Why are laws of physics stable?
Recently I've been thinking about why we live in a world with stable laws of physics, out of the plethora of all possible worlds. Why does the sun rise every day, why is the intensity of the Earth's gravitational field constant, why do causal relations ("the constant conjunction between causes and effects", as Hume put it) persist in time?
It seems to me that there are two principles that might jointly explain why the laws of physics in our world are stable: anthropic principle and Solomonoff induction.
The anthropic principle states the self-evident truth that we can only live in such a world that has conditions suitable for life of our kind (we would not be able to live in those worlds that lack such conditions). These conditions would probably include regularities such as stable laws of physics because otherwise the world would be too random or chaotic to support the formation of living organisms, let alone conscious beings such as us, with our evolved predictive neurological apparatus. That would explain why the laws of physics in our world have been stable for billions of years: it is the kind of world in which we would necessarily find ourselves, as we couldn't have evolved in an unstable world. (The anthropic principle also explains why our world is not too regular, since life and consciousness seem to require some amount of complexity.)
But the anthropic principle doesn't seem to explain why we should expect that the laws of physics will continue to be stable in the future. In fact, it may seem that such a stability is very unlikely because there are many ways our world could be in the future but only one way in which it would be a deterministic extension of the world it has been until now. This is where the Solomonoff induction comes in, which seems to imply the opposite: it is more likely that laws of physics will continue to hold. Why? Because given the way our world has been until now, this world is more simple if its regularities (such as laws of physics) continue than if they are discontinued, and simple worlds are more likely (more frequent in the collection of all possible worlds) than more complex worlds. (A simpler set of properties is instantiated in more possible worlds than a more complex set of properties.) Such a deterministic world is fully defined by some initial conditions and laws of physics, while a world whose regularity is discontinued at some point would need an additional property that would define the discontinuation and thereby make the world more complex. Maybe the Solomonoff probability of the laws of physics in our universe remaining stable became overwhelmingly high even long before any living or conscious organisms appeared here. Solomonoff induction is basically a mathematical formalization of Occam's razor.
https://arbital.com/p/solomonoff_induction
Anyone familiar with this explanation or any alternative explanations of the stability of laws of physics?
It seems to me that there are two principles that might jointly explain why the laws of physics in our world are stable: anthropic principle and Solomonoff induction.
The anthropic principle states the self-evident truth that we can only live in such a world that has conditions suitable for life of our kind (we would not be able to live in those worlds that lack such conditions). These conditions would probably include regularities such as stable laws of physics because otherwise the world would be too random or chaotic to support the formation of living organisms, let alone conscious beings such as us, with our evolved predictive neurological apparatus. That would explain why the laws of physics in our world have been stable for billions of years: it is the kind of world in which we would necessarily find ourselves, as we couldn't have evolved in an unstable world. (The anthropic principle also explains why our world is not too regular, since life and consciousness seem to require some amount of complexity.)
But the anthropic principle doesn't seem to explain why we should expect that the laws of physics will continue to be stable in the future. In fact, it may seem that such a stability is very unlikely because there are many ways our world could be in the future but only one way in which it would be a deterministic extension of the world it has been until now. This is where the Solomonoff induction comes in, which seems to imply the opposite: it is more likely that laws of physics will continue to hold. Why? Because given the way our world has been until now, this world is more simple if its regularities (such as laws of physics) continue than if they are discontinued, and simple worlds are more likely (more frequent in the collection of all possible worlds) than more complex worlds. (A simpler set of properties is instantiated in more possible worlds than a more complex set of properties.) Such a deterministic world is fully defined by some initial conditions and laws of physics, while a world whose regularity is discontinued at some point would need an additional property that would define the discontinuation and thereby make the world more complex. Maybe the Solomonoff probability of the laws of physics in our universe remaining stable became overwhelmingly high even long before any living or conscious organisms appeared here. Solomonoff induction is basically a mathematical formalization of Occam's razor.
https://arbital.com/p/solomonoff_induction
Anyone familiar with this explanation or any alternative explanations of the stability of laws of physics?
Comments (56)
The notion that reality may contain the potential for many different sets of laws is indeed fascinating.
Sounds like false vacuum decay, an event very unlikely in our universe, considering it hasn't happened here for billions of years.
https://en.m.wikipedia.org/wiki/False_vacuum_decay
Quoting Foghorn
From the perspective of reality as a whole there is no difference between potential and realized laws, as everything that is possible is also realized. But we can use the anthropic principle and maybe also probability over possible worlds to explain some properties of our world.
Do we? Aristotle said that bowling balls fall down because bowling balls are like the earth and things tend to go to like things. Fire is like air, that's why fire goes up. Good a theory as any, two thousand years worth of mindshare.
Newton said bowling balls fall down because [math]F = \frac{m_1 m_2}{r^2}[/math].
Einstein said that bowling balls fall down because spacetime is curved by mass and bowling balls are just following a geodesic in spacetime near the earth.
Multiverse theory says that bowling balls fall down because we happen to live in a universe where bowling balls fall down. In some other universe, people don't go bowling. Or something. I'm not actually sure what the multiverse says about bowling balls.
But the larger point is clear. The laws of physics are historically contingent ideas made up by people.
But perhaps by "laws of physics" you mean the "ultimate" laws of physics that our contingent theories are only approximations to. But what makes you think that (1) there are any such things; and (2) even if there are, that they don't change over time? Those are two metaphysical assumptions, not supported by empirical proof.
As a more striking example, consider dark matter. One theory of dark matter is that it's made of particles that interact with the gravitational field but not with electromagnetism, so that we can't possibly ever see them. Nevertheless they affect the rotational speed of galaxies.
Another fascinating approach to dark matter is called modified Newtonian dynamics, which is the speculative idea that we haven't got gravity quite right, and that we need to make subtle corrections to the theory to account for the observed rotational speed of galaxies.
So you say we live in a world of stable laws of physics, but that claim is highly questionable. First, if by "laws of physics" you mean our human-created theories as delineated by the likes of Aristotle, Galileo, Newton, and Einstein, those clearly are not stable, but rather change over time.
But if by laws of physics you mean some sort of ultimate rules that the world must obey, I have to ask you one, why you think there are any; and two, what makes you think they don't change all the time? I don't doubt that you could give a decent argument as to why these two things are the case; but I hope you'd at least concede that you are making metaphysical assumptions that lie far outside the limits of science.
Your point of view has a name, Scientific realism. It is a metaphysical stance, not an established fact. The Wiki link has arguments for and against it; and the SEP article has lots more.
The first question I have is whether the 'plethora of possible worlds' is simply a figment of the imagination. That there might be 'other worlds' or 'other universes' seems like the most idle of idle speculation - what light does it cast, what explanatory advantages does it provide? And what evidence could there ever be for it? Why is that such speculation is regarded as scientifically respectable, when, for instance, speculation about any form of higher intelligence is inevitably dismissed as creationist?
Second point is to consider that the stability of the laws of physics are conditions for the existence of anything whatever, and questioning them is tantamount to questioning why two plus two equals four, and not an elephant. They are simply so, in order that anything might exist whatever.
As is well-known, there are a small number of fundamental constants that seemed to be poised according to minute tolerances, in the absence of which a Universe would not have been formed in the first place (per Lloyd Rees, Just Six Numbers.)
But I have to say that overall, I think such questions as that posed by the OP are basically empty, in that they can be answered by the simple observation that unless it were so, there would be nobody around to ask the question. What is implied by that I'm not sure, but I can see how it can be evaded.
Look at the following random sequence of numbers generated using the software in RANDOM.ORG:
{3, 5, 10, 4, 2, 9, 8, 7, 6, 1}
Despite the fact that the sequence is completely random, we see a pattern (bolded for emphasis) in the sequence which is 9, 8, 7, 6.
The laws of nature are simply patterns in the way matter & energy interact and that they've been as they are now for quite some time (how long I'm not certain. read Hume's problems of induction) may simply mean that we're in a certain phase/stage in what is actually chaos and the stability of the laws of nature we're observing could be nothing more than temporary patterns in chaos. So, take a deep breath, strap yourselves in because the so-called laws of nature (the order/ the pattern) could devolve into utter chaos at any time.
This just popped into my head and seems relevant: We know human history is marked by both peace (laws in effect) and wars (laws suspended) but, interestingly, we can't seem to be able to tell whether wars (laws suspended) interrupt the peace (laws in effect) or peace (laws in effect) interrupts the wars (laws suspended). Are we peaceful creatures (wars disrupting the peace) or are we warlike (peace only to recover our strength to wage more war)? Insofar as the OP's point is concerned, is the universe chaos with periods of order or is the universe order with periods of chaos? Hume might know!
"Physical laws" are features of physical models and not the universe itself. Our physical models are stable, therefore "physical laws" are stable. If in current scientific terms new observations indicate that aspects of the universe have changed, then, in order to account for such changes, we will have to reformulate our current (or conjecture new) physical models which might entail changes to current (or wholly different) "physical laws". E.g. Aristotlean teleology —> Newtonian gravity —> Einsteinian relativity.
It's a genuine field of research: are physical constants really constant or just slow? can other universes have different constants, qualitatively and quantitatively?
To build structures like atoms places extreme limits on what physical constants can be, but there's no reason why a universe ought to have atom-like structures. The maximum speed of light was perhaps different in the brief period when we had no structures of charges (this difference might even allow that this period was not so brief).
Quoting litewave
A change in a law would raise the question, what changed it? In all other things, inertia is a sign of being left the hell alone: change suggests something driving that change.
True, physical laws are manmade. But colloquially we use the term to describe the thing that our physical laws approach in their representations (if indeed they are approaching anything).
Yet in your example with objects falling down, all the historical theories from Aristotle to Einstein say that objects consistently fall down rather than up or in random directions. The later theories give more accurate predictions than earlier ones but from all of them it seems that the phenomenon of objects falling down is highly stable. How do you explain that if not by a stable regularity in the world?
Quoting fishfry
There are obviously persistent regularities in the world that we know have been observed for millennia and have been used to make successful predictions. This doesn't mean that the regularities cannot change but they are obviously highly stable.
Quoting fishfry
Yes but I don't know of a better alternative. Realism explains that our theories work because they correspond to reality while Instrumentalism offers no explanation why our theories work.
In other words, strictly speaking, the answer is 'none of our business'. Maybe positivism has something useful to contribute after all.
I agree that theories supported by sensory evidence are more convincing and useful than theories supported by logic alone but one might also ask whether sensory experiences are not simply figments of the imagination. My reason for taking other possible worlds as real is that I don't know a difference between "real" and "possible" worlds. I can't even imagine what such a difference could be. The only way to show that a world is not real would be to show that it is not logically consistent, but such a world would not be possible either.
Quoting Wayfarer
It can provide a clear logical explanation why some properties of our world are the way they are when it seems that they could have been different, for example the stability of the laws of physics or the values of some constants that seem fine tuned for the existence of living organisms. There may be no other explanation than that we happen to live or are likely to live or necessarily live in one of those worlds that have certain properties.
Quoting Wayfarer
I don't know. If any interaction with other worlds is in principle impossible then we can never have direct observational evidence of such worlds. But we may get indirect supportive observational evidence if we observe things in our world that can be better explained or predicted by a theory that assumes the existence of other worlds.
Quoting Wayfarer
Multiverse theories are more logically transparent than obscure theories of God, and their more limited versions are also closely connected with theories of physics that are well supported by observational evidence, for example theories that postulate that there are worlds beyond the horizon of our observable universe (beyond the Hubble volume) but still within our universe, inflationary multiverse, string theory multiverse, or quantum mechanical multiverse.
Quoting Wayfarer
Something can exist also in a world with unstable laws of physics, namely the unstable laws themselves and various unstable or random objects that are compatible with those laws.
Quoting Wayfarer
A world without these constants would be very different from ours but that doesn't mean it would be nothing. Worlds without time seem to be possible too, basically any consistent mathematical structure, and in the extreme also completely empty worlds that are identical to an empty set.
That seems explained by the anthropic principle: we could have evolved only in a world where the laws have been stable for a long time.
Quoting TheMadFool
Solomonoff induction seems to show that this is very unlikely.
Still, all these theories describe a stable phenomenon of objects falling down (rather than up or in random directions), although later theories give more accurate predictions than earlier ones.
I am not sure we can view it this way. If the structure of a world is random instead of deterministic does it mean that the world is not being "left alone"? It does mean that randomness increases (Kolmogorov) complexity of a world.
We need to understand one thing before we shoot our mouths off. We're talking about cosmic-level events. Time, to be precise duration, may need to be adapted to the so-called Cosmic Perspective (Neil deGrasse Tyson). You know, like astronomical distances need to be measured in lightyears, astronomical units, parsecs.
A similar effect is observed for time - geological time, aeons, deep time, so on and so forth.
Let's not forget the Hindu idea of Kalpa (aeon) which in the Western world would be Conformal Cyclic Cosmology or some such.
Quoting litewave
Quoting TheMadFool
It would depend on how long an orderly phase in the chaos lasts (billions/trillions of years) and where we are, temporally, in the ongoing non-chaotic part of the kalpa.
For instance, the quantum vacuum is random, but we know why: it is a property of 4D waves that they are only well-defined on large timescales. There are an infinity of possible modes and therefore at any given time some of those modes must briefly exist (because 0 is well-defined).
Ultimately, the reason for anything to exist is the same: that it is possible (logically consistent). But some properties are simpler and therefore more likely than others. A world where the speed of light randomly changes is less simple than a world where it is constant (all other laws and initial conditions being equal).
Nah, it's probably you that has changed, not the law of gravity.
I agree.
It only makes sense to invoke the anthropic principle for properties that are necessary for the existence of human life and that vary across a collection of different possible worlds (the so-called fine-tuning for human life), and it may be difficult to identify such properties or their combinations. It may be a very general principle that can't describe a detailed structure of our world since there may be many different worlds that can support human life. And while it may explain why the laws of physics have been stable for a long time it doesn't seem able to explain why we should expect that they will remain stable in the future.
If you go to the moon, the gravitational acceleration is different than on earth, And I took the trouble in my post to give the striking example of dark matter, which shows that we still don't understand gravity. If you deny that human-created physics is historically contingent, you must not be familiar with the history of science. "Bowling balls fall down" is not a law of physics, it's an empirical observation. We STILL don't fully understand the underlying law, if in fact there is one.
Quoting litewave
Yes indeed. The Ptolemaic system that placed the earth at the center of the solar system fit all known observations and was accepted for millennia. In fact Ptolemy's system actually fit the observed data better than the new heliocentric system of Copernicus, because Copernicus thought the orbits were circles with the sun directly at the center. Showing that "obvious persistent regularities" can be flat out wrong, and overthrown in a historical instant. Once Kepler and Newton showed up, it was all over for Ptolemy. But he had a nice 1600 year run. Made extremely successful predictions.
Quoting litewave
"Obviously" is not a scientific principle, it's an anti-scientific one. Newton's ideas were obvious. Einstein's are much less so.
Quoting litewave
That can only be because you didn't bother to read the Wiki and SEP articles I linked. Refusing to read the counterarguments then saying you don't know them doesn't prove anything except that you prefer not to learn about what you don't know.
Quoting litewave
Now you're just making an argument for scientific realism. Which is fine, it's a perfectly good idea. It's just not provable. It literally lies outside of science. It's a metaphysical assumption.
I'm not arguing for the falsity of scientific realism; only noting that it's a metaphysical stance and not a scientific one. It's not even necessary to assume in order to do science. Whether there's really a consistent reality "out there" or only seems that way due to our highly limited observational experience, is not something we can know for sure. After all others have noted in this thread that the latest theories suggest that perhaps the only reason our laws of nature are the way they are is that we just happen to live in this particular branch of the multiverse; and that nature could be quite different in other ones. Even the scientists don't believe in "obvious persistent regularities" anymore, since the ones we observe are just random rolls of the dice and could have been different. You know, the famous six constants that someone linked.
Gotta go, it's feeding time in my vat.
From a different the philosopher Joseph Margolis: "The Flux of History and the Flux of Science": https://publishing.cdlib.org/ucpressebooks/view?docId=ft6t1nb4gf;query=;brand=ucpress
These laws were different immediately at and immediately after the Big Bang. They also have problems inside black holes. And who knows if they apply to all of the universe? We can't detect all of it, in principle.
I think natural habits would've made more sense or a tendency to behave in such a such a way under X circumstances.
A law that doesn't change also raises questions: why doesn't the law change?
Still, the apple falls down on Earth, similarly like it did 2000 years ago.
Quoting fishfry
Do you expect that once we understand dark matter the apple will stop falling down on Earth?
Quoting fishfry
I really don't know why the apple falling down on Earth would be a historically contingent, human-made regularity.
Quoting fishfry
The planets were moving with the same regularity at the time of Ptolemy as they are now. Ptolemy just invented a very cumbersome way of describing their motions, by choosing to describe them in relation to Earth. Copernicus later found that it was much more simple to describe them in relation to the Sun, Kepler found it was more accurate to approximate their orbits with ellipses rather than circles, Newton postulated a universal law of gravity, from which the elliptical orbits followed as a logical consequence, and Einstein improved the accuracy of description by introducing a curved spacetime. Still, the planets were moving with the same regularity at the time of Ptolemy as they are now.
Quoting fishfry
Obviously the apple falls down on Earth like it did 2000 years ago. The planets move in the same way too.
Quoting fishfry
Actually I gave a look at the Wikipedia article, which confirmed that realism offers a sensible explanation of why our theories work while instrumentalism offers none.
Quoting fishfry
Ok.
Quoting fishfry
Indeed. I can't even be sure that you are not just a figment of my imagination. But I am pretty sure that whatever you are, you are what you are and not what you are not. In other words, you are a consistent object, identical to itself. To assume otherwise would be a nonsense which would lead to a logical explosion that would make discussion, science and understanding meaningless.
Quoting fishfry
Sure, that was also the basis of my OP.
Agreed. We adopt scientific realism for pragmatic reasons. I suspect we are in agreement.
And FWIW, bowling balls always seem to fall down, but electrons are detected as spin up or spin down randomly. And so we invent a contingent theory to "explain" that, using an explanation that nobody really believes or understands. The 20th century was not kind to the Newtonian worldview of a consistent reality "out there," can we agree on that?
:lol:
:100:
Quoting litewave
:up:
"Everything should be made as simple as possible, but no simpler." ~Albert Einstein
Later physical theories consist of better, more comprehensive, less ad hoc explanations than earlier physical theories. From this comes more precise predictions (i.e. experimental tests) and unforeseen fecund problems (i.e. inquiries for research). Predictions in science matter only to the degree they are derived from, as David Deutsch says (following Karl Popper) "good explanations".
Smolin has a theory of cosmological natural selection in which the laws or constants may change when a new universe is born from a black hole in the preceding universe. So if I understand him right, he doesn't propose that such a change has happened in our universe since it was born.
But apparently the known laws of physics (regularities) have been stable for billions of years.
Sure, they're pretty stable all right.
Still, there are exceptions: the Big Bang and Black Holes. Things break down at these levels. I'm thinking that when James Webb Space Telescope gets going - hopefully it will be in space this year - we might find considerable surprises where such laws break down.
I have nothing against these laws at all, but I like Art Hobson's idea of thinking about these in terms of "habits" or tendencies. Not a big deal though, it's still impressive.
Isn't that just the currently contingent theory, subject to revision in next week's Physical Review Letters? You have no actual evidence for such a proposition. It's an idea based on a mathematical model in a highly speculative area. It's a lot different than noticing that bowling balls fall down, isn't it? If these "known laws" -- which have become known only in the last few decades -- are changing in subtle ways, we'd be the last to know about it. Not so?
Right. But apparently the regularities of the world have not changed much, if at all. Scientists just found more accurate descriptions of them.
We don't know what is going on in these special cases but outside of them the regularities seem stable.
I guess physicists have a lot of evidence that points to the stability of the known laws?
In other words, better general descriptions from which more detailed descriptions can be derived. This is possible when there is an even more general/universal regularity than the ones we knew before.
Several decades at least. Of course Ptolemy had evidence too. Isn't this just Hume's problem of induction? Old philosophical conundrum. Like the turkey said on Thanksgiving morning, "The farmer's been good to me every day this year ..."
:up:
I don't understand any of this or his work sufficiently well enough to say one way or the other, and if I did say it was one way or another it would probably not be for the right reason.
The laws according to Smolinare contingent not necessary and changeable in time. I don't know if he says one way or the other whether he things they are invariant in this universe.
Here is a short clip from an interview: https://www.closertotruth.com/interviews/3649
From another interview:
https://www.edge.org/conversation/lee_smolin-think-about-nature
Because our sciences substitute idealized abstractions for a more immediate and intricate experiencing of our world. The way we have carved up the world rigs the deck by forcing our experience into over-generalized channels such as objective causation and universal natural lawfulness. Then we mistake the peculiar constraints our models impose with the world itself.
Yes. I wonder why Solomonoff's solution to the problem of induction is not mentioned in that article.
I usually understand "explanation" as a derivation of something particular from something more general. Like, why does the apple fall down? Because of the law of gravity.
So if our sciences didn't do this, we would not experience the apple as falling down every time we drop it?
It’s not that the sciences are wrong, it’s that they operate with blinders on. So rather than just defining what is in front of me as this object moving in space according to a mathematical rule, we could embed this restrictive formulation within a much wider, richer and more interconnected experience that recognizes the contribution of my subjective schemes to what appears in front of me , and acknowledges the contribution of an intersubjectivity community of the construction of what we call ‘external’ reality.
It may be hard to see how this way of seeing improves physics , but it makes a profound difference to the social sciences and psychology, which have suffered under the rule of the methods of the hard sciences.
Ok, but the apple would still fall down every time we drop it, no? So the regularity would still exist even if we took our blinders off and the question would remain why the regularity persists.
The seemingly mathematically exact regularity would make way form a more complex pattern, What’s most valuable in the relation between the apple and gravity is that two previously unconnected phenomena were unified via an empirical explanation , not that a certain number (the gravitational constant ) resulted. In and of it self the constant, this ‘law of nature’ is not connected to anything. It is arbitrary. The most profound progress in science isn’t about arriving at arbitrary constants but showing the interconnectness of the world. As an example, a multiple universe hypothesis that makes the gravitational constant in this universe just one point in a spectrum of evolving universes, each with their own constants, changes the constant from an isolated and arbitrary number to part of an interconnected process of development. It could be that in a hundred years or so the physical laws have been replaced by a probabilistic , process -oriented physical model that puts unidirectional time at this core . We already are hints of this thinking with Lee Smolen and Ilya Prigogine.
So there are many kinds of models of regularities, and each has its drawbacks as well strengths. A strictly mathematical ‘lawfulness’ has as a drawback that everything that doesn’t fit into the model is rendered as chaos, randomness and chance. Prior to chaos theory, many physical behaviors were treated
that way ( cloud and smoke formations).
Thanks for the reference, I'll take a look at that.
Now if the physical laws as we know, were allowed to change in a random non deterministic way, then the overall result is that there is no physical laws at all, because a law that is subject to change at any moment, cannot predict anything with certainty.
Now, in a universe in which there is no law at all, is highly improbable that stars and planets arise, let alone life.
Example 1: sum of absolute value of two velocities is greater than or equal to either one of them, I. Newton. Oops, relativity. Correction: for velocities quite a bit less than c, law approximately applies.
Example 2: heavenly bodies move in circular orbits, circles being perfect. Sorry, they don't. Oh, ok then, old observations chucked, revise the law.
"Law" being another way to say "stable.". If we're in a stream where everything is moving at the same speed, everything around us will be stable. To realize that we're moving, we'd have to imagine a fixed vantage point. I think that's what you're saying? That to know that physical laws are changing, we'd have to have something fixed to compare the change to.
Are there uncaused events? Possibly. We could not know the difference between an uncaused event and an event with a cause we were unable to detect.
Balance
Two equally opposing forces in perfect eternal balance.
Nothingness and somethingness,
Light and dark,
Up and down,
Reality is a paradox.