Saturday Night Science: The Evolutionary Psychology Behind Politics

“The Evolutionary Psychology behind Politics” by Anonymous ConservativeOne of the puzzles noted by observers of the contemporary political and cultural scene is the division of the population into two factions, (called in the sloppy terminology of the United States) “liberal” and “conservative”, and that if you pick a member from either faction by observing his or her position on one of the divisive issues of the time, you can, with a high probability of accuracy, predict their preferences on all of a long list of other issues which do not, on the face of it, seem to have very much to do with one another. For example, here is a list of present-day hot-button issues, presented in no particular order.

  1. Health care, socialised medicine
  2. Climate change, renewable energy
  3. School choice
  4. Gun control
  5. Higher education subsidies, debt relief
  6. Free speech (hate speech laws, Internet censorship)
  7. Deficit spending, debt, and entitlement reform
  8. Immigration
  9. Tax policy, redistribution
  10. Abortion
  11. Foreign interventions, military spending

What a motley collection of topics! About the only thing they have in common is that the omnipresent administrative super-state has become involved in them in one way or another, and therefore partisans of policies affecting them view it important to influence the state’s action in their regard. And yet, pick any one, tell me what policies you favour, and I’ll bet I can guess at where you come down on at least eight of the other ten. What’s going on?

Might there be some deeper, common thread or cause which explains this otherwise curious clustering of opinions? Maybe there’s something rooted in biology, possibly even heritable, which predisposes people to choose the same option on disparate questions? Let’s take a brief excursion into ecological modelling and see if there’s something of interest there.

As with all modelling, we start with a simplified, almost cartoon abstraction of the gnarly complexity of the real world. Consider a closed territory (say, an island) with abundant edible vegetation and no animals. Now introduce a species, such as rabbits, which can eat the vegetation and turn it into more rabbits. We start with a small number, P, of rabbits. Now, once they get busy with bunny business, the population will expand at a rate r which is essentially constant over a large population. If r is larger than 1 (which for rabbits it will be, with litter sizes between 4 and 10 depending on the breed, and gestation time around a month) the population will increase. Since the rate of increase is constant and the total increase is proportional to the size of the existing population, this growth will be exponential. Ask any Australian.

Now, what will eventually happen? Will the island disappear under a towering pile of rabbits inexorably climbing to the top of the atmosphere? No—eventually the number of rabbits will increase to the point where they are eating all the vegetation the territory can produce. This number, K, is called the “carrying capacity” of the environment, and it is an absolute number for a given species and environment. This can be expressed as a differential equation called the Verhulst model, as follows:

Verhulst equation: \frac{dP}{dt} & = & rP(1-\frac{P}{K})

It’s a maxim among popular science writers that every equation you include cuts your readership by a factor of two, so among the hardy half who remain, let’s see how this works. It’s really very simple (and indeed, far simpler than actual population dynamics in a real environment). The left side, “dP/dt” simply means “the rate of growth of the population P with respect to time, t”. On the right hand side, “rP” accounts for the increase (or decrease, if r is less than 0) in population, proportional to the current population. The population is limited by the carrying capacity of the habitat, K, which is modelled by the factor “(1 − P/K)”. Now think about how this works: when the population is very small, P/K will be close to zero and, subtracted from one, will yield a number very close to one. This, then, multiplied by the increase due to rP will have little effect and the growth will be largely unconstrained. As the population P grows and begins to approach K, however, P/K will approach unity and the factor will fall to zero, meaning that growth has completely stopped due to the population reaching the carrying capacity of the environment—it simply doesn’t produce enough vegetation to feed any more rabbits. If the rabbit population overshoots, this factor will go negative and there will be a die-off which eventually brings the population P below the carrying capacity K. (Sorry if this seems tedious; one of the great things about learning even a very little about differential equations is that all of this is apparent at a glance from the equation once you get over the speed bump of understanding the notation and algebra involved.)

This is grossly over-simplified. In fact, real populations are prone to oscillations and even chaotic dynamics, but we don’t need to get into any of that for what follows, so I won’t.

Let’s complicate things in our bunny paradise by introducing a population of wolves. The wolves can’t eat the vegetation, since their digestive systems cannot extract nutrients from it, so their only source of food is the rabbits. Each wolf eats many rabbits every year, so a large rabbit population is required to support a modest number of wolves. Now if we go back and look at the equation for wolves, K represents the number of wolves the rabbit population can sustain, in the steady state, where the number of rabbits eaten by the wolves just balances the rabbits’ rate of reproduction. This will often result in a rabbit population smaller than the carrying capacity of the environment, since their population is now constrained by wolf predation and not K.

What happens as this (oversimplified) system cranks away, generation after generation, and Darwinian evolution kicks in? Evolution consists of two processes: variation, which is largely random, and selection, which is sensitively dependent upon the environment. The rabbits are unconstrained by K, the carrying capacity of their environment. If their numbers increase beyond a population P substantially smaller than K, the wolves will simply eat more of them and bring the population back down. The rabbit population, then, is not at all constrained by K, but rather by r: the rate at which they can produce new offspring. Population biologists call this an r-selected species: evolution will select for individuals who produce the largest number of progeny in the shortest time, and hence for a life cycle which minimises parental investment in offspring and against mating strategies, such as lifetime pair bonding, which would limit their numbers. Rabbits which produce fewer offspring will lose a larger fraction of them to predation (which affects all rabbits, essentially at random), and the genes which they carry will be selected out of the population. An r-selected population, sometimes referred to as r-strategists, will tend to be small, with short gestation time, high fertility (offspring per litter), rapid maturation to the point where offspring can reproduce, and broad distribution of offspring within the environment.

Wolves operate under an entirely different set of constraints. Their entire food supply is the rabbits, and since it takes a lot of rabbits to keep a wolf going, there will be fewer wolves than rabbits. What this means, going back to the Verhulst equation, is that the 1 − P/K factor will largely determine their population: the carrying capacity K of the environment supports a much smaller population of wolves than their food source, rabbits, and if their rate of population growth r were to increase, it would simply mean that more wolves would starve due to insufficient prey. This results in an entirely different set of selection criteria driving their evolution: the wolves are said to be K-selected or K-strategists. A successful wolf (defined by evolution theory as more likely to pass its genes on to successive generations) is not one which can produce more offspring (who would merely starve by hitting the K limit before reproducing), but rather highly optimised predators, able to efficiently exploit the limited supply of rabbits, and to pass their genes on to a small number of offspring, produced infrequently, which require substantial investment by their parents to train them to hunt and, in many cases, acquire social skills to act as part of a group that hunts together. These K-selected species tend to be larger, live longer, have fewer offspring, and have parents who spend much more effort raising them and training them to be successful predators, either individually or as part of a pack.

K or r, r or K: once you’ve seen it, you can’t look away.”

Just as our island of bunnies and wolves was over-simplified, the dichotomy of r– and K-selection is rarely precisely observed in nature (although rabbits and wolves are pretty close to the extremes, which it why I chose them). Many species fall somewhere in the middle and, more importantly, are able to shift their strategy on the fly, much faster than evolution by natural selection, based upon the availability of resources. These r/K shape-shifters react to their environment. When resources are abundant, they adopt an r-strategy, but as their numbers approach the carrying capacity of their environment, shift to life cycles you’d expect from K-selection.

What about humans? At a first glance, humans would seem to be a quintessentially K-selected species. We are large, have long lifespans (about twice as long as we “should” based upon the number of heartbeats per lifetime of other mammals), usually only produce one child (and occasionally two) per gestation, with around a one year turn-around between children, and massive investment by parents in raising infants to the point of minimal autonomy and many additional years before they become fully functional adults. Humans are “knowledge workers”, and whether they are hunter-gatherers, farmers, or denizens of cubicles at The Company, live largely by their wits, which are a combination of the innate capability of their hypertrophied brains and what they’ve learned in their long apprenticeship through childhood. Humans are not just predators on what they eat, but also on one another. They fight, and they fight in bands, which means that they either develop the social skills to defend themselves and meet their needs by raiding other, less competent groups, or get selected out in the fullness of evolutionary time.

But humans are also highly adaptable. Since modern humans appeared some time between fifty and two hundred thousand years ago they have survived, prospered, proliferated, and spread into almost every habitable region of the Earth. They have been hunter-gatherers, farmers, warriors, city-builders, conquerors, explorers, colonisers, traders, inventors, industrialists, financiers, managers, and, in the Final Days of their species, WordPress site administrators.

In many species, the selection of a predominantly r or K strategy is a mix of genetics and switches that get set based upon experience in the environment. It is reasonable to expect that humans, with their large brains and ability to override inherited instinct, would be especially sensitive to signals directing them to one or the other strategy.

Now, finally, we get back to politics. This was a post about politics. I hope you’ve been thinking about it as we spent time in the island of bunnies and wolves, the cruel realities of natural selection, and the arcana of differential equations.

What does r-selection produce in a human population? Well, it might, say, be averse to competition and all means of selection by measures of performance. It would favour the production of large numbers of offspring at an early age, by early onset of mating, promiscuity, and the raising of children by single mothers with minimal investment by them and little or none by the fathers (leaving the raising of children to the State). It would welcome other r-selected people into the community, and hence favour immigration from heavily r populations. It would oppose any kind of selection based upon performance, whether by intelligence tests, academic records, physical fitness, or job performance. It would strive to create the ideal r environment of unlimited resources, where all were provided all their basic needs without having to do anything but consume. It would oppose and be repelled by the K component of the population, seeking to marginalise it as toxic, privileged, or exploiters of the real people. It might even welcome conflict with K warriors of adversaries to reduce their numbers in otherwise pointless foreign adventures.

And K-troop? Once a society in which they initially predominated creates sufficient wealth to support a burgeoning r population, they will find themselves outnumbered and outvoted, especially once the r wave removes the firebreaks put in place when K was king to guard against majoritarian rule by an urban underclass. The K population will continue to do what they do best: preserving the institutions and infrastructure which sustain life, defending the society in the military, building and running businesses, creating the basic science and technologies to cope with emerging problems and expand the human potential, and governing an increasingly complex society made up, with every generation, of a population, and voters, who are fundamentally unlike them.

Note that the r/K model completely explains the “crunchy to soggy” evolution of societies which has been remarked upon since antiquity. Human societies always start out, as our genetic heritage predisposes us to, K-selected. We work to better our condition and turn our large brains to problem-solving and, before long, the privation our ancestors endured turns into a pretty good life and then, eventually, abundance. But abundance is what selects for the r strategy. Those who would not have reproduced, or have as many children in the K days of yore, now have babies-a-poppin’ as in the introduction to Idiocracy, and before long, not waiting for genetics to do its inexorable work, but purely by a shift in incentives, the rs outvote the Ks and the Ks begin to count the days until their society runs out of the wealth which can be plundered from them.

But recall that equation. In our simple bunnies and wolves model, the resources of the island were static. Nothing the wolves could do would increase K and permit a larger rabbit and wolf population. This isn’t the case for humans. K humans dramatically increase the carrying capacity of their environment by inventing new technologies such as agriculture, selective breeding of plants and animals, discovering and exploiting new energy sources such as firewood, coal, and petroleum, and exploring and settling new territories and environments which may require their discoveries to render habitable. The rs don’t do these things. And as the rs predominate and take control, this momentum stalls and begins to recede. Then the hard times ensue. As Heinlein said many years ago, “This is known as bad luck.”

And then the Gods of the Copybook Headings will, with terror and slaughter return. And K-selection will, with them, again assert itself.

Is this a complete model, a Rosetta stone for human behaviour? I think not: there are a number of things it doesn’t explain, and the shifts in behaviour based upon incentives are much too fast to account for by genetics. Still, when you look at those eleven issues I listed so many words ago through the r/K perspective, you can almost immediately see how each strategy maps onto one side or the other of each one, and they are consistent with the policy preferences of “liberals” and “conservatives”. There is also some rather fuzzy evidence for genetic differences (in particular the DRD4-7R allele of the dopamine receptor and size of the right brain amygdala) which appear to correlate with ideology.

Still, if you’re on one side of the ideological divide and confronted with somebody on the other and try to argue from facts and logical inference, you may end up throwing up your hands (if not your breakfast) and saying, “They just don’t get it!” Perhaps they don’t. Perhaps they can’t. Perhaps there’s a difference between you and them as great as that between rabbits and wolves, which can’t be worked out by predator and prey sitting down and voting on what to have for dinner. This may not be a hopeful view of the political prospect in the near future, but hope is not a strategy and to survive and prosper requires accepting reality as it is and acting accordingly.

Anonymous Conservative [Michael Trust]. The Evolutionary Psychology Behind Politics. Macclenny, FL: Federalist Publications, [2012, 2014] 2017. ISBN 978-0-9829479-3-7.

18+
avataravataravataravataravataravataravataravataravataravataravataravataravataravataravataravatar

Author: John Walker

Founder of Ratburger.org, Autodesk, Inc., and Marinchip Systems. Author of The Hacker's Diet. Creator of www.fourmilab.ch.

14 thoughts on “Saturday Night Science: The Evolutionary Psychology Behind Politics

  1. John Walker:
    Humans are not just predators on what they eat, but also on one another.

    You’ll think of something.

    – Jim Taggart to his sister Dagny

     

    Thanks for the thought-provoking review.

    7+
    avataravataravataravataravataravataravatar
  2. Thank you, John. This is indeed interesting and plausible. Having considered these divisions and their possible origins for  quite some time, I have tended to think at a simpler level, an environmental or epigenetic one, perhaps. My simple-minded division is between people who, on the one hand, wish to be adults and accept responsibility for themselves, their families and their communities; on the other, those who prefer to remain children and conjure a big mommy and daddy substitute somewhere (the state) which can right all wrongs and impose “fairness” for those who won’t fend for themselves. Since God is no longer relevant to large swaths of society, the state stands alone in loco parentis.

    9+
    avataravataravataravataravataravataravataravataravatar
  3. Fascinating concepts. Anonymous Conservative and his r/K model of political thought were big influences on me and on the fictional universe of my Hidden Truth novels.

    You might also want to check out Anonymous Conservative’s Blog, for his daily news summaries. In addition, he wrote an excellent and very practical book on the psychology of narcissists: How to Deal with Narcissists: Why They Become Evil, How They Think, and Strategies and Techniques to Take Control.

    Thanks, John, for sharing and for providing this excellent summary.

    9+
    avataravataravataravataravataravataravatar
  4. Thanks for this lead. Questions that arise immediately relate to ‘changing individual human behavior’ and, given that possibility, to what purpose. As a matter of fact, it seems any effort at the former would depend on convincing rationale for the latter.

    1+
    avatar
  5. Since this is a post about science and we have some with that interest here, what does this subject matter tell us about the ‘scientists’ lining up with great concern over the effects of ‘climate change’ versus those who just consider this business as usual (there must be some)?

    2+
    avataravatar
  6. I see people like Chuck Schumer as “K”s posing as “r”s in order to use them to enslave the rest of the “K”s.  That is a strategy that can extend the duration of r dominance to the benefit of their camouflaged K leadership.

    Ripping off the camouflage is necessary to avoid enslavement (or minimize the duration, worst case).

    2+
    avataravatar
  7. Bob Thompson:
    Since this is a post about science and we have some with that interest here, what does this subject matter tell us about the ‘scientists’ lining up with great concern over the effects of ‘climate change’ versus those who just consider this business as usual (there must be some)?

    I’m not sure there’s much of a connection.  The key thing about the whole “climate change” agenda is that aligns almost precisely with the agenda of the collectivist slavers before anybody ever heard the term “climate change”: reduced expectations for most of the population and especially the global poor, more top-down government control with rationing, punitive taxation, and mandates all around, and suppression of technologies (nuclear fission and fusion, space solar power satellites, etc.) which might provide a fix to the problem rather than empowering the ruling class to forever “manage” it.

    I don’t see a clear mapping of the r/K model into climate change, except that the proposed solutions (“Green New Deal”) are pure catnip to an r population, since they’re all about sticking it to the wealthy energy producers and spreading the bounty around to consumers and cronies in on the scam (just look at wind farm subsidies in the UK for a fine example).

    4+
    avataravataravataravatar
  8. John Walker:

    Bob Thompson:
    Since this is a post about science and we have some with that interest here, what does this subject matter tell us about the ‘scientists’ lining up with great concern over the effects of ‘climate change’ versus those who just consider this business as usual (there must be some)?

    I’m not sure there’s much of a connection.  The key thing about the whole “climate change” agenda is that aligns almost precisely with the agenda of the collectivist slavers before anybody ever heard the term “climate change”: reduced expectations for most of the population and especially the global poor, more top-down government control with rationing, punitive taxation, and mandates all around, and suppression of technologies (nuclear fission and fusion, space solar power satellites, etc.) which might provide a fix to the problem rather than empowering the ruling class to forever “manage” it.

    I don’t see a clear mapping of the r/K model into climate change, except that the proposed solutions (“Green New Deal”) are pure catnip to an r population, since they’re all about sticking it to the wealthy energy producers and spreading the bounty around to consumers and cronies in on the scam (just look a wind farm subsidies in the UK for a fine example).

    I guess I was thinking along the lines that ‘real’ scientists would look more towards technological solutions, not just giving up long standing traditions for survival,  even if we are facing man-made climate change. The significant percentage of scientists joining the chorus is surprising.

    2+
    avataravatar
  9. I don’t understand his claim that the r strategists would necessarily welcome an influx of fellow r strategists, unless he means as mediated by a middle-term political calculus to preserve the voting bloc which in turn preserves the r subsidy.

    The rabbits would not welcome an influx of some other strain of rabbit.  Those might as well be wolves.

    That is to say, the claim certainly holds for the ultimate argument, but it does not follow from the analogous rabbits used to map r/K onto politics.

    But I agree with his broad claims.

    1+
    avatar
  10. Haakon Dahl:
    I don’t understand his claim that the r strategists would necessarily welcome an influx of fellow r strategists, unless he means as mediated by a middle-term political calculus to preserve the voting bloc which in turn preserves the r subsidy.

    There is only one page in the book explicitly about immigration, but the essence of his argument is as follows:

    Within the immigration debate, we see two positions.  One holds that America has plentiful resources, and there is no reason to prevent fellow citizens of the world from coming here, and helping themselves to our bounty.  To this psychology, resources will always be freely available, conflict or denial of resources to anyone else is wrong, and there is no real out-group or in-group to either show loyalty to, or withhold loyalty from.

    Additionally, in seeking to give foreigners the right to elect leaders who will dictate policy to our native citizens, this strategy also sets the stage for conflict between individuals who are conflict-predisposed, in the future.  If r and K are traits, then the r-strategy is setting the stage for K-trait carriers (both domestic and foreign) to become embroiled in a depopulating fight, which the r-strategists will try to avoid and hide from, instinctually.  This would be a brilliant competitive strategy for a conflict-averse trait to engage in.

    One psychology sees a world of limited resources, where competition is wrong and immoral.  The other sees a world of limited resources, and groups which need to be fought off to secure them.  If conditions turn K-selective, there is no doubt which psychology would quickly dominate the scene.

    The perceived (or actual presence) of unlimited resources essentially the definition of an r-selected population: it is not constrained by carrying capacity.  You hear this echoed in the common left-wing trope, “There’s plenty of money for everything.  It’s just in the wrong hands.”

    This is, however, one of the places where the book’s trying to connect the r and K strategies to specific policy preferences on contemporary issues strikes me a bit like “When all you have is a hammer, everything looks like a nail.”  The reason for support for immigration among the r population may be much simpler: since immigrants are predominantly r, they and their progeny will eventually tilt the demographic in electoral balance decisively in their favour.

    Haakon Dahl:
    The rabbits would not welcome an influx of some other strain of rabbit.  Those might as well be wolves.

    In an r-constrained environment (for example, where wolf predation kept the number of rabbits well below K), introduction of another species of rabbit would not deprive the existing population of food supply.  It would simply introduce other targets for predation.  One pattern among r-selected species is that they avoid conflict: if there are abundant resources, why waste time (and possible loss of your genes) fighting when you can be breeding?

    3+
    avataravataravatar
  11. Lefty hates the r/K life history concept because it’s racist… or something. Ed Dutton has argued that abortion has depressed the fertility of r-strategists because, although they live fast, they have no compunctions about killing their offspring. Thus, we have fewer spiteful mutants* than one might expect. Ireland was the exception since they were late in allowing abortion. Dutton attributes the precipitous decline of religiosity in Ireland to prodigious breeding by the spiteful mutants that was uninhibited by abortion prohibition.

    *Dutton’s coauthor Michael Woodley of Menie coined the term spiteful mutants to refer to ethnomasochists who promote dysgenic, self-destructive policies.

    4+
    avataravataravataravatar
  12. Bob Thompson:
    I guess I was thinking along the lines that ‘real’ scientists would look more towards technological solutions, not just giving up long standing traditions for survival,  even if we are facing man-made climate change. The significant percentage of scientists joining the chorus is surprising.

    It’s not surprising at all. Funding for science is overwhelmingly provided by national and supranational entities. Scientists are no different from anyone else: they need to put food on the table. If the ruling class likes global warming climate change because it centralizes authority so as to better control the populace, that’s what they will pay scientists to “discover.”

    It soon got to the point where expressing any heretical thoughts concerning the climate orthodoxy was career-ending. A good example is Henk Tennekes, coauthor of one of my favorite books (I have two copies!) and distinguished fluid dynamicist. Soon after he published an article critical of climate modeling, he was pushed out of the directorship of the Royal Dutch Meteorological Institute.

    This is why the critics are mostly older scientists. I can’t blame younger folks for keeping their heads down. There are plenty of stories like Tennekes’s, as well as milder versions.

    6+
    avataravataravataravataravataravatar

Leave a Reply