David Gould

I have just read something interesting which attacks the way the 2LOT (2nd law of thermodynamics) is used by some scientists.

Basically, what I read suggested that although it is known that the 2LOT applies to the short range force interactions (collisions) of closed systems when observations of gravitational forces appear to contradict the second law many 'theory fitting' twists and turns are taken in order to force gravity (a long range force interaction of a completely different type) to fit the 2LOT.

They also suggest that there may be a similar thing in biology, perhaps caused by creationist attacks. For example, if something appears to violate the 2LOT in biology, creationists leap on it as proof of God because such a law is inviolable in nature. Scientists then attack the appearance of violation rather than the second part of the creationist's statement.

There is no real evidence that the 2LOT is inviolable in nature. Rather, it appears to be a statistical law that says it is very rare for a closed system to decrease in entropy.

If the 2LOT was not touted as such an inviolable law then it could not be used by creationists at all.

After all, when we come up against something in nature that goes against one of our theories, shouldn't we adjust the theory to fit nature rather than try to force fit nature into the theory?

What do people think of their argument?

fando

The law is correct if we consider only classical physics. Briefly, it states that entropy never decreases with time. The picture is much more complex in modern physics. For one thing, the Heisenberg Uncertainty Principle allows for violations of the Principe of the Conservation of Energy so long as the violations occur on very short timescales. So yes, the law is correct only in context. When you start talking about Quantum Mechanics, M-Theory and the extremes of General Relativity, you start to require more sophisticated views on the nature of these laws.

<a href="http://www.theonion.com/onion3631/christian_right_lobbies.html" target="_blank">A relevant piece of humor.</a>

Jimmy Higgins

Entropy is extremely difficult to link to evolution, I think atleast. One of the interpretations of 2LOT (thank "god" for abbrev.) is that perpetual motion machines are impossible. This can be tested, by starting a machine and measuring energy loss, aka friction and etc...

I don't see how this is even possible with mutations. A mutation happens. How do you quantify that as energy gain? How much did it gain? I mean, energy was required for the mutation to occur, as it was a process.

I'd love to see a representation showing how one single mutation created more energy or order than disorder through quantification. How can creationists be allowed to make an argument without having any numbers?

Actually, entropy was also linked to the Clinton Administration.
<a href="http://the_champions_mls.tripod.com/AP/press/entropy.html" target="_blank">Entropy Increase Linked to Clinton</a>

Which is from <a href="http://the_champions_mls.tripod.com/AP/The_Aardvark_Press.html" target="_blank">The Aardvark Press</a>

Tim Thompson

DG: What do people think of their argument?

Which people? I suspect the ones around here don't think much of it. I think the creationist argument is pretty stupid, but then that's what I think of most creationist arguments anyway.

DG: There is no real evidence that the 2LOT is inviolable in nature. Rather, it appears to be a statistical law that says it is very rare for a closed system to decrease in entropy.

As developed in classical thermodynamics, 2LOT is indeed alleged to be "inviolate". However, the transition to statistical mechanics, and the theory that gases were made of "molecules" changed all of that. The formulation of 2LOT from Boltzmann (Vorlesungen über Gastheorie, Ludwig Boltzmann, 1896 & 1898; Lectures on Gas Theory, University of California Press 1964, Dover Publications 1995) is indeed statistical. The likelihood of returning to a previous state is extremely small, because the analysis has to be done in a multiparameter phase space, as opposed to a simple coordinate space (which represents the physical volume of gas). Simply doubling the physical volume of a gas can increase the volume of phase space by a whopping factor of 10^(3x10^18). With that kind of volume to play around in, you can see why the repetition of a previous pattern (reduced entropy) is really unlikely. Not really impossible, but darned near to it.

Creationists actually use a kind of misdirection or "bait & switch". They start out by claiming the privilege of the 2LOT, but when the open vs closed system aspect is pointed out, they switch to the requirement for some kind of directing program. Not in itself a totally silly idea, but no longer relevnt to 2LOT either, a point quietly overlooked. 2LOT does not refer to any enabling program, but simply to the basic physical essence of the system.

As for short vs long range forces, the Boltzmann formulation for entropy may indeed be incorrect, in the sense that Newtonian physics is "incorrect" because it is only a subset of a more general model offered in special & general relativity. Likewise there is probably a more general formalism for entropy. The idea is only really gaining ground in the last few years, mainly under the pressure of Brazilian physicist <a href="http://www.santafe.edu/sfi/publications/Bulletins/bulletinFall00/features/tsallis.html" target="_blank">Constantino Tsallis</a>. Tsallis, and the Hungarian mathematician <a href="http://www-gap.dcs.st-and.ac.uk/~history/Mathematicians/Renyi.html" target="_blank">Alfréd Rényi</a> have reformulated the mathematics of entropy, Tsallis for statistical systems, and Rényi for fractal systems. There is some reason for believing that the Tsallis entropy in particular is better suited to many physical systems than is the usual Boltzmann entropy (<a href="http://xxx.lanl.gov/abs/cond-mat/0206078" target="_blank">Stability of Tsallis antropy and instabilities of Renyi and normalized Tsallis entropies: A basis for q-exponential distributions</a>, Sumiyoshi Abe, June 2002; <a href="http://xxx.lanl.gov/abs/cond-mat/0204315" target="_blank">Gravothermal Catastrophe and Tsallis' Generalized Entropy of Self-Gravitating Systems II. Thermodynamic Properties of Stellar Polytrope</a>, Taruya & Sakagami, April 2002; <a href="http://xxx.lanl.gov/abs/cond-mat/0201368" target="_blank">On the thermodynamic stability conditions of Tsallis' entropy</a>, T. Wada, January 2002; <a href="http://xxx.lanl.gov/abs/cond-mat/0108184" target="_blank">The world according to Renyi: thermodynamics of fractal systems</a>, Jizba & Arimitsu, August 2001; <a href="http://xxx.lanl.gov/abs/cond-mat/0107494" target="_blank">Gravothermal Catastrophe and Generalized Entropy of Self-Gravitating Systems</a>, Taruya & Sakagami, July & November, 2001). It may well be that the "theory fitting" DG mentions, to account for the difference between entropy for collisions & gravity, is in fact a reference to the development of Tsallis entropy.

In any case, whether one chooses Boltzmann or Tsallis or Rényi entropy, the creationist argument remains just as bad.

[ June 19, 2002: Message edited by: Tim Thompson ]</p>

KeithHarwood

The energy required for a mutation is miniscule. The normal thermal motions of the molecules is more than enough. Remember these are normal chemical processes. You have billions of molecules inside a cell all batting about at high speeds, bouncing off each other and sometimes sticking. The clever bit isn't the creation of defects, it's correcting the errors afterwards. A mutation is a defect that has managed to get past the error correction. But the error correction is itself a chemical process and the energy required for its operation is again small.

The change in entropy required for a mutation is vanishingly small compared to the change in entropy of the organism wherein the mutation occurs.

And don't confuse energy with entropy.

Jimmy Higgins

My bad for making it sound that confusing. My question remains as to actually quantitatively showing how mutations (or apparantly only good mutations as far as creationists are concerned) create more order than disorder. I don't think I've ever seen the numbers from a single creationist. They just play statistics.

You take a pot of water at 60 degrees Celcius. It has a certain disorder. You evaporate the water, you can calculate that disorder increase. The water condenses, you can calculate the disorder increase to the environment. Actually calculate the physical number. Its not done statistically! That is how evolution is being attacked. Isn't this not fair?

David Gould

And isn't it possible that the temperature in the pot might vary slightly upwards at some point - in other words, that there might be a slightly odd configuration of water molecules. How do you account for the possibility of a slight entropy decrease in your calculation? Sure, the chances are small of it happening but it can happen.

Poincaire cycles, i think they are called. Any closed system of things bumping around will come into an ordered state every so often.

Your calculation is simply an estimate of the total entropy increase at that point in time - a bloody good one, as it is unlikely an configuration that decreases entropy by a noticeable amount will have occurred in the time you take to do the measurement but an estimate none the less.

An estimate based on statistics.

Or am I getting something very wrong here?

Jimmy Higgins

David Gould: What I'm asking is for actual numbers from creationists. While it may be possible, though highly improbable that disorder turns into order, I'm not interested in that line of argument. It is simply too easy to refute that through ignorance.

What I'm wondering is whether there is any scientific backing to saying evolution is against the 2LOT on the grounds that they have no grounds to say that. It is just conjecture.
1) Mutations are brought about processes that create disorder, energy entering the system. Just as taking two halves of a deck and putting them together like a dealer in Vegas. The end product has no disorder, however, the process of getting the product there did.
2) You can't really (can you?) numerically specify just how much entropy is introduced by a mutation. Therefore, how can one say a mutation is in violation with 2LOT when they can't even quantitatively saying what amount of order was created.

Am I wrong with this line of thinking?

Bane

I had thought that creationists had given up the 2lot argument but it would seem that they still use it.

This guy is using it here:

<a href="http://boards.theforce.net/message.asp?topic=6682132&page=18" target="_blank">Theforce.net</a>

Here's an excerpt from the article he posted:



Wouldn't a seed or a fertilized egg go against the creationist version of the 2lot? Wouldn't the ability to pass on our DNA get around this problem?

scigirl

Yes, by their definition. If the 2nd law prevents a single cell from evolving into a human in 4 billion years, than it sure as hell prevents a single cell from growing into a human in a mere 9 months. I've asked YECS that same question, and they just mumble about "well the DNA is there." But it doesnt matter - only final and initial states matter in physics! The
the sequence and length of the DNA has absolutely nothing to do with the amount of entropy you have to change to go from zygote to baby.

So clearly, there must be a special creation event every time a baby develops. You'd think then, that this God could do something about crack babies, or fetal alcohol syndrome (that rat bastard).

scigirl