For some reason Nature News was inspired to tweet about my old blog post on Seriously, The Laws of Physics Underlying Everyday Life Are Completely Understood. Which I mentioned on Facebook, which led to an interesting comment, which I then mentioned on Google+… but now it’s substantive enough that I feel like I am neglecting our loyal blog readers! So here is a copy of my G+ comment, and a lament that I suck at proper use of the internet.
Not sure what brought this back to life. Like the Lord of the Rings, this is part of a trilogy; don’t miss the first installment, or the exciting conclusion.
As Michael Salem points out on an alternative social-media site (rhymes with “lacebook”), some of the resistance to this really quite unobjectionable claim comes from a lack of familiarity with the idea of a “range of validity” for a theory. We tend to think of scientific theories as “right” or “wrong,” which is hardly surprising. But not correct! Theories can be “right” within a certain regime, and useless outside that regime. Newtonian gravity is perfectly good if you want to fly a rocket to the Moon. But you need to toss it out and use general relativity (which has a wider range of validity) if you want to talk about black holes. And you have to toss out GR and use quantum gravity if you want to talk about the birth of the universe.
Just because there is something we don’t understand about some phenomenon (superconductivity, cancer, consciousness) does not imply that everything we think we know might be wrong. Sometimes we can say with confidence that certain things are known, even when other things are not.
Not only do theories have ranges of validity, but in some cases (as with the Standard Model of particle physics) we know what the range is. Or at least, we know where we have tested the theory and where we can be confident it is valid. The Standard Model is valid for all the particles and interactions that constitute our everyday existence.
Today we think of ourselves and the stuff we see around us as made of electrons, protons, and neutrons, interacting through gravity, electromagnetism, and the nuclear forces. A thousand years from now, we will still think precisely that. Unless we destroy the planet, or are uploaded into computers and decide that the laws of physics outside the Matrix aren’t that interesting any more.
Julien, it’s a moral and political question, and as such doesn’t have any “right” answer. Should we provide citizenship to a smart AI? To my big gray parrot (looks smarter than a dog and speaks!) ? How about 12-year olds? Clinical imbeciles? Criminals in jail?
I don’t know … I think when they appear, smart computer programs, the world will change big time.
I think the smart AI will not need our concept of “citizenship”. Besides, how can you infringe on its “rights” ? Good to know what are those in the first place 🙂 Take the right to exist: we have hard time dealing with stupid computer viruses! AI will have no problem making and storing as much copies of itself as it fancies.
@ Physicalists
1. The idea that our mind is a computation.
2. Physicalism means two physically identical systems share all the same properties. Supervenience means that some properties are the sames despite not physically identical. Do you agree with that?
3. “Maybe” more than “obviously”. To me this is an experimental question for when we will be able to test it.
4. If we accept supervenience, which is part of the standard physicalist point-of-view, then we have to decide what system are conscious, as we know we are. But at this point there is just no experiment we can do to test it. This why this is hard. Maybe one day we will understand this better, and provide a description of consciousness as an emergent properties. I would bet on that. Or maybe this will requiere to include some modification of QM, as Penrose think. I don’t know, but the lack of experimental ways to test for consciousness -even in principle- makes this problem fondamentally different from a simple lack of knowledge.
@Loki
Sorry, I did meant to ask for a legal issue. My question should have been: would you consider human any program winning a Turing test, or just those that are constructed to emulate humans both from the point-of-view of behavior and inner mecanisms?
@ Julien,
1. But there’s no reason that accepting Effective Field Theories would imply that mental processes are nothing but computational processes. As I said above, one could accept that life and the mind are nothing but QED processes and still hold that something that is computationally equivalent (e.g., a computer simulation) is not mentally equivalent. This is because the QED descriptions of computers differs from the QED description of our brain.
2. It seems like your account of supervenience is a bit fuzzy.
(a) To say that some higher-level properties supervene on lower-level properties is just to say that the higher level properties cannot be different unless the lower level properties are different. Equivalently, if the lower level properties are the *same*, then the higher level properties have to be the same too.
(b) Multiple realization is the claim that higher level properties can be the same even if the lower level properties are different.
(c) The standard account of physicalism allows for multiple realization by saying that all higher level properties supervene on physical properties (it does not require that the higher level properties be *identical* to the lower level properties).
3. Well, if the property dualists are right and consciousness has no physical effects whatsoever, then it’s going to be pretty hard to test. Fortunately, we have no reason to believe that they are right, and we can safely say that conscious states just are the brain processes that result in my saying, “There’s a tickling feeling in my left knee.”
4. We do face a problem in that we don’t have a good science of mentality right now. But this set of problems is what Chalmers calls the “easy problem” of consciousness. Once we have an account of belief, desire, representation, memory, psychological awareness, etc., there won’t be some extra “hard problem of phenomenal consciousness” left over.
(And Penrose’s reasons for thinking that quantum gravity comes into play in conscious thought are unconvincing, to put it mildly.)
Essentially this post states that current physics is true and complete at the scale of everyday life. I think this is an uncontroversial statement. Put differently, are there any advances in theoretical physics since 1980 [edit] that add to what previous theories held about everyday life? I’m not sure that there are any. [If there are, that would be interesting to know]
It could probably also be said that the pace of advances in theoretical physics, in terms of theories that are testable or falsifiable, has been minimal since 1980. Perhaps, we’ve already reached the point of diminishing returns in terms of advances in new theories, that can tested by experimental evidence. [a speculative statement, and certainly one to be made cautiously, since Lord Kelvin’s famous quotation, paraphrasing, “there’s nothing left to discover.”]
[I am not suggesting that something akin to the “End of Physics” has occurred]
since 1970 : chaos,
sorry you just changed your cutoff to 1980
i’ll get back to you if i think of anything
@jpd: That seems like a reasonable example. Still it may be that since Mandelbrot’s work was accepted in the scientific community (~early 1980s?), there have been no advances since then in theoretical physics as applied to everyday life, and no apparently testable advances.
@ physicalist.
1. You have no choice to accept computationalism once you accept current EFT.
2. Sure. It contredicts your previous point, thought.
3. As I said, we have experimental evidences that this issue is not as obvious as you may think.
4. Well, thanks for the discussion, and let’s agree to disagree.
@Justin
Decoherence maybe?
@Julien: At the moment, decoherence is a competing hypothesis, but unproven. So it’s not yet an established theoretical advance. Perhaps it has a better chance of experimental testability, though.
@Justin: How would you define domain of “everyday life physics”? How did the set of questions about “everyday life physics” changed since 1900’s? Were any new questions added since then (don’t mind the answers) ?
@Oleg: I think I’m just using the definition that Sean is using which is the domain of our scales of experience, i.e. essentially conceivable scientific applications in everyday life. Even if some alternative to relativity or quantum theory were developed, my understanding of Sean’s argument is that at the scale of our experience, such as engineering applications, chemistry, biology, and electronics, the predictions of that new theory would be empirically identical to existing theories. At different scales, a new theory would more accurate.
I believe that was the intent of the analogy that Newtonian physics is true at a certain scale, and not true, clearly at relativistic speeds. That’s my interpretation.
Of course, it is conceivable that some new theory could have an impact at our scale of experience, but that impact would likely not be significant.
[Of course, predicting the course of science is a bit like predicting the weather 1,000 years from now]
The only three articles you ever need to understand, its all numbers.
To tired to discuss though.
http://en.wikipedia.org/wiki/Number_theory
http://en.wikipedia.org/wiki/Complete_set_of_commuting_observables
http://www.scholarpedia.org/article/Second_quantization
@Justin: How does second law of thermodynamics relate to all this? Imagine you already know quantum theory, standard model of particles and theory of relativity, but you don’t know thermodynamics. Will you still need second law to explain “everyday life” physics? And how can you a priori figure out what’s missing?
[some amateur thoughts on your question]
@Oleg: Well, we don’t see a broken egg reassemble itself; we don’t see high entropy states spontaneously converting into low entropy states. So, the other problem I’ve read about is that there is no arrow of time in quantum theory, the standard model, or relativity. We have experience of a succession of events, of remembering the past, and not remembering the future. If we want to account for our experience of time in this sense, the second law of thermodynamics has been invoked as an explanation, since the direction of events is from low entropy to progressively greater entropy. From my reading, there’s no arrow of time in quantum mechanics. In relativity theory there’s a block universe, and so the experience of the flow of time is an illusion, and the standard model, according to Baez, http://math.ucr.edu/home/baez/time/time.html, has time reversal symmetry. So, as I read it, the arrow of time can’t be derived from the standard model or relativity theory.
The textbook derivation of the 2nd law was by Boltzmann, but, apparently, the 2nd law can also be derived from quantum mechanical systems, which also increase their entropy over time [yet are time symmetrical].
Yes, the second law is certainly required to explain “everyday life” physics or to explain why we don’t see a gas remain condensed in a small region of that space rather than spread throughout a given space, or why we perceive the flow of time from past to present to future.
I believe you’re also asking how to derive the 2nd law? Well there’s a set of observations that a gas spreads throughout a given volume, that objects decay over time, etc. Without the second law it wouldn’t be possible to explain why don’t see high entropy -> low entropy changes, such as spontaneously growing younger.
[It’s a little puzzling as to why it seems that the 2nd law can be derived from quantum mechanics and yet quantum mechanics is itself time symmetrical. I’m not familiar with this issue.]
EDIT: I just realized that the arrow of time is apparently a primary area of expertise for the author of this blog. Caveat: this is definitely not my area of expertise.
I think it’s wonderful news that the laws of physics of everyday life are completely understood, but the reason science still fails, and religion wins, is pretty simple: physical laws of fields and particles don’t offer any narrative or guidance for everyday life — they’re utterly devoid of purpose.
Atheism will never eradicate religion because religion provides a higher myth to live by, which is what people really crave. Until science can find an inspiring story, maybe something like what Carl Sagan offered but more overtly religious, it will continue to lose to the religious myth-makers.
ref: Apparently Dr. Charles Bennett believes that quantum mechanics ” helps resolves the paradox or puzzle of the origin of the second law of thermodynamics”, see lecture: http://mitworld.mit.edu/video/500.
After embarrassing myself with speculations predicated upon long-lived neutrons, I have to ask, IS there a standard model explanation for the stability of the neutron in an atomic nucleus?
To me, there seems to be a scale gap in knowledge separating QCD and nuclear physics. I recently did editing work on a science paper that involved the accurate calculation of the half life of a certain popular element that was done on a ginormous computer. What was interesting to me was that no mention of the standard model was made by the authors, despite it being a fundamental study and despite it involving a number of calculations that is so stunningly high I would break my non-disclosure agreement with the authors’ institutions by mentioning it. The authors reasonably assumed that quarks and gluons are at such a high energy scale they would never affect the nuclear physics of atoms.
But, I had to go only one step further in asking myself if the standard model accurately predicts the lifetime of the neutron outside as well as inside the nucleus.
I fuzzily checked Wikipedia, and it doesn’t say for sure how the lifetime outside the nucleus is calculated in the standard model. Note also that measurements of the precise lifetime seem to have significant error associated with them, so I wonder if the standard model’s predictions are considered to be accurate enough in this regard. A quick glance seems to indicate to me that the standard model should be able to give a number for the lifetime.
On the other hand, regarding the stability of the neutron inside the nucleus, all I could get from Wikipedia is “When bound inside of a nucleus, the instability of a single neutron to beta decay is balanced against the instability that would be acquired by the nucleus as a whole if an additional proton were to participate in repulsive interactions with the other protons that are already present in the nucleus” (citation needed).
The above explanation neither mentions the standard model nor sounds very convincing. Another source mentioned that the Pauli Exclusion principle is the reason for the stability of the neutron in the nucleus. More convincing, to me, is when one calculates the energies of free particles making up the nucleus and compare that total with the energy of the bound system.
But I have to contrast this cookbook binding energy calculation with the standard model picture. Does something in the standard model prevent a quark changing from down to up and undergoing subsequent beta decay when it is inside one nucleus, say a deuterium nucleus, but not another, say iodine 131?
More generally, can the standard model account for the magic numbers of nuclear physics?
@Baby Bones: Try posting that question at Physicsforums: http://www.physicsforums.com/. There’s a good mix of advanced graduate students and a few professors under pseudonyms there.
@jpd and Justin Loe: chaos isn’t fundamental, though. It’s emergent. It doesn’t involve any new fundamental theories of physics. (Not that it’s not interesting – most people, even most physicists, spend their lives studying things that are emergent.)
@Oleg and Justin Loe: thermodynamics is reasonably easily derived from statistical mechanics, which is just an application of your underlying dynamics (classical or quantum). It’s only this last level Sean is calling fundamental. (The quantum version, anyway – the classical version can be derived by taking hbar -> 0. Except, sadly, general relativity.) Sean’s book about time/entropy/etc. is a pretty good read.
@Baby Bones: it’s been a long time since I’ve taken nuclear physics, but as I recall, you can derive the magic numbers from the nuclear shell model. This was approximate the way we did it in undergrad, but presumably the actual hardcore method is better. A quick google scholar search turned up this review, although that may not be much use to you if you aren’t at a university that provides a subscription to that journal, and don’t have a PhD in physics, perhaps. It does at least tell you (in the abstract) that we have a fair but imperfect understanding of this.
And on the lifetime of the free neutron, this article has an equation for the Standard Model prediction. (Which they say is from this paper, though a little manipulation is involved.) There are parameters in this equation that have to be measured experimentally, and others that can be calculated to high precision. The Standard Model has a lot of input that it needs… can’t predict everything. (Although note that this is not my area of expertise – I only care about electrons, and, well, I guess protons too. And dark matter. But forget neutrons.)
@Charon: What’s your opinion on the fact that the arrow of time is only found in the 2nd law of thermodynamics but not relativity, or quantum mechanics or the standard model? Admittedly, this wasn’t really the topic of Sean’s post, but is it correct to state that the arrow of time, i.e. the experience of time that we have, is currently only a result of the 2nd law as derived from statistical mechanics but not the three fundamental theories? (that is, the three fundamental theories all exhibit time reversal symmetry?
Essentially, then is the 2nd law of thermodynamics required to account for our experience of the flow of time, or is the flow of time found in fundamental theories?
Or, is it possible that the flow of time is itself an illusion as Julian Barbour, http://en.wikipedia.org/wiki/Julian_Barbour, has advocated?
Thanks for the responses.
EDIT:
Barbour’s take on the arrow of time is to dismiss that the universe formed from a “special” low entropy state, and to suggest that, instead, it is eternal, and has passed thru an infinite number of states, and that accounts for the particular low energy state that gave rise to our particular universe, at least as I read his interpretation.
Julien, sorry for late reply, i live in a wildly different time zone.
Off course, i won’t consider “human” any AI, whether it emulates human behavior or not. No more than my parrot is human when it pronounces “Nice bird Pushy!”. It is a different animal anyway 🙂
Doesn’t mean i’d feel ok to offend it. No more than hit a dog.
@Charon: Let me summarize, briefly: is it troubling that all the fundamental theories, which seem to explain all of “everyday” life, do not explain the arrow of time, and our experience of it?
[I had another edit on the above post, but I ran out of time…]
EDIT of previous post:
Apparently this question has already been asked as Loschmidt’s paradox: “objection that it should not be possible to deduce an irreversible process from time-symmetric dynamics. This puts the time reversal symmetry of (almost) all known low-level fundamental physical processes at odds with any attempt to infer from them the second law of thermodynamics which describes the behaviour of macroscopic systems. ”
http://en.wikipedia.org/wiki/Loschmidt's_paradox (for some reason the link isn’t working, but I’m just referencing the wikipedia article on Loschmidt’s paradox)
@Justin: There are two points which are troubling me.
The first point is ambiguity of the dividing line that separates laws of physics from mere boundary conditions. For example, it is clear that exact value of the distance between the Earth and the Sun is not a law of physics: the Earth’s orbit could be 1 million km closer or 1 million km farther from the Sun, and no law would be violated. I cannot see how the fact that the Big Bang had very low entropy differs from billions of other facts about our Universe. Still it is necessary to know boundary conditions of our universe in order to explain the arrow of time (and second law). The similar thing is dark energy versus matter/antimatter gravitational repulsion (http://en.wikipedia.org/wiki/Gravitational_interaction_of_antimatter) explanations of expansion of the Universe. How can we discriminate between the “law of physics” (lambda-CDM model) and “boundary condition” (living in a Universe where matter and antimatter have already clustered in separate galaxies) before measurments of gravitational interaction of antimatter? Of course, this is far beyond the realm of “everyday life” physics, but I wonder, how can we prove that laws of physics known today are not an emergent properties of our specific Universe?
The second thing that is troubling me is that once we know all reversible laws of dynamics, we don’t really need second law of thermodynamics. We could have said that we already understand the laws of physics underlying everyday life, and this statement would be correct. Moreover, the question “why a table is solid?” has a simple, plain and correct answer “because it is made of wood”, so we don’t really need relativity, standard model of particles and quantum mechanics to generate correct explanations of “everyday life”. In general, I think that Sean’s statement cannot be falsified.