The Demographic Theory of Aging

Aging destroys fitness.  How could aging have evolved?  Below is my answer to this question.  This is mainstream science from peer-reviewed journals [Ref 1, Ref 2, Ref 3, but it is my science, and as Richard Feynman warned us*, I’m the last one who can be objective about the merits of this theory.

Too fit for its own good

In 1874, a swarm of Rocky Mountain Locusts descended on the American midwest. They covered the sky and shadowed the earth underneath for hundreds of miles. A single cloud was larger than the state of California. Once on the ground, they ate everything that was green, leaving behind a dust bowl. The earth was thick with egg masses, ready to renew the plague the following year.

Laura Ingalls Wilder wrote in her childhood memoir (in the third person)

Huge brown grasshoppers were hitting the ground all around her, hitting her head and her face and her arms. They came thudding down like hail. The cloud was hailing grasshoppers. The cloud was grasshoppers. Their bodies hid the sun and made darkness. Their thin, large wings gleamed and glittered. The rasping, whirring of their wings filled the whole air and they hit the ground and the house with the noise of a hailstorm. Laura tried to beat them off. Their claws clung to her skin and her dress. They looked at her with bulging eyes, turning their heads this way and that. Mary ran screaming into the house. Grasshoppers covered the ground, there was not one bare bit to step on. Laura had to step on grasshoppers and they smashed squirming and slimy under her feet.

The locusts returned in several more seasons, but the last reported sighting of a Rocky Mountain locust was in 1902. There are preserved specimens in museums and laboratories today, but no living locusts. Entomologists interested in the locust’s rise and fall travel to the glaciers of Wyoming, mining hundred-year-old ice for carcasses that they might study.

Where did they go?  The Rocky Mountain Locust drove itself to extinction by overshooting its sustainable population.

Every animal species is part of a food web, and depends on an ecosystem to survive. If the ecosystem collapses, it takes down every species and every individual with it. Because of their mobility, the locusts were able to devastate many ecosystems, denuding one landscape, then flying hundreds of miles to deposit their children in a fresh location.  Animals that can’t fly become victims of their own greed much more quickly than the locust. If the lions killed every gazelle on the Serengeti, how long would it be before the lions were gone, too?

Evolution of Individuals and Groups

How would an evolutionary biologist describe this situation? Were the locusts too fit for their own good? To capture this story, you have to distinguish between individual fitness and collective fitness. Individually, these locusts were super-competitors. Collectively, they were a circular firing squad.  The science of individual fitness and collective fitness is called Multi-level Selection Theory, and it has been spearheaded by David S Wilson of Binghamton University, based on theoretical foundations by George Price.  MLS is regarded with suspicion by most evolutionary biologists, but embraced by a minority as sound science.

Selfish organisms that consume as much of the available food species as possible may thrive for a time. They may crowd out other individuals of the same species that compete less aggressively.  But as soon as their kind grows to be the majority, they are doomed – they wipe out the food source on which their children depend.

Animals are evolved to be “prudent predators”†.  Species that have exploited their food sources too aggressively, or that have reproduced too fast have become extinct in a series of local population crashes.  These extinctions have been a potent force of natural selection, counterbalancing the better-known selective pressure toward ever faster and more prolific reproduction.

How did Evolutionary Theory go Wrong?

This is an idea that has common-sense appeal to anyone who thinks logically and practically about evolutionary science. In order not to to appreciate this idea, you need years of training in the mathematical science of evolutionary genetics. Evolutionary genetics is an axiomatic framework, built up logically from postulates that sound reasonable, but the conclusions to which they lead are deeply at odds with the biological world we see. This is the “selfish gene” theory that says all cooperation in nature is a sort of illusion, based on a gene’s tendency to encourage behaviors that promote the welfare of other copies of the same gene in closely-related individuals.

The “selfish gene” is an idea that should have been rejected long ago, as absurd on its face. Yes, there is plenty of selfishness and aggression in nature.  But nature is also rich with examples of cooperation between unrelated individuals, and even cooperation across species lines, which is called “co-evolution”.  Species become intimately adapted to depend on tiny details of the other’s shape or habits or chemistry.  Examples of this are everywhere, from the bacteria in your gut to the flowers and the honeybees.  In the same way, predators and their prey (I’m using this word to include plant as well as animal food sources) adapt to be able to co-exist for the long haul.  It is obvious to every naturalist that there is a temperance in nature’s communities, that when ecosystems are out of balance they don’t last very long.

It makes good scientific sense that extinctions from overpopulation are a powerful evolutionary force, and it is part of Darwin’s legacy as well. Natural selection is more than merely a race among individuals to reproduce the fastest. The very word “fitness” came from an ability to fit well into the life of the local community.

But beginning some forty years after Darwin’s death, mathematical thinking has led the evolutionary theorists astray. They have forgotten the first principle of science, which is that every theory must be validated by comparing predictions from the theory to the world we see around us. Predictions of the selfish gene theory work well in the genetics lab, but as a description of nature, they fail spectacularly.

Understanding Aging based on Multi-level Selection

If we are willing to look past the “selfish gene” and embrace the science of multi-level selection, we can understand aging as a tribute paid by the individual in support of the ecosystem.  If it weren’t for aging, the only way that individuals would die would be by starvation, by diseases, and by predation.  All three of these tend to be “clumpy” – that is to say that either no one is dying or everyone is dying at once. Until food species are exhausted, there is no starvation; but then there is a famine, and everyone dies at once. If a disease strikes a community in which everyone is at the peak of their immunological fitness, then either everyone can fend it off, or else everyone dies in an epidemic.  And without aging, even death by predation would be very clumpy.  Many large predators are just fast enough to catch the aging, crippled prey individuals.  If this were not so, then either all the prey would be vulnerable to predators, or none of them would be.  There could be no lasting balance between predators and prey.

Aging helps to level the death rate in good times and bad. Without aging, horde dynamics would prevail, as deaths would occur primarily in famines and epidemics. Population would swing wildly up and down. With aging comes the possibility of predictable life spans and death rates that don’t alternately soar and plummet.  Ecosystems can have some stability and some persistence.

Aging is plastic, providing further support for ecosystem stability

This would be true even if aging operated on a fixed schedule; but natural selection has created an adaptive aging clock, which further enhances the stabilizing effect. When there is a famine and many animals are dying of starvation, the death rate from old age is down, because of the Caloric Restriction effect.  In times of famine and other environmental stress, the death rate from aging actually takes a vacation, because animals become hardier and age more slowly.

When we ask “Why does an animal live longer when it is starving?” the answer is, of course, that the ability to last out a famine and re-seed the population when food once again becomes plentiful provides a great selective advantage.  This may sound like it is an adaptation for individual survival, consistent with the selfish gene.  But we might ask the same question conversely: “Why does an animal have a shorter life span when there is plenty to eat?” When we look at it this way, it is clear that tying aging to food cannot  be explained in terms of the selfish gene.  In order to be able to live longer under conditions of starvation, animals must be genetically programmed to hold some fitness in reserve when they have plenty to eat, and this offers an advantage only to the community, not to the individual.

Hormesis is an important clue concerning the evolutionary meaning of aging. This word refers to the fact that when an individual is in a challenging environment, its metabolism doesn’t just compensate to mitigate the damage, but it overcompensates. It becomes so much stronger that it lives longer with challenge than without. The best-known example is that people (and animals) live longer when they’re underfed than when they’re overfed. We also know that exercise tends to increase our life expectancy, despite the fact that exercise generates copious free radicals (ROS) that ought to be pro-aging in their effect.

Without aging, it is difficult for nature to put together a stable ecosystem. Populations are either rising exponentially or collapsing to zero. With aging, it becomes possible to balance birth and death rates, and population growth and subsequent crashes are tamed sufficiently that ecosystems may persist.  This is the evolutionary meaning of aging:  Aging is a group-selected adaptation for the purpose of damping the wild swings in death rate to which natural populations are prone.  Aging helps to make possible stable ecosystems.

___________

“ The first principle is that you must not fool yourself, and you are the easiest person to fool.” – R P Feynman (from the Galileo Symposium, 1964)

† Here “predator” can mean herbivore as well as carnivore.  This is the common usage in ecology.

14 thoughts on “The Demographic Theory of Aging

  1. Nicely put. We have to remember that almost all species are now extinct, but we are special, for each of us, none of our ancestors died before leaving offspring. None.

    And if success is measured by not being extinct, then both personal fitness and group fitness are absolutely required to accomplish this. But evolution is not consciously directed (until we started agriculture 10,000 years ago) so all we see is those accidental mutations that happened to provide advantages in fitness within the ecosystems in which they lived.

    • Thanks, Dick! I like the way you think. It makes me realize that I’ve broken a 4-billion-year perfect record by not having biological children of my own. I console myself that I haven’t contributed to the circular firing squad that is human overpopulation.

  2. John,

    I love the artwork showing the animals playing a board game. Can you please identify the source, and anything you may know about it?

    Thanks,

    Peter Kraus

  3. Hey Josh!

    You know I read all of your articles in the H+ Magazine and I usually like them but I can’t see how you can disregard the “selfish gene” theory so bluntly while making some assumptions that are not necessarily true like:

    “With aging comes the possibility of predictable life spans and death rates that don’t alternately soar and plummet. Ecosystems can have some stability and some persistence.”

    “Without aging, it is difficult for nature to put together a stable ecosystem.”

    Really? Here you totally ignore that the stability of an ecosystem has not necessarily anything to do with the minima and maxima of its variables. There are many ecosystems who thrive on short bursts of both maxima and minima say fruit trees (and their respective eaters), say yeast, say ebb and tide related organisms. If there was no aging we would most likely still live in a “stable” ecosystem, even if it would not look very stable from what we know.

    Nature has many subtle ways to regulate overpopulation in specie. Where – for example – whales lose their “navigation system” for a short time, drift away from their flock and end up in a different pasture. When their original flock dies out because of overfeeding the survival of the species isn’t in danger. This is a perfect example for both: A stable ecosystem that can handle overpopulation and the selfish gene theory.

    “Why does an animal have a shorter life span when there is plenty to eat?”

    Have you ever considered the argument from Aubrey de Grey’s book where he says – I am paraphrasing – that your mitochondria get destroyed faster the more you eat just because they have to process more and therefor the process described in his book is accelerated?

    When I first read the selfish gene I was blown away by it’s ingenuity because it’s explaination for aging is as simple as logic and I have yet to see one example that does not fit the theory.

    I am sorry Josh – I don’t know if the selfish gene theory is always true or not – but I know there are more than a couple fallacies in your article some that I haven’t even mentioned. Why should starvation and predation be “clunky” causes of death in a stable ecosystem? How can “everyone die at once” when populations are living in different locations?

    But keep up your good work with this blog :-).

    Cheers,
    Stefan

    • Hi, Stefan –
      Thanks for coming over here to my primary blog site to comment – and it’s a good reminder for me to monitor the H+ site more regularly. I don’t usually look at the comments over there.
      You raise good and interesting points. This is just the kind of dialog I am hoping to get going. My blog post last week outlined a theory without any of the background or motivation for it, and I realize that it requires a leap into new ideas for people like you who have already thought about aging in an evolutionary context.
      But this was not a “hit and run” piece. Rather, it was an invitation into the main body of what I have thought about and written about the last decade. I’ll respond briefly and refer you to that writing.

      your mitochondria get destroyed faster the more you eat just because they have to process more and therefor the process described in his book is accelerated

      This sounds reasonable on the surface, and it’s a hypothesis well worth exploring. But embedded in a larger context, this theory requires that evolution has looked for possible ways to solve the problem of free radical damage, but just cannot manage to do so in the presence of an abundance of food. Stated in this way, it’s just not plausible. The best way to see this is to remember that muscular activity – exercise – involves more mitochondrial energy generation and more free radical damage than anything else we do. And yet the body is programmed to handle this damage, and even to overcompensate so that exercises causes people (and animals) to live longer.
      This and other evidence has convinced me that the caloric restriction effect and aging in general cannot be explained on the level of physiology; that they demand an evolutionary explanation.
      Over the course of several years, I came to the Demographic Theory of Aging in two steps. First, I became convinced that aging is an adaptation, selected in a Darwinian process for its own sake. The arguments for this are laid out at a technical level in a book chapter which I’ve offered free on-line: http://tinyurl.com/7m53ogd. There is a more accessible version coming out in my book, Suicide Genes, which will be published next year. If you write to me privately, I’ll share with you a preview of the appropriate chapter.
      The second step is to note that the selection of aging as an adaptation is deeply at odds with the “selfish gene” version of evolutionary theory. What is the least radical change to evolutionary theory that would accommodate adaptive aging? The Demographic Theory is my answer to that question. References to theoretical papers and computer models are included at the beginning of my blog post.
      Thanks for taking time to respond, Stefan, and I hope that you’ll follow up and read some of these references and write again.
      – JJM

  4. Hi Josh,

    This is a nice idea which I haven’t seen before. But it’s hard to imagine that damage can be completely taken care of, were it for the right genes. Every complex system around us fails after sufficient time when the accumulated damage is too much to bear, whether it is living or non-living. Biology is not my area, so please excuse the inaccurate language, but here are some issues:

    1. Is aging the only possible solution to maintain ecological stability? And how come it is so pervasive, ranging from microorganisms to big animals? Intuitively it feels like this one solution should not fit (or be the best solution for) all creatures across all ecosystems, nor should the ecological stability problem exist for all creatures.

    2. I understand that the notion that caloric restriction (CR) extends lifespan is quite popular. But is it established conclusively? Let’s say there are 2 animals in the wild, one is small, weak, and vulnerable due to CR, while the other gets plenty to eat. I suspect that despite all the advantages in signalling, the CRed animal will on average die earlier due to weak immunity, predation, etc while also being less successful in reproduction. (then how to explain CR effect in lab? Perhaps it’s just less damage from eating less. Secondly it’s not fully established: CR didn’t increase average lifespan in wild mice according to one study (pubmed PMC2923404), and the benefit is questionable in the well known NIA rhesus monkeys study).

    As for exercise, I don’t know if it really increases lifespan or if it’s just the presently popular magical pill purported to be a solution for every problem. Maybe it’s just obesity avoidance, or maybe the people who exercise have healthy habits.

    3. According to this theory, if the ecological balance problem didn’t exist, then organisms would have evolved mechanisms to withstand damage. This may be a stupid question, but how would you handle the increasing load of mutations in this case? I had read somewhere that there exists a mechanism in the cells to fix errors in the genome, but errors during this recovery step become permanent (obviously).

    • This is a nice idea which I haven’t seen before. But it’s hard to imagine that damage can be completely taken care of, were it for the right genes. Every complex system around us fails after sufficient time when the accumulated damage is too much to bear, whether it is living or non-living.

      Perhaps it would, eventually. But that is not what causes aging. Aging is programmed self-destruction.

      Biology is not my area, so please excuse the inaccurate language, but here are some issues:
      What is your field?

      1. Is aging the only possible solution to maintain ecological stability?

      No – there are many adaptations that evolution has used to maintain ecological stability. Perhaps the most important is sensing crowded conditions and adjusting reproduction accordingly. Even very simple animals
      (worms, microbes) can sense crowding.

      And how come it is so pervasive, ranging from microorganisms to big animals? Intuitively it feels like this one solution should not fit (or be the best solution for) all creatures across all ecosystems, nor should the ecological stability problem exist for all creatures.

      There are a few animals and many plants that do not age. The reason non-aging is so much more prevalent in plants is that they are at the base of the ecosystem, and don’t have to worry about preserving the species underneath them.

      2. I understand that the notion that caloric restriction (CR) extends lifespan is quite popular. But is it established conclusively?
      I refer you to a huge literature in this area, tens of thousands of articles

      Let’s say there are 2 animals in the wild, one is small, weak, and vulnerable due to CR, while the other gets plenty to eat. I suspect that despite all the advantages in signalling, the CRed animal will on average die earlier due to weak immunity, predation, etc while also being less successful in reproduction. (then how to explain CR effect in lab? Perhaps it’s just less damage from eating less.

      You can “suspect” what you like, but to make it science you have to do an experiment.

      Secondly it’s not fully established: CR didn’t increase average lifespan in wild mice according to one study (pubmed PMC2923404), and the benefit is questionable in the well known NIA rhesus monkeys study).
      Yes there are rare exceptions, but life extension from CR is robust.

      As for exercise, I don’t know if it really increases lifespan or if it’s just the presently popular magical pill purported to be a solution for every problem. Maybe it’s just obesity avoidance, or maybe the people who exercise have healthy habits.

      Why do you think obesity shortens life span? It’s a fact we’ve gotten used to, but it is by no means “expected”.

      3. According to this theory, if the ecological balance problem didn’t exist, then organisms would have evolved mechanisms to withstand damage. This may be a stupid question, but how would you handle the increasing load of mutations in this case? I had read somewhere that there exists a mechanism in the cells to fix errors in the genome, but errors during this recovery step become permanent (obviously).

      Error correction is as efficient as it needs to be given the life span of the organism. Trees that live 3,000 years have impressively efficient error correction. Worms live only a few days, and they keep the same cells, so error correction is quite crude.

      But it is not the error correction that determines the life span – rather it is the life span that determines the error rate that can be tolerated. We know this because no animal or plant has a life span limited by somatic mutations.

      ——–

      Thanks for your interest in this area. 19 years ago, I started out with such questions as yours, and I have done a lot of reading in that time to satisfy my curiosity. I wish you the same good fortune I have had.

      -Josh

      • Thanks for your instructive comments Josh. You clearly have thought this through 🙂

        “What is your field?”

        My background is computer science; it’s just that I stumbled upon CR and thereafter got interested in aging itself.

  5. Hi Josh, thanks for your theory of aging. I do want to express my appreciation for your work, even though I may not always agree with you; so don’t take my disagreements personally.
    In this case though I do want to correct an error that I have read in at least one previous essay: The production of ROS does not increase during exercise. Just the reverse. ROS production goes down to almost zero; because the electrons follow the path of least resistance and simply flow through the electron chain as they are supposed to do, eventually leading to the production of ATP, which is being used up rapidly during exercise.
    However, I am not suggesting that the reduction of ROS during exercise is responsible for the life extending properties of exercise. Rather, I believe exercise activates repair mechanisms responsible for life extension. 🙂

  6. Linda Partridge of University College London has a great paper summarizing the research on how a single mutation of an insulin like growth hormone gene can dramatically slow down the aging process in worms. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2842712/

    I came across this article after reading your blog and as a result it seemed obvious that this mutation was a disruption in the pre-programmed aging process brought on by evolution. But the paper seemed to hold on to the old dogma of evolution as a vehicle for reproductive fitness alone. Naturally, the paper then struggles to answer why this specific gene (in the non-mutated form) is so highly conserved across so many different forms of life (from mammals to yeast to the aforementioned worms).

    After citing papers and asserting that group selection couldn’t be correct, the paper then states, “And the realization that it is an evolutionary side effect, rather than an adaptive process, led to the widespread assumption that mutations in single genes were unlikely to be capable of slowing down ageing. Furthermore, it seemed improbable that mechanisms of ageing would be the same in different kinds of organisms”

    But of course a single mutation does seem to be capable of slowing down aging and the mechanisms of aging do not seem to variable as one would expect if the randomness of genetic drift were to take over. The paper then cannot explain “why is the mutant not the wild-type?” It decides that it must cause weaknesses not seen in the lab.

    It seems painfully obvious to me that evolution wants organisms to die for the good of the population, and has both selected and conserved genes that ensure it happens. I think the demographic theory of aging might be the best theoretical explanation of why evolution would want this.

    I have two questions: 1) It seems to me that VERY early life would have less predation because there was less diversity and therefore be MORE susceptible to the boom and bust that make aging a requirement? Do you think there is any truth to that?
    2)Do you think that if an organism is living, it’s not really helping the species evolve (bacteria aside)…..and hence that species is losing the evolutionary arms race if it doesn’t include aging?

      • I was thinking that the first type of life to emerge from the primordial soup that was not capable of making their own food would be the ideal candidate to evolve programmed aging. In my head, this would be among the first type of life that would have a limited food supply and it seems reasonable that this life would also have a very limited number of predators or diseases to control populations. In other words, perhaps aging evolved before many of the predators and diseases that now serve to keep populations in check. You have made the point that aging is essential to the development of a balanced ecology…..I just think that perhaps it was EVEN MORE important at the onset of life.

        Anyway, great blog! I hope your ideas win out because it seems clear to me the rest of evolutionary biology has veered off on a tangent to nowhere on this one, and it may be preventing actual cures from being found.

Leave a Reply

Your email address will not be published. Required fields are marked *