As someone who is studying actuarial sciences ... if you use some real data against the Gompertz you see that it doesn't really fit that well for very small and very large ages. Which is why this guy Makeham did this
As a student with a mathematics background (and boy do we love 'simplifications'). I can never stop theorizing and thinking about what would happen if we would just insist on a maximum age of lets say 80. It will probably never happen, but just think about it. It's both morbid and fascinating in it's own sense.
I remember reading an IEEE Spectrum article about death rates over 100, and while obviously very high, apparently they didn't increase much. I can't seem to find it unfortunately as it was a good article.
That's an unwarranted claim. Various quantities of interest of complex systems more often than not follow certain statistical distributions. This is especially true about natural systems.
Flip 100 coins, call it one trial. Perform such trials a large number of times, a few hundred let's say. For each trial count the number of heads. In the end create a histogram (frequency of N-head trials, N ranging from 0 to 100). It'll look like a gaussian distribution! The emergence of such distribution feels strange, even though we pretty much know everything about the system. (Incidentally this is related to the Central Limit Theorem). Similar idea applies to human mortality distributions.
The best you could say is that these are very counter-intuitive observations/facts.
In that case we understand what causes the Gaussian to emerge - lots of individual coin flips. However, in human mortality, the underlying "coin flip" - i.e. the death-causing event - isn't so easily found. Particularly when you consider that apparently the underlying cause seems to happen more frequently as you age, which is kind of interesting when you consider that the physical laws haven't changed - there's obviously some kind of emergent process going on - but what and why? That's the question.
It wouldn't suprise me to learn that it's basically planned obsolescence - i.e. that there's an evolutionary advantage to a species that cycles through individuals as rapidly as possibly (while leaving enough time to produce a lot of progeny).
It's less about planned obsolescence that's built into multi celled organisms. The choice is how much energy you spend fighting it.
EX: A fruit fly needs far less cancer defenses than a squirrel. And, 2 fruit flies can become 2,000 fruit fly's a lot faster than 2 squirrels can become 2,000 squirrels. But, fruit fly's can't eat nuts so as long as trees produce nuts there is a nitch it takes something with at least the mass of a squirrels to fill.
Iron and heavy metals accumulate in tissues over time. The tissue membranes accumulate an increasing share of polyunsaturated lipids, which in turn interfere with efficient metabolism. Many ways that biological tissue wears out over time are well understood. It's not just some mystery.
That doesn't really ring true - after all, they're going to be passed to newborns too. "Gunk" collects in tissues that are never replaced, but why does that occur in the first place? Clearly, growing new tissue isn't a biologically hard problem - when there's evolutionary pressure to do so, several species evolved means to replace surprisingly large parts of their body. Some insect metamorphose; crustaceans rebuild complete skeletons - energetically, it's almost certainly hell of a lot cheaper to occasionally replace bad bits than it is to procreate and lose the previous generation. There's something else going on there.
The programmed senescence model fits with this observation quite well. It argues that aging isn't the result of a breakdown or accumulated damage, but rather a deterministic process based on our genetic programming.
Evolution programmed us to make a human, but it didn't program us to stop developing. We're like a cake left in the oven. No one told the concrete suppliers to go home, so they just keep dumping it everywhere on our completed skyscraper.
As someone with a biology background I find the conclusions a little too general. Mortality curves are not only species specific, but particular for that environment. A curve using saltwater crocodiles or lobsters would look quite different with very high mortality for the young and limited mortality for the very old.
The curve for humans can (and has) been adjusted by human action to give it a better shape - especially at extreme ages. So I think it should be emphasized that the curve is not some mathematical unalterable fact of life.
"their expected remaining lifetime does not decrease with age." (emph REMAINING) implies that two lobsters, at 1 and 10 years, each have the same probability to live for some number X years in the future for some fixed X. This seems to me a certainly false statement, where you claim otherwise. Can you elaborate?
Theoretical work by Fisher, Medawar, and Hamilton during the last century pretty much explained why we see super-exponential decay in mortality distributions. Their theories center around using the "Euler-Lotka Equation", which estimates the number of descendants from an individual considering that each member ages. You can then show that the effect of aging on fitness is marginal, and hence not selected against.
I think the real issue is that when it comes to complex animals, it's much easier to build a brand new one than to repair existing ones. Our genes encode how to grow a human from scratch, but now how to deal with the myriad of problems that can occur as a result of normal wear and tear. Repairing an ageing body is a completely different problem from growing one.
We do have some DNA that deals with maintenance processes, but I think the issue is that these processes inevitably can't cope. It's like trying to patch up an old house. You can keep doing little fixes here and there, but at some point, serious renovations are needed, and we're just not genetically equipped to do that.
You have to wonder what it is nature could do anyway, if humans living to be extremely old had been selected for. Old human bodies are so broken, we'd need to be able to grow entirely new ones and shed body parts, even regrow brain tissue. We'd need a metamorphosis on the level of what butterflies go through.
Matt Ridley's The Red Queen also talks about the effects of parasites, which also maps in to why there are genders (more accurately sexual reproduction). The problem is the mismatch between lengths of generations. A parasite that has one generation every week gets 52 shots a year at an unchanging host. As time passes the probability of success for the parasite increases. The human with ~20 years per generation has to withstand that long before making a new one that is different and (partially) resets the clock.
Gender comes in because asexual reproduction would produce offspring virtually identical to their parent which means a successful parasite would have a large number of hosts to exploit. By combining DNA from two different individuals (note how much we avoid individuals from the same family with substantially similar DNA) things are mixed up enough to give a head start against the parasites.
> Natural selection seems to have little regard for longevity.
I'd argue the opposite, that natural selection depends entirely on death, not just to weed out the less fit, but also because Without death, there would be no evolution. This is by definition, at least for an environment that can only support a limited population. Corollary: The maximum rate of evolution is proportional to the rate of generation turnover, that is to say it is inversely proportional to life span. (http://qr.ae/L5v6z)
Mother Nature did not build us to last, but quite "intentionally".
Second, there is no contradiction between the two theories. According to r/K selection, long life-span is a typical characteristic of K-selected species, while short lifespan is typical for the r-selected. The theory I espouse would simple add that the continuum between K-selected and r-selected species would tend to correspond to a continuum of evolutionary rates.
If you still disagree, please follow the link and point out flaws in the argument.
>The theory I espouse would simple add that the continuum between K-selected and r-selected species would tend to correspond to a continuum of evolutionary rates.
Looking at the last 2B years who has shown higher "global" evolutionary rate - the line of living matter leading to humans or the line leading to fish?
While of course it is obvious that the "local" evolutionary rate - rate of producing and churning through minor changes - is higher for shorter lived.
In short - giving the result the human rate of evolution
>You have fallen for the false assumption that evolution is teleological
it isn't false, it is true - the evolution leads to exponentially more complex systems. There is a very simple mathematical reason for that - evolution changes are small continuous deltas and a delta-changes to a more complex system cover more volume in the parameter space. Very rough - an organism with 2 times more types of cells, limbs, other morphological and behavioral features, etc.. will produce 2 times more of survivable variations.
>humans are the farthest along that teleological path.
mammals are farthest down the complexity path and humans just a bit further as it seems that we have a bit more complex brain.
Your Harward link is wrong on at least 2 metrics of biological systems both of which shows clear [exponential] increase as result of evolution - complexity and entropy increase integrated over given biological system's life-path in the space-time (that ability to maximize entropy beyond what can be achieved by following local gradient is the main differentiator between live and regular matter)
Natural selection doesn't work how you think it does. It's about fitness for making to the point where you have reproduced. Dying after that point doesn't come into it much, except in that surviving after is a result of other properties that are also good.
Nothing I said contradicts your description of natural selection, that "fitness for making to the point where you have reproduced." But you are missing my point that the rate of evolution produced by natural selection is dependent on the turnover of the population. The parents need to die to make room for more fit offspring once the carrying capacity of the eco-system has been reached.
If you still disagree, can you follow the link and point out flaws in the logic there?
That seems to explain why evolution hasn't done much to stop aging, but it doesn't describe the mechanisms of aging itself, which cause mortality rates to increase over time with this particular curve.
I think I can do better. Michael Rose breed fruit flies selecting for longevity. I believed his work showed that it was possible to extend lifespan simply by introducing the right selective pressure. If this doesn't demonstrate that aging is the result of Evolutionary processes, I don't know what will.
No one reasonably disputes that aging is the result of an evolutionary process. But that's a far cry from explaining exactly "why we see super-exponential decay in mortality distributions".
Man, reading this gave me such a strange sensation. I began to read it over lunch, and I was just hit with this wall of sadness. I looked around at the people at the shop I was in, thought about everyone I know, and thought, "We're all going to die."
It's weird, we all pretend like it won't happen. We do our best to live without thinking about the fact that we have a X% chance of not making it to our next birthday.
I thought about my friend from college who died of cancer a year or so ago. I thought about my parents, who are quickly approaching those scary 1-in-1XX odds of dying every year.
It's like there's this scary axe-murderer out there, who's killing people left and right, and we all just do our best to ignore him. Because there's no stopping him. He'll come for you... He'll come for all of us. The only thing you can do is try not to think about it.
I believe that the best thing to do is to think deeply about it and explore those feelings. Because it's inevitable, we need to accept and embrace it.
Picture any time before you were born. If it's not scary or sad to think about that, then thinking about a time after you die shouldn't be any different.
If you're worried about the actual moment of transition, it's probably going to be either too abrupt for you to really react to it (accident or violence), or you'll be super tired/unconscious and it'll basically be a relief, like sleep. This isn't a guarantee, but it's pretty likely.
As far as I'm concerned, as long as I can avoid a slow, painful death, it's all good. This is why torturous murders/executions and outlawing euthanasia are the most horrible things that people can inflict on each other-- you're ruining a person's only shot at a good death.
Immortality is such an awful goal. What is life if there is no beginning middle and end? I can't begin to think of the resource constraints a population would have if none of them have to die. I can only think life evolved to die from immortality and should not go in reverse.
That's your opinion. I think it's a mind bogglingly bizarre one, but sure. But what isn't an opinion is the "resource constraints" idea. That's just wrong and unrelated. There would need be no resource constraints. In fact, without a workforce that gets old and dies, certain resources would be even more plentiful. Certainly labor. Economies would have to adapt somewhat.
It is many people's (and my) opinion that death is one of the most tragic facts of life we currently face. All that knowledge and personality and life that makes up a person just disappears? What an idiotic process. If we can put an end to it we should.
> I can't begin to think of the resource constraints a population would have if none of them have to die.
population growth is caused by procreation which is just other side of the coin to mortality. If (big "if" actually) species of immortals (who thus wouldn't need any significant procreation) would have higher evolution speed than mortal&procreation species then immortals would take over. We'll see this split (and resulting competition between subspecies) in human species in the coming decades.
We might take better care of Earth, if we thought we would be around in 5000 years. Turns out that people tend to care more about themselves than theoretical great-great-grandchildren - no matter how much they protest otherwise.
So I don't think your argument from resource constraints and carrying capacity is self-evident.
Aging is a disease, not a wear-out problem. Now that most of the early-killer problems have been solved, the remaining lifespan curve makes that clear. There are specific evolved mechanisms built in to get the old people out of the way so the species can progress.
> Shown above are the results from a simulated world where “lightning bolts” of misfortune hit people on average every 16 years, and death occurs at the fifth hit.
I think that graph would look almost perfect if the decay was exponential rather than linear. A better hypothesis would be every time you are hit by a lightning bolt, you are twice as likely to be restruck every year. So if you are hit and it was a 1/40 chance, next year would be a 1/20 chance until you are hit 5 times.
Obviously it isn't really working well yet, so newer versions with better DNA are released and old ones get retired regularly. ;-) Perhaps we could have fewer such DNA "updates" if we didn't mess with our environment (= living conditions) so much that frequent adaptations are needed.
I'd really like to see a study that examines people's life expectancy when they move (very early) from a modern environment to one that resembles their parents' living conditions 10-20 years ago, the hypothesis being that their genes would have adapted to cope with those conditions and not with more recent changes, so they might live longer and healthier on average. Usually we do the opposite, many people move from rural to urban areas, so they need to adapt to more changes than those happening at an already fast pace in one area.
(apologies for half-serious unscientific techie ramblings)
You know how the amoeba reproduces, right? It splits into two. So every amoeba* alive today has been alive for hundreds of millions, perhaps billions, of years.
*And all the other organisms that "reproduce" this way.
As someone who have never had to deal with the question of mortality. this is really disturbing !
The frustrating part is if people throw enough money at the problem it can be solved ! its not like the problem of time-travel or anti-gravity. Ageing in just a computation problem that can be solved with enough data !
I wished my fellow humans spent a large part of the economy trying to solve it.
Um, Last I checked, not many human machines can continuously operate up to 100 years. Most robots and machines like cars are considered "long life" if they last 30. Seems to me the human body is built to last.
That was exactly my point. thank you. Humans last longer than machines because we self-repair and acquire our own energy. Last I checked, human-made machines do neither of these things. Not sure why I was down voted for pointing that out?
You were saying that the human body is built to last because it barely lasts longer than machines. But comparing the human body to a machine is fruitless; the ability to self repair and collect our own energy should allow us to last almost indefinitely, not a mere 80 years. 80 years is more like you would expect from a machine.
Those capabilities -- self repair and energy collection -- should allow many orders of magnitude greater longevity. The fact that they haven't indicates that those capabilities are not optimized for longevity. Ie., we were not 'built to last.'
We were optimized for species-longevity through reproduction, not individual longevity. The fact that we've existed as a species for hundreds of thousands of years is more like what you should expect from something that is built to last.
Not really, but seems to be basic statistics to me with some obvious trickery in place. My complaint is that he kinda glosses over (cut outs?) high infant mortality which would bend the curve in a not so aesthetically pleasing way. I imagine elderly mortality would do the same towards the end of the curve. People in their 80s and 90s aren't doubling mortality rates in 8 years, they're doing so practically yearly.
wikipedia: The Gompertz–Makeham law of mortality describes the age dynamics of human mortality rather accurately in the age window from about 30 to 80 years of age.
So, he cuts off the high risk childhood, teen, and 20s and everyone over 80? Yeah, that's how you get such a smooth graph. So not exactly pseudoscience, but definitely dishonest. Seems to me there are a lot of people who think nice clean graphs are 'beautiful' and try to fit in data and act pretentious about some 'mystery' link to math or the universe or somesuch. The same way every hack pop-science writer fits everything to the Fibonacci sequence. Uh, no, all this seems to say is that if youre healthy enough to hit age 30, you probably have a nice normalized mortality rate from then on to about 80.
Skipping the first 30 years and anything over 80 is the only way this "beautiful" math and "mystery" works. When we include all ages, its a pretty predictable graph:
http://en.wikipedia.org/wiki/Gompertz%E2%80%93Makeham_law_of...
As a student with a mathematics background (and boy do we love 'simplifications'). I can never stop theorizing and thinking about what would happen if we would just insist on a maximum age of lets say 80. It will probably never happen, but just think about it. It's both morbid and fascinating in it's own sense.