved out in them to prevent the things that make life good from being optimized away.

My perspective is slightly different than yours; I suppose I believe that every child has the right to enjoy their first few weeks of life in as peaceful an environment as can practically be arranged and that all parents deserve time to physically and mentally recover from childbirth and have time to adjust to the reality of a new child.

I think this is absolutely beneficial to the child and I still maintain that in the steady state it is not actually redistributive because everyone receives the benefit and everyone pays for every other person to get it.

I wonder how many kids end up costing "several dollars" down the line because they had good, responsible parents who just had to work and couldn't be there as much as they wanted because they had to put food on the table. Versus kids who had absentee parents who didn't even want to be there in the first place, or abusive parents.

Basically, I don't think it's a given that there will be huge dividends down the line if you just provide free child care to people.

> Versus kids who had absentee parents who didn't even want to be there in the first place, or abusive parents.

While always a risk the fact remains many parents who would like to do better for their kids cannot. Finances are a huge part of that. Taking off some of that pressure means the would-be-better parents can be better.

And for those who would game the system or prove unfit there is CPS. Obviously CPS cannot be perfect itself. Still we should try our best to get as close as we can; individually, as institutions, and as a society.

It is not just whether you are there x amount of time. It is also whether when you are there you lash out of stress or whether you are happy calm stable.

It is also whether you are so overworked and tired that you cant think of kids problems vs you had time to think through kids behavior and figure out there is an issue. It is whether you are isolated in bubble of own head or whether you get to meet variety of people to understand how society works and have experience with it.

It is whether you are high confidence and mentally ok to discuss issues with kids friends parents, or whether you are low confidence on end of rope generally so you cant (or botches it).

There is no such thing as 'paid leave', other than in exceptional circumstances they all come out of your paycheck.

You can't make an employer pay for your leave. Market competition is already ensuring that all the money he could give you, he is giving you through competition. Making a 'paid leave' law does not reduce his profit margin.

If you mean FMLA then keep in mind that's a new development. And as an unpaid furlough of sorts it's only marginally better than nothing.

The “paid leave” money comes from paying you less when you’re not on leave. Little industry secret I guess.

Not true. Well I'm not sure of the US implementation but generally the state pays parental leave. The amount can be tied to a persons salary up to a point. Salary has to be competitive for those that don't want kids anyway.

Isn't that literally the job of a parent? Knowing better than the child what is best for them?

Obviously that should include letting the young person stretch, learn and exercise their own judgement, but within bounds set by the parent and with the parent there to help with any consequences.

I was a full-time parent (homemaker) for a lot of years. I always operated as much as possible on the assumption that my kids had a great deal of local knowledge about their life that I would never know and when they were little they were incapable of articulating it but still knew it.

An event that helped cement my commitment to respecting their boundaries as much as possible is where I made my son eat lunch because he wasn't eating and he tended to be skinny and I worried that I would end up charged with neglect for not feeding him enough. About thirty minutes later, he threw up all over my jacket which I actively encouraged to keep it off the cloth truck seat.

After that, I doubled down on trying to make sure there was food available that he liked and that I felt was sufficiently healthy, but I left it up to him to decide to eat.

He likely has two conditions that can each lead to requiring hospitalization to treat aversion to eating by mouth. He's never developed any such issues.

So, unbeknownst to me, I had some serious challenges to deal with. Respecting his boundaries paid off.

Kids like mine frequently end up seriously abused because the parents just keep increasing their attempts to control the kid and force the kid to do as they are told rather than coming at the issue from another angle as I chose to do.

I am on my third parenting blog, still trying to figure out how to talk at folks about such things in a way that is helpful and doesn't sound too accusatory. The intent is to offer options, not criticism per se, for people dealing with challenging children.

You can respect their boundaries while still knowing better than them.

You would stop them running into the road, as an extreme example. Good parents know well enough to communicate with their child, and provide an environment in which they can learn and grow, and also when to intervene because there is danger.

It's overly simplistic to say that parents don't have "know better" as part of the job description.

No. The job of a parent is to raise a child so they can know what's best for themselves and be able to handle problems without you. Most adults think with little evidence that they know what's best for their child. I've talked to many adults who think that their child should go to Harvard, or should go into a trade school, or shouldn't hang out with risk-taking friends, or shouldn't be gay, or should go to church. Outside of the very obvious situations where immediate safety is at risk, I don't see the evidence that most parents know what's best for their child. It's just arrogance. You should be trying to raise a capable adult rather than be protecting a foolish child for 18 years.

I remember reading somewhere that job of a parent is to teach the kid how to swim. It is a terrifying proposition since I am not always sure I know how to.

I mostly agree with you. But I also know there is a blurry line there. For example, I did some stupid stuff as a kid and it is a good things parents were there to intervene. I would not call their concern for my risk taking in that instance to be arrogant today.

But that is my perception of my individual upbringing. I have no real knowledge how parenting looks 'on average'. I know I was lucky in some ways; unlucky in others.

This is where it gets complicated, because the advice we get for parenting is a generic advice for a typically specific issue.

I am writing this as an expecting parent so take my musings as just that.

> The job of a parent is to raise a child so they can know what's best for themselves and be able to handle problems without you.

Yes, it is part of it, absolutely. But in the early years there's more to it that that.

When a four year old wants to eat ice cream all day and nothing else, are you asserting that they know better?

> I don't see the evidence that most parents know what's best for their child

That's not my assertion, my assertion is that it's their job, not that they're always good at it.

Especially young children aren't always able to explain.

Like a kid will come and say "I need a stick", and the parent is "what, you'll just hit someone", and refuse it. And the kid will cry.

Later you discover the kid needs to make a school project, with a stick, and without the stick they simply did not do the project, and failed.

And they were just not able to explain this to you.

(This isn't a real story, but it's similar to the kind of thing that happens.)

You also get "I need a dollar", and you give it to them, and later discover they had some kind of bet going with another kid (which you would have never approved of), and needed the dollar to pay the other kid.

Upshot: You have to ask the kid what's up, there is a lot of knowledge they have that you don't.

Heck, I saw this happen. The default assumption was that they wanted to beat each other or do something shady.

Another way to look at this may be something like:

Odds of an adult knowing what's best for another adult: Low to Medium

Odds of an adult knowing what's best for a child: Medium to High

Odds of an adult being able to talk with the child in an informative way in a non-emergency situation and help them make a better decision for themselves: Extremely high.

You don't have to boss the kid around and treat them like a puppet you control to help them make better decisions.

Obviously (and it should go without saying, but I will say it anyway): Emergency situations are an exception. You can and should stop a child from sticking their hand in the fire without trying to nicely and at length talk them out of it while they are in the midst of doing so anyway.

You will also de facto be making a fair number of decisions on their behalf when they are below a certain age. Infants can cry to let you know something is wrong, but they can't tell you they need to be fed, etc. Adults have to do their best to figure out what the issue is and address it.

Good parents typically don't have a policy of "Oh, just let them cry it out." They typically feel that a crying child requires parental intervention to solve whatever their problem is.

(Exception: It's okay for them to just cry for emotional reasons. I never tried to convince my kids to stop crying about being told "no" or whatever. If you know they are crying because small kids have big feels, let them cry. No big deal.

But babies don't typically cry for emotional reasons. They cry because they have a problem that needs to be addressed.)

Odds of an adult knowing what's best for another adult: Low

Odds of an adult knowing what's best for a child: Medium

Why do you think that?

Odds of knowing what's best. Approximately 0 in both cases.

Odds of knowing better than the adult/child in question though, sure, those ranges sound right.

There's a deep line between "crappy things" and "traumatic abuse" though. It's up to you how rare it is though.

Many years ago, I read an article about childcare practices in some country that valued boy children over girl children.

First born boys routinely got little "extras." They got their favorite food just because it was their favorite. The got taken to the doctor a bit quicker, like tonight instead waiting until morning to see how they felt. That sort of thing.

This was not article document horrifying abuse of little girls. It just documented boys, especially first-born boys, getting a little extra care and attention.

The difference could be measured in mortality. Boys, especially first-born boys, had lower death rates than girls.

I don't think there is a deep line at all between the two things. I think if there is a line between them at all, it's pretty darn fuzzy.

You are right, I should have said "big difference" instead of "deep line". I tried to focus on how meaning is conveyed while neglecting my own language. And there's a big difference between "big difference" and "deep line".

Back to the original topic, yours also serves as an example of how easily damage can be delivered through being systemic. I'm confident it's not hard to produce many cases that qualify as systemic traumatic abuse, but it's an exercise I'd rather not do today.

This is also the primary motivation in censorship, from social media companies to authoritarian regimes. You believe that you know better and must censor wrong information so that stupid people don't consume it and get hurt because of it.

This. Just, so fucking much this. Child abuse is so common and so accepted that most people justifiably can't even recognize it anymore.

A short documentary called “Pick of the Litter” focuses on the training of service dogs for the vision impaired, and it first introduced me to this concept of Intelligent Disobedience.

No longer on Netflix, but you can find it on Amazon Prime here: https://amzn.to/3gttRxt

If you are a dog lover, this movie will tug at your heart strings. If you are not into dogs, it’s still a wonderful character study in perseverance & human-animal symbiosis.

For programmers, you’ll start to realize that even reaching dog-like AI would be quite an accomplishment and open up a lot of possibilities for humanity.

On a lighter note, whenever I hear the term Intelligent Disobedience now, I always also think of one of my favorite organizational psychology terms, Malicious Compliance.


This is a fascinating concept, and I always appreciate wiki reads like these whenever they come through HN.

One thing I wonder, though, what's going on in this part:

>The animal understands that this contradicts the learned behavior to respond to the owner's instructions: instead it makes an alternative decision because the human is not in a position to decide safely.[5] The dog in this case has the capacity to understand that it is performing such an action for the welfare of the person.[6]

I think that's a very bold and categorical claim to make. Not necessarily because it's wrong, but animal cognition is a charged subject where it's easy to become motivated to make claims, and I think it's fair to say it's a sphere where untrue claims abound.

It wasn't clear to me that the source for [6] is sufficiently authoritative. It seems to be derived from interviews with dog owners. Here's a quote from [6] on the cited page:

>"The competent guide dog can recognize dangerous situations and, even when commanded to engage in a articular action, can decide to disobdy in order to protect the owner's welfare.. All of the interviewees spoke of their dogs as regularly involved in behaviors that were not understandable if one were only to see dogs merely as automatons responding to instinct or behavioral conditioning."

Anyway, I bring this up because that sentence stuck out to me as something that felt highly motivated, and in general it's a feature I think you see sometimes in wiki articles - bold claims, almost vulgar in their simplicity, that just go ahead and declare a truth and don't try to couch the language in terms of earned institutional understanding that fully merit the claim. Again, not saying it's wrong, it just stuck out.

You do realize this is Wikipedia, right? There is no bar to editing it.

The rules of the game there are basically, if you can source something to a published book or newspaper, even if idiotic; it's game. Trying to change anything in case of a conflict takes time, effort and allies. And the corporation that owns Wikipedia not being hostile ("Framgate" has been publicized about).

And this is one of the few good Wikipedia's, where at least some important articles will receive due care from capable and benevolent editors. Try checking out the "Croatian Wikipedia", or some of the other Wikipedia's relevant to a smaller country.

Thanks for explaining to me how the fifth most visited english language website in the entire world works. A site that I've been visiting on a near daily basis for past 15 years, and to which I've contributed numerous edits and where I've become way too familiar with weirdly specific controversies (my favorite being probably the 'Verifiability Not Truth' debate). But wow, you told me that Wikipedia can be edited, which blew my mind. So thanks to your extremely helpful insights, I'm all up to speed now. But... did you have anything constructive to add that related to my comments on this particular article?

In the example cases given, the animal's behavior seems to be related to its own survival. Training adds more stimulated links which also aid in its survival standing. We just label this intelligent in our human/social sense.

I think that's just a need to introduce conciseness to the situation. It's less about comprehension and more about "despite an express general request, the device (in this case, a dog) responds differently in this specific situation, unless receiving a countermanding specific command".

i.e. it's not about cognition, it's about the tool (in this case, the dog) having certain behaviour.

For instance, consider a hypothetical system with pop up dialog boxes requesting confirmation (yes undo is better, etc.). A user hits Del signaling they wish to delete an entry from a list, and a popup is displayed requesting confirmation. The user hits Enter within 10 ms of the appearance of the popup but the system ignores the input. One might describe this as "The program understands that the user did not actually confirm since they did not have sufficient time to do so" and not actually mean "The program acquired sapience within 10 ms and proceeded to overrule the human".

Teaching a dog to not run into traffic, even if ordered to do s (a) is quite doable, and (b) doesn't the dog to have any notion of the assisted person's mindset.

That's my point. It's just tool behaviour being described and not cognition.

The article briefly mentions teaching this principle to children, and references this article, which I found interesting: https://blinkthinkchoicevoice.com/resources-and-tips-for-tea...

"Blink, Think, Choice, Voice" is a pretty poor mnemonic though. They compare it to "Stop, Drop, and Roll", but those instructions make sense even before you read anything else about it.

It's weird and forced. "Blink, Think, Choose, Voice" would be much more consistent, but I'm guessing someone really wanted it to rhyme.

"Blink, Think, Choose, Refuse"

"Blink, Think, Make a Stink"

or, the classic, "Question Authority"

"Refuse" implies an outcome.

I also think "blink" in the first place is a bad term. I understand what they're going for, but it's not an intuitive instruction.

When I was a child I watched a blind man repeatedly kick his service dog while screaming obscenities because it would not let him cross the street outside of a cross walk. I think about that poor animal often.

Yeah, I saw a CCTV video on Australian news where a train stopped short of the platform and the blind owner dragged his poor dog off the edge. That sucks for everyone.

Some people seem invested in the idea that animals are not intelligent, do not have feelings, are not persons etc, despite so much evidence to the contrary. Why is that?

Assuming you're genuinely asking. I believe it's because feelings and intelligence would mean that animals suffer pain like we do, which would imply that we are cruel for eating or using them as we do.

The best analogy I can come up with that many Americans would relate to would be slavery, when slave owners would insist on dehumanizing them. They were both financially and emotionally invested in the idea that slaves could be used and abused.

I think this article describes a quality of communications that many discussions of artificial intelligence seem to miss.

Intelligent disobedience is effectively a formalization of the way informal language-interactions often work. When you ask a person to do something, the response is can easily be a request for clarification, a comment on possible negative results, some suggestions about alternative approaches and so-forth. Often, you get a decision after a few rounds of this. Basically, a good portion of language interactions involve a bargaining and clarification process.

Now, consider the average "AI goes wrong" argument. The classic scenario is someone asks a general AI to "build a lot of paperclips" and, like Disney's Sorcerers Apprentice, the AI converts the entire earth into a paperclip factory. Here, the interactions between AI and human fail to be anything like human-to-human informal interactions. And this hypothetical scenario seems implausible just given that the scenario also posits vast understanding in the AI, an understanding which would seem to encompass language understanding such that AI could do that back-and-forth bargaining approach (the human might ask for such behavior to be avoided but theoretically we're talking the human that invented the AI and also has this sort of meta-understanding).

I think your missing the point of the paper clip maximiser example. All it’s for is to show how complex reasoning and intelligence are, and that strong AI has to be about a lot more than just solving problems.

Arguing that a strong AI would have to be smart enough not to make a mistake like that is really the purpose of the example.

The paperclip AI knows exactly what you want, it just doesn't care. Any naive way of trying to make the AI care will fail because we don't know how to design good measures that can't be hacked.

I think back-and-forth bargaining is an important and valuable concept. And while I think you are right, I suspect that the capability to bargain in response to commands would be fraught with its own category of concerns and potentials for undesired outcomes.

I also think the paperclip concept could be rehabilitated and treated with a charitable/steelman interpretation, where it's regarded as a toy example of unanticipated adverse outcomes.

Reminds me of the Milgram experiment https://youtu.be/rdrKCilEhC0 , where people were followed up with, to ask if there was anything the man they were electrocuting could have said to get them to stop electrocuting him, no matter what the investigator told them to do (36:30 is an example of one)

> In 2012 Australian psychologist Gina Perry investigated Milgram's data and writings and concluded that Milgram had manipulated the results, and that there was "troubling mismatch between (published) descriptions of the experiment and evidence of what actually transpired." She wrote that "only half of the people who undertook the experiment fully believed it was real and of those, 66% disobeyed the experimenter".[23][24] She described her findings as "an unexpected outcome" that "leaves social psychology in a difficult situation."[25]

Tangentially, I'm reminded of something a high school teacher mentioned to me. The context was military service, but of course this applies in any organization.

'The way to get rid of an undesireable superior is to do only what you are ordered to do.'

The supervisor's task will fail in some way, and new leadership will be needed.

I've kept this in mind, not as a way to undermine leadership, but as a lesson in what good leaders need to know — that getting things done requires everyone to be be empowered and act in some degree at their own inititive.

Reminds me of Asimov's Second Law of Robotics[1].

[1] https://en.wikipedia.org/wiki/Three_Laws_of_Robotics

Dogs are beautiful creatures

small anecdote, saw couple years ago - a woman with a dog leaves the dog park and pulls the dog to follow her toward the parking, the dog pulls into opposite direction and is pretty insistent, after some moments of mutual pulling the woman finally "Oh! you're right! I parked there today!" and follows the dog to the other parking.

I have been fortunate to work under a few managers who value this trait.

Those who don't often get its evil twin - malicious compliance.

I'm in a unique position where I work for one division, but I'm occasionally loaned to another area for a special project. Sometimes it's fine, but when my attempts at persuasion have failed, malicious compliance was what they got: with fully documented explanations regarding the limitations of what was delivered. "This is not an adequate tool for the intended purpose" was my comment on one such project. They used it for a single "business cycle" before they came around to my way of thinking.

Also known as "give'em rope". I did that to a few former "bosses". After a few months they were let go, hi hi.

Great reddit sub to read btw :)

It has the flavor of sticking it to the man and it has great comedy potential (like 70% Dilbert is about it). That said, you can only be passive-agressive for so long before you get jaded and unhappy.

> you can only be passive-agressive for so long before you get jaded and unhappy

I think you’re putting the cart before the horse here

Service animals do the best for the person they're trying to help!

I feel like this is obvious, and, therefore, if you agree with me, pls don't upvote. I feel like recently there's been an emphasis on individual responsibility which is at the very least not sustainable in the long term. I'd wrap my point up but I feel the original poster hasn't really put in the effort :-)

A different framing of this would be multi-heuristic decision making.

Can have applications in AI / AGI alignment.

Really good documentary on Disney+ about the training dogs go through, and how they reinforce this type of behavior is really interesting

"This behavior is a part of the dog's training and is central to a service animal's success on the job."

Now get back to work.


I thought of it the other way:

Essential service workers are the service animals and the politicians are the disabled.

Intelligent disobedience may be required to prevent the politicians from walking us all off the edge of a cliff. They're clearly far too busy yelling into their respective echo chambers to see the cliff coming.

Agreed, except the point of them being unable to see the cliff. For me it's almost as if they consciously decided to test their freshly-baked algorithms of gainful cliff falling...


I agree, it is much broader. Let's settle on the literal meaning: a medium is a carrier through which ideas propagate.

You're getting downvoted, but you have two points.

Almost everybody believes a sizable part of the population to be brainwashed, they disagree on what part it is however.

And I do very much believe that it is a discussion worth having for society at large. Do we believe that we are generally aware of the dangers of things we do? Is outlawing oversized sodas the right thing to do? Would it be okay to allow them only if you can use a code word that signals that you do, in fact, understand the risk? Do we not like that approach at all, or do we just not want to deal with deciding who understands the risk and is able to walk down the stairs and who isn't?

Let those who don't understand the risk learn the hard way. Free will is not only about doing right, it's also about making mistakes. Now it's obvious that the trend is biasing towards enforcement of the right-doing regardless of consequences and impact on individual freedoms. Almost all writing of Stanislav Lem, for example, is dedicated to this one theme. And extrapolations he made are only becoming more and more relevant as the time goes.

Having said that, the freedom to fail should only apply to individuals. Corporate or government entities should never be treated as humans and be allowed to fail the hard way. First, because the magnitude of consequences are incomparable, and second, because the strings of corporate responsibilities are entangled in such a way so as to lead nowhere. They can always lose a head and grow another one.

As to crossing of one's individual freedom into another's, that's what the service men are for. But it was alright while they operated under presumption of innocence. Now they are clearly trying to render people guilty until proven otherwise. That was one of the points in my first reply.

One large issue is that you'll often not have a chance to learn from your mistakes and do better in the future. If you jump off of a building, you die. If you drink the big gulp for 40 years, you get diabetes. Sure, you may have learned from it, but you can't start over.

I'm still in the camp of "let them", but it's a mixture of "it's hard to figure out who really knows the risks", "I don't want to live in a world that's optimized for perfect safety and takes away all freedom to achieve it" and "we want people to take risks, even giant risks, even when they clearly have no idea how large the risk is, because we'll advance much quicker because of it, we just don't want all people to take those risks at the same time".

If we let them crash and burn.

Shouldn't welfare and free will go directly against it each other?

If someone crashes and burns, society has to pay for it through taxes where all members of the society suffer because one of them has crashed and burned.

Meanwhile, if you remove this person's agency to crash and burn, you've saved society from scarifice but you removed that person's free will.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact