7 ms·

If it's entirely determined by physical stuff thats a contradiction. It may be very complicated, but its still physical.

by foxes 8y ago

bordercases 8y ago
No, because the reduction in-principle would fail if the system's description language could not support explaining the phenomena in the higher level. And going the other way, it could be impossible in-principle to conduct the experiments that get a complete description because the tools required break how it was originally organized. It's also not just about adding more variables because "it's complicated". Biology for example seems to be highly reflexive, that sort of behaviour can't really be accounted for in basic causal models. At best we have models which resemble fluid dynamics, and fluid dynamics is only as good as it is because we don't reduce the phenomena down to every molecule.
- ssadler 8y ago
  See also: https://www.scottaaronson.com/blog/?p=3294 https://www.scottaaronson.com/blog/?p=3294
- sooheon 8y ago
  And yet nobody claims that fluids are supernatural and above the laws of physics. It's our ability to understand, explain, and model it that is deficient, not the phenomena that is supernatural.
  - bordercases 8y ago
    Maxwell's Demon might have a thing or two to say to you. The fallacy is to believe that reduction is a total property of the territory, rather than a property of some of the territory that lends to a certain kind of map. We know that there are phenomena in the world with strong boundaries and there are phenomena with weaker boundaries. It may be that at low scales things can be easily separated, but that doesn't have to translate to how easy it is to tease-apart macro-phenomena. Heck below atoms, which lend themselves most strongly to the reductionist intuition, we have quantum waves. The wave-particle duality should at least simultaneously give reduction and non-reduction some credence.
    - sooheon 8y ago
      I won't argue that some phenomena may be resistant to reduction beyond human capability to reduce. My main problem with the thesis is that whatever irreducible phenomena that is, is still part of the natural, physical world. Because we lack the knowledge to fully map thoughts to underlying atomic events does not mean thoughts are suddenly supernatural and non-physical. They are a different category of physical event, perhaps. Maxwell's Demon is new to me, and I'm not quite sure how to interpret it in this context. At first read, the claim that the observing, calculating, acting Demon, manipulating a physical trapdoor is not part of the thermodynamic system seems... laughable.
      - bordercases 8y ago
        > Maxwell's Demon is new to me, and I'm not quite sure how to interpret it in this context. At first read, the claim that the observing, calculating, acting Demon, manipulating a physical trapdoor is not part of the thermodynamic system seems... laughable. That's the immediate refutation a lot of people propose, but then to model that demon plus the system is to also become another demon. So you can't escape this problem as long as you are not an abstract ideal observer, which none of us are in practice. (The fact that it's laughable is correct, because it could never happen, including to us.)
      - bordercases 8y ago
        > I won't argue that some phenomena may be resistant to reduction beyond human capability to reduce. My main problem with the thesis is that whatever irreducible phenomena that is, is still part of the natural, physical world. You don't have to posit the supernatural. People conflate believing in a single objective reality with having to believe in strong materialism (or worse, atomic materialism, when QM goes deeper), when all you have to do is believe that reality is consistent. You might not be able to create a total consistent model of reality, but if reality is consistent then you don't have to propose extra-physical forces due to conservation of energy or other such constraints. You can exclude contradictory hypotheses even if you can't settle on the "actual" one, even in-principle. In which case you can have models in one domain of science, and models in another domain, which may not connect with each other, but they don't have to contradict each other either, and you can do this all while not believing in the supernatural or denying an objective reality.
  - bordercases 8y ago
    To see why you can't confuse reductionism in-principle with reductionism in-practice, you could read this paper: "More is Different: Broken symmetry and the hierarchical nature of science" http://robotics.cs.tamu.edu/dshell/cs689/papers/anderson72more_is_different.pdf http://robotics.cs.tamu.edu/dshell/cs689/papers/anderson72mo... Science divides into different disciplines precisely because the units and laws of one model do not close over the phenomena of another.
    - sooheon 8y ago
      Science divides into disciplines because human compute power and lifespans are limited. Anderson is right to point out that reduction is easier than construction. I deny him the implication that construction is fundamentally impossible, only that it is out of reach of current human knowledge. Nothing prevents a Demon from starting with first principles of natural laws and matter, sitting and pondering for infinity, and writing a treatise on any number of emergent phenomena, say: realpolitik between countries made up of parties made up of people made up of organic matter made up of atoms made up of... We know this must be possible because it must have happened at least once, in the mind of the demon we call reality. I get the distinction you're making between in-principle and in-practice, but this is an accident of our capabilities. Any gaps between the sciences is due to our ignorance. Sadly it seems human scientists desire to claim their corner of science is strong, independent and just as "fundamental" as any other's, and like playing up these gaps in our knowledge. I don't understand this desire to draw permanent boundaries based on arbitrary human limitations, and so quickly declare non-overlapping magesteria.
      - bordercases 8y ago
        Nothing of what I said implied that the boundaries should always be expected to remain permanent for any typical pair of disciplines (although there will be disciplines where that holds). There are instances where boundaries have been productively crossed and fields merged. And it's certainly more productive to try to get simpler explanations using past hypotheses between disparate phenomena – as long as no information is lost. But that doesn't remove the limitations, and the only desire here is to understand and know these limitations. That's as scientific an endeavour as any. If it's a matter of the framing that makes you sombre, consider the alternative where being a finite observer with the capacity for self-reflection such as ourselves is put in a situation where we can navigate alternative descriptions of reality, never having to terminate with a final explanation, yet always growing in personal understanding... this is what your Demon would actually be doing, for if it were as powerful as you imply, it would not be a scientist at all. It would simply be reality, with these observers the lesser of it.
  - goatlover 8y ago
    Why mention the supernatural? The question is whether all of nature is reducible to physics, or whether there are strongly emergent/irreducible phenomenon. It's all nature, it's just a question of ontology. It's always logically possible that the world has non-physical properties. Physicalism is an assumption.
    - simonh 8y ago
      To my mind what matters at a physical level is something I’ll call causal significance. Is the property in question a cause of anything physical such as a behaviour? If so, then that property must have a physical basis in reality even if it is simply a physical encoding in cells or chemicals. A for loop in a computer program is a property. It causes a behaviour. It can only do that because it is encoded in the materials of the computer. The question is, are there any properties a biological system can have that are not encoded in its physical constituents? Anyone answering ‘yes’ to that question really needs to provide a very clear answer as to how they think that might be possible and how else that property might arise.
      - bordercases 8y ago
        I would be interested in seeing if two neural network models with equivalent topology and weights could always be said to 'have' the same subject matter. (It's a better example than for-loops, because I think both brains and computers can embody what it means to have a neuron in a close-enough fashion). I would expect the answer would be "no". You would have to know about the environment that trained them, which blows the door open for something which is irreducible without knowing about environmental conditions, and biology having too rich of an environment to be as precisely reducible as computers might ever be without them being also a computer system. In my other posts in this thread I emphasize the "in-principle" and "in-practice" distinction because I don't believe it is the "in-principle" part of reductionism that ought to be at stake. It's if we can actually do it in a way that doesn't invite approximations.
      - goatlover 8y ago
        > The question is, are there any properties a biological system can have that are not encoded in its physical constituents? Anyone answering ‘yes’ to that question really needs to provide a very clear answer as to how they think that might be possible and how else that property might arise. I believe the answer is that there are properties which only come into existence once you have the right physical arrangement, but these properties can't be predicted in advance from the physical constituents. As to how it could be possible that truly novel properties come into existence, that's not really different than how it can be possible that QM can have indeterminate properties, or anything that's physically fundament for which we have no explanation. Something happens to exist, and we don't know whether there is answer to why anything exists. There's also other possibilities, like the existence of abstract and universal entities, such as numbers or universals. How is it that every electron has the exact same charge and mass? Is electron a universal? Why does math work so well for physics? Are there laws of nature that govern the world (make it be a certain way as opposed to just being regularities)? That sort of thing.

If it's entirely determined by physical stuff thats a contradiction. It may be very complicated, but its still physical.

by foxes 8y ago

If it's entirely determined by physical stuff thats a contradiction. It may be very complicated, but its still physical.

bordercases 8y ago

No, because the reduction in-principle would fail if the system's description language could not support explaining the phenomena in the higher level. And going the other way, it could be impossible in-principle to conduct the experiments that get a complete description because the tools required break how it was originally organized. It's also not just about adding more variables because "it's complicated". Biology for example seems to be highly reflexive, that sort of behaviour can't really be accounted for in basic causal models. At best we have models which resemble fluid dynamics, and fluid dynamics is only as good as it is because we don't reduce the phenomena down to every molecule.

ssadler 8y ago
See also: https://www.scottaaronson.com/blog/?p=3294 https://www.scottaaronson.com/blog/?p=3294
sooheon 8y ago
And yet nobody claims that fluids are supernatural and above the laws of physics. It's our ability to understand, explain, and model it that is deficient, not the phenomena that is supernatural.
- bordercases 8y ago
  Maxwell's Demon might have a thing or two to say to you. The fallacy is to believe that reduction is a total property of the territory, rather than a property of some of the territory that lends to a certain kind of map. We know that there are phenomena in the world with strong boundaries and there are phenomena with weaker boundaries. It may be that at low scales things can be easily separated, but that doesn't have to translate to how easy it is to tease-apart macro-phenomena. Heck below atoms, which lend themselves most strongly to the reductionist intuition, we have quantum waves. The wave-particle duality should at least simultaneously give reduction and non-reduction some credence.
  - sooheon 8y ago
    I won't argue that some phenomena may be resistant to reduction beyond human capability to reduce. My main problem with the thesis is that whatever irreducible phenomena that is, is still part of the natural, physical world. Because we lack the knowledge to fully map thoughts to underlying atomic events does not mean thoughts are suddenly supernatural and non-physical. They are a different category of physical event, perhaps. Maxwell's Demon is new to me, and I'm not quite sure how to interpret it in this context. At first read, the claim that the observing, calculating, acting Demon, manipulating a physical trapdoor is not part of the thermodynamic system seems... laughable.
    - bordercases 8y ago
      > Maxwell's Demon is new to me, and I'm not quite sure how to interpret it in this context. At first read, the claim that the observing, calculating, acting Demon, manipulating a physical trapdoor is not part of the thermodynamic system seems... laughable. That's the immediate refutation a lot of people propose, but then to model that demon plus the system is to also become another demon. So you can't escape this problem as long as you are not an abstract ideal observer, which none of us are in practice. (The fact that it's laughable is correct, because it could never happen, including to us.)
    - bordercases 8y ago
      > I won't argue that some phenomena may be resistant to reduction beyond human capability to reduce. My main problem with the thesis is that whatever irreducible phenomena that is, is still part of the natural, physical world. You don't have to posit the supernatural. People conflate believing in a single objective reality with having to believe in strong materialism (or worse, atomic materialism, when QM goes deeper), when all you have to do is believe that reality is consistent. You might not be able to create a total consistent model of reality, but if reality is consistent then you don't have to propose extra-physical forces due to conservation of energy or other such constraints. You can exclude contradictory hypotheses even if you can't settle on the "actual" one, even in-principle. In which case you can have models in one domain of science, and models in another domain, which may not connect with each other, but they don't have to contradict each other either, and you can do this all while not believing in the supernatural or denying an objective reality.
- bordercases 8y ago
  To see why you can't confuse reductionism in-principle with reductionism in-practice, you could read this paper: "More is Different: Broken symmetry and the hierarchical nature of science" http://robotics.cs.tamu.edu/dshell/cs689/papers/anderson72more_is_different.pdf http://robotics.cs.tamu.edu/dshell/cs689/papers/anderson72mo... Science divides into different disciplines precisely because the units and laws of one model do not close over the phenomena of another.
  - sooheon 8y ago
    Science divides into disciplines because human compute power and lifespans are limited. Anderson is right to point out that reduction is easier than construction. I deny him the implication that construction is fundamentally impossible, only that it is out of reach of current human knowledge. Nothing prevents a Demon from starting with first principles of natural laws and matter, sitting and pondering for infinity, and writing a treatise on any number of emergent phenomena, say: realpolitik between countries made up of parties made up of people made up of organic matter made up of atoms made up of... We know this must be possible because it must have happened at least once, in the mind of the demon we call reality. I get the distinction you're making between in-principle and in-practice, but this is an accident of our capabilities. Any gaps between the sciences is due to our ignorance. Sadly it seems human scientists desire to claim their corner of science is strong, independent and just as "fundamental" as any other's, and like playing up these gaps in our knowledge. I don't understand this desire to draw permanent boundaries based on arbitrary human limitations, and so quickly declare non-overlapping magesteria.
    - bordercases 8y ago
      Nothing of what I said implied that the boundaries should always be expected to remain permanent for any typical pair of disciplines (although there will be disciplines where that holds). There are instances where boundaries have been productively crossed and fields merged. And it's certainly more productive to try to get simpler explanations using past hypotheses between disparate phenomena – as long as no information is lost. But that doesn't remove the limitations, and the only desire here is to understand and know these limitations. That's as scientific an endeavour as any. If it's a matter of the framing that makes you sombre, consider the alternative where being a finite observer with the capacity for self-reflection such as ourselves is put in a situation where we can navigate alternative descriptions of reality, never having to terminate with a final explanation, yet always growing in personal understanding... this is what your Demon would actually be doing, for if it were as powerful as you imply, it would not be a scientist at all. It would simply be reality, with these observers the lesser of it.
- goatlover 8y ago
  Why mention the supernatural? The question is whether all of nature is reducible to physics, or whether there are strongly emergent/irreducible phenomenon. It's all nature, it's just a question of ontology. It's always logically possible that the world has non-physical properties. Physicalism is an assumption.
  - simonh 8y ago
    To my mind what matters at a physical level is something I’ll call causal significance. Is the property in question a cause of anything physical such as a behaviour? If so, then that property must have a physical basis in reality even if it is simply a physical encoding in cells or chemicals. A for loop in a computer program is a property. It causes a behaviour. It can only do that because it is encoded in the materials of the computer. The question is, are there any properties a biological system can have that are not encoded in its physical constituents? Anyone answering ‘yes’ to that question really needs to provide a very clear answer as to how they think that might be possible and how else that property might arise.
    - bordercases 8y ago
      I would be interested in seeing if two neural network models with equivalent topology and weights could always be said to 'have' the same subject matter. (It's a better example than for-loops, because I think both brains and computers can embody what it means to have a neuron in a close-enough fashion). I would expect the answer would be "no". You would have to know about the environment that trained them, which blows the door open for something which is irreducible without knowing about environmental conditions, and biology having too rich of an environment to be as precisely reducible as computers might ever be without them being also a computer system. In my other posts in this thread I emphasize the "in-principle" and "in-practice" distinction because I don't believe it is the "in-principle" part of reductionism that ought to be at stake. It's if we can actually do it in a way that doesn't invite approximations.
    - goatlover 8y ago
      > The question is, are there any properties a biological system can have that are not encoded in its physical constituents? Anyone answering ‘yes’ to that question really needs to provide a very clear answer as to how they think that might be possible and how else that property might arise. I believe the answer is that there are properties which only come into existence once you have the right physical arrangement, but these properties can't be predicted in advance from the physical constituents. As to how it could be possible that truly novel properties come into existence, that's not really different than how it can be possible that QM can have indeterminate properties, or anything that's physically fundament for which we have no explanation. Something happens to exist, and we don't know whether there is answer to why anything exists. There's also other possibilities, like the existence of abstract and universal entities, such as numbers or universals. How is it that every electron has the exact same charge and mass? Is electron a universal? Why does math work so well for physics? Are there laws of nature that govern the world (make it be a certain way as opposed to just being regularities)? That sort of thing.