Afghan

I came across this interesting article on the BBC website about some Australian scientists who have, supposedly, disproved the Second Law of Thermodynamics. It can be found <a href="http://news.bbc.co.uk/hi/english/sci/tech/newsid_2135000/2135779.stm" target="_blank">here</a>.

I am not convinced it is quite as exciting as it sounds but it reminded me of a point that I've been wondering about for some time. I don't know a good deal about entropy so you'll have to forgive if I'm being dense. The 2nd Law says, I think and paraphrasing crudely, that the current state of affairs is more likely to develop into a more disordered state in the future than an more ordered one.

Does this also imply that the current state is more likely to have developed from a more ordered state than a more disordered one? Simplistically, it seems as though it should but it occurs to me that of the possible previous states there are a hell of lot more disordered ones than ordered ones. Conditional probability and whatnot would seem then to imply that it was more disordered state that the current one evolved from, if you see what I mean.

So my question boils down to this, does entropy work backwards as well as forwards?

Abacus

My understanding of the 2LOT was that it was believed to be a statistical law, not an absolute, dynamical, physical law. I don't know why that article says otherwise.

That being said, this is the first time that an entropy decrease has actually been observed.

There is another thread on this somewhere. I'll post a link to it if I find it.

Here it is:
<a href="http://iidb.org/cgi-bin/ultimatebb.cgi?ubb=get_topic&f=58&t=001094" target="_blank">2LOT</a>

[ July 18, 2002: Message edited by: Random Number Generator ]</p>

Steven S

I posted this same thing on the evo vs. creation section but got some flack for doing so. Anyway, no this doens't invalidate the 2nd law. The 2nd law is statistical in nature and Evans and company apparently showed theoretically under what time scale you can have systems decreasing their entropy rather than increasing. This was then tested experimentally which is what the BBC article describes. As I'm somewhat inebriated right now I'll say no more.

Steven "still trying to prove you can constructively think about physics under the influence" S.

Afghan

My gut feeling was the 2nd Law held but my query is a little more general. Do the laws of probability and thermodynamics predict that the current state of affairs are more likely to have developed from a low entropy state or a high entropy one?

It seems to me that there are a whole host of possible states that the present one could have developed from and that most of them will be high entropy ones. Whilst it is unlikely that a low entropy state will develop from high entropy one it is also not particularly likely that this specific state will develop from a low entropy state.

I suppose what I'm asking is whether entropy is time symmetric or not.

Bill

Yes, the Second Law is statistical, and that is a very important point to keep in mind. As Stephen Hawking emphasizes in his book <a href="http://www.secweb.org/bookstore/bookdetail.asp?BookID=848" target="_blank">The Theory of Everything: The Origin and Fate of the Universe</a>, there is no barrier in place to prevent all of the parts of any given system from randomly aligning into a state of order. And since it is a statistical law, the smaller you make the system, and the shorter you make the time frame for ascertaining whether or not "order" exists, the more likely it is that some random system state will be in a state of greater "order" than before on at least some measurements.

The bottom line is that I would not call this a revolutionary discovery by any means, and in fact, the Second Law is NOT disproved by this scientific experiment. Frankly, the scientists who claim that it is disproved have now destroyed every ounce of credibility that they might have enjoyed due to their making this clearly ludicrous claim. No, we observe entropy decreasing all the time. It is the nature of evolutionary life that it represents a local decrease in the entropy of the area within which it evolves. The key is that we all recognized that no such system is in any way "closed."

And this raises the real issue: what is the proof that the "beads in solution" are actually a truly "closed" system (within the meaning of the Second Law)? I could think of several reasons why the system might not be "closed" at all, thus explaining the apparent decrease in entropy.

In the classical example of entropy decreasing on Earth due to the evolution of life, the Second Law is not violated because of the vast quantity of "input power" transmitted from the Sun to the Earth. You have to take that power into account if you are performing an analysis of the entropy of our Solar System. The Earth cannot be examined in isolation!

And if the entire Earth cannot be examined in isolation, then why should I believe that some bead that has been zapped with a laser and which is floating in a liquid is in some way "isolated" from its surroundings? Was there a quantum physicist on the staff of this experiment? Could the variations they saw be explained by the Heisenberg Uncertainty Principle? Could there be some other electromagnetic effect at work here that makes these scientists believe that entropy is decreasing even when it is not?

To make a claim of this sort and not have "all your ducks in a row" is the height of irresponsibility. And it appears to me that they knew all along how controversial their claim would be, and yet they would appear to have failed to adequately support their claim. While I am generally in favor of giving the benefit of the doubt to science, in this case, I believe there is strong reason for some healthy skepticism.

== Bill

Shadowy Man

This is the same kind of exaggerated journalism as when the "speed of light was broken" a few months ago. Science reporting always get overblown.

Mike H

As Shadowy Man noted, science news always gets overblown. Having read the BBC article, in all fairness to the group from ANU, they were quoted only once and the quote was nothing along the lines of "We've violated the second law of thermodynamics." The few press releases I've read have had no such claims by the ANU group.

Since work was been a bit slow today, I dug up some of the references. Seems interesting enough, nothing too incredible being claimed that I found on a brief, perfunctory skim. The webpage of the senior author (along with various papers, including the current one) can be found <a href="http://rsc.anu.edu.au/~evans/" target="_blank">here</a>.

Michael Ledo

I have my doubts. First off the system is so small, it is difficult to measure, and the "effect" only lasts for a tenth of a second. There is of course no proof anything on the atomic scale has or can act this way, which is really what the whole creationist debate is about.

Since we are dealing with small particles, than any energy that can be but into a system would make the experiment null. In this case something like cosmic radiation or a neutrino might invalidate the claim, since their presence could change the outcome in a small system. Note larger sytems where cosmic radiation would not be a major factor does not exhibit the quality.

Abacus

I'm curious now. Has anybody ever calculated the entropy change due to life? Or do they just look at a full grown Chimpanzee and say, "Well, obviously it is more complex than a zygote."

Perhaps what I should have said is that this is the first time that I have heard of an entropy decrease being calculated in a real system. And it must be calculated, because there is no such thing as an entropymeter.

tronvillain

As I said on the other threads, since the second law is statistical, this is not a surprising result. In an small system over a short time period, entropy may decrease measurably. Something similar could happen in a large system over a long time period, but the chances of it happening are infinitesimal.

Afghan:
The laws of probability, and hence the laws of thermodynamics, preducts that the current state of affairs are far more likely to have developed from a low entropy state.

While it is not particularly likely that this specific state will develop from a low entropy state, is even less likely that this specific state will develop from a high entropy state. Well, that's not necessarily true, since it is entirely possible that the probability of this specific state developing from the previous state was one. Anyway, the odds of entropy decreasing on a macroscopic scale are so vanishingly small we ignore the possibility.

Well, entropy tends to increase over time, if that helps.

[ July 18, 2002: Message edited by: tronvillain ]</p>