Time's Arrow

[deezelboy]

This is very interesting - Lorenzo Maccone has come up with a rather fascinating hypothesis regarding time flowing backwars (or more properly, entropy decreasing).

From his abstract:

The arrow of time dilemma: the laws of physics are invariant for time inversion, whereas the familiar phenomena we see everyday are not (i.e. entropy increases). I show that, within a quantum mechanical framework, all phenomena which leave a trail of information behind (and hence can be studied by physics) are those where entropy necessarily increases or remains constant. All phenomena where the entropy decreases must not leave any information of their having happened. This situation is completely indistinguishable from their not having happened at all. In the light of this observation, the second law of thermodynamics is reduced to a mere tautology: physics cannot study those processes where entropy has decreased, even if they were commonplace.

http://arxiv.org/abs/0802.0438

The slightly easier-to-understand Guardian piece (worth reading in full for a more measured approach):

Briefly, the problem is that while our laws of physics are all symmetrical or "time-reversal invariant" – they apply equally well if time runs forwards or backwards – most of the everyday phenomena we observe, like the cooling of hot coffee, are not. They never seem to happen in reverse.

We have a statistical law that describes these everyday phenomena called the Second Law of Thermodynamics. This law tells us that the "entropy" or degree of disorder of a closed system never decreases. Roughly speaking, a process in which entropy increases is one where the system becomes increasingly disordered. Windows break, thereby increasing disorder, but they will not spontaneously unbreak. Gases will disperse but not spontaneously compress.

However, entropy describes what happens with large numbers of particles. We presume that it must arise from what happens with individual particles, but all the laws that govern the behaviour of individual particles are time-reversal invariant. This means that any process they allow in one direction of time, they also allow in the other.

So why will your coffee spontaneously cool down, but not heat up?

Maccone's solution is to suggest that in fact entropy-decreasing events occur all the time – so there is no asymmetry and no associated mystery about the arrow of time.

He argues that quantum mechanics dictates that if anyone does observe an entropy-decreasing event, their memories of the event "will have been erased by necessity".

Maccone doesn't mean that your memories will never form in the first place. "What I'm pointing out is that memories are formed and then are subsequently erased," he tells me.

When you observe any system, according to Maccone, you enter into a "quantum entanglement" with it. That is, you and the system are entangled and cannot properly be described separately.

The entanglement, Maccone says, is between your memory and the system. When you disentangle, "the disentangling operation will erase this entanglement, namely the observer's memory". His paper derives this conclusion mathematically.

While we cannot remember our cups of coffee re-heating, and hence cannot study them, Maccone thinks that entropy-decreasing events like that must happen.

"If transformations that increase the entropy do occur – and we know that they do – by symmetry we should expect also transformations that decrease the entropy – but we cannot see them."


http://www.guardian.co.uk/science/blog/2009/aug/26/entropy-time-arrow-quantum-mechanics?commentpage=2

(There's also a piece from New Scientist here, metioning the Many Worlds Interpretation to get round the micro/macro problem of quantum entanglement.)

If I've got Maccone right, it means that we're incapable of observing, either directly or indirectly, whole chunks of physical phonemena.  Or rather, we do observe it, but it is then erased.  Which is very, very interesting (not in the least because Maccone's derived the conclusion mathematically).

Don't think we've got a 'time' thread on the science board yet, so use this to talk about time travel or any other theories of time as well, if you're interested!

[SM]

He argues that quantum mechanics dictates that if anyone does observe an entropy-decreasing event, their memories of the event "will have been erased by necessity".

Maccone doesn't mean that your memories will never form in the first place. "What I'm pointing out is that memories are formed and then are subsequently erased," he tells me.

Sorry, but whenever I read stuff like this it sounds like "A wizard did it".

Like "We've got this theory, but it's got some holes, so we'll make stuff up to plug the holes".  Seem to recall Hawking doing something similar about information loss at event horizons...

While we cannot remember our cups of coffee re-heating, and hence cannot study them, Maccone thinks that entropy-decreasing events like that must happen.

Isn't there a dude with a razor who says "No it just cooled down".  Explain that and don't make stuff like "No it did reheat and you just forgot that it reheated".

Makes me brain 'urt, so it does 'Enry 'Iggins.

[Neltharion]

Is it just me, or does this theory have a religious sense to it. As in, you can't prove me wrong therefore I am right, yet I can't be proven right sorta thing?

[deezelboy]

Isn't there a dude with a razor who says "No it just cooled down".

Yes, but it's there that the problem begins (and this has been a long problem, going back to the 19th Century): why does it just cool down?

What Maccone has demonstrated is that if it heats up again, we would not be able to observe it doing so (not in any recordable way, at least!), so we would end up seeing it as just cooling down.  This is, as far as I'm aware, the first time anybody's tried to plug the hole with a rigourous definition of why things just cool down.  Essentially Maccone's wielding the razor here.

Explain that and don't make stuff like "No it did reheat and you just forgot that it reheated".

That may be a major problem if you wanted to explain something at the quantum level.  From my layman's point of view: things happen, and the cause of that thing happening is then backdated retrospectively to fit the thing that happened.  So with the CI and Many World's interpretations of theSchrodinger's cat thought experiment, when you look in the box and see the cat alive, the cat immediately loses all memory of being in a half-alive/half-dead superposition, and instead remembers being alive during the whole of the experiment.  So it is thought to happen all the time in quantum mechanics - sorry about the possibility of head-explosion, SM (but draw some comfort that if you're still reading on, you've forgotten that your head exploded).

From a classical standpoint, it's a lot easier to explain, but requires some wizadry: the universe starts in a state of low entropy which then increases.  The increase in entropy is what gives us the concept of time.  Why entropy should always be seen to be increasing in the future and decreasing in the past is down to a wizard making it so - there's nothing in our understanding of the laws of nature that says it should be this way, except the continuing observation that it is.

Is it just me, or does this theory have a religious sense to it. As in, you can't prove me wrong therefore I am right, yet I can't be proven right sorta thing?

You can disprove Maccone's hypothesis, though. 

Firstly, you could observe entropy decreasing. 

Secondly, you could prove that entanglement can only occur at the micro- level, and couldn't affect things as large as brains.  To be fair to Maccone, he's made it abundantly clear that this is the major presumption of his paper.

But basically, what Maccone's done is given a proof as to why entropy isn't seen to decrease, and which doesn't violate quantum theory. Whether the universe works this way is anybody's guess, but it can now be said that this is one way it could work, and doesn't violate any known observations, and is supported by known observations and theories.  Bit like all of science, really.

[Pandorag]

I need a babelfish to understand.

[beckmen]

I need a new brain to understand.

[Neltharion]

But you can see entropy decrease. In a small sense, how ever over all Entropy will always decrease, in the grand scale.

Fertilization and Photosynthasis (second one not so sure on though) are shown as decreasing entropy.


The Religious sense comment was the overall tone of the article really.



With the Second Law of Thermodynamics comment, and it becoming a Tautology. A tautology means that all the logical outcomes are true (Mathematical logic at least). So therefore, it confirms it right?

Or did I miss read the article?

[deezelboy]

But you can see entropy decrease. In a small sense, how ever over all Entropy will always decrease, in the grand scale.

Yes, in the small sense - which is to say that entropy is not really being decreased, but shifted outside of one system and into another.  So if I turn on the air conditioning, while the entropy of the room I'm in is being decreased, I'm increasing entropy in another area (for instance, the ultimate energy source of the air conditioner itself).  This is vastly different to the room spontaneously cooling by itself - which should be able to happen, by our understanding of physics, but doesn't.

The Religious sense comment was the overall tone of the article really.

With the Second Law of Thermodynamics comment, and it becoming a Tautology. A tautology means that all the logical outcomes are true (Mathematical logic at least). So therefore, it confirms it right?

Not sure I get what you're saying. 

Yes, it confirms the Second Law, and it reduces it to a tautology - we observe entropy increasing into the future simply because we are unable to observe it decreasing into the future, not because it can't happen that way.  Is that what you mean?

[Neltharion]

Ah alright, I guess I miss read it, cause I thought he was trying to disprove the Second Law.

Yes, in the small sense - which is to say that entropy is not really being decreased, but shifted outside of one system and into another.  So if I turn on the air conditioning, while the entropy of the room I'm in is being decreased, I'm increasing entropy in another area (for instance, the ultimate energy source of the air conditioner itself).  This is vastly different to the room spontaneously cooling by itself - which should be able to happen, by our understanding of physics, but doesn't.

Oh I know that over all entropy will always increase (still a scary thought). Still, when you think about it, the fact we can decrease Entropy with our biological processes is pretty amazing.



But yea, what you said is generally what I was getting at

[SM]

why does it just cool down?

Cos there's nothing to keep it warm?

Sorry, the entropy in my brain just increased and I forgot what we were talking about.

[the_philanthropy]

Oh I know that over all entropy will always increase (still a scary thought). Still, when you think about it, the fact we can decrease Entropy with our biological processes is pretty amazing.

If I understand it correctly...
Entropy always increases, in every single instance. There is no such thing as decreasing entropy. Our "biological processes" may seem like they have a low entropic level, but every living thing produces heat, which in itself is an increase of entropy.

What entropy really describes has not been discussed here yet (from what I can tell). Actually it simply means that when entropy increases, things become more symmetrical (yeah, sounds weird). What I mean by that is this: There are fewer states in which things will be asymmetrical. Consider a gas that expands in a room, starting from a single point. While in the beginning of the process of expanding, changes are very easily noticable ==> asymmetry. But once it has expanded, and is filling the whole room, every single atom may change positions a lot, but it will hardly make a difference for the appearance of the gas cloud ==> symmetry.

And since our laws of physics are symmetrical, wouldn't that also mean, that even if we were to look back in time, entropy would still increase? After all, considering special relativity, every perspective in spacetime is equally valid; meaning: there's no such thing as a definite, universal, "now". Things that are yet to happen from our viewpoint, might already have happened from another viewpoint. Yet, entropy would still have to increase... _{<getting a headache from being too damn nerdy...>}

And then there's quantum physics. Well, what can I say? The whole thing hardly matters in real life. In a controlled environment, e.g. in a lab, it may be possible to isolate single photons or electrons and observe quantum states in them (superposition, etc...). But in real life, these particles are not isolated. They interact with their environment all the time, and therefore the vast majority of stuff in the universe is not really affected any more by quantum states. (This is by the way the only lead to date as to why the arrow of time seems to have a real direction. But I can't confirm that right now and would have to re-read it...)


I think Maccone simply got it all wrong. Sure, if you consider Quantum stuff.... well, let me rant on:
Consider a photon that has travelled 10 light-years to reach a detector. Quantum physics says, that the photon is in a superposition; it has both the attributes of a wave and a particle. But as soon as you measure it, the photon takes on either the state of a wave or a particle, but never both. The photon travelled 10 years to reach the detector, and then takes on either state (due to being measured). Now the big question is: What happened in the last 10 years? Symmetry dictates, that its history must change accordingly. The photon was always(last 10 years) either wave or particle. So, yeah. Considering this, the photon does infact "leave a trail behind", in contrast to what Maccone thinks.

Yes, before the Big Bang our universe was in an unusual state of minimal entropy. But ever since, entropy has increased, and always will. The only thing he's right about is that we cannot remember what happened before the BB, which is a non sequitur. Science can't answer that question anyway, nor can anyone else (at least not in this universe).

Maccone is just a poser, who wants to get some attention by explaining something that never happened by invoking some mysterious BS.

As was stated before, the brain cannot be in a state of superposition. It's too complex a system. Maybe Maccone's brain is special, though. Maybe it really is either dead or alive. We just don't know yet...

[Neltharion]

If I understand it correctly...
Entropy always increases, in every single instance. There is no such thing as decreasing entropy. Our "biological processes" may seem like they have a low entropic level, but every living thing produces heat, which in itself is an increase of entropy.

As I said, Entropy increases always. But the fact we can create order from disorder (which is basically reversing Entropy, yet not a net decrease in it) biologically is an amazing thing.

[the_philanthropy]

As I said, Entropy increases always. But the fact we can create order from disorder (which is basically reversing Entropy, yet not a net decrease in it) biologically is an amazing thing.

well, think of it this way: every time you move you spent energy. The energy does not get lost. It gets "translated" into heat, if minimal. Every time you, or, for that matter anything else, dead or alive, moves there's a certain friction; again heat. Entropy describes change. Without entropy, there would be no time to measure, because nothing would change.
What I'm trying to say is that it's not really a net amount we're talking about here. Every thing that assembles order (as perceived) actually creates more entropy, all the time, or it would not be able to exist in the first place. I know it's hard to wrap your head around it. It is a weird thought, after all... 

[deezelboy]

Considering this, the photon does infact "leave a trail behind", in contrast to what Maccone thinks.

That's not what Maccone's saying.  He's saying that within a quantum mechanical framework, all phenomena which leave a trail of information behind (and hence can be studied by physics) are those where entropy necessarily increases or remains constant.

If you take your example with the photon, the energy of the photon leaves a mark on a detection screen, which involves an increase of entropy.  Thus it leaves an information trail - the mark on the detection screen itself (which lets us know the position of the photon at the point of impact, although not its trajectory).

If entropy decreased in the above example, the energy that caused the mark on the detection screen would become a photon, and the mark would necessarily disappear.  Without the mark, we don't have an information trail, and thus we have no way of knowing whether this had happened.

As far as I can see from his paper, Maccone goes one step further, and states that if we consider that quantum systems can be extended to the macroscopic world, it would follow that all observations of that mark on the detection screen would need to 'disappear' as well.  So even though we saw the mark at 5:00pm, and the entropy of the system decreased in such a way as to cause the mark to turn into a photon at 5:15pm, we'd have no memory of seeing the mark after 5:15pm.