If I could potentially live for a thousand years, would I see my thousandth birthday?
July 18, 2009 2:04 PM Subscribe
I think building immortal people is just on the edge of scientific possibility. I can imagine it happening in the future. However, it's impossible to build people who are impervious to accidents (for that you'd need magic). So assuming a world pretty similar to ours, except that doctors can cure all diseases, how long would the average person live? In other words, as I walk down the street every day, what are the chances of a safe dropping on my head?
To make this simpler, let's not worry about socio-political-environmental effects of longevity. Somehow we've solved what would, in reality, be a massive over-population problem.
Also assume we're still living in a world with standard, cars, airplanes, buildings, etc.
And, because I imagine one's chance of being in a fatal accident differs depending on where one lives (e.g. country or city), let's focus on people living in a large, urban environment, some place like NYC or Chicago.
In my counterfactual, doctors can cure cancer, AIDS, heart disease, etc. However, they can't bring people back from the dead. So if you get shot, stop breathing and aren't revived quickly via CPR or something, you're a goner; If you've head gets chopped off, you're doomed; If your heart stops beating and it isn't started again very quickly, you're dead meat; if you bleed out and don't get a transfusion, you're toast.
I'm thinking that we (in the real world) have enough statistics about fatal car accidents, murders, etc. to know what my odds are of living through a day in NYC. So assuming I have longevity treatments, what are my odds of living for 100 years? 200 years? 500 years? 5000 years?
In my fantasy world, would it be common for people to live for 800 years or would that be relatively rare?
Let's assume that most people don't take extreme measures to protect their bodies. People, as a rule, don't hole up in padded cells. They go about their business -- riding trains, eating in restaurants, crossing the street -- the way most of us do now. How long, on average, will most people live?
To make this simpler, let's not worry about socio-political-environmental effects of longevity. Somehow we've solved what would, in reality, be a massive over-population problem.
Also assume we're still living in a world with standard, cars, airplanes, buildings, etc.
And, because I imagine one's chance of being in a fatal accident differs depending on where one lives (e.g. country or city), let's focus on people living in a large, urban environment, some place like NYC or Chicago.
In my counterfactual, doctors can cure cancer, AIDS, heart disease, etc. However, they can't bring people back from the dead. So if you get shot, stop breathing and aren't revived quickly via CPR or something, you're a goner; If you've head gets chopped off, you're doomed; If your heart stops beating and it isn't started again very quickly, you're dead meat; if you bleed out and don't get a transfusion, you're toast.
I'm thinking that we (in the real world) have enough statistics about fatal car accidents, murders, etc. to know what my odds are of living through a day in NYC. So assuming I have longevity treatments, what are my odds of living for 100 years? 200 years? 500 years? 5000 years?
In my fantasy world, would it be common for people to live for 800 years or would that be relatively rare?
Let's assume that most people don't take extreme measures to protect their bodies. People, as a rule, don't hole up in padded cells. They go about their business -- riding trains, eating in restaurants, crossing the street -- the way most of us do now. How long, on average, will most people live?
This post was deleted for the following reason: "But I am assuming that. It's my fantasy and I can play it however I like it." = hypothetical chatfilter. sorry. -- jessamyn
I've heard estimates of the average longevity of a clinical immortal to be from 600 to 10,000 years.
posted by bunnytricks at 2:18 PM on July 18, 2009
posted by bunnytricks at 2:18 PM on July 18, 2009
Best answer: From what I can make of this, the three leading causes of injury death (motor vehicles, firearms, and poisoning) have rates of 14.7, 10.3, and 10.0 per 100,000 people (respectively). That means (maybe?) that together, they account for deaths of 35 people per 100,000 every year.
That's the same as 0.035% chance of dying of one of those injuries each year. Your odds of still being alive after x years are 1 - 0.00035x, which gets to be 1/2 at about 1429.
So, assuming my math is right -- which is a HUGE assumption -- it would be very common for people to be alive at 800 (and 1000, 1200, and 1400) years old.
I look forward to seeing what someone more stats-minded will come up with.
posted by cider at 2:23 PM on July 18, 2009
That's the same as 0.035% chance of dying of one of those injuries each year. Your odds of still being alive after x years are 1 - 0.00035x, which gets to be 1/2 at about 1429.
So, assuming my math is right -- which is a HUGE assumption -- it would be very common for people to be alive at 800 (and 1000, 1200, and 1400) years old.
I look forward to seeing what someone more stats-minded will come up with.
posted by cider at 2:23 PM on July 18, 2009
Response by poster: For one thing, you seem to be assuming that in such a world, people's behavior wouldn't change compared to ours, and that's almost certainly not valid.
But I am assuming that. It's my fantasy and I can play it however I like it.
Here's a less fantastic way of asking the question: assume the real world, New York City. I get up every week day, take the subway to work, walk a few blocks in the busy streets, work in a office, walk the streets again at lunch time, and then, after work, take the subway home. What are my chances, each day, of dying in a horrible accident? I assume the changes are pretty small.
Okay given this routine, what are my chances of dying in a horrible accident over the course of one year?
Over the course of 50 years?
Over the course of 200 years....?
If you and I BOTH had the capacity to live for 1000 years, would you bet all your money, double or nothing, that I would shake hands with you 800 years from now? Or do you think it's likely that over that time randomness would pelt me with a fatal accident?
posted by grumblebee at 2:25 PM on July 18, 2009
But I am assuming that. It's my fantasy and I can play it however I like it.
Here's a less fantastic way of asking the question: assume the real world, New York City. I get up every week day, take the subway to work, walk a few blocks in the busy streets, work in a office, walk the streets again at lunch time, and then, after work, take the subway home. What are my chances, each day, of dying in a horrible accident? I assume the changes are pretty small.
Okay given this routine, what are my chances of dying in a horrible accident over the course of one year?
Over the course of 50 years?
Over the course of 200 years....?
If you and I BOTH had the capacity to live for 1000 years, would you bet all your money, double or nothing, that I would shake hands with you 800 years from now? Or do you think it's likely that over that time randomness would pelt me with a fatal accident?
posted by grumblebee at 2:25 PM on July 18, 2009
Response by poster: cider's answer is the kind of thing I'm looking for.
posted by grumblebee at 2:27 PM on July 18, 2009
posted by grumblebee at 2:27 PM on July 18, 2009
This seems to say that the odds of dying of any injury in a given year is 1 in 1,681, which is 0.00059 -- not too far off from what I had above. 1 - 0.00059x gets to be 1/2 when x = 847. (Am I doing this right? The more I look at this, the more I'm not sure.)
In any case, this would indicate, again, that it would be very common for someone to live to 8000 -- but he'd only have an 18% chance of living to 1400.
posted by cider at 2:28 PM on July 18, 2009
In any case, this would indicate, again, that it would be very common for someone to live to 8000 -- but he'd only have an 18% chance of living to 1400.
posted by cider at 2:28 PM on July 18, 2009
Response by poster: Just to try to cover all my bases, let's assume the world stays pretty much in stasis. No death by global warming, etc.
If you're STILL troubled by my counterfactual not taking changes to culture, etc. into account, imagine it's a simulation. We simulate an UNCHANGING early-21st-century NYC, run the sim for 4000 sim years, give the inhabitants cures for all diseases, and see how long the average sim-citizen lives.
posted by grumblebee at 2:30 PM on July 18, 2009
If you're STILL troubled by my counterfactual not taking changes to culture, etc. into account, imagine it's a simulation. We simulate an UNCHANGING early-21st-century NYC, run the sim for 4000 sim years, give the inhabitants cures for all diseases, and see how long the average sim-citizen lives.
posted by grumblebee at 2:30 PM on July 18, 2009
I would disagree with one of your premises, that it's impossible to build people that are impervious to accidents. Presumably at some point in the future we will be able to make a "backup copy" of ourselves. If a piano drops out of the sky and crushes me, I can just be uploaded to a new body. Assuming the backup copy is reliably stored, I would be impervious to accidents.
posted by whiskeyspider at 2:31 PM on July 18, 2009
posted by whiskeyspider at 2:31 PM on July 18, 2009
Here are numbers for the US from 2005
Your one years odds are: 1,681:1
Though that includes the odds of self harm and legal execution, among other things you might not want considered.
posted by pseudonick at 2:31 PM on July 18, 2009
Your one years odds are: 1,681:1
Though that includes the odds of self harm and legal execution, among other things you might not want considered.
posted by pseudonick at 2:31 PM on July 18, 2009
What chocolate pickle said. If immortally was possible in the manner you describe, you can bet that a cult would evolved that would consist of its members locking themselves indoors in an attempt to live forever.
In short, the framing of whatever question you're trying to ask will probably get in the way, as you're trying to control variables that can't really be controlled.
And is this chatfilter?
posted by Brandon Blatcher at 2:32 PM on July 18, 2009
In short, the framing of whatever question you're trying to ask will probably get in the way, as you're trying to control variables that can't really be controlled.
And is this chatfilter?
posted by Brandon Blatcher at 2:32 PM on July 18, 2009
You could use the tables of current statistical causes of death [pdf, see Table D] and remove all those which involve disease or organ failure to come up with a number. (According to that table, accidents and suicide together account for about 8 percent of all deaths in the US.) Or there's this, which further subdivides the causes of accidental death in case you wanted to rule any of them out too (would death from surgical complications still happen in this world?)
It wouldn't be a meaningful number, of course, because there's no way human behavior wouldn't change dramatically given the proposed situation. Even in a simulation.
posted by ook at 2:33 PM on July 18, 2009
It wouldn't be a meaningful number, of course, because there's no way human behavior wouldn't change dramatically given the proposed situation. Even in a simulation.
posted by ook at 2:33 PM on July 18, 2009
List of causes of death by rate, from Wikipedia. This gives a classification of different kinds of deaths; in 2002 "unintentional injuries" make up 6.23% of them and "intentional injuries" make up 2.84%. There were 57 million deaths that year, against a world population of 6.2 billion; thus about 5 million of them died of such deaths. So in a typical year, about one in 1,200 people die of injury; thus life expectancy, if injury-related deaths stayed at the same rate and disease-related deaths were entirely eliminated, would be about 1,200 years. (But I'm starting with different assumptions than you -- basically letting the data I could find quickly dictate my definitions for me -- so this is VERY rough.)
posted by madcaptenor at 2:36 PM on July 18, 2009
posted by madcaptenor at 2:36 PM on July 18, 2009
Cider, we'd also have to account for things like homicide, war, sanctioned killings (eg, police, self-defense, etc.), and suicide.
posted by Conrad Cornelius o'Donald o'Dell at 2:36 PM on July 18, 2009
posted by Conrad Cornelius o'Donald o'Dell at 2:36 PM on July 18, 2009
I think that that last link that I posted included those, Conrad -- if you scroll down it talks about "Assault", etc, and I think it's all folded into that 1,681 number. But I haven't looked at it super-closely, so I might be wrong.
posted by cider at 2:40 PM on July 18, 2009
posted by cider at 2:40 PM on July 18, 2009
cider, your math is wrong. Your odds of still being alive after x years is (1-.00035)^x, which pretty closely approximates your result for small x but diverges wildly for large x. (And 1429 is large.) Basically, your math would be right except that once you die, you can't die again.
posted by madcaptenor at 2:40 PM on July 18, 2009
posted by madcaptenor at 2:40 PM on July 18, 2009
I think Aubrey de Grey has probably crunched the numbers somewhere. I'd say you'd get your LD50 of accidents around a thousand years, give or take a couple of centuries.
This has been tackled in some sci-fi, which describes societies and individuals becoming progressively more risk averse. At that point, deaths are not merely accidental, but freak accidents.
posted by adipocere at 2:43 PM on July 18, 2009
This has been tackled in some sci-fi, which describes societies and individuals becoming progressively more risk averse. At that point, deaths are not merely accidental, but freak accidents.
posted by adipocere at 2:43 PM on July 18, 2009
Also, are you Ray Kurzweil?
If you aren't have you been totally influenced by him?
If not...I think you should meet up with your soulmate and help him on his quest.
Kurzweil is more on the side of A.I., though he does invest himself heavily in the idea that immortality will be achieved (hence the [literally] hundreds of supplements he takes). Aubrey de Grey, co-founder of the Methuselah Foundation, is a better soulmate for this way of thinking.
posted by tybeet at 2:46 PM on July 18, 2009
If you aren't have you been totally influenced by him?
If not...I think you should meet up with your soulmate and help him on his quest.
Kurzweil is more on the side of A.I., though he does invest himself heavily in the idea that immortality will be achieved (hence the [literally] hundreds of supplements he takes). Aubrey de Grey, co-founder of the Methuselah Foundation, is a better soulmate for this way of thinking.
posted by tybeet at 2:46 PM on July 18, 2009
But I am assuming that. It's my fantasy and I can play it however I like it.
In that case this is chatfilter.
posted by Chocolate Pickle at 2:46 PM on July 18, 2009 [1 favorite]
In that case this is chatfilter.
posted by Chocolate Pickle at 2:46 PM on July 18, 2009 [1 favorite]
Thank you, madcaptenor -- I had a feeling I was doing something wrong.
So let's see -- assuming I'm not making some other error, the probability of being alive after 800 years (using (1-0.00059)^x) would be 0.62. After 1400 years it would be 0.44.
(Also, in my comment above, I meant 800, not 8000. Oops. I'm too excited about the meetup to think straight!)
posted by cider at 2:48 PM on July 18, 2009
So let's see -- assuming I'm not making some other error, the probability of being alive after 800 years (using (1-0.00059)^x) would be 0.62. After 1400 years it would be 0.44.
(Also, in my comment above, I meant 800, not 8000. Oops. I'm too excited about the meetup to think straight!)
posted by cider at 2:48 PM on July 18, 2009
Response by poster: I don't think this is chatfilter, but I'm not the best person to make that call. I think there are right and wrong answers to my question. In my view, many of you are over-complicating it. That's my fault for phrasing my question as a sci-fi story.
What if my question was this: if I put $100 in a bank account that paid 5% interest and never touched the money, how much interest would I have accrued in a million years?
Would people be saying, "Umm... It's really unlikely we'll have the same banking and monetary system in a million years?"
We probably won't, but that's not the point of the question.
Here's one more try: I am a sim. You drop me in a simulated 2009 NYC and let me live my life for 1 year. During that you, you don't control the sim. It runs by itself, accidents happen, etc. After a year, you stop the sim. What are my odds of having survived the year.
Okay, now you rerun the sim. It's the same sim, but randomness can cause different events to happen. The empy street I might have crossed the during the first run might be full of cars during this run.
You rerun the sim 500 times. What are the odds that the sim-me will never had died in an accident during any of those runs?
What about after 5000 runs? 10,000 runs?
posted by grumblebee at 2:53 PM on July 18, 2009
What if my question was this: if I put $100 in a bank account that paid 5% interest and never touched the money, how much interest would I have accrued in a million years?
Would people be saying, "Umm... It's really unlikely we'll have the same banking and monetary system in a million years?"
We probably won't, but that's not the point of the question.
Here's one more try: I am a sim. You drop me in a simulated 2009 NYC and let me live my life for 1 year. During that you, you don't control the sim. It runs by itself, accidents happen, etc. After a year, you stop the sim. What are my odds of having survived the year.
Okay, now you rerun the sim. It's the same sim, but randomness can cause different events to happen. The empy street I might have crossed the during the first run might be full of cars during this run.
You rerun the sim 500 times. What are the odds that the sim-me will never had died in an accident during any of those runs?
What about after 5000 runs? 10,000 runs?
posted by grumblebee at 2:53 PM on July 18, 2009
Best answer: I concur with madcaptenor. If your probability of dying within the span of a year is x, then the probability that you'll be alive at the end of the year is 1-x, and the probability that you'll be alive after n years is (1-x)^n (assuming that the odds of you dying are constant and are independent from year to year). So if you want to know when your chances of still being alive dip below, say 50%, you can use n = log(.5)/log(1-x). If x is 1/1681, as mentioned above, then n is about 1165.
posted by epimorph at 3:00 PM on July 18, 2009
posted by epimorph at 3:00 PM on July 18, 2009
Sorry, cider - this thread moved way faster than I anticipated, and preview failed me (or perhaps, I failed preview).
posted by Conrad Cornelius o'Donald o'Dell at 3:03 PM on July 18, 2009
posted by Conrad Cornelius o'Donald o'Dell at 3:03 PM on July 18, 2009
Response by poster: In that case this is chatfilter.
It depends on your definition of chatfilter.
In my mind, chatfilter is a question in which everyone's answer is equally valid because answers are totally subjective. "What's the best flavor of ice cream?" is chatfilter, because if I say "vanilla" is neither right nor wrong. The concept of "best answer" is meaningless.
My question definitely has wrong answers and I suspect it has right ones. "You'd have a 80% chance of dying in the 1st year" is wrong.
If, on the other hand, you think questions about counterfactuals (that could never happen) are chatfilter, then my question is definitely chatfilter. I will leave that distinction up to the mods and not complain about their decision. Feel free to flag.
One thing I'd point out is that this is a counterfactual:
Let p = (a and b) or (a and c)
if a and c are true and b is false, is p true or false?
Would that question be chatfilter?
posted by grumblebee at 3:04 PM on July 18, 2009
It depends on your definition of chatfilter.
In my mind, chatfilter is a question in which everyone's answer is equally valid because answers are totally subjective. "What's the best flavor of ice cream?" is chatfilter, because if I say "vanilla" is neither right nor wrong. The concept of "best answer" is meaningless.
My question definitely has wrong answers and I suspect it has right ones. "You'd have a 80% chance of dying in the 1st year" is wrong.
If, on the other hand, you think questions about counterfactuals (that could never happen) are chatfilter, then my question is definitely chatfilter. I will leave that distinction up to the mods and not complain about their decision. Feel free to flag.
One thing I'd point out is that this is a counterfactual:
Let p = (a and b) or (a and c)
if a and c are true and b is false, is p true or false?
Would that question be chatfilter?
posted by grumblebee at 3:04 PM on July 18, 2009
epimorph's math points to something interesting: with the data that's been thrown around, the mean lifespan is 1681, but the median lifespan is significantly lower at 1165. In general the median lifespan is (ln 2) times the mean lifespan. (ln 2 = .693..., for those of you who aren't blessed/cursed with the ability to remember mathematical constants.) Why is there this difference? Because there's a good number of people who live much longer than the median or the mean. One in a hundred people would live to (ln 100)*1681 = 7741 years, which is 4.6 times the average lifespan; nobody lives to 4.6 times the average lifespan in the real world.
Less quantitatively, there's not some characteristic time at which people tend to die. In general a lot of planning in the real world implicitly assumes that people live for, say, "three score years and ten"; there's no analogous number in this disease-free world.
posted by madcaptenor at 3:25 PM on July 18, 2009 [1 favorite]
Less quantitatively, there's not some characteristic time at which people tend to die. In general a lot of planning in the real world implicitly assumes that people live for, say, "three score years and ten"; there's no analogous number in this disease-free world.
posted by madcaptenor at 3:25 PM on July 18, 2009 [1 favorite]
We could change the question to "what would life expectancy be if tomorrow a universal cure was found for all disease". That would start us at a fixed point and eliminate any intervening social, political and cultural changes.
However, the premise makes the current statistics invalid in the future. If death rates are cut in half, does starvation become a more frequent cause of death? More cars on the road (with more elderly drivers), do cars create more deaths?
You could ask what is the current rate of deaths due to causes other than disease. That has a correct answer.
The premise makes this a very interesting, yet unanswerable question.
posted by 26.2 at 3:40 PM on July 18, 2009
However, the premise makes the current statistics invalid in the future. If death rates are cut in half, does starvation become a more frequent cause of death? More cars on the road (with more elderly drivers), do cars create more deaths?
You could ask what is the current rate of deaths due to causes other than disease. That has a correct answer.
The premise makes this a very interesting, yet unanswerable question.
posted by 26.2 at 3:40 PM on July 18, 2009
This thread is closed to new comments.
For one thing, you seem to be assuming that in such a world, people's behavior wouldn't change compared to ours, and that's almost certainly not valid.
posted by Chocolate Pickle at 2:17 PM on July 18, 2009 [1 favorite]