Tyler Cowen: How recognizable will humans be in five hundred years? Alex reports: Tyler and I argued recently about whether or not humans will be recognizably human in 500 years. Let us assume that scientific progress continues. My view is that parents don’t so much like “difference,” unless it is very directly in their favor. Using technology, parents will select for children who are taller, smarter in the way that parents value, better looking, and perhaps also more loyal to their families. The people in the wealthy parts of the world will look more like models and movie stars, but they will be quite recognizable. … People will in various ways be cyborgs, but more or less invisibly from the outside at least.
"Darwinian evolution" is a rather vague term. Evolution itself will still be in progress by the end of the century - since we won't have reached our limits by then.
In the next 500 hundred years a great change will explode and technology is everywhere. Maybe nobody will bear fetus in mother’s womb because there will be technologies to develop a child outside the uterus or even more than that.
Absolutely. Either very baby hungry, very careless (i.e., accidental pregnancies), or very needful of feeling dominant. Maybe not even the middle one. Essentially, if technological growth continues at all, its hard to think of any biological change (e.g., allergies or resistance to birth control pills) that couldn't be cracked if people wanted to. So the evolutionary changes that will result will be changes in what people want.
Surely by "Darwinian evolution will not be in effect" he meant "the impact of genetic evolution will be severely reduced"And yes, Darwinian evolution could be eliminated if we wanted to. Just ban normal reproduction and make everybody a clone. Read "Brave new world".
I know if I were genetically engineering children I'd probably do my best to make them extremely intelligent, to an extent asocial (in the sense that Aspergers is asocial) and unecessarily physically powerful. While a majority herd may go for super-herd animals, you can't discount the formation of subspecies from people with more Nietzschean tendencies.
So explain to me, why should our preferences for body form change in just 500 years of evolution, just because we are much richer? Selection will emphasize some features we already desire such as height.
Even then, there is no runaway, parents will tinker with the actual height of their offspring than with their desire for even taller mates or offspring. That desire might catch up via old-fashioned evolution, but in 500 years time an 8" man would still look weird, even if 6'2" looks short and sweet.
Only fundamental misunderstanding of evolution allows claims that human Darwinian evolution will somehow be eliminated. Evolution cannot be stopped. The only things that can happen is a change in the adaptive landscape.
I kind of hope human nature (genetic intrusion to solipsist preferences) proves difficult or immensely unattractive to engineer.
I think within the Transhumanist community there is an assumption that hacking will be bottom-up, but history shows a durable urge to top-down engineer, which has generally ended quite catastrophically (c.f. Mao, Pol Pot).
So I think engineering human nature, like AI, is a bit of a Pandora's Box.
So long as genetic manipulation is not free, so not practiced by some fraction of the poor, manipulation among higher income humans will only increase the range of human forms. Humans will remain recognizably human to us. Of course humans today will have the appearance of low-status 'unselected' human forms. (So if you're going to be frozen, you might think twice about the signal your low-status-looking body is going to send.)
But I think on the higher ends, we'll get something like the effects of fashion that Chip Morningstar mentions in comments above. Today's parents want their children to be (male) tall, smart, handsome alpha male athletes or (female) tall, smart, beautiful models. But in a world in which all of the (other high-status family) children are like this, then competition pushes parents to want a bit more. If all male children are 6'2", then you want your son to be 6'4"; if all female children are 6', you want yours to be 6'1". Views of beauty will change, and genetic manipulation will allow humans to pursue these views much faster than the current speeds of sexual selection.
I think the result would look humanoid to us, like many of the aliens in the Star Trek TV series, but we might not immediately guess Earth-sourced human if we saw photos of these creatures today.
Why should humans look any different in 500 years time? Most humans and dogs look the same as their ancestors 500 years (albeit a tad fatter). Most changes over the last 50 years has been superficial (except electronics of course). After all, we already (sort of) live in the "future" because we enjoy some things that were once the preserve of science fiction however sexual selection hasn't budged. Women don't want nerdy brainiacs, they still want the badboys. Likewise men don't want brainy nerdettes, they still want the hot babes.
Then again who's to say technology will continue to go up over the centuries. The opposite could happen and everyone in 500 years time could be living the way people 500 years did. Heck, it ought to happen because it's regression to the mean especially as human sexual selection continues to favour traits in both sexes that really only make sense in a low tech world (brawny men and wide-hipped women).
Technological singularity in 500 years? We will just have to think tweets rather than typing them and we will do research on the internet and gain knowledge by mere thought. Internet with full accessibility through thought so to say.
And maybe we will even have reached some form of non-competitive individualism, totally different from the one we have today.
Consider how much clothing styles change over very short time periods. Then consider what happens once our technology enables us to apply the same kind of design plasticity to the human form, which I expect is coming within a generation. After 500 years of that I can easily imagine there being human phenotypes that span the range from people who would be completely recognizable to us today to entities that we would not only not recognize as human but would be unable to categorize at all.
Also, tech will let us change existing people radically. So change in the design of people doesn't have to wait for or funnel through reproduction cycles. That is, design features can have a shorter reproductive cycle than people. In a world where most young adults-- including the parents-- have, say, visible techie mods, will the parents still think that lack of visible techie mods will get their kids better mates? Besides, if the original design doesn't work out, you or the kid can always change it later. Seems to me pressure for conservatism out of caution is reduced.
But while such control might eventually be possible... it surely is not true... It might be trivially true if... Should "possible" be substituted for "true" there, or is there more to it?
Up to that point I think you're saying that we don't have enough control over evolution to hold the future to our present preferences. But in the last paragraph,
Yes with new techs change no longer need follow DNA mutation and crossover, and yes this allows for more rapid change. But variation and selection should continue, no matter where designs are stored.
Here I think it would help to include Tyler's sentence: "Future science is more likely to reverse the boost in variance than to support it," as context.
In general, I think Tyler's main mistake is to focus on the kinds of parents' choices we can foresee, rather than
o the extent to which parents' preferences are functions of trends in the world around them,
o design details whose effects the parents didn't think about,
o especially details with effects on the kids preferences about kids,
o selection effects on the kids after they are born (or produced),
o and as you mention, forgetting compound interest.
Tyler could come back with a "Friendly Genetic Engineering" argument where, once parents and engineers see what might happen otherwise, they will design children who will only ever want children who will only ever want children... much like they want.
But I'm guessing Tyler is making a simpler mistake, probably slipping and treating design as the opposite of evolution at some points in his thinking. Also, as I said, focusing on some evolutionary dynamics and neglecting just how many there are.
p.s. "We can suture the future / shut like a cut we can / replicate structures which replicate us." --Milemarker
It's really hard to make a prediction like this, but the one thing I'm sure of is that whoever or whatever is left in 500 years will want kids. It's no longer enough to want sex, since birth control decouples it from baby-making, and lots of people aren't making babies. 500 years is enough time to largely get rid of them.
Emryonic genetic screening combined with invitro-fertilization will allow parents to select children with very specific genetic profiles from perhaps hundreds of embryos. Falling costs of genetic screening, increasing knowledge of the genome, and rising standard of living in the developing world will make these technologies widely available in probably twenty years. It won't be long before not only individual lineages, but the entire human genome is being actively engineered.
I think this kind of highly directed selection is likely to radically impact the rate of change in the human genome.
"Darwinian evolution" is a rather vague term. Evolution itself will still be in progress by the end of the century - since we won't have reached our limits by then.
In the next 500 hundred years a great change will explode and technology is everywhere. Maybe nobody will bear fetus in mother’s womb because there will be technologies to develop a child outside the uterus or even more than that.
Absolutely. Either very baby hungry, very careless (i.e., accidental pregnancies), or very needful of feeling dominant. Maybe not even the middle one. Essentially, if technological growth continues at all, its hard to think of any biological change (e.g., allergies or resistance to birth control pills) that couldn't be cracked if people wanted to. So the evolutionary changes that will result will be changes in what people want.
Surely by "Darwinian evolution will not be in effect" he meant "the impact of genetic evolution will be severely reduced"And yes, Darwinian evolution could be eliminated if we wanted to. Just ban normal reproduction and make everybody a clone. Read "Brave new world".
I know if I were genetically engineering children I'd probably do my best to make them extremely intelligent, to an extent asocial (in the sense that Aspergers is asocial) and unecessarily physically powerful. While a majority herd may go for super-herd animals, you can't discount the formation of subspecies from people with more Nietzschean tendencies.
So explain to me, why should our preferences for body form change in just 500 years of evolution, just because we are much richer? Selection will emphasize some features we already desire such as height.
Even then, there is no runaway, parents will tinker with the actual height of their offspring than with their desire for even taller mates or offspring. That desire might catch up via old-fashioned evolution, but in 500 years time an 8" man would still look weird, even if 6'2" looks short and sweet.
Only fundamental misunderstanding of evolution allows claims that human Darwinian evolution will somehow be eliminated. Evolution cannot be stopped. The only things that can happen is a change in the adaptive landscape.
I kind of hope human nature (genetic intrusion to solipsist preferences) proves difficult or immensely unattractive to engineer.
I think within the Transhumanist community there is an assumption that hacking will be bottom-up, but history shows a durable urge to top-down engineer, which has generally ended quite catastrophically (c.f. Mao, Pol Pot).
So I think engineering human nature, like AI, is a bit of a Pandora's Box.
So long as genetic manipulation is not free, so not practiced by some fraction of the poor, manipulation among higher income humans will only increase the range of human forms. Humans will remain recognizably human to us. Of course humans today will have the appearance of low-status 'unselected' human forms. (So if you're going to be frozen, you might think twice about the signal your low-status-looking body is going to send.)
But I think on the higher ends, we'll get something like the effects of fashion that Chip Morningstar mentions in comments above. Today's parents want their children to be (male) tall, smart, handsome alpha male athletes or (female) tall, smart, beautiful models. But in a world in which all of the (other high-status family) children are like this, then competition pushes parents to want a bit more. If all male children are 6'2", then you want your son to be 6'4"; if all female children are 6', you want yours to be 6'1". Views of beauty will change, and genetic manipulation will allow humans to pursue these views much faster than the current speeds of sexual selection.
I think the result would look humanoid to us, like many of the aliens in the Star Trek TV series, but we might not immediately guess Earth-sourced human if we saw photos of these creatures today.
Why should humans look any different in 500 years time? Most humans and dogs look the same as their ancestors 500 years (albeit a tad fatter). Most changes over the last 50 years has been superficial (except electronics of course). After all, we already (sort of) live in the "future" because we enjoy some things that were once the preserve of science fiction however sexual selection hasn't budged. Women don't want nerdy brainiacs, they still want the badboys. Likewise men don't want brainy nerdettes, they still want the hot babes.
Then again who's to say technology will continue to go up over the centuries. The opposite could happen and everyone in 500 years time could be living the way people 500 years did. Heck, it ought to happen because it's regression to the mean especially as human sexual selection continues to favour traits in both sexes that really only make sense in a low tech world (brawny men and wide-hipped women).
Technological singularity in 500 years? We will just have to think tweets rather than typing them and we will do research on the internet and gain knowledge by mere thought. Internet with full accessibility through thought so to say.
And maybe we will even have reached some form of non-competitive individualism, totally different from the one we have today.
Consider how much clothing styles change over very short time periods. Then consider what happens once our technology enables us to apply the same kind of design plasticity to the human form, which I expect is coming within a generation. After 500 years of that I can easily imagine there being human phenotypes that span the range from people who would be completely recognizable to us today to entities that we would not only not recognize as human but would be unable to categorize at all.
Also, tech will let us change existing people radically. So change in the design of people doesn't have to wait for or funnel through reproduction cycles. That is, design features can have a shorter reproductive cycle than people. In a world where most young adults-- including the parents-- have, say, visible techie mods, will the parents still think that lack of visible techie mods will get their kids better mates? Besides, if the original design doesn't work out, you or the kid can always change it later. Seems to me pressure for conservatism out of caution is reduced.
Editing issue in the 2nd-last paragraph:
But while such control might eventually be possible... it surely is not true... It might be trivially true if... Should "possible" be substituted for "true" there, or is there more to it?
Up to that point I think you're saying that we don't have enough control over evolution to hold the future to our present preferences. But in the last paragraph,
Yes with new techs change no longer need follow DNA mutation and crossover, and yes this allows for more rapid change. But variation and selection should continue, no matter where designs are stored.
Here I think it would help to include Tyler's sentence: "Future science is more likely to reverse the boost in variance than to support it," as context.
In general, I think Tyler's main mistake is to focus on the kinds of parents' choices we can foresee, rather than
o the extent to which parents' preferences are functions of trends in the world around them,
o design details whose effects the parents didn't think about,
o especially details with effects on the kids preferences about kids,
o selection effects on the kids after they are born (or produced),
o and as you mention, forgetting compound interest.
Tyler could come back with a "Friendly Genetic Engineering" argument where, once parents and engineers see what might happen otherwise, they will design children who will only ever want children who will only ever want children... much like they want.
But I'm guessing Tyler is making a simpler mistake, probably slipping and treating design as the opposite of evolution at some points in his thinking. Also, as I said, focusing on some evolutionary dynamics and neglecting just how many there are.
p.s. "We can suture the future / shut like a cut we can / replicate structures which replicate us." --Milemarker
It's really hard to make a prediction like this, but the one thing I'm sure of is that whoever or whatever is left in 500 years will want kids. It's no longer enough to want sex, since birth control decouples it from baby-making, and lots of people aren't making babies. 500 years is enough time to largely get rid of them.
Emryonic genetic screening combined with invitro-fertilization will allow parents to select children with very specific genetic profiles from perhaps hundreds of embryos. Falling costs of genetic screening, increasing knowledge of the genome, and rising standard of living in the developing world will make these technologies widely available in probably twenty years. It won't be long before not only individual lineages, but the entire human genome is being actively engineered.
I think this kind of highly directed selection is likely to radically impact the rate of change in the human genome.