Demography is the statistical study of population age structures, or the study of aging and fertility through population statistics. It’s not deep math, but it definitely attracts people who love numbers. The world’s formost demographer is James Vaupel, an American who has been working at the Max Planck Inst in Germany most of his career. Annette Baudisch is a brilliantly creative protege of Vaupel, who has come into her own in the last decade*.
The point of this is that when such people declare that evolutionary theory doesn’t work, we ought to be listening. We the evolutionary theorists, we the gerontologists, and we who simply seek a path to a longer, healthier life, and we who have been influenced, perhaps unawarely, by tacit assumptions about evolution.
Essay in Science a year ago, Survey in Nature last month
Last month, these two world-class demographers published a cross-species study correlating fertility with the rate of aging, and they report results that are deeply at odds with the predictions of evolutionary theory. They are not shy about saying so.
The classic evolutionary theories of aging provide the theoretical framework that has guided aging research for 60 years. Are the theories consistent with recent evidence?
At the heart of the theories lies the observation that the old count less than the young: Unfavorable traits are weeded out by evolution more slowly at higher ages; traits that are beneficial early in life are selected for despite late life costs; and resources are used to enhance reproduction at younger ages instead of maintaining the body at ages that do not matter much for evolution. The decline in the force of selection with age is viewed as the fundamental cause of aging. It is why, starting at reproductive maturity, senescence—increases in susceptibility to death and decreases in fertility—should be inevitable in all multicellular species capable of repeated breeding. Yet, this is not the case. Increasing, constant, and decreasing mortality (and fertility) patterns (see the figure) are three generic variants that compose the rich diversity of life trajectories observed in nature. For vertebrates, reproductive trajectories are commonly hump-shaped, and death rates may start rising much later than reproductive maturity. Thus, a new view on the fundamental causes of aging is needed to explain the clash of theory and data. [from a 2012 Science essay by the same authors]
This is the kind of condensed academic prose that makes scientific communication so efficient to specialists and so opaque to anyone on the outside trying to figure out what’s going on. They have captured the core of three evolutionary theories that are currently considered acceptable. (If you are already familiar with them, you will recognize what they’re talking about; if you would like more detailed accounts of the three theories and their failures, I have written about that here.) They go on to say that aging in nature has a richness and diversity that these simple theories do not begin to address. “theories to explain the ultimate evolutionary causes of the varieties of ageing…are in their infancy.”
Baudisch pioneered the study across species of different shapes of life history curves, regardless of the time scale on which they unfold. In other words, some species have life plans that unfold over days, and others over decades. Let’s ignore that and stretch out the time axis so that they all fit in the same plot. Some species (e.g. modern man) have a high rate of survival right up to the end, and then everyone dies in a narrow range of old age; while other species tend to die at a steady rate, unrelated to age, and for some they are actually less likely to die the older they get. This last case seems strange to us. Baudisch and Vaupel coined the term “negative senesence”, but it doesn’t have to look like Benjamin Button. Just think of a pine tree that gets larger and stronger over the decades, and thus more resistant to a drought or a fire or a windstorm, thus less and less likely to die with each passing year.
(You can blow up the figure below to view details.) The blue line plots fertility over a lifetime: how many offspring are produced per unit time; the red line plots mortality: what is the probability of an individual dying before the next year or the next day? These many different plots represent the diversity of different patterns of aging in nature. The graphs are all stretched out or compressed in time so that each box contains one lifetime, whether that be a day or a decade.
The top row codifies the life plan that is most familiar to us, because it is ours. Fertility peaks in early life, then declines. For females, it declines to zero. Mortality is modest for a long while, then it climbs steeply and everyone dies. This is the story we take for granted. It is the form of aging shared by humans, guppies, and certain sea birds.
The next line shows life plans that are similar, but where mortality rises more slowly, so that age of death is spread out over time. This row contains some familiar mammals like deer, lions, and orcas, but it also contains water fleas and bdelloid rotifers, microscopic creatures famous (at least to biologists) for their chastity.
Further down are stranger and less familiar life plans, including those where fertility rises as mortality falls through most of the life span. According to evolutionary theory, metabolisms aren’t supposed to be able to do this. The whole reason for aging is (according to theory) the necessity for compromise. Some organisms can have their cake and eat it. (Would it be mixing metaphors to call this an evolutionary free lunch?) When theory accounts for aging in other species as a sacrifice of longevity for fertility, the story rings hollow. Why are some species but not others compelled to this compromise?
Of special significance is post-reproductive life span. Human females go right on living after they have lost their fertility. This is supposed to be explained by the need to care for her grandchildren. But in this chart are several other species that also outlive their fertility, including elephants and ground squirrels but also worms and guppies that don’t care for their young at all, let alone their grandchildren.
This poses a big problem for evolutionary theories of aging, because maintaining the body through an extended life span is always presumed to be costly in one way or another – that’s why the body skimps on the job, and the body is permitted to deteriorate with age. Why, then, would the body take the trouble to preserve itself for a time when it was unable to reproduce, useless to the species and invisible to evolution? “There should be little or no postreproductive period in the normal life-cycle of any species.”, predicted George Williams in his seminal paper (1957) which has inspired most modern thought on the subject. (I have written about this topic, suggesting that the post-reproductive segment of the population provides a stabilizing buffer in times of famine, helping to guard against extinction.**)
The bottom line
The bottom line is that nature has been able to do pretty much anything she wants with the metabolism of aging, and the trajectory of mortality that comes from that metabolism. Most of the evolutionary theory on the subject is based on the idea that natural selection could never affirmatively choose aging if there were a choice. Aging is bad for the fitness of the organism that suffers aging, and theory says that natural selection should always work against aging. If most living things suffer decline with age, leading to death, this must have taken place despite natural selection. There must be some genetic constraints, or physical limitations, or conditions beyond control of the genome.
This survey of Baudisch and Vaupel tells a different story.
For fifteen years now, I’ve been saying that the evolutionary theory of aging doesn’t work, and that it’s high time to stop making excuses for the old theory, to adopt a new theory that fits better with empirical reality. I have argued from experiments and from field surveys and from general knowledge and from logic, and I have found many people who agree, and maybe a few who have been turned around by my presentation. But I don’t have the stature that would attract a large readership, or compel anyone to take my word for it. Maybe they’ll listen to Baudisch and Vaupel.
———–
* I first became aware of Baudisch in 2004. One of the most common of ancient human follies is to imagine that the way things are is the way things must be. William D Hamilton – a biologist smart enough to know better – published in 1966 his “proof” from fundamental precepts of evolution that all living things must decline with age. In a paper provocatively titled, “negative senescence”, Baudisch and Vaupel surveyed a number of animals and plants that don’t decline with age – quite the opposite, they continue to get larger, stronger, and ever more fertile. “Negative senescence”. They even included their own “proof” that aging was impossible, and that all living things ought to grow ever stronger and more fertile.
[back to text]
** In this article, I have proposed that population stability is a major part of fitness in nature. Populations that swing too wildly up and down are in danger of extinction, and it makes sense that such extinctions are a form of Darwinian selection, and they would have left their mark on the genome. In the case of post-reproductive life span, here’s how it would work: Sometimes there’s plenty of food, and the population is expanding on a trajectory that’s going to lead to overpopulation and a crash. Then it’s a good thing to have these older, infertile adults around, because they eat up some of the food, but they don’t contribute to population growth, while the population is growing too fast already. At other times, food may be scarce and the population is shrinking. Then the post-reproductive population will be the first to go, because they are old and weak. They are expendable, from a demographic perspective, because when they disappear there’s no loss to the population’s potential for growth, but there is a benefit when competition for scarce food is reduced. This is the sense in which a post-reproductive population can provide a buffer, protecting against steep population fluctuations and potential extinction. [Link to article in Oikos] [back to text]
Discover more from Josh Mitteldorf
Subscribe to get the latest posts sent to your email.
Aging is a complex phenomenon. for eg. all mammals evolved from shrew like ancestors ( which had a shorter life span than we do today). even among mammals there is a significant difference in life expectancy. futhermore even among apes gorillas and chimps have a much lower life expectancy than us but significantly longer than mice. This can only mean that life expectancy itself is a variable that evolves with time.
a whole host of factors – some known, like rate of electron leak in the mitochondria, racial (Caucasian vs African) (recent evolution close to poles vs close to equator)(fecundity vs longevity), pollution etc and some which we are only scratching the surface (the microbiome of the individual) play a role.
truth is we fully don’t understand why we die.
“They are expendable, from a demographic perspective, because when they disappear there’s no loss to the population’s potential for growth, but there is a benefit when competition for scarce food is reduced.”
But that is an explanation, not a mechanism. The question is not, “Why is the death of some members of a population good for that population?”. The question is, “Why is the death of some members of a population better for their offspring than for the rest of the population?”.
Yes – conventionally, the only recognized mechanism of evolution is the difference among individuals in how many offspring they leave behind. But I (and others before me) have proposed that local population extinctions are common, and that they, too, leave their imprint on the genome. The mechanism by which shorter life spans and curtailed fertility come to be selected is that those communities with higher fertility and longer life spans overshoot their sustainable population and wipe themselves out.
That doesn’t answer my objection. If mortality were the norm, and immortality a freak mutation, your answer would explain why it inevitably takes over the whole population, then wipes it out. But you have not explained how a mortality gene could spread through an immortal or long-lived population, which is what it would have to do to be “selected”. Nor why a mortality gene inherited by a population would not be outcompeted by mutants that had lost it.
Put it this way; if mortality is a positive trait — meaning that there is a gene or genes that code for it — then there are humans who lack it. They walk among us!
Thanks for commenting and thanks for persisting with this. This is a debate that I have invited, but rarely has the willingness to debate been reciprocated.
For many evolutionary biologists, the mechanism of gradual increase or decrease in gene frequency is not a “model of Darwinian selection”, but rather the only possible model, or rather, they regard it as a logical translation of Darwin’s thinking into a modern, quantitative form.
I ask you to expand your consideration to include other possible mechanisms. The Standard Model holds when the total population of a species remains constant, while the proportions of gene frequency within that total change. But what if the population is expanding or contracting on the same time scale as the gene frequency is changing? This opens the possibility of a qualitatively different evolutionary dynamic.
Perhaps you would find this computer model convincing, published 2012 in Oikos.
If you are the Jerome Berryhill from Eugene Oregon, I send my best to Lake Waldo, where I camped out the summer of 1973, while a grad student in the UCBerkeley Physics Dept.
I will take a look at the papers you cite.
For an update on Waldo Lake, take a look at WaldoCats.org.
I read the paper. I note this sentence; “Simply allowing the life history genes to evolve was not an option, because this leads to extinction.”
What that is saying, is that the existence of stable populations is incompatible with your model of evolution. That might be evidence for the existence of God, or it might be evidence that your model is too simple. What it is not, is license to arbitrarily eliminate the strongest individual competitors from your trials. Which is what you seem to have taken it to be.
Unless you can describe the mechanism that suppresses those competitors, the detailed behavior of the model when they are suppressed is beside the point.
Thinking about this further, it seems to me that the same issue arises with regard to fertility. Why don’t more women have twins? That would seem to be an obvious path to reproductive superiority, a good deal more obvious than reproductive senescence. The obvious answer, is that in the conditions under which we evolved, a woman’s resources did not suffice to raise two children at once. That is, the population may have the carrying capacity for more children, but the individual does not, and that is the deciding factor. One might suppose that the advent of agriculture and the era of surplus would have changed that, but it does not appear to be so.
I believe you have quoted that sentence out of context. My thesis is that the force of individual selection favors higher fertility and longer life spans, but that the collective consequence of this is a population that grows too fast and crashes too hard. So there is a series of local extinctions, a form of group selection that counterbalances individual selection and keeps growth rates in check. Aging is best understood in this context.
I believe I understand that aspect of your approach. Namely, you believe that the “naive” model of evolution leads to a boom followed by a bust, which is to say, extinction. Since that is not what we observe, that model must be wrong. That is fine. But it appears that you are content to postulate some mechanism which prevents those outcomes, and then to discuss the options which remain, without describing the mechanism. Because boom and bust does not suppress only the fertile immortals. Unless you are going to suggest that evolution somehow “knows better” than to allow fertile immortals to arise, simply pointing out that their existence would ultimately lead to extinction does not explain why they should not prosper in the short term.
This is why I said, in jest, that perhaps you have found evidence for the existence of God. If it is truly the case that no evolutionary model can be devised that does not lead immediately to extinction, then a logical person would have to conclude that something other than evolutionary theory is required to explain the situation we observe. My certainty that this will not occur makes me doubt that evolutionary theory, as currently practiced, is a science. I do not believe it is falsifiable.
Perhaps it is possible what you say but, how probable is it? It’s unlikely aging is regulated by just a few genes that could give rise to immortals from random mutations alone and the more mutations required the less probable it gets until it is vanishingly small.
In fact, it probably DID happen on the remote past — probably MANY times — and, every time it did, it was selected out by crashing, which also means there are probably parallel mechanisms in place to ensure any surviving immortal population would die otherwise even if they could survive long enough to pass on those mutations. (or, alternatively to dying, curbing fertility to zero)
Or if you allow me the poetics, there must be fierce guardians to the grail and we should thread carefully when trying to “hack” them down.
Moreover given how old aging is (excuse the pun =)) it’s bound to be on a more stable site of the genome, diminishing that probability even further.
Theory can only take you so far in this game, because Nature is full of surprises. But yes, I agree, there is powerful individual selection against aging, so aging must be held in place by, as you say “guardians of the grail”. We might expect these to be of three kinds:
1) Redundancy, so that if you don’t die of one thing, you’ll die of another. “If the ‘gators don’t get ya then the ‘skeeters will.”
2) Pleiotropy: this is a genetic linkage such that when the body ages more slowly, it becomes less fertile, or less resistant to something.
3) Hiding the genes that cause aging, so that their mutation rate is low.
I agree that we should challenge our existing scientific ideas, no matter how widespread or accurate they seem to be. Evolutionary theory as it stands does not properly account for the great variability in the lifespans of organisms and even more so the occurrence of post reproductive life span. Ageing is not rigorously explained, but we do know that there are a great many factors that influences the age of organisms. I tend to agree with the author in that organisms past their reproductive stage do still provide some benefit to the group, so perhaps some alternate version of altruism could be the basis for long life spans to be selected. Factors other than reproduction would be responsible for long lifespans. Another factor might be the age at which an organism first is able to reproduce. Evolutionary features selected to ensure an organisms survival until they reproduce might be sufficient to cause a lifespan longer than what is necessary to reproduce, if the age of reproduction is sufficiently high.
Thomas Thorburn (14012350)
Student at University of Pretoria
Thanks, Thomas. I tend to think life span is too important for Evolution to have been looking the other way when “other factors” took over. A great deal of genetics and signaling seems to be devoted to regulating life span, and it has been so for a billion years. So I think it’s worthwhile to look at the individual effects and the demographic effects of life span and try to understand many ways in which life span is adaptively regulated.
On a fundamental level, since resource availability is variable over time, can there ever be a “stable” population size?
Since resources for any population are finite, if individuals are immortal they would need to cease effective reproduction (i.e., the offspring can also or at least have the opportunity to reproduce) once carrying capacity is reached or overshoot would result (boom-bust with frequent extirpation). The younger members of the population would never have the opportunity for reproduction at all if the population is completely stable, since they would be the individuals dependent on the last available resources and there would be no free resources available for their offspring.
But then how would the population of immortals respond to changing conditions? Because the environment fluctuates over time, carrying capacity also fluctuates. But a population of immortals should, all else being equal, have fewer generations than mortals. This is simply because in those times when resource conditions are unchanged there should be only minimal turnover of individuals. This in turn presents less opportunity for adaptation, because fewer generations equals fewer mutations that may be adaptive in the future. In other words, mortals have > generations = more opportunity for mutation = greater capacity for adaptation.