Schneier on Security
A blog covering security and security technology.
« Zero-Tolerance Policies |
| The Problems with Unscientific Security »
November 4, 2009
Fear and Overreaction
It's hard work being prey. Watch the birds at a feeder. They're constantly on alert, and will fly away from food -- from easy nutrition -- at the slightest movement or sound. Given that I've never, ever seen a bird plucked from a feeder by a predator, it seems like a whole lot of wasted effort against not very big a threat.
Assessing and reacting to risk is one of the most important things a living creature has to deal with. The amygdala, an ancient part of the brain that first evolved in primitive fishes, has that job. It's what's responsible for the fight-or-flight reflex. Adrenaline in the bloodstream, increased heart rate, increased muscle tension, sweaty palms; that's the amygdala in action. And it works fast, faster than consciousnesses: show someone a snake and their amygdala will react before their conscious brain registers that they're looking at a snake.
Fear motivates all sorts of animal behaviors. Schooling, flocking, and herding are all security measures. Not only is it less likely that any member of the group will be eaten, but each member of the group has to spend less time watching out for predators. Animals as diverse as bumblebees and monkeys both avoid food in areas where predators are common. Different prey species have developed various alarm calls, some surprisingly specific. And some prey species have even evolved to react to the alarms given off by other species.
Evolutionary biologist Randolph Nesse has studied animal defenses, particularly those that seem to be overreactions. These defenses are mostly all-or-nothing; a creature can't do them halfway. Birds flying off, sea cucumbers expelling their stomachs, and vomiting are all examples. Using signal detection theory, Nesse showed that all-or-nothing defenses are expected to have many false alarms. "The smoke detector principle shows that the overresponsiveness of many defenses is an illusion. The defenses appear overresponsive because they are 'inexpensive' compared to the harms they protect against and because errors of too little defense are often more costly than errors of too much defense."
So according to the theory, if flight costs 100 calories, both in flying and lost eating time, and there's a 1 in 100 chance of being eaten if you don't fly away, it's smarter for survival to use up 10,000 calories repeatedly flying at the slightest movement even though there's a 99 percent false alarm rate. Whatever the numbers happen to be for a particular species, it has evolved to get the trade-off right.
This makes sense, until the conditions that the species evolved under change quicker than evolution can react to. Even though there are far fewer predators in the city, birds at my feeder react as if they were in the primal forest. Even birds safe in a zoo's aviary don't realize that the situation has changed.
Humans are both no different and very different. We, too, feel fear and react with our amygdala, but we also have a conscious brain that can override those reactions. And we too live in a world very different from the one we evolved in. Our reflexive defenses might be optimized for the risks endemic to living in small family groups in the East African highlands in 100,000 BC, not 2009 New York City. But we can go beyond fear, and actually think sensibly about security.
Far too often, we don't. We tend to be poor judges of risk. We overreact to rare risks, we ignore long-term risks, we magnify risks that are also morally offensive. We get risks wrong -- threats, probabilities, and costs -- all the time. When we're afraid, really afraid, we'll do almost anything to make that fear go away. Both politicians and marketers have learned to push that fear button to get us to do what they want.
One night last month, I was awakened from my hotel-room sleep by a loud, piercing alarm. There was no way I could ignore it, but I weighed the risks and did what any reasonable person would do under the circumstances: I stayed in bed and waited for the alarm to be turned off. No point getting dressed, walking down ten flights of stairs, and going outside into the cold for what invariably would be a false alarm -- serious hotel fires are very rare. Unlike the bird in an aviary, I knew better.
You can disagree with my risk calculus, and I'm sure many hotel guests walked downstairs and outside to the designated assembly point. But it's important to recognize that the ability to have this sort of discussion is uniquely human. And we need to have the discussion repeatedly, whether the topic is the installation of a home burglar alarm, the latest TSA security measures, or the potential military invasion of another country. These things aren't part of our evolutionary history; we have no natural sense of how to respond to them. Our fears are often calibrated wrong, and reason is the only way we can override them.
This essay first appeared on DarkReading.com.
Posted on November 4, 2009 at 7:12 AM
• 64 Comments
To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
Unlike hotel fires, predators adapt to their prey. A change in sparrow risk policy that concludes that the odds of being attacked at the feeder is too low to bother would not go unnoticed by the hawk.
Got a cat that has jumped up to a feeder (over 5' from the ground, on a pole) and snagged birds. Before he got too fat and slow.
So it does happen, sometimes.
did what any reasonable person would do under the circumstances
I dunno. I once did that, and then I heard fire trucks and smelled smoke. I almost waited too long.
Second example: squirrels crossing roads now are more likely to run straight across, rather than the (seemingly suicidal) halfway across, reverse, then reverse again.
Traffic is killing more squirrels than hawks, it seems, and the behavior has evolved in response in a matter of decades. (okay, probably 10 squirrel generations).
So saying that humans are saddled with behavioral traits determined 100k years ago isn't quite the argument-ender that one might expect.
Also, consider the people at the WTC who returned to their desks because they thought it was a false alarm.
Whoa. HTML markup no longer allowed?
ital underline bold
Nope. No more html. Too bad.
"Unlike hotel fires, predators adapt to their prey. A change in sparrow risk policy that concludes that the odds of being attacked at the feeder is too low to bother would not go unnoticed by the hawk."
On an evolutionary scale, both predator and prey adapt. On an individual-life scale, neither adapt. Just as sparrows don't change their habits if there are fewer hawks around, hawks don't change their hunting habits if there are more agile sparrows around.
"So it does happen, sometimes."
Of course. Reduced risk is not the same as eliminated risk.
On the contrary: I spent this past weekend in Shenendoah National Park, where the wildlife (deer, raccoons, even a bear) aren't the least bit afraid of people. The deer will let you come almost close enough to touch them. They have learned that, contrary to what their instincts tell them, we are not a threat there.
One thing I find very interesting about the animal reactions to threats is that their behavior obviously has evolutionary value, yet most of time they are reacting to nothing. The point, for birds and other animals at least, seems to be not so much about only reacting to REAL threats as it is to react in a reasonable way.
The problem with humans isn't that we react when most of the time there is no threat, it's that we react in ways that are highly disruptive. If we were birds, when startled at the bird feeder, we'd fly away and never return to that particular feeder, even if finding another source of food was difficult.
Rather than our large brains being an asset in security thinking, they almost seem like a detriment. Birds might constantly be on the alert, but that's all instinct. A bird would never sit on his couch and really THINK about how scared he is of predators, or plan how best to avoid them. Because of our larger brains, we're able to come up with hugely complicated and invasive ways of responding to threats. The fear may be a primitive reflex, but the fact that we respond to it with invasive and insulting airport security, warrantless wiretapping and waterboarding detainees is all too human a reaction.
Just as Bruce's individual observations of bird feeders are statistically meaningless, so are those of the commentators above. Please don't be fooled into thinking your extremely limited personal observations of something are any real measurement to make assumptions about the world by. What if you have the only cat in the world that attacks bird feeders? Alternatively, what if birds get eaten at the feeder 99.9% of the time but Bruce has only been present for the other 0.1%?
Using the 9/11 WTC example as a counter-point was very disappointing. You're demonstrating fear based on one VERY VERY rare event that still has you in its grasp. How many times in the history of smoke detectors do you think there have been false alarms compared to actual threats? And what are you basing that assumption on, personal experience?
If I know I don't have the actual numbers, I try not to embarrass myself by pretending like I can make a logical decision. It's like when you hear a news report about someone getting hit by a car while walking home. Are cars suddenly a growing threat to pedestrians? How many people walked home that day without getting hit by a car? That week? That year? Funny, the news didn't report that.
Assume his example is just that, an example. Maybe it's accurate, maybe it isn't, but if you can get past that distraction and consider the point of the essay you might actually learn something.
This is nitpicking, but on the hotel fire alarms: there's another factor in the trade-off besides your own safety and sleep. If there is a real fire then someone will have to rescue you, possibly at considerable risk to themselves, certainly with an opportunity cost in terms of time available to search for or rescue other people.
Have you ever seen a sparow hawk in action?
They take small birds off feeders in a way that is startalingly amazing. Not common but I've seen a couple just looking out the window.
Likewise humans don't override the "monky brain" that well think about driving a car at dusk, most drivers have got to the point where driving is not done by the concious mind but the good old "monkey brain", they see a shadow and swerve and outside where I live it happens quite a bit at this time of the year hence the first aid box by the door...
@mut: "there's another factor in the trade-off besides your own safety and sleep. If there is a real fire then someone will have to rescue you, ..."
In Bruce's terms, that would probably be an 'externality'...
We've had birds grabbed from feeders by hawks. The feeders are then strangely empty for hours.
Bruce, I assumed your post wasn't really about sparrows and hawks. People can adapt on sub-evolutionary time scales, short enough to matter to individuals. (Whether they actually do is another matter, but those in black hats probably do.)
"Also, consider the people at the WTC who returned to their desks because they thought it was a false alarm."
IIRC, in the South Tower (I think), the evacuees were *told* to return to their desks by building security. That was after the North Tower was hit (but before the South Tower was)
For captive birds (or birds at a feeder), it still might make sense to flee at the slightest provocation as the cost in calories is lower (they have an easy source of food).
Though I am not fully sure I understand the point you are trying to make (that we are poor judges of risk because of our evolution?), I agree with some of the other commentators that your arguments are flawed.
For instance, the animal (bird) example may need further thought. It is firstly possible to domesticate animals. In other words learn them that some risks are not as they would perceive from their evolution. Secondly, indeed, in an urban environment birds may still face predators (domesticated cats, other birds, some humans) or entirely new risks (cars). Do animals really not change their habits if their are no predators in sight?
I think your hotel anecdote is downright dangerous and you apparently have not understood the point of procedural safety. It is to take away your discretion of reacting in a situation in which you cannot judge the true risk. You should go out. Others will evaluate the efficacy of the alarm systems, and I bet commercially operated hotels will not tolerate frequent false alarms.
As Mut rightly commented people like you would endanger first responders for whom the math works out differently. What, BTW, if they would ignore alarms because they could be (and are often) false? To take your reasoning to the extreme, alarms
that have a certain false positive rate are useless, and should be abandoned. I don't really believe that is what you want to say.
In fact, since you may have (as a well known blogger) a certain impact, and people may do as you say, you should realize that your statement statistically will kill people. Some may follow your suggestion and stay in bed, which they otherwise would have not done. Statistically a small percentage of these people will either suffocate or turn into charcoal. Absolutely, this will be the number of people you have killed.
Birds don't have sophisticated alarm systems like we do. But they react on their eyes and ears which have a high false positive rate. We have far better systems and that means we should ignore them?
Not evacuating when the alarm sounds is illegal in many areas. You might be at risk of a fine. The theory being that if it had been a real emergency, a firefighter would have risked his/her life to save your doubting ass.
Major hotel fires may be rare, but so are fire alarms. The real question is: what's the rate of false alarms? I've never heard one.
(This is one of my brother's most hilarious college stories, but it probably loses a lot as text.)
My brother was in a real zoo of a dormitory, full of drunks and vandals, with grafitti on every surface and all the carpets layered with broken glass and cigarette butts. In the middle of the night he was woken by the fire alarm, and promptly rolled over and tried to get back to sleep. A minute later, with the alarm still going, there was a loud banging on his door. He got up, opened the door and saw clouds of smoke rolling down the hall and one of his neighbors standing there and shouting to him over the alarm: "YOU GOT ANY MAYONNAISE?"
"Also, consider the people at the WTC who returned to their desks because they thought it was a false alarm."
Consider in what way? I'm not sure how examples are useful in a discussion of risks, costs, and probabilities.
And -- at least in one tower -- people were told to return to their desks. Still irrelevant.
"This is nitpicking, but on the hotel fire alarms: there's another factor in the trade-off besides your own safety and sleep. If there is a real fire then someone will have to rescue you, possibly at considerable risk to themselves, certainly with an opportunity cost in terms of time available to search for or rescue other people."
It's not at all nitpicking. It's an externality: it's a cost of my decision that I don't bear. This is why, as another commenter mentioned above, that some jurisdictions will fine me for making that decision; it's an attempt to force me to internalize that cost.
Interesting. About a month ago, we had a small fire in my building, but didn't evacutate when the fire alarm sounded. The alarm didn't sound/flash (audio and visiual) in our wing, and being new to the building we didn't know the weak distant warbling we heard was the alarm. In our wing, the alarm system had shorted out and caused a small short lived electrical fire. Our non-action was a result of ignorance, not evolution. No, they didn't repair the alarm system, just removed the offending circuit.
More often than not, our politicians, law enforcement, and corporate leaders waste too many resources on the threats du jour without ever bringing reason and logic into the equation. It is usually the least likely and illogical fears that strike the deepest terror in our minds, and consequently a disproportionate of resources trying to reduce or eliminate these "threats." Unfortunately, this means that the threats that are "less fearsome" but that are more likely to occur and have a profound impact are usually ignored or downplayed signifcantly as there aren't enough resources available.
What we all need to do is to step-back from a situation and evaluate a threat objectively, logically, and sensiblly. By allowing our fears to control our internal (or business) risk management processes we will always be looking over our shoulders for the threat that will likely never come.
Forgetting the whole externality thing..
I would think that the false alarm rate of a given hotel is an unknown and to risk being trapped in a fire (unless you are on a low floor and can jump to safety) over a few minutes in the cold is not a favorable risk assessment. Is that really any different then those who say: We haven't been hacked and therefore don't need to worry about it?
@Matt -- Did that cause excessive psychological suffering from the fear of not knowing if there really was a fire? ;-)
I recently did the same as Bruce (chose to not react to -- could not ignore -- a hotel fire alarm). But only based on me being on the ground floor (same thinking as @AppSec)
Excuse me? Far fewer predators in the city?
House cats kill birds by the millions. Any bird that thinks it's safe in a suburban back yard isn't going to last long.
"defenses are mostly all-or-nothing; a creature can't do them halfway"
So evolution has given us zero-discretion as a model?
"also have a conscious brain that can override those reactions. ...tend to be poor judges of risk. We overact to rare risks, we ignore long-term risks, we magnify risks that are also morally offensive. We get risks wrong -- threats, probabilities, and costs -- all the time. "
Can but don't. Some studies show that numeracy (literacy with numbers you all) tend to make people better at risk judging.
So what? teach math and the foundations of why things are or are not risky? Is that a winning stategy? We are here only because the subject interests us. Is it possible to interest lit majors or soccer moms in something as obscure as actuarial analysis?
Right now wacko's are running around saying DON'T get your flu shot. It'll do to you what it did to that cheerleader. The news (that I saw covering the indicident) didn't put that into context of the shot has been implicated in X number of harmful events in 10,000,000 doses (or however).
Then there are the people, philosophers who don't care about the payoff... herd immunity vs some individual damage. "I don't CARE if 50,000 people will live if there's ANY risk to me I won't do it." These rugged individualists are usually big fans of Ayn Rand.
You have too small a sample size. My mother (~80yo) has been feeding birds so much for so long that I swear new genii have evolved that do not know how to find food that isnt piled high for them. She loses probably 1 cardinal, dove or finch every 3 days to hawk(s) at the feeders.
What I find even more fascinating is that hummingbirds, which must burn more energy per pound than any other creature, will spend hours fighting each other for dominance of the feeders, despite the fact that. I dont see how they can get enough nutrition to offset that. For that matter, I dont see how a creature can afford to hover while sipping nectar, seems like it would burn more energy simply hovering than it gets back from the nectar (nevermind the travel cost of going bloom to bloom). On the other hand they are beneath the hawk radar...
From this data, it looks like the US is far more prone to hotel fires than the rest of the world, with about half the world's fires.
Of course, that's not at all likely, but it goes to show that we don't know the risk terribly well.
Has an interesting point. From my personal experience 99% of fire alarms I've heard (including drills and testing) were false alarms.
But my training from kindergarden on was fire alarm line up, grown up opens door march out and gather in the appropriate spot. Once in 24 years of schoolin' we had actual smoke, alarm and _nobody_ panicked. We just left the building in a reasonably orderly fashion (for high schoolers) my disreputable friends and I just kept going but that was something not related to the fire.
Some controls are aimed at crowds of people not the individual.
Oh there was a partial evacuation on a client site a couple of weeks ago. They emptied the first 3 floors but we didn't hear any alarm on the 8th 9th or 10th floors. People were pissed that they weren't evacuated too. (Me and my disreputable co-workers would have just kept walking to the nearest mccormick and schmidt but that's something else)
We keep our hawk feeding station constantly filled with birdseed. Several times a week we'll see a nice hawk (broad-winged or redtail we get both) sitting on our fence or in a tree. We often find piles of feathers or other leftovers in the yard. Generally, the smaller quicker birds can make it to the trees in time. It's the slower city pigeons that get eaten.
Can you explain mocking birds then?
They will attack Hawks and other birds of prey if they get too close to their babies. They almost never get eaten, and most of the time drive the bird of prey out of the area.
Why have other birds not learned this tactic?
Whenever we have what I perceive to be a fire alarm (many modern alarms have some weird tone/volume/pulse rate such that they do not enter my consciousness as a fire alarm - I wish they would emulate a loud metal gong at 2c/s instead of a video game, a cellphone ringtone or a cheap battery-powered alarm clock), I always treat it as if it were real in order to train myself to an instinctive reaction to lock my stuff, grab my things and leave. Which will probably get me killed when the real alarm goes off and it turns out to be a workplace shooter at the front door pulling the alarm (we have signs on the doors prohibiting law-abiding people from carrying weapons so only criminals are encouraged to bring in weapons to work).
The problem as I see it is that the tendencies toward under-reaction and over-reaction are largely kept in balance in nature, whereas in government they are not. In nature, while over-reaction might keep you safe from predators, it frequently brings with it an elevated risk if not of actual starvation, then likely of malnourishment, leading to a partial or complete loss of ability to compete for mating rights and hence to a corrective effect on the gene pool.
Contrast that with government. In many instances, the costs of over-reaction do not feed back to a Chicken Little. Firstly the (increasing!) 'us and them' attitude of many in government leads to a dismissal/underestimate of the costs: it's not them waiting in line for hours to miss their flight, nor losing income from the tourists that switch to a more agreeable destination etc. Second, the feedback mechanism--voters get pissed off enough and in large enough numbers that they vote for/demand political change--is so slow that Mr/Ms Little is quite likely to have moved on to a different--and likely more senior/influential--job anyway before the feedback arrives. Thirdly, the odd focus that many recruiters have on the size of the budget someone managed gives direct incentive toward scaremongering over objective risk analysis. Fourthly in nature in a 'catastrophic under-reaction', it's likely to be all over quickly whereas in government, a lifetime of public vilification/humiliation likely awaits. Fifthly in government, the Hawk himself has a vested interest in over-reaction: who cares how much the peasants were inconvenienced/endangered as long as the royal/presidential/senatorial personage is unharmed?
They've had several dozen false fire alarms at my mom's retirement community building. The residents pretty much ignore them now and many have taped over the sirens to disable them.
I'd say the problem with those hotel alarms is that what the smoke detector detects is a situation that needs attention by someone (and that is almost certainly harmless if properly cared for), while the alarm suggests a situation that needs immediate escape of everyone.
I assume that it's indeed the lone smoke detector that rings the alarm for the whole building.
It would perhaps be beneficial to have a universal escalation procedure, where an automated alarm alerts people (e.g., so they can choose to get dressed and be attentive to any signs of an immediate danger) and gives the fearful a chance of early flight, with a second level reserved for emergencies that have been confirmed by a human being.
I tend to ignore unspecific alarms at hotels as well, but I have to admit that I've never considered them as an opportunity for picking up women. Does it work ? :-)
The problem with the Zoo analogy as compared to the realworld is of course that the birds are not free so flight is not possible, the predators know where to target the caged birds so it is not a safe heaven and the owner of the cage has conflicting interest so he is not friendly.
Meaning instead of accepting getting slowly boiled the birds should flee at any oppourtunity .. but it is too late as they were likely born to captivity.
Remember, Bruce, when the fire alarm goes off, you only get to be wrong once for evacuation to be the right choice.
Both you and the birds are getting pretty-much-free food, so it seems to me the rational response is to take as little risk as possible--energy usage shouldn't be a factor (and so you were being less rational than the birds, particularly if you then went to the gym the next day!).
I also wonder if you're underestimating the birds in another sense. Isn't it possible that they're judging the birdfeeder as somewhat "too good to be true?" Do I give them too much credit to imagine that they might treat free food with some suspicion?
In regard to fire trucks attending a hotel. I believe that in most places it is the policy to send the fire brigade whenever someone trips a hotel fire alarm. The potential result of a real hotel fire not getting immediate attention from the fire brigade is too awful - and besides the firemen are bored most of the time anyway.
Also just running out of the building is not the correct thing to do. If there is smoke outside your door then you are better off keeping the door closed and sealed with wet towels while waiting for the fire brigade. Apparently some people who leave their rooms die of asphyxiation in the hallway while their luggage survives...
"Apparently some people who leave their rooms die of asphyxiation in the hallway while their luggage survives..."
Well, yes. Luggage is much harder to asphyxiate. :-p
scote: Fair point. But obviously if the hotel was damaged by flame it would affect the luggage.
If you have a window that can be opened (or smashed if necessary) and there is no flame in your room or immediately nearby then you can last for a long time.
If you are below the 10th floor then a city fire brigade should be able to get you with a ladder.
You are doing one better than I. I did not even wake up but slept right through the alarm.
We have an Eastern Rozella that visits our balcony. He gets preferential treatment because he is constantly chased away by the more aggressive Lorikeets. Being shy and over alert because of predators can have (and oviously does) its advantage on the evolutionary scale of things.
On the topic of primordial hard-wiring, I once had an extreme visceral reaction to a hotel fire alarm.
Awakened from a deep sleep by the fire alarm directly above my bed, I screamed something gutteral, stood on the bed, and proceeded to hit the alarm with my fists for 30 seconds or so before coming to full consciousness.
I then dressed and walked downstairs and outside like a civilized humanoid.
I noticed the same with hedgehogs - once they hit the asphalt, they go straight across, and literally run for their life, even if there are no cars. Then, when reaching the other side of the road, they lay on the grass and take quite a while to recuperate.
Any predictable behavior might create a risk.
Ex.: sound piercing alarm - wait for people to assemble somewhere/get out of the room - get them there.
One way to get security is by dispersion, multiplying unpredictable possibilities.
In a way, the same is true for attackers.
The more creative they are, the less will predictable countermeasures be a defense.
So it's a question of creating a bottleneck to get at somebody, and a question of outsmarting and dispersion to overcome a bottleneck.
"Ex.: sound piercing alarm - wait for people to assemble somewhere/get out of the room - get them there."
There have been a number of terrorist bombings using this technique. You place a big bomb at a point of "apparent safety" and you put one or more small bombs in a shoping center or night club etc. You phone a warning through shortly before the first bomb goes off likewise for any other small bombs. The effect is to use the small bombs like sheap dogs to herd people to your chosen point for your big bomb...
"One way to get security is by dispersion, multiplying unpredictable possibilities."
This is basic security advice almost the first thing you get told is "take a different route to work each day". The problem is humans are lazy creatures of habit by and large and quickly drop back into predictable behaviour.
However there is a flip side to this when defending a location or strong point it is best to have one where your oponent is given little or no choice about how they attack, thus alowing the deployment of resources where they are most likley to be of use.
But the best security is never to be where you can be attacked by anything other than chance encounter.
Sometimes being "light weight and agile" is the best for both offence and defense. It is one of the principles of certain types of warfare and over several thousand years we have not found an effective means of defending against it.
Which is why snipers are so effective and hated, the actual physical damage they do is minimal, however the fear they induce is almost debilitating, which is why senior field commanders will use totaly disproportionate force to remove them.
"dispersion" is also the reason why "security by obscurity" DOES work. In fact, every password is nothing else than security by obscurity.
The problem with security by obscurity arises precisely when the obscure is uncovered (by knowing the source code of the obscuring mechanism, or by discovering the password).
So the generalized rejection of "security by obscurity" is just a confusion of words.
"Over there in the back yard, near the cat feeder."
"Don't you mean bird feeder."
"You don't know my cats."
The utility difference between 30 minutes lying in bed listening to the fire alarm,* and 30 minutes standing around outside seems trivial compared to the chance of either losing the rest of one's life, or even spending a few weeks in the burn ICU. The quote from a friend who was in one that sticks with me is: "There's a REASON that they have bars on the windows in here."
*'cause you're not sleeping through it.
Re: Fire alarms
In the complex I work at the fire alarm signal is followed by a verbal announcement of the location and severity of the fire. The procedures for responding depend on the severity and how close you are.
Re: Staying in the Hotel
Bruce, I suspect that you stay in many more hotels than I do. I've never had a fire alarm during a hotel stay. I suspect that your assessment of the risks of a Hotel fire are biased through familiarity. My grandfather worked on the railroad and wouldn't wear a seatbelt because the only accidents he saw that people lived through were the ones in which people were not using them.
@David, RE: small birds harrassing predator birds.
This isn't that unusual and is pretty easily explained. At the moment when a bird spots a potential predator then the immediate response is to fly. Once the predator is spotted, however, the predator loses its advantage. The predator is not faster or more agile than the smaller birds, it is more stealthy. So if it is worth the effort (either because of a food supply or protecting the young) the small birds can drive off the predator either by harrassment or by raising a ruckus and making sure that it isn't going to get an easy meal.
A bird must avoid predators 100% of the time to live long enough to reproduce. A predator can have a 1% success rate, or even lower, and still thrive.
Fires: Consider how bad a fire has to get before an alarm goes off. It could be somebody smoking, a short circuit, or just a false alarm. Or it could be really bad.
Early on, it's easy to walk out. After flashover, or when the air becomes poisoned... Well, how long can you hold your breath?
When you're on the first or second floor, you can smash out a window and walk away. On the 21st floor -- did you bring wings?
(It's not usually the heat that kills you. It's the toxic fumes, especially carbon monoxide, and the lack of oxygen that get you.)
The Pauline Kael theory of security -- all the birds Bruce knows voted for McGovern, so he must have won, right? How much time have you spent watching birds at feeders? More than a few seconds, total, in your life? And you reach a conclusion about the efficacy of millenia of response? I think the birds are smarter about their own security than you are. If they're wrong once, they're dead. That sort of ups the ante for a correct response, and reduces the attraction of taking a shot on staying put.
It is dangerous to claim that evolution has created a particular trait for a particular purpose. Just because a trait has a security function does not mean the trait is "for" security.
This is a major difference between our approach to security and nature's approach to security. We humans treat security as a specialty. Nature integrates security as a continuous part of the whole.
Two general comments.
The observations about evolution, energy cost and so are very interesting, particularly as I have recently been making observation of the various defensive behaviours of birds around my new, semi-rural neighbourhood.
One thing that evolutionary biologists have recently come to appreciate more is that evolutionary adaptation can be extremely fast. Adaptations that require development of completely new structures may well take tens of thousands or even millions of years, but adaptations that modulate the performance of existing structures can take place in a few tens of generations (which for many species, means a decade or so.) This is particularly apparent with behaviour modification in intelligent species; rural crows don't take very long at all to adapt to urbanisation.
On a completely different note, I have recently been studying the problems with human estimation of risk, from the perspective of project management. In this field, poor risk estimation annually causes loss and wastage on the order of billions, so much research has been done -- and the result of much of this research is that Bruce is wrong.
True, we greatly overestimate the risk of rare events in comparison to common ones. But in absolute terms, the problem is not that we overestimate the risk of rare events; we actually get those about right. The problem is that we are real gamblers with common disasters. (Interestingly, depressives do much better than psychological "well" people; the estimates by depressives are not pessimistic as one might think, they are accurate ... )
The reason is not really known, but in evolutionary terms it makes sense so long as the population is continuously expanding; in that case, excessive gambling with common hazards is a winning strategy for the *species* (rather than the individual gambler) because of the Martingale effect. But as soon as the population stabilises, it becomes a losing strategy.
A such, we now either need to find new worlds to conquer, or become a species of old maids.
(Roger who, incidentally, will not be responding to follow-ups this weekend because he is about to head off and do a half-marathon in a remote area in the mountains in blazing heat ... )
I've seen a hawk grab a bird off our feeder - pretty cool unless you are the prey ;)
i have seen a hawk pop a finch off of our bird feeder a few years back. plus i've seen a hawk pop a pigeon from underneath an overpass. small birds are nervous for a reason. to a fault?
Our neighbors have a bird feeder and hawks nest in the area. The other day I watched a hawk go after the feeder. Its tactics were interesting--it perched on a chimney above the feeder, in plain sight but without moving much, for a good 30 minutes. After a while, life at the feeder returned pretty much to normal, give or take a few extra-cautious birds that winged over, spotted the hawk, did a nifty mid-air 180 and went elsewhere to dine. Then the hawk dove on one of the birds (on the ground--I'm guessing a perched bird would be a tougher target). It exploited the birds' habituation to its presence. Rather like humans and certain risks we face, such as the death toll from automobiles, I suppose.
There are many stop signs, traffic lights where I'm sure I have 99+% probability of no accident or ticket if I don't stop or if I run the red light. However, 99% with about 300 trials a year would mean about 3 unpleasant situations per year. Even 99.9% is not high enough for me to take the risk, so I don't. Maybe those sparrows are smarter than you think.
I visited my local aquarium a few days ago. They have sharks and several species of fish in the same pool - all of them are well fed.
Some of the small fish have taken to ALWAYS swimming near the sharks. They have learned that a well fed shark is not going to hurt them. But the Snapper in the tank that would eat the small fish have not learned this. One of the small fish has the habit of swimming directly in front of a shark, but the shark still doesn't eat it. Note that these fish were all bred elsewhere as the water filtration system in the tank prevents fish from breeding in there - the fish did not evolve to learn that the sharks are safe.
This shows that some fish can learn different responses to dangerous situations quickly while others (such as Snapper) can't.
As an aside the sharks do eat whiting when they are well fed so there is only 1/20 whiting remaining in the tank. There is a possibility that the sharks might have more desire to eat Snapper than the small fish in question, but this is doubtful. Given the confined space the sharks would surely have caught a few Snapper if they cared to do so.
Schneier.com is a personal website. Opinions expressed are not necessarily those of Co3 Systems, Inc.