Reproduction Control
TLDR: To reduce r-promoting reproductive strategies from dominating the future, Utopia restricts the flow of democratic influence and social support in a way that attempts to be fair and nonviolent.
Prerequisites: Personhood, Futarchy
Utopia, as I’ve described it up to now, has a problem. In my conception of Utopia, many non-human animals are seen as people, and are granted all the rights that a human would. And while Utopia leans on futarchy to set policy, this is still backed with democratic voting on priorities. Finally, Utopia supports all people (including non-humans) with a basic income, such that nobody goes hungry.
I’m not entirely clear where the line of personhood should lie, but it seems plausible to me that dogs should be considered to be people. They may need guardians their whole lives, but that doesn’t mean that the law shouldn’t extend the same protections to dogs as it does to humans. If dogs are considered to be people, they possess the right to collect basic income and have (their guardians express their) priorities in the Utopian futarchy.
The Power of Puppies
Not only would this give significantly more resources to people who take care of dogs, making dog-caretaker a lucrative and powerful position in society, but puppies become extremely valuable. A guardian that watches over a hundred dogs could easily live purely off the basic income, and would be incentivized to have their dogs breed and breed and breed to accumulate ever more power and influence. If breeding dogs constituted as abuse, their guardianship could be challenged and taken away, but it seems very likely to me that if given the opportunity to comment on the situation, most dogs would say that they’d rather breed than not breed.
The situation gets worse if we consider that varieties of dog that are cheaper to care for and which breed faster would be the most lucrative. In the race to have as many puppies as possible, breeds like the Great Dane that are costly to feed and house and only start being able to have puppies at one year of age would lose out to Chihuahuas who are small and start being able to become pregnant around 6 months.
Chihuahuas have about three pups per litter, and it takes about two months to carry them to term and another two months for their body to become able to breed again. Dogs can continue breeding their whole lives, but after a certain point they become much less fertile. Let’s say that happy female chihuahuas stop being able to have puppies at 12 years old.
If we model time as a discrete series of chunks, we can model a population of chihuahuas as a vector in ℝⁿ (i.e. a list of numbers), where each entry represents the quantity of chihuahuas who are a certain age. The change in a population vector with each timestep can then be modeled as a matrix. I won’t go into the linear algebra in depth here, but the short story is that if we consider a vector that retains the same demographic balance over time (“an eigenvector”) we can ask how much the population grows with each time step (“the eigenvalue”) and use that to find the doubling-time of our population. If you’re curious about the math, see the footnotes1. The take away is that a chihuahua population that's encouraged to breed will double in size every 4.2 months or so. (Interestingly, at this extreme rate of growth, a majority of the dogs at any given time will be puppies!)
Even if there are only 4 million chihuahuas alive when this madness begins, within just four years there will be ten BILLION chihuahuas. If this trend was able to continue for twenty years the combined mass of the chihuahuas would almost exactly equal that of the earth! ALL HAIL OUR NEW DOG OVERLORDS PLANET!
Principled Policy
Growth by itself isn’t the issue here. This is not some Malthusian catastrophe2. The issue is differential growth rates. In a peaceful, abundant world — like the kind we want to have — the law of natural selection will specifically promote organisms that reproduce quickly and have many children. This pressure will, in the absence of reproduction controls or scarcity, result in fast-breeders being everywhere and drowning out everyone else.
While I use the example of dogs, this concern is not limited to our canine friends. Any society with differential breeding rates is vulnerable to the dynamic. If one genetic population of humans is prone to breeding quickly and often and another is slower and more careful, in a post-scarcity society with no reproduction controls, the slow-breeders will disappear in the long-run.
This might seem like a fake concern. Current projections indicate that the human population will level-out this century. And groups of humans with reproductive-focused ideologies such as the extremely religious don’t seem to be taking hold. If anything, the number of extremely religious people has been dropping.
But a few centuries is nothing in the grand scheme. Natural selection is operating right now, whether it’s obvious or not. If it continues to operate unchecked over millions of years, the decedents of modern humans will think very differently about things like birth control and adoption.
More imminently, we are entering the age of thinking machines. It seems very plausible that within this century we will have digital people who have all the capacities of a flesh-and-blood human. In a world where having another person is as simple as building a new computer, we could enter a new regime of population growth where flesh-and-blood people are soon the minority, even ignoring the possibility that artificial intelligence will simply defeat everyone.
But ultimately, I think the issue is one of principles. It seems good and natural for groups to grow. With that growth will almost certainly come power. But that power should come from new people putting in the work to change the world, rather than simply being entitled to a share of the pie by way of being born. In other words, a good society should have a story for why having as many kids as possible is not a valid way to automatically dominate society. Establishing a clear principle that prevents this exploit is an important part of having a mature civilization.
Ethical Control
Alas, there’s an immediate obstacle to fixing the exploit. When most people hear “government control of (human) reproduction,” they’re likely to associate it with tyranny, and rightly so. The one-child policy in communist China is the most prominent example of reproductive control at the state level, and this was pretty obviously a gross violation of human rights and probably bad for China as a whole. More extreme examples of reproductive controls also loom in the shadows, such as the forced sterilizations in Nazi Germany and other countries. Whether done as an act of genocide or simply to shape the population, compelling someone to give up their natural ability to reproduce seems to me to be unacceptably violent.
Thus we find ourselves, as policy-makers, in a double bind. Reproductive control seems necessary to avoid having to yield to the pressures of evolution, but it also seems ethically monstrous to dictate to individuals whether they’re allowed to have children. What can we do?
We can use incentives!
For instance, we might pay some people (e.g. humans) to have children, and pay others (e.g. dogs) to not have children. Since everything is done with payments, in theory there’s no coercion being applied (though one wonders how the taxes were collected). Economic incentives aren’t the only kind, either. If a community can coordinate to praise those who are having children at the desired rate and shame those who are breeding too quickly, this might also allow some control.
Incentives aren’t necessarily ethical, however. Paying someone to commit a crime certainly isn’t, for instance. And rewarding people of one species not to have children while doing the opposite for another species is, well, speciesist. But by thinking of “control” less in terms of force and restriction and more in terms of guidance and incentivizing, I think it’s at least possible to find something acceptable.
The key, in my eyes, is in holding up a principle of equality, such that no particular kind of person is discriminated against. If we could somehow do that while offering incentives that smooth out population growth (or at least prevent it from being one weird trick for taking over society) we’d be golden.
Utopian Reproduction Control
As I envision it, incentives around reproduction in Utopia are primarily introduced at the level of the world government. Even though local governments can apply incentives of their own, if there are no top-level controls, the reproductive pressure dynamics will still come into play.
In order to future-proof society for future technologies such as cloning, uplifting, uploading, and de-novo digital people, Utopia frames reproduction not as a strictly biological process, but as anything which brings a new person into the world. In the limit, laws around reproduction should not get philosophically confused about continuity of personhood — someone who copies themselves into a new body and destroys the original should be treated approximately the same as if the original had just changed bodies gradually (modulo questions of murder).
When considering reproduction, each person in Utopia has two numbers associated with them. The first is called a “civilizational influence score” (CIS). Anyone with a CIS of 1 or more gets full basic income and full weight when voting on civilizational priorities. But someone with a CIS greater than 1 receives no additional benefit; this is called “excess CIS.” People with a CIS value that’s less than 1 have their basic income and vote-weight scaled down by their CIS. Thus someone with 0.35 CIS would get 35% of the normal influence when voting and only 35% of the standard basic income. CIS can’t be negative.
The second number that citizens of Utopia have is a CIS Generation Coefficient (CGC). Based on the (changing) priorities of the people, the futarchy selects a base rate of CIS growth (BRCG). The BRCG is essentially a way for society to encourage or discourage people from having more children. Each moment, each person's CIS increases by that rate times their personal CGC. For instance, let’s say the BRCG is 0.05 CIS/year and someone has a CGC of 0.5 — they’d then gain 0.025 CIS each year.
Whenever anyone has a child (or otherwise reproduces), they and any co-parent(s)3 collectively pay 1 CIS to their offspring. In most situations this means that each parent pays 0.5 influence. If the parents can't pay the full CIS, then sibling CIS is split evenly to make up as much of the difference as possible (half-siblings split half their CIS). In a somewhat similar way, whenever a child is born, CGC is (separately) split between each parent and their children, including the new child (naturally causing CGC to flow between siblings).
Let’s consider an example to help illustrate. Let’s say that Alice is born, and gets a full 1 CIS from her parents, and a CGC of 0.4 from her parents and siblings. The BRCG will be fixed at 0.05 CIS/year for this example, so by the time she’s 22 years old she’ll have accumulated an additional (0.05×0.4×22=) 0.44 CIS, bringing her total up to 1.44. Now let’s say that she meets Bob, who also got 1 CIS from his parents, but was blessed with a CGC of 0.9, and happens to be 25 years old. Bob thus has a CIS of (1 + 0.05×0.9×25=) 2.125.
CIS CGC
Alice | 1.44 | 0.4
Bob | 2.125 | 0.9Alice and Bob have a child: Carol. Alice and Bob each pay Carol 0.5 CIS. Alice now has 0.94 CIS, Bob has 1.625, and baby Carol has 1. Alice will personally get slightly less basic income and have less ability to steer civilization’s priorities, as a result of her <1 CIS, but as a guardian of Carol, she’ll collect (some of) Carol’s basic income, and have additional overall influence on policy by voting on Carol’s behalf. Alice and Carol split their CGC between them, and so do Bob and Carol. Alice then has a CGC of 0.2, Bob has a CGC of 0.45 and Carol has a CGC of 0.65.
CIS CGC
Alice | 0.94 | 0.4/2 = 0.2
Bob | 1.625 | 0.9/2 = 0.45
Carol | 1 | 0.4/2 + 0.9/2 = 0.65Five years pass, and Alice and Bob have another child: Dave. Alice and Bob agree that Bob should pay all of the CIS this time. The CGC calculations can be done by forming a pool for each parent based on what their CGC’s were before Carol was born, and splitting each pool three-ways.
Pre-Dave CIS New CIS
Alice | 0.94 + 0.05×0.2×5 = 0.99 | 0.99
Bob | 1.625 + 0.05×0.45×5 = 1.7375 | 0.7375
Carol | 1 + 0.05×0.65×5 = 1.1625 | 1.1625
Dave | | 1
Alice's CGC Pool "A" = 0.2 + (0.2/(0.2+0.45))*0.65 = 0.4
Bob's CGC Pool "B" = 0.45 + (0.45/(0.2+0.45))*0.65 = 0.9
New CGC
Alice | A/3 = 0.133333
Bob | B/3 = 0.3
Carol | A/3 + B/3 = 0.433333
Dave | A/3 + B/3 = 0.433333CIS can be traded (often with a license, to reduce exploitation), but CGC can not. In the event of someone’s death, both their CIS and CGC usually transfer to their family, prioritizing descendants. Because excess CIS is useless, it’s common for people to sell their excess CIS unless they’re saving up for having a kid. To help facilitate trades, people can register to automatically auction off their unused CIS, and it’s not uncommon for people to sell off their CIS right up until they want to have a child, at which point they (jointly) buy 1 CIS at auction.
CGC, on the other hand, is much more precious, since it’s a fixed resource and effectively represents a share over future growth. If it could be traded there's a chance that a subpopulation could end up buying more and more of it up over time, using the increasing market price of CIS to fund their acquisition, similar to how land ownership can create inequality in a world without sensible tax policies.
By scaling the BRCG, the futarchy can exert pressure on how fast Utopia grows, but populations with a fixed CGC have a limit on the votes and income that can be gained by having lots of children. For instance, if dogs start off with 1/100th the total CGC, they might be able to have lots and lots of babies, but they’re unlikely4 to collectively get more than 1% of the vote, regardless of how many babies they have. Thus, even if dogs are considered people in Utopia, their guardians don't have much more influence or income, and have only a small incentive to breed them.
If a new species is added to the collection of people, a hard decision must be made of how much CGC to grant them at the outset. I don’t know of how to decide the answer to that question in general, by my guess is Utopia works hard to guess at how much diversity the newcomers add, and how harmoniously they’ll mesh with the existing vision of Utopia held by those making the decision. Species with more harmony and more diversity should probably receive a greater share of the future than those who are similar and/or unaligned. If a-priori principles are held in a fair and unbiased way, Utopia can perhaps manage to avoid being speciesist, and instead make room in the future for a wide range of people, spread out across the galaxy.
We could model a dog population by splitting all individuals into uniformly-spaced age-buckets, but because we’re only interested in the eigenvalue we can simplify the problem by pretending that there are only four types of dogs: female puppies, pregnant dogs, nursing mothers, and elderly females. Let’s use a two-month timestep. Elderly dogs either die or stay elderly; since old-age lasts 36 months on average, we can say that 2/36ths die each timestep. All pregnant dogs become nursing dogs on the next timestep, and nearly all nursing dogs become pregnant on the next timestep. The only exception is that a small minority of nursing chihuahuas become elderly; since maturity lasts 138 months on average, and half of that time is spent nursing, 2/69ths of nursing mothers become elderly each timestep. Finally, since puppies reach maturity in about 6 months, we can say that 2/6ths of them will become pregnant each timestep. Lastly, we add 3/2 female puppies every time a dog transitions from pregnant to nursing, since average litter size is 3 and half of puppies are female.
2/3 3/2 0 0
1/3 0 67/69 0
0 1 0 0
0 0 2/69 17/18This matrix gives us four eigenvalues, but two of the corresponding eigenvectors involve negative quantities of dogs, and one involves a simple declining population of elderly dogs. The only interesting eigenvector (≈v_1≈(44.3, 21.4, 15.4, 1)) has an eigenvalue of ≈ 1.39.
Doubling time can be then calculated as 2months × ln(2)/ln(1.39) ≈ 4.2months.
I do believe that Malthusian concerns are worth attending to in the long-run, and the policy proposals in this essay can work as a way to nudge things in the right direction, there. That said, I believe that a broader framework is needed than the simple proposal made here to find the right balance between over and underpopulation, once civilization becomes fully mature, and stops growing economically and technologically. There’s a decent chance I’ll write about this if I ever get around to considering technologically-advanced, future Utopias.
In contexts where one of the parents is unknown the remaining parent pays all of the CIS. This is sad, but necessary in order to prevent exploitation of the system.
There are two ways in which CGC can move between groups. First, people from the two groups can interbreed. Second, the victim of a crime such as murder, manslaughter, or involuntary sterilization is often awarded the CGC of their wrongdoer (and then often this flows to their family).