I fought the law

May 31, 2006 5:37 AM

Is there any evidence that physical laws have changed, or could possibly change over time? For example, has the speed of light - since the beginning of the universe - been the same, and will it always remain the same into the future?

i.e., Is it possible conditions were so different for the first billion years of the universe that this 'constant' was different than the speed we now measure?

I am interested in the answer to this question as it might apply to any law of physics or 'constants', light is just an example. (Also, I'm aware of experiments that show, for example, that light may be 'slowed' under certain singular conditions...I'm more interested if these laws could change or have changed on a universe-wide scale)

posted by extrabox to science & nature (23 comments total)

Since the meter is the distance light travels in 1/299,792,458 of a second in vacuum, the speed of light by definition is constant-- so in this case the question is "is distance constant."

But I understand what you mean and your question is a valid one and is addressed here:

http://math.ucr.edu/home/baez/physics/ParticleAndNuclear/constants.html
posted by justkevin at 5:45 AM on May 31, 2006

http://www.newscientist.com/article/dn6092.html

"The speed of light, one of the most sacrosanct of the universal physical constants, may have been lower as recently as two billion years ago - and not in some far corner of the universe, but right here on Earth."
posted by martinrebas at 5:52 AM on May 31, 2006

I went to a lecture by Rupert Sheldrake (renegade scientist who came up with the hypothesis of Morphic Resonance) where he talked about the evidence against the speed of light being constant. Very interesting.
posted by BobsterLobster at 6:19 AM on May 31, 2006

Isaac Asimov's The Gods Themselves comes up with one scenario where the physical laws of a universe could change and what the consequences might be. It's obviously speculative, but he does drop some pretty good science.
posted by PinkStainlessTail at 6:21 AM on May 31, 2006

On Rupert Sheldrake's website:
Incidentally, he was saying this at least 15 years ago.
posted by BobsterLobster at 6:22 AM on May 31, 2006

Grr, here's the link! http://www.sheldrake.org/experiments/constants/
posted by BobsterLobster at 6:23 AM on May 31, 2006

It's often somewhat tricky to figure out if the speed of light has changed or something else has. For this reason physicists normally try to figure out if some dimensionless parameter has changed instead. This is why one often hears of experiments or observations to determine if the fine structure constant has changed (this being e²/2ε₀hc it has the speed of light in its definition).

The constraints that are mentioned in the above links are pretty strong though - the speed of light hasn't changed very much if at all even over cosmic timescales.
posted by edd at 6:26 AM on May 31, 2006

Oh and I wouldn't use Sheldrake as a source even though much of what he says might be right.

For starters, I don't know anyone who would think Avogadro's number is fundamental (although given it depends on the mass of the carbon-12 atom it could be read that way, although this mass then depends on several of his other parameters) and Boltzmann's constant is so far from being fundamental it's not even funny.
posted by edd at 6:32 AM on May 31, 2006

Something else that I think should be pointed out is the difficulty of the notion of how much is a little or a lot and the related implications. What I have in mind is the simple statement that a lot of the basic constants could not be changed "very much" at all before we find a universe very unlike this one and, even more easily, one inhospitable to us.

But I mention the problematic relativism of "very much" because the statement I make above appeals to our intuitions in direct proportion to how much we agree on "very much" and that such a relative quantification makes any sense. Our intution demands that such a statement makes sense, but I spend a lot of time these days being skeptical of this. This may seem esoteric, but there's a lot of interesting things that rely upon this intution, most specifically the anthropic principles implied by it (with which I strongly disagree, independently and prior to my discomfort with this intuitive quantification).
posted by Ethereal Bligh at 6:53 AM on May 31, 2006

Previous Ask.
posted by meehawl at 7:08 AM on May 31, 2006

Well, you ask about both changes in constants and changes in laws. I'll address the possibility of the latter change since most of the existing posts address the former.

From a philosophical perspective there is strong reason to suppose that natural laws (the laws of science) cannot change. To see this we can explore one of two conceptions of what a law is: an inviolate causal determiner; or a human function.

If laws are causal determiners then they by definition they cannot be contravened. If a law changes, then at T2 (time at an instant after an instant, T1) it is different from the law at T1. Thus the law at T2 contradicts the law at T1. Thus the law is contravened by itself, a logical contradiction. Now, you can (technically) derive anything you want from a logical contradiction, but we'll be responsible and just conclude that there is something wrong with the argument. So, laws, if they are causal determiners, cannot be violated.
(Alternatively you could argue (assuming you don't think laws are true for all time), that the law at T1 is not a law at T2 and so no law has changed. One law simply ceases to be causally effective.)


If natural laws are Humean functions then they are really nothing more than a generalization of observed phenomena. In this case the change in the law at T2 would just be incorporated into the function. Thus, rather than a contradiction we just get a more conceptually complex law. This does not constitute a change in the law because no law can be known until all of the data is in (i.e. the universe has ceased to exist and we have a complete range for the function). Of course, once there is no longer an possibility of change in values there is no longer a possibility of change in generalizations, so laws cannot possibly change.

What can change in both cases is what we might mistakenly believe to be laws, e.g. the Universal Law of Gravitation. But really these are just approximations that we hope are fairly accurate, but that we know are not the complete laws (for one thing most so-called laws are given ceteris paribus).
posted by oddman at 7:13 AM on May 31, 2006

What can change in both cases is what we might mistakenly believe to be laws, e.g. the Universal Law of Gravitation. But really these are just approximations that we hope are fairly accurate, but that we know are not the complete laws (for one thing most so-called laws are given ceteris paribus). from oddman

That is one hell of an excellent articulation, oddman.

All these numbers, relationships, interactions are just reality filtered through our limited observational and verbal capabilities. Quantification is not the thing quantified, it is its symbol.
posted by FauxScot at 7:27 AM on May 31, 2006

As someone who worked for about two years analyzing spectra (which is how scientists determined that the fine structure constant might be changing), I can tell you that they're incredibly noisy and error-prone. It's amazing just how much information you can get out of them even still; but if one seems to show that fundamental constants are changing, I'm more inclined to believe that there was some noise, or poor calibration in the detector. I have to say that I (and all the astrophysicists that I've talked to) am somewhat confused by the idea that the fine structure changing = speed of light changing. The speed of light can be (and often is) formally set equal to 1 in the appropriate system of units. This cannot be done with the fine structure constant ---it really is a free variable in physics (at the moment).

As for physical laws changing, I think that most physicists would say that if a "law" is not applicable over the whole history of the universe, then it is not really a law. If it is applicable over a large range of time, then it may be a good approximation to a true law that we don't know. Apparent violation of physical laws indicates that we have a flaw in our understanding (either our laws are wrong, or our interpretation of data is wrong).

Oh, and the term "renegade scientist" is less applicable to Rupert Sheldrake than the term "crackpot".
http://en.wikipedia.org/wiki/Rupert_Sheldrake
posted by Humanzee at 9:33 AM on May 31, 2006

Oddman has given an excellent answer to your question.

If certain physical constants do change, what explains this? Two answers seem available: (1) There are other, less or unchanging laws governing what we previously thought were unchanging laws (a hierarchy view in which the laws at the very top don't change). Or (2) nothing explains this, in which case the changes in natural laws are random -- which doesn't make them much like laws.

Personally, I think physicists should be a little more careful with their "changing laws" way of talking about explaining results that don't conform so well with what they expect.
posted by ontic at 9:34 AM on May 31, 2006

Time has definitely changed, almost on a daily basis. Right now it is going a helluva lot slower than it was last week.
posted by JJ86 at 10:33 AM on May 31, 2006

ontic, you missed (3) God changed the rules.

I have a friend for whom the idea that the laws change is an important possibility for showing the earth to be 6,000 years old. Unfortunately, there's no way to prove or disprove any claim about what the laws were in the past. IMO Occum's Razor suggests we should assume the laws we experience today are very similar if not exactly the same as the laws for the past 12 billion years.

He, on the other hand, thinks my idea (aka uniformitarianism) is an assumption that should be challenged.
posted by knave at 10:34 AM on May 31, 2006

Following on from what Humanzee said - I've personally analysed one quasar spectrum in an effort to constrain the change in the fine structure constant α and managed only to get a 1σ significance result that was an order of magnitude larger than the current best results and the wrong side of zero to them too. Calling it a headache would be an understatement.

That said, the spectra are very carefully calibrated, they use some very clever techniques to get the best result possible, and they use a lot more than one spectrum and often more than one pair of lines in the spectrum to try to get a result. And the teams do check their errors thoroughly.

I'd call the current non-zero results interesting, but I wouldn't call them strong evidence of a varying α yet. I should probably go back over the most recent work and see what the numbers actually support.
posted by edd at 10:40 AM on May 31, 2006

I remember news reports from a couple of months ago of someone reporting evidence that Einstein's universal constant had changed over time. This was, needless to say, controversial.

I believe some elements of string theory allow for the possibility of changes to things that other branches of physics consider to be constant.
posted by alms at 11:13 AM on May 31, 2006

I think this delves into the scope of existence and even religion.

Many of the foundations of our very life are based upon constants.

For example, the very foundations of Eucledian geometry and through that, many of our concepts of physics and science (which have sublimated into modern technology) are based upon simplistic concepts like right triangles (a^2+b^2 ALWAYS equals c^2). To change this property would change the very fabric of existence.

In my sort of secular/scientific/pseudo-deist view of religion, I believe that God is the entity which forges these constants; the very fabrics of our universe. Imagine an Earth, in a paralell universe where a^2+b^2 does not equal c^2? Their society/progression/existence would be completely different...
posted by stratastar at 1:16 PM on May 31, 2006

knave: Actually, I think your friend's (3) is covered by my (1). God (or God's plan) would simply be the highest, unchanging law in the hierarchy. I think that adopting the position that the laws of the universe change in order to show that the earth is 6000 years old is a bit like adopting the position that we were created yesterday with all of our memories in order to explain why one shouldn't be punished for a crime committed last week. Unless, of course, there are other independent reasons for believing uniformitarianism isn't true.

One of the other key debates encountered here is the nature of possibility. If laws can change, nomological possibility becomes a very weird notion.
posted by ontic at 1:31 PM on May 31, 2006

stratastar, hate to break it to you, but a^2 + b^2 != c^2 is one way we determine that a region of spacetime is curved. This happens around any object with mass, the curvature due to gravity.
posted by knave at 1:42 PM on May 31, 2006

okay, but imagine a paralell universe (as I stated above) where gravitational properties are different. This was my point; our existence is predicated on basic constants in what we are able to percieved as our universe. Minor tweeks to these basic foundations would yeild catastrophically different results
posted by stratastar at 6:49 PM on May 31, 2006

All very interesting answers and links too.

Despite the criticisms of Sheldrake, I thought that was an overall interesting article/link that really didn't seem too controversial (although I am by not at all in a position to be appalled at things like Avogadro's number & Boltzmann's constant being declared fundamental as edd so clearly was).

Without asking this question (or having seen the earlier ask me - thanks meehawl) I would have never come across the fascinating oklo phenomena (there's a good movie based on that if someone wants to write it) or learned about Dirac's Coincidences.

Oddman's response was interesting, but didn't really get to the root of what I was asking. In a sense, that response seemed more about the words used to express concepts, than the concepts themselves. Really, I was wondering if our "approximations" known by such names as "The Speed of Light" could have been different on a universe wide basis at different times in the universe.

In general, the "constant" I can most philosophically throw my support behind is that "everything changes". That's why it would surprise me if there weren't changes over time to these "approximations". Further, I believe all science is open to amendation at a later date as additional empirical data comes in.

Thanks again for the interesting thoughts and answers.
posted by extrabox at 7:40 PM on May 31, 2006