Recently I’ve said that it seems obvious that our descendants will soon be artificial, as will be any advanced aliens we meet. And I’ve complained of futurists and astrobiologists who ignore this possibility. But I’ve realized that even though I may “know it when I see it”, I don’t have a good definition of “artificial”.
Like Robin, I am pleasantly surprised that this idea has such an ancient and esteemed pedigree. Very cool!
A fancy term for self-made or self-creating is autopoiesis.
One interesting aspect about artificial vs organic or non-self made vs selfmade is that the latter is usually regarded as having an intrinstic and often intangible/difficult to articulate advantage over the former.e.g. the preference for 'organicly grown' food, various forms of environmentalism, small family businesses, respect for primitive societies, etc etc
There are various Hansonian explanations for this phenomenon already but one explanation might be that 'self-made' things are subject to a natural selection while nonselfmade things are not.
In other words self-made things are subject to a mostly hidden optimizing process.This means that 'self-made' things may be more robust in hard to understand ways and would explain the seemingly irrational preference of many folks for self-made and organic things/culture/art/food etc.
One is reminded of Joe Henrich's "Secret of our succes".
Wow, that's great. This is such an old subject I doubt I could be original on it, so the best to hope for is to have converged with high quality others. :)
Your answer is similar to Aristotle's: he defines "natural" as meaning having an internal source of change and rest, whereas artificial things have an external source of change and rest: the craftsman (or, more precisely, the form of the craft object in the soul of the craftsman). So a natural thing is the cause of its own changes, literally making itself: guiding its own processes of growth and reproduction and maintenance.This is the subject of Physics II, here's an excerpt. https://uploads.disquscdn.c...
Those words suggested a direction toward a definition, rather than a final one. "Incorporating intelligent design" seems clearer than "self-made", as all life-like replication involves a notion of the self that gets replicated.
But maybe the real challenge is in delineating the entities from their reproduction infrastructure, because there wouldn't be much debate whether e.g. the colony ship itself is artificial. The question is to what extent the inhabitants of the ship are artificial.
And again, my intuition is that the design history of the entities would be the crucial factor. Such a debate is playing out right now in real time on Twitter. #LabLeak proponents are arguing that SARS-CoV-2 shows evidence of either design or at least directed evolution. Opponents say that this virus could have evolved completely naturally. Both could be right! Thus in a weird new form of convergent evolution, identical viruses could be both completely natural and at least partly artificial, even though functionally identical in terms of how they "make themselves".
So the path dependence of how much intelligent design was historically involved seems critical. And thanks to creationists, the concept of intelligent design has been extensively discussed. For example, it surely has been noted that, while peahens in some sense "designed" peacock feathers, their design did not rely on intelligence.
"Why not just say that life is artificial to the extent that the replicating entity itself embodies intentional artifice / skill / design / technology" those last few words seem to me less clear than is "artificial". Definitions need to define vague words in terms of clearer ones.
The normal scifi scenario with lots of little bots is that they are capable of making copies of themselves rather than leaving that to a separate factory bot. That way a single one reaching a planet full of resources can make lots of copies of itself by itself.
A definition I've used for life: functional organization for sustaining self via active use of energy and sustaining kind via information replication. I apply this to a taxonomy of replicators (fire, viruses, memes, mules, symbionts, parasites, gametes, etc.) at http://humanknowledge.net/T...
But I don't think "self" is the key for distinguishing artificial life. Why not just say that life is artificial to the extent that the replicating entity itself embodies intentional artifice / skill / design / technology?
The goal here of course is to capture the difference between 1) biological organisms in colony ships and 2) ems/robots in such ships (or that ARE such ships). The distinction should still apply even if the biological organisms have become obligate mutualists with their colony ships, such that neither could replicate without the other.
P.S. Saying they "ignore this possibility" is polite, as it seems more like an inevitability.
Also, many of our descendents will want to embrace differing levels of artificiality. And some will probably prefer to remain baseline humans. On the other hand, it's fair to assume a higher chance of artificiality enhancement among the people who travel far from Earth and may meet other species, if only because of practicality.
I suspect one should ask "how artificial" rather than "whether or not artificial". To an early human (prehistoric or early historic), modern society would be distinctly alien, and it would be very reasonable to consider us "artificial" - much of our experiences are crafted or designed, rather than occurring without human guidance. If a Roman considered a Victorian lifestyle, much of that would still be true, though to a smaller level of incomprehension.
Like Robin, I am pleasantly surprised that this idea has such an ancient and esteemed pedigree. Very cool!
A fancy term for self-made or self-creating is autopoiesis.
One interesting aspect about artificial vs organic or non-self made vs selfmade is that the latter is usually regarded as having an intrinstic and often intangible/difficult to articulate advantage over the former.e.g. the preference for 'organicly grown' food, various forms of environmentalism, small family businesses, respect for primitive societies, etc etc
There are various Hansonian explanations for this phenomenon already but one explanation might be that 'self-made' things are subject to a natural selection while nonselfmade things are not.
In other words self-made things are subject to a mostly hidden optimizing process.This means that 'self-made' things may be more robust in hard to understand ways and would explain the seemingly irrational preference of many folks for self-made and organic things/culture/art/food etc.
One is reminded of Joe Henrich's "Secret of our succes".
Wow, that's great. This is such an old subject I doubt I could be original on it, so the best to hope for is to have converged with high quality others. :)
Your answer is similar to Aristotle's: he defines "natural" as meaning having an internal source of change and rest, whereas artificial things have an external source of change and rest: the craftsman (or, more precisely, the form of the craft object in the soul of the craftsman). So a natural thing is the cause of its own changes, literally making itself: guiding its own processes of growth and reproduction and maintenance.This is the subject of Physics II, here's an excerpt. https://uploads.disquscdn.c...
Those words suggested a direction toward a definition, rather than a final one. "Incorporating intelligent design" seems clearer than "self-made", as all life-like replication involves a notion of the self that gets replicated.
But maybe the real challenge is in delineating the entities from their reproduction infrastructure, because there wouldn't be much debate whether e.g. the colony ship itself is artificial. The question is to what extent the inhabitants of the ship are artificial.
And again, my intuition is that the design history of the entities would be the crucial factor. Such a debate is playing out right now in real time on Twitter. #LabLeak proponents are arguing that SARS-CoV-2 shows evidence of either design or at least directed evolution. Opponents say that this virus could have evolved completely naturally. Both could be right! Thus in a weird new form of convergent evolution, identical viruses could be both completely natural and at least partly artificial, even though functionally identical in terms of how they "make themselves".
So the path dependence of how much intelligent design was historically involved seems critical. And thanks to creationists, the concept of intelligent design has been extensively discussed. For example, it surely has been noted that, while peahens in some sense "designed" peacock feathers, their design did not rely on intelligence.
Yes, on short time scales.
A bot made by a bot of similar form would count as "self-made" and thus not "artificial" in your definition, right?
These words of mind clearly signal I'm not thinking of a binary concept: "Thus how 'self made' something is depends"
"Why not just say that life is artificial to the extent that the replicating entity itself embodies intentional artifice / skill / design / technology" those last few words seem to me less clear than is "artificial". Definitions need to define vague words in terms of clearer ones.
Yeah but such sci-fi authors just haven't thought this through.
The normal scifi scenario with lots of little bots is that they are capable of making copies of themselves rather than leaving that to a separate factory bot. That way a single one reaching a planet full of resources can make lots of copies of itself by itself.
A definition I've used for life: functional organization for sustaining self via active use of energy and sustaining kind via information replication. I apply this to a taxonomy of replicators (fire, viruses, memes, mules, symbionts, parasites, gametes, etc.) at http://humanknowledge.net/T...
But I don't think "self" is the key for distinguishing artificial life. Why not just say that life is artificial to the extent that the replicating entity itself embodies intentional artifice / skill / design / technology?
The goal here of course is to capture the difference between 1) biological organisms in colony ships and 2) ems/robots in such ships (or that ARE such ships). The distinction should still apply even if the biological organisms have become obligate mutualists with their colony ships, such that neither could replicate without the other.
P.S. Saying they "ignore this possibility" is polite, as it seems more like an inevitability.
Also, many of our descendents will want to embrace differing levels of artificiality. And some will probably prefer to remain baseline humans. On the other hand, it's fair to assume a higher chance of artificiality enhancement among the people who travel far from Earth and may meet other species, if only because of practicality.
I suspect one should ask "how artificial" rather than "whether or not artificial". To an early human (prehistoric or early historic), modern society would be distinctly alien, and it would be very reasonable to consider us "artificial" - much of our experiences are crafted or designed, rather than occurring without human guidance. If a Roman considered a Victorian lifestyle, much of that would still be true, though to a smaller level of incomprehension.