For all our years of toil in machine learning research, we still only have one really usable model of intelligence—the mammalian brain. At first glance, it makes sense that a really complicated multicellular organism would want a control centre that can function faster than turning on and off genes or transmitting hormones, and we have examples of nervous systems (such as the one in the nematode Caenorhabditis elegans) that do nothing other than steer the creature semi-randomly towards possible food sources. So how the heck did we end up with the wheel, wars, New York, and so on?
In the last part of this series, we looked at what it takes to create reproductive life, and a key real-world example of borderline living phenomena, the transposon. If you haven't read that part, now's a great time, since this part builds on it.
We generally regard transposon activity as harmful. Left unchecked, the spontaneous rearrangement of pieces of the genome into not-quite-completely predictable configurations has great potential to cause harm to an organism. A great many animals, both vertebrate and invertebrate, seem to be aware of this, as we have special genetic elements called piRNAs and miRNAs, many of which serve to suppress transposon activity. These are particularly active in the gonads (ovaries and testes), when unexpected genome changes have a high chance of affecting the delicate meiosis process (sperm and ovum production.) The story would be nice and neat if that was all it was; a host and its parasite in an eternal arms race.
But it's not that simple.
For a couple of years now, it has been clear that retrotransposons, a subtype of transposons, are actually important for human brain function. While the electric brain is busy transmitting signals back and forth in an adaptive manner which we are certain is powerful enough to approximate any mathematical relationship, this retrotransposon activity is providing something more. What it does, exactly, we can't say yet, but exactly what it does in humans has a very high chance of being unique, as the regions of the genome occupied by these transposons, and their sequences themselves, differ greatly between species.
We do know that this activity affects how proteins are regulated, however; proteins that are relatively unchanging across the animals. Even between us and our closest surviving relative, the chimpanzee, the greatest obvious difference is in regions of non-coding RNA, such as in HAR1. In computing terms, it is as if we have a very elaborate set of shell scripts, batch files, or compiler macros busy at work. It seems that as far as evolution is concerned, the key step to successful attainment of a higher intelligence is less about what building blocks are available and more about how the organism has adapted to use them. Proteins perhaps half a billion years old have contributed toward our rare (if not unique) intellectual development in perhaps a hundredth of that time, if not even more quickly.
So what the heck were we competing against? Why were humans driven so hard to develop this extraordinary intelligence, when there are plenty of other animals in Africa who are comfortably at the top of their respective food chains?
The answer is that genetic evolution is nearly obsolete. We have gained the ability to evolve something much more dynamic and adaptive than random genetic mutation can provide: culture.
Culture, in the evolutionary sense, is a hugely broad concept that encompasses all learned behaviour passed from generation to generation. It is something that is unique to animals, but can be found in a wide variety of vertebrates, from the intricate calls of parrots to how chimpanzees raise and care for their children. In all such cases, the animal is disadvantaged by being deprived of these learning opportunities, and indeed such behaviours can even direct genetic evolution if they induce a change in the environment.
Most likely, the majority of the pressure to develop our intelligence to an arms race between groups of humans as a result of these cultural adaptations. This seems like an easy thing to swallow—one group develops fire, another group figures out how to get it in order to fight back, eventually the group with the most intelligence is able to attain the most, so the cycle keeps going. It does, however, leave a couple of loose ends.
First, why the heck are we so smart? If evolution is guided by necessity and the human genome has barely changed in the past few thousand years, why do we have the capacity for developing such an advanced civilization? Why can we learn advanced mathematics when such complex abstract reasoning skills do not appear to have been applicable to our environments in the past?
And second, why do we see other species being able to do the same? Chimpanzees can carve spears. Gorillas can learn sign language. These are faculties that appear initially discontiguous with their previous needs.
The answer to both questions is that we need more than capacity for learning: we need speed, too. If every tribe has the intelligence necessary to develop a weapon, then the tribe who can figure it out first—or reverse-engineer it first—will have the upper hand. For perhaps millions of years, humans lived far below their intellectual capacities, because our intelligence was only being used to solve day-to-day problems in a more rapid fashion. That's what survival takes.
Now, this being said, another factor that massively increases the development of the mind is socialization: when an organism (usually a complex animal) lives in the vicinity of a group for long enough, they pool resources to improve overall survival, and hence must communicate their needs—and the group's needs—to each other. This prompted the creation (or at least expansion) of many psychological concepts, which authors should take note of: completely solitary animals have no need for language (growls will do), the ability to judge self-worth (there is no one around to replace them), a conscience (there is no society to judge them), discrimination (no social roles or tribal purity to defend), or love other than between parent and child (lust is sufficient for mating purposes in sexual species.)
They will understand beauty, however: it is critical to discerning the rightness of food and potential mates. Beauty is an interesting phenomenon because it is more or less our sense of 'normalness'; averaging out many photographs of human faces, for example, removes the exceptional unique features of each face and produces what most would call a perfect middle. Developed societies tend to be less accepting of dramatic physical shifts in appearance, which they will almost inevitably reject as imperfect.
All of this means that highly developed species will tend to retain the physical forms they had when they attained a functioning civilization. It may be gaunt and weary, it may be overfed, and it may be slightly stylized according to some quirk that was in fashion once a thousand years ago, but it will be the same bag of meat and bones on the outside. Classic portrayals of human evolution into cheesy 60s sci-fi aliens are resoundingly unlikely; in half a million years, all we have managed to do is eliminate the sloped brow of the Neanderthal. If anything is likely, it is that humans will get taller to enable higher physical brain capacity, not big-headed, as this is already a natural selective pressure at work on the population. (But, mostly, height is linked to the availability of nutrients, so keep in mind that what is average under natural conditions may not be anywhere near the species' civilized peak.)
And that, more or less, is what it takes to create a people from scratch. Stay tuned for more Syngenesis.
In the last part of this series, we looked at what it takes to create reproductive life, and a key real-world example of borderline living phenomena, the transposon. If you haven't read that part, now's a great time, since this part builds on it.
The brain is probably more complicated than our current models permit.
We generally regard transposon activity as harmful. Left unchecked, the spontaneous rearrangement of pieces of the genome into not-quite-completely predictable configurations has great potential to cause harm to an organism. A great many animals, both vertebrate and invertebrate, seem to be aware of this, as we have special genetic elements called piRNAs and miRNAs, many of which serve to suppress transposon activity. These are particularly active in the gonads (ovaries and testes), when unexpected genome changes have a high chance of affecting the delicate meiosis process (sperm and ovum production.) The story would be nice and neat if that was all it was; a host and its parasite in an eternal arms race.
But it's not that simple.
For a couple of years now, it has been clear that retrotransposons, a subtype of transposons, are actually important for human brain function. While the electric brain is busy transmitting signals back and forth in an adaptive manner which we are certain is powerful enough to approximate any mathematical relationship, this retrotransposon activity is providing something more. What it does, exactly, we can't say yet, but exactly what it does in humans has a very high chance of being unique, as the regions of the genome occupied by these transposons, and their sequences themselves, differ greatly between species.
We do know that this activity affects how proteins are regulated, however; proteins that are relatively unchanging across the animals. Even between us and our closest surviving relative, the chimpanzee, the greatest obvious difference is in regions of non-coding RNA, such as in HAR1. In computing terms, it is as if we have a very elaborate set of shell scripts, batch files, or compiler macros busy at work. It seems that as far as evolution is concerned, the key step to successful attainment of a higher intelligence is less about what building blocks are available and more about how the organism has adapted to use them. Proteins perhaps half a billion years old have contributed toward our rare (if not unique) intellectual development in perhaps a hundredth of that time, if not even more quickly.
So what the heck were we competing against? Why were humans driven so hard to develop this extraordinary intelligence, when there are plenty of other animals in Africa who are comfortably at the top of their respective food chains?
The answer is that genetic evolution is nearly obsolete. We have gained the ability to evolve something much more dynamic and adaptive than random genetic mutation can provide: culture.
Culture, in the evolutionary sense, is a hugely broad concept that encompasses all learned behaviour passed from generation to generation. It is something that is unique to animals, but can be found in a wide variety of vertebrates, from the intricate calls of parrots to how chimpanzees raise and care for their children. In all such cases, the animal is disadvantaged by being deprived of these learning opportunities, and indeed such behaviours can even direct genetic evolution if they induce a change in the environment.
Most likely, the majority of the pressure to develop our intelligence to an arms race between groups of humans as a result of these cultural adaptations. This seems like an easy thing to swallow—one group develops fire, another group figures out how to get it in order to fight back, eventually the group with the most intelligence is able to attain the most, so the cycle keeps going. It does, however, leave a couple of loose ends.
First, why the heck are we so smart? If evolution is guided by necessity and the human genome has barely changed in the past few thousand years, why do we have the capacity for developing such an advanced civilization? Why can we learn advanced mathematics when such complex abstract reasoning skills do not appear to have been applicable to our environments in the past?
And second, why do we see other species being able to do the same? Chimpanzees can carve spears. Gorillas can learn sign language. These are faculties that appear initially discontiguous with their previous needs.
The answer to both questions is that we need more than capacity for learning: we need speed, too. If every tribe has the intelligence necessary to develop a weapon, then the tribe who can figure it out first—or reverse-engineer it first—will have the upper hand. For perhaps millions of years, humans lived far below their intellectual capacities, because our intelligence was only being used to solve day-to-day problems in a more rapid fashion. That's what survival takes.
Now, this being said, another factor that massively increases the development of the mind is socialization: when an organism (usually a complex animal) lives in the vicinity of a group for long enough, they pool resources to improve overall survival, and hence must communicate their needs—and the group's needs—to each other. This prompted the creation (or at least expansion) of many psychological concepts, which authors should take note of: completely solitary animals have no need for language (growls will do), the ability to judge self-worth (there is no one around to replace them), a conscience (there is no society to judge them), discrimination (no social roles or tribal purity to defend), or love other than between parent and child (lust is sufficient for mating purposes in sexual species.)
They will understand beauty, however: it is critical to discerning the rightness of food and potential mates. Beauty is an interesting phenomenon because it is more or less our sense of 'normalness'; averaging out many photographs of human faces, for example, removes the exceptional unique features of each face and produces what most would call a perfect middle. Developed societies tend to be less accepting of dramatic physical shifts in appearance, which they will almost inevitably reject as imperfect.
All of this means that highly developed species will tend to retain the physical forms they had when they attained a functioning civilization. It may be gaunt and weary, it may be overfed, and it may be slightly stylized according to some quirk that was in fashion once a thousand years ago, but it will be the same bag of meat and bones on the outside. Classic portrayals of human evolution into cheesy 60s sci-fi aliens are resoundingly unlikely; in half a million years, all we have managed to do is eliminate the sloped brow of the Neanderthal. If anything is likely, it is that humans will get taller to enable higher physical brain capacity, not big-headed, as this is already a natural selective pressure at work on the population. (But, mostly, height is linked to the availability of nutrients, so keep in mind that what is average under natural conditions may not be anywhere near the species' civilized peak.)
And that, more or less, is what it takes to create a people from scratch. Stay tuned for more Syngenesis.
