Hacker News Reader

Comment by afhof | Hacker News Reader

Its too bad that most people don't have a basic appreciation of statistics or probability. Its too bad Schneier only spends a tiny paragraph on it: "Even though this will be in the news for weeks, we should recognize this for what it is: a rare event".

The chance that a person will die from a bombing, or be shot, or their plane crash, or a hurricane kill them is so small that there is almost NO reason to be afraid of it. It is far more likely that a person will die of Heart disease or cancer than any of the above. What will almost certainly kill us, we accept. What is almost inconsequential to us, we abhor.

View on HN

People don't react to terrorism because they're afraid of being blown up personally. They do it because they're afraid of the precedent that might be set by not responding.

It's statistically unlikely that you'll be the victim of gang violence, but does that mean it's irrational prosecute and punish gang members? No, because the activity threatens social stability vastly out of proportion with the probability of any given person being victim to it. In general, we punish crimes not because it's likely that we'll be the victims of one, but because if we don't the problem might grow out of hand and undermine civil order.

#citationneeded

Really now. I don't need to go much beyond the evening news to know that people spend a lot of time being afraid of things that aren't likely to happen. (What don't you know about your toothbrush that could kill you? Probably very little.)

Your model of humanity is one that's much more rational and thoughtful than what I've come to expect from my neighbors. Nobody is scared of a decline in social stability; they're scared of getting shot or robbed or blown up. Usually by people who look different than they do.

Listen to Obama's speech on the issue. It's targeted to calm the average American. Listen to the word he uses to characterize the administration's response: justice. What does "justice" mean in this context? It means nothing other than the government's response to activity that threatens social stability and psychologically remedy the injuries created by actions that upset social stability.

What does the War in Iraq or having to remove one's shoes before boarding an aircraft have to do with justice?

I don't think anyone is making the argument that there should be no justice. Prosecuting the criminals who commit murder for the purpose of terrorism is clearly a "no reasonable person could disagree" sort of a thing.

But the question remains, how is justice won? In theory we could impose martial law and suspend elections, or permanently shut all the roads and trains and prohibit gatherings of more than five people in the same place, or nuke the entire middle east. Perhaps doing those things could bring justice to more terrorists.

Well before we reach that point, we come to a line we should not cross. We come to a choice that will cost us more of our humanity than it gains us in justice. And whipping the public into a frenzy is how popular support for crossing the line we should not cross is manufactured.

All that is really irrelevant to my point, whixh is that the nature of the word "justice" shows that people are concerned about social order, not just the individual likelihood of being killed. One can argue about the extent to which any given response helps maintain order.

That's why people care about "justice" for murderers even though more people are killed by auto accidents. Murder upsets the social order. A car accident doesn't.

This might explain the political response to terrorism, but I'm not sure it applies to the individual.

Fear of setting a bad precedent which could lead to the undermining of civil order seems to be a more complex feeling than that of being the victim of an attack. It's been my experience that the most common fears of an individual are related to sudden/unexpected pain and/or death, not concern for the social good.

And if people are truly concerned over the long term social effects of a threat, why does this not also apply to large percentages of the population routinely dying of preventable causes? Surely this is just as relevant to the well-being of society.

Because people aren't just worried about death. They're worried about social order. Yes, not just governments but individuals worry about social order. It is a luxury of growing up in the west that allows people to say stuff like only governments worry about social order. Highway accidents don't have the potential to undermine social order the way terrorism does.

I think it's more that they fear random death: crazy snipers on rooftops, bombs at the park. They don't fear things that they (mistakenly) think they have a lot more control over.

heart disease: "But I eat healthily ... usually" traffic accidents: "I'm a good driver! I won't get in an accident."

It's a combination of poor risk assessment and Dunning-Kruger mischaracterization's of one's skills. I'm sure there's more to it than that.

That does not explain why people stopped flying after 9/11.

People are terrible at estimating risk.

I flew less after 9/11 because it made rational sense to do so.

Before I could turn up at the airport 10 minutes before my plane took off and go the the gate and check in.

After 9/11 I was in a queue for ages and had to arrive really early. So it was quicker to just drive.

Also deadlier.

Do you have statistics showing that it's more deadly to fly post-9/11 than it was before? Are more airliners crashing now?

Parent is probably referring to the fact that it is safer to fly than drive a car, statistically speaking.[0]

[0] - http://reason.com/archives/2006/08/11/dont-be-terrorized

That is indeed what I was referring to. Deaths due to traffic accidents increased measurably in the months after 9/11 as more people opted to drive instead of flying.

"People don't react to terrorism because they're afraid of being blown up personally."

I would disagree. My experience is that people view dying in a terrorist act as an especially horrific way to die and hence especially scary. I have had this specific conversation with people where they say it would be much worse to die in a terrorist attack than to die in a car accident.

To me it doesn't make sense. They both kill you but a lot of people see it differently.

> They do it because they're afraid of the precedent that might be set by not responding

What precedent? Isn't the whole purpose of terrorism to provoke a response -- to force a change?

most people have a better grasp of probability than you suggest. What they're scared of is not so much the remote possibility of being a bombing victim, but the complete lack of control they would have in such a situation. there's a much higher chance of being in a car accident than in a bombing, but you can mitigate your risk of a car accidnet by paying attention to your own and others' driving, and in most cases you'd have the opportunity to take some evasive action. Likewise you can mitigate your risk of getting mugged by not going into dangerous neighborhoods or suchlike. And so on.

If you're the victim of a bombing, on the other hand, you probably won't have any warning or any time to do anything about it. By the time you figure out what's happened, you're already maimed/ dying/ dead, or one of your loved ones is, or whatever. Remote though the risk is, the randomness and immediacy of it are a lot scarier. I've been in close proximity to a car bomb; it's shocking because it's a lot bigger than you are, and whether you are a victim or not has absolutely nothing to do with your skills or reaction time, it's purely a matter of luck.

It's like getting killed by a falling tree or a sinkhole opening up under you or something. The risk of that actually happening is tiny, but what's scary is how helpless you would be if it did. Unlike car accidents or confrontations with wild animals or people, you probably won't get the opportunity to think/talk/fight your way out of danger in the former situation.If you think about movies, TV etc., people are always getting in jeopardy but worming their way out of it somehow or another. We find it much more shocking when we see someone get killed without having any time to react to the danger they're in.

Yes, and that's the GPs point; that paralysing fear outweighs the logic of the situation. Indeed, I'd suggest your own comment about being able to mitigate a car accident is a prime example of our difficulty with assessing the probability of risk (you may have more control of the situation; but that does not reduce the probability of you being killed by a car).

Many many things could kill us suddenly and without warning (even heart failure!). We are disproportionately scared of some things (eg flight) because they "seem" scary, and are blasé to actually risky things because they seem familiar.

And excellent book on this is "Risk; the science andpolitics of fear" which uses case studies and real data to examine this phenomenon.

If you think that, I didn't explain myself very well. What I'm suggesting is that people are (moderately) well aware of the relative probabilities, but fear a predictable lack of control. This is entirely rational; you can be scared of something without necessarily expecting it to happen.

so to take the driving example, it's very possible that you could be hit without warning, but you can also conceive of many situations in which you do have some warning and can swerve to avoid danger or suchlike. People may even underestimate the risk by being overconfident in their ability to react, but the fact is that if you drive a lot then after a while you'll have experienced a variety of close calls or even been in some minor accidents, and you can weigh the risk of many different driving situations.

On the other hand, most people have no flying experience as pilots. So if you're a passenger in a plane and it becomes apparent that the plane is going to crash, there's nothing you can do about it. No matter how remote the risk of your plane crashing, if it does fall out of the sky then you are almost certainly fucked and there's nothing you can do to improve your chances. I know that if I fly somewhere the journey to and from the airport on the ground is statistically much more risky than the journey through the air, as I will be in much greater proximity to many other vehicles on the ground, my vehicle will have to go through numerous complex maneuvers and so on, and if freeway travel is involved an accident has a much higher likelihood of being fatal due to the speed involved.

But if I had to be in a crash, I'd far rather be it be in a car than a plane; I've been in/at the scene of car crashes before and they strike me as rather more survivable. So I'm not more scared of flying than driving, but I am more scared of a plane crash than a car crash.

Plenty of people survive plane accidents; in fact if you are involved in a plane accident in the US you have a greater than 90% chance of surviving.

Which highlights the point; it is fear that drives your viewpoint (that car accidents are preferable plane crashes) rather than statistics :)

EDIT: I don't have my copy of Risk to hand, but from memory the numbers work out something like.. probability of being in an air crash and it killing you is 1 in several million. Probability of being in a car crash and it killing you is 1 in several thousand (or ten thousands, can't remember). Either way - an order of magnitude more risky :)

I m not sure people can so freely control their emotional responses - their prefrontal cortex might tell them such and such an event is pretty rare and low risk, but their lizard brain will pretty much override that. Unless you have disciplined training, i doubt you can truly get rid of that irrational fear of certain things.

Well, I mentioned the specific example of passenger airliner falling out of the sky because the odds of survival on that are so law, as opposed to takeoff/landing accidents (like the one in Bali the other day) where the odds of surviving are much higher. Also, I am perfectly aware of the disparity in the odds. you don't need to explain that to me.

You're right but that's precisely the point -- what matters to your safety is the probability, not how much control you feel you have. Yes, you can do more to mitigate the risk of being in a car accident than to mitigate the risk of being a victim of a bombing attack. But even if you do all you can, you are still more likely to be in a car accident than in a bombing (and thus, rationally, should be more concerned about the former than the latter).

Let me put it another way; if I'm facing or even in a car accident, at least I know what to do about it in general terms. I've been in a few already and on the scene to help in several others. While the probability of car accidents is high enough to be a worry, I have many options for mitigating that risk.

Now, I'm very unlikely to be in a bombing attack, thank heavens. But I have been in one already, when a car bomb went off right outside my office just as I was getting ready to leave (I wasn't a target or anything, it was a car bomb parked outside a subway station in London that I happened to work next to). What happened was that there was a huge bang and I watched the windows swell inwards like they were covered in plastic sheeting instead of glass. Lucky for me, they didn't shatter, and lucky for everyone else it was late at night and there were no people on the street, so nobody died.

But at the time of the explosion, I was keenly aware that if the windows had burst inwards, the glass would have hit me faster than I could have ducked down behind a piece of furniture. It's this lack of opportunity to alter the outcome that bothers people; it's a situation they'd understandably rather not be confronted with at all. Helplessness is unpleasant because it negates your normal defense mechanisms.

I don't think it's a matter of statistics or probability: it's just easier said when you're not directly affected by the events.

For me personally, I completely agree when reading this article but I'm here, on the west coast, not feeling threatened directly and knowing my friends are now safe.

But what if you were passing the finish line as these devices exploded. You might well be the most rationale person in the world, you would not think about probabilities. What if you were there in the public, wondering why your life has been spared because you decided to stand further away from that point.

I agree with you that refusing to be terrorized is borderline impossible for the people who got hurt in the blast or their closest family and friends. The article is aimed at the rest, the majority who can potentially either panick or just carry on.

There's nothing in-between? Nobody's panicking, as far as I can tell, but an awful lot of people seem ready to describe any response as panic-inspired.

This wasn't just a few hundred people being injured or killed: This event, by its targeted and spectacular nature, negatively affected hundreds of thousands of people. Add together the costly, chaotic disruptions for all those individuals, and you don't need many of these "rare" events to make terrorism a much more significant problem than you're making it out to be.

Sorry, but I think you don't understand probability that much and you also underestimate how much people do probability subconsciously in the right way. Also, even experts and scientists studying probability theory are pretty confused when it comes to reasoning correctly with probability.

Look at it this way: the probability of me dying from heart disease is extremely low even though a lot of Americans have a high probability for that.

There are quite genuine concerns in AI that people have messed up uncertainty, probability and statistics.

http://www.agroparistech.fr/mia/agi-workshop/

So please stop the pop sci version of probability and statistics.

Carnap is probably spinning in his grave right now :)

Any good book on probability will show that there a half dozen ways to do probability from first principles and all of them are sort of equally valid but differ greatly.

http://books.google.com/books/about/Universal_Artificial_Int...

More people died on black friday every year. Shit more people die every year falling out of bed.

Reply on news.ycombinator.com