Entropy and evolution

This "dispersal", as Boltzmann and others showed, always ends as heat. No matter how you do it, the end is always heat unavailable for work.

"The Second Law of Thermodynamics says that (again in a closed system) energy goes from a usable form to a less usable form. Things run downhill, flywheels slow to a stop, useful kinetic energy becomes useless heat. No physical process, no machine, is ever 100% efficient. Energy is not lost, but for all practical purposes, it is lost, as it is no longer usable."

http://www.jimloy.com/physics/2ndlaw.htm

If so, then it would be impossible for mountains to be lifted up, or plants to grow, or babies to be born, or populations to evolve, or even new information to evolve within individuals.

And we see all of those things going on. Local and temporary (on the scale of billions of years, sometimes) decreases in entropy are everyday occurances.

You can relate information to entropy, by treating it like Boltzmann, as a statistical problem.

Statistical entropy is a measure of uncertainty or ignorance as a probability; information measures the reduction of uncertainty.

What Shannon showed was that, within the constraints of the channel, one can reduce uncertainty as much as one choses, and thereby reduce the "entropy" of the system.

It does not rule out increases in information by natural selection. Even random events can produce an increase in information.

 

Yep. It make take billions of years for a photon to zip across space and finally hit a body that absorbs it and emits the energy as heat, but that's what happens.

Only because it spreads out. The photons lose none of their energy, unless they are absorbed by some substance and turned to heat.

Light is absorbed by objects, which become warmer thereby.

No.

Just making an observation about entropy.

Tell me how entropy works in terms of information theory.

Barbarian asks:
If so, then it would be impossible for mountains to be lifted up, or plants to grow, or babies to be born, or populations to evolve, or even new information to evolve within individuals.

Reduction in entropy. Remember what entropy is. A lack of useful energy. Potential energy is quite useful. It takes a reduction in entropy to make mountains.

And yet, in making them, potential energy is increased. How can this be so?

Just pointing out that orogeny is one of many natural processes that result in a decrease in entropy.

I know you believe it, but the sticking point is an absence of evidence. Are hurricanes also designed to use energy? It looks like it's explainable in very simple physical laws, and het we see a highly complex, fractal structure to hurricanes.

Even if God created your life by natural means, it doesn't mean that He didn't create you. Our bodies are natural, but our soul is not. The Wonder is not in some little fertility demiurge, but rather in the creation of a universe in which nature so fully serves the purposes of God, at His direction.

Let's see... bacteria evolving additional genes for various things, humans evolving new alleles while keeping the information of the old ones (duplicate genes). That sort of thing. From where did you think that added complexity comes?

A new gene, the old retained. That's pretty obviously new information.

Barbarian observes:
And we see all of those things going on. Local and temporary (on the scale of billions of years, sometimes) decreases in entropy are everyday occurances.

Hurricanes are alive? I don't think so. And yet, we see a decrease in entropy there.

Nothing could be more natural than life. God does almost everything by natural means in this world.

Biologically complex molecules don't last long when bacteria are about. But since God tells us in Genesis that the earth and waters brought forth life, I believe it.

And yet we see mountains thrown up, and hurricanes, and many other non-living systems showing a decrease in entropy. Just for the record, local decreases in entropy are perfectly normal and often observed in non-living things.

(Barbarian discusses the application of entropy to information)

Yes, entropy in information is independent of meaning. It is onlyi related to the uncertainty of the message. It could be gibberish, and yet low in entropy.

I suppose you could argue that information was not relevant to living things. But it appears that you have misunderstood what "information" is, and how it applies to living things.

Nope. For example, radio telescopes can pick up a great deal of information about distant objects by the bits coming over the channel. And yet there is no intelligence involved in sending them.

Not unless you think God personally starts and shuts down hurricanes.

In the long run, the sun will dim, and then grow, and vaporize the Earth, and then finally burn out to a cold cinder of matter. And then evolution is over. But as long as local decreases in entropy happen, it works just fine.

Barbarian observes:
It does not rule out increases in information by natural selection. Even random events can produce an increase in information.

But as Shannon observed, deleting specific bits can actually reduce the entropy of a signal. You're confusing bits with the message.

At that point, you are not talking about information, but something else. It is not necessary for information to be meaningful.

 

This thread gotten off to a rather silly start.

Helen, it is pickayune for you to say that Galatian is wrong to say that entropy is about heat. While it is true that one can define entropy from a microscopic stanpoint without doing so in terms of heat, our discussion here isn't from a microscopic point of view. Even if you did want to speak from a microscopic point of view you'd be flat wrong (not just a little bit off point) to say that a definition of entropy not having to do with heat per e has to do with energy. From a microscopic point view, if for each microstate of the system i there is probability p_i that the system occupies that microstate, then the entropy of the system is simply k_B * (Sum over i) - p_i ln(p_i).

That's all. No energy mentioned. It is true that at thermal equillibrium all microstates having the same energy are required to have the same probability of being filled, but in general that is not so.

The usual macroscopic definition of entropy (which is equivalent to the microscopic one, albeit it's hard to show) is that an infinitesimal change in entropy of a system undergoing a reversible process is qiven ty dS = DQ/T, where DQ is the invinitesimal amount of heat absorbed by the system and T its absolute temperature.

While one may approximately and vaguely say that entropy has to do with energy dispersal, that isn't its definition.

It is not correct to say that the higher the temperature of storage of energy the lower the entropy. For examplee, a system consisting of a powerful chemical explosive forming nearly perfect crystals at a temperature near absolute zero has lots of energy, but very little entropy. What you might be trying to articulate here is that a given amount of heat flowing from a high temperature reservoir causes less entropy flow than the same amount of heat flowing from a lower-temperature source. The difference between storage and flow of energy is important.

It is absolutely not true that it is natural processes that are characterized by dispersal of energy. If we adopt Adkins' and your approximation of entropy as sorta-kinda dispersal of energy, then ALL processes result in nondecrease of net entropy and thus nonnegative net dispersal of energy. That in no way means that all parts of a system must disperse energy in either natural or nonnatural processes. (Thermodynamics knows absolutely no distinction between "natural" and "nonnatural" processes. Such terms are entirely foreign to the vocabulary of thermodynamice.)

There is, of courxse, no conflict between evolution and thernodynamics. Even if one were to concede that evol ing organisms somehow concentrated, rather than dispersed energy, the 2LOT would be violated only if there was negative net dispersal of energy. That is, organism(s) plus surroundings would have to decrease in total entropy. That isn't required in any way by evolution.

Galatian: You should not say that the end result of dispersal of energy is "heat." Heat is a process, not a form of energy or substance. When an object is "heated" its internal energy may increase (depending on how much work it does or how it may gain and lose energy in other ways.) But internal energy is not heat, although people who shoud know better make this same mistake. A "hot" body has more internal energy than a cool body of similar composition; but it has not more heat. For all we know its higher internal energy might have been the result of much work being done on it.

Itwould be OK, Helen, to speak of a role for entropy larger than that given by the second law of thermodynamics if you were discussing some nuances of microstates of systems (statistical mechanics.) You're not doing so. The only known rule for nondecrease of entropy over time is the second law of thermodynamics. You might wish there were some other super law. None is known to physics. The law of nondecreasing entropy IS the second law of thermodynamics. There is no vague, feel-0good fuzzy spirit of the law extension. It's all there in thermodynamics or (from a microscopic perspective) in statistical mechanics.

Helen, you say that natural selection deletes information via death. While there is a connection between information and entropy it cannot possibly be discussed here or anywhere with you until you have a better basic grasp of what entropy is.

In summary, Galatian has made a minor slipup. Helen receives a grade of F for her poor essay. I would suggest to Helen that she read Adkins with particular view of understanding the mathematics of entropy and thermodynamics, rather than concentrating on his prosaic description of entropy. It isn't uncommon for students to wonder what the physical meaning of entropy is, and the mathematical definition(s) can be at first off putting. Nevertheless, as a student of thermodynamics matures he or she needs to realize that the definition of entropy IS in the mathematical definition(s) and that prosaic descriptions like Adkins' "energy dispersal" are at best literary pictures to help build, however imperfectly, students' physical intuition.

 

(Barbarian observes that all radiation will eventually reduce to heat)

Actually, it does; the energy is just changed. Energy is never lost or gained. This is the way the greenhouse effect works, for example. Think of it in the same way you would consider potential and kinetic energy to be essentially the same thing, just in a different form.

As in all processes, much energy is wasted. However, that is beside the point. As you now see, orogeny is a local decrease in entropy, and all perfectly natural.

C'mon, Helen. If we have time to build mountains over tens of millions of years (as in the Himalayas), there is time for evolution. Fact is, the Earth has been around for about 4.5 billion years, with local entropy decreasing on that scale (largely from the Sun's input, but also from radioactive decay in the Earth)

Belief is one thing; evidence another. That's what's missing.

Mountains and hurricanes are living and designed? I don't think so.

You were the one who said such decreases in entropy could only happen in living systems. That's demonstrably wrong.

Destroy a system and then make it work? That makes no sense, Helen.

I can't see where you got that idea. Perhaps you could be a little more specific?

Hmm... weren't you just the one who said I was arguing for evolution to support evolution?

Why would anyone want to do that? From time to time, though, a new protien evolves in an organism that does something useful. I gave you a few examples. None of the things you predicted happened in those cases.

And yet, the bacteria, being too stupid to know it's impossible go merrily on, evolving new information, such as instructions for metabolizing nylon, and making irreducibly complex enzyme systems.

But then humans, who should know better, are doing it too. One of them evolved a new enzyme that makes them resistant to arteriosclerosis, and passed it on to his descendents. Foolish man, not realizing he had done the impossible.

Franceschini G, et al. (1980) "A-IMilano apoprotein. Decreased high density lipoprotein cholesterol levels with significant lipoprotein modifications and without clinical atherosclerosis in an Italian family." J Clin Invest. 66, 892-900.

(assertion that information, in the scientific sense, is not what living things do)

No, that's wrong. There's been a great deal of study on information in living things, and how it works. I would suggest Jeremy Campbell's "Grammatical Man", which explains it in a very clear and non-technical way.

Barbarian observes:
Nope. For example, radio telescopes can pick up a great deal of information about distant objects by the bits coming over the channel. And yet there is no intelligence involved in sending them.

Wouldn't matter. For example, the distribution of radiation over a wide band, falling on the earth, provides the information for hurricanes. Hurricanes aren't intelligent, or even alive, but they can use that information to form, grow, and move in a directed way.

(more assertion that Helen's version of "information" is somehow different)

There's not much use in this line of thinking. The only information needed for systems like living things, is Shannon information. Adding your own conditions just makes it irrelevant to information and living things.

It helps when Helen is so far in disagreement with the scientific community as to the nature of "information". That's pretty much a gimme as far as refuting it.

Would you say that the new enzyme that evolved in those bacteria to metabolize nylon oligomer was "intrinsically meaningful"? If not why would it matter if it was or not?

We know you believe that. But since we have examples of new information evolving, and since we have seen that all sorts of living and non-living systems show examples of reduced entropy over long periods of time, it's clear that evolution is not ruled out by thermodynamics. Boltzmann, BTW, was a supporter of Darwin. I'm sure he knew more about thermodynamics than both of us combined.

Barbarian observes:
But as long as local decreases in entropy happen, new enzymes will evolve, mountains will form, and so on.

Hmm... moving goal posts, again. They were examples of decreases in entropy creating complex systems. However, I don't see how you think the evolution of a useful new enzyme cannot be anything but what it is.

It's still that lack of evidence holding you back.

 

Helen,

Those experts who write textbooks on thermodynamics disagree with you about your claim that thermodynamics is a barrier to evolution. I wonder if you would comment on how your judgment is better than the three experts whose comments appear on the following site:

http://www.ntanet.net/Thermo-Internet.htm

 

Helen, your beliefs, and those of Gitt are diametrically opposed to the findings of almost all scientists who have studied these things.

I would, however, ask you if the new enzymes that evolved in those bacteria, that permitted them to metabolize nylon oligomers, indicated the evolution of new "information". If not, from where did the information come?

 

>>>>>>Entropy has a wider general definition which is not necessarily connected with heat per se. It has to do with energy.<<<<<

No Helen. You need to look at the definition of entropy in thermodynamics. It is defined precisely as a function of heat.

>>>>>>>Second, I did not argue about local decreases in entropy; I argued about the universal trend toward increased entropy mitigating against evolution.<<<<<

Well the authors of textbooks on the subject disagree with you on that.

 

>>>>>>You then started discussing the Second Law of Thermodynamics. I was not discussing that. I was discussing the larger 'rule' of entropy.<<<<<

The second law and entropy are closely connected -and there is no "larger rule of entropy" as you claim. It is a figment of your imagination.

 

Gitt, as you might know, insists that in order to qualify as information, it must be "useful".

This odd requirement puts Gitt at odds with information theory, and veritually every other scientist who studies it.

He is, as you might imagine, a creationist.

 

I>>>>>>>t was also Atkins who stated that "if it [energy] is stored at a high temperature, then its entropy is relatively low, and its quality is high. On the other hand, if the same amount of energy is stored at a low temperature, then the entropy of that energy is high, and its quality is low." (p. 38). As I said, please argue with him, I just referenced him as an expert so I could make my point.<<<<<<<

Atkins' statements as quoted above come directly from the definition of entropy, where the equation has energy divided by absolute temperature. I think it is important to note that Atkins, along with virtually all experts on thermodynamics, does not accept your argument that entropy or the 2nd law of thermodynamics is a barrier to evolution. As far as I can see, the statements that you quoted from Atkins are entirely consistent with the views of other experts in thermodynamics and are also consistent with the view that there is no conflict at all between thermodynamics and entropy on one hand and evolution on the other hand. If you think Atkins supports your views, then I think you are mistaken.

[ June 04, 2003, 02:24 AM: Message edited by: Peter101 ]

 

Oh Goody! Back to Entropy! Well, why not?

Now I know its a matter of dispute that evolution theory is enough to account the variety of life as we know it. Be that as it may, it is also alledged by some that the second law of thermodynamics - that is, entropy must increase - somehow precludes evolution from working. So lets look at the basic elements of evolution theory and see where 2Lot stops evolution in theory. (if it does).

1. A given species reproduces over time and begets a new generation.

Hmmm. Nothing against 2lot there. At least, we know that happens.

2. Mutations sometimes occur in the genetic structure.

Hmmm. We know THAT actually happens.

3. Some of the mutations hinder reproductive success, some don't make any difference in reproductive success, and some help reproductive success.

Hmmm. That's all the possibilities with regard to reproductive success, all right. Nothing against 2lot there, anyway.

4. Reproduction occurs and the mutations help or hinder the situation according to what they are.

Hmmm. Seems like plain common sense so far.
Helping genes help, hindering genes hinder, and there's more helping genes in the next generation, fewer hindering.

5. In this manner, helpful genes accumulate until there are enough changes to make a noticeable evolutionary change.

How can they help but accumulate?

OK Now regardless of whether or not this theory actually works in real life to allow evolution to occur, boys and girls, your challenge for this exercise is to pinpoint somewhere in these steps wherever there is a violation of the second law of thermodynamics.

Of course, there isn't any such violation anywhere along the way. The 2Lot argument against evolution is wrong.

AND NOW FOR A SHORT DISCUSSION ABOUT INFORMATION.

Evolution requires a gain in information over time in some sense. How does it happen? Here's how.

A given form of life has a given amount of information - the genetic code - which describes how to make the next generation.

The genetic code gets altered randomly from time to time by mutations.

So far, that is not a gain in information. That is a degredation of information.

BUT - we go ahead with reproduction for a few generations. NOW we go back and survey the mutations. Some of them have been reduced or even eliminated in the struggle for existence. Others - a very few that by chance had the property of facilitating reproductive success - have become more common by virtue of that property of facilitating reproductive success.

And now, at this stage, we have gained information. We have new information in the genetic code about what will work to better facilitate reproductive success.

This is how information can accrue over time in the genetic code.

 

I'm going to bed. Thanks all, for a really stimulating discussion tonight, especially Helen.

 

>>>>>Mine and Atkins' and the people at the journals who recommended his book....<<<<

The above was posted by Helen about whether or not there is a "larger rule of entropy". I don't think there is Helen. That is, I think entropy is defined by the second law and there is no rule more general or expansive than that provided in the second law. If you can quote a bona fide expert to the effect that there is some larger rule, as you call it, then do so. Be prepared to show in detail why it prohibits evolution. I know of three textbooks authors, on thermodynamics, who are on record as refuting your ideas, so maybe you can find some textbook authors who support your ideas - maybe. Remember that the 2nd law provides what amounts to a loophole for living systems, which allows them to take in energy from the outside and to, at least temporarily, decrease in entropy if necessary. Your problem all along has been that you try to make the Second law, and entropy, more sweeping than it actually is. Mainstream science, including many experts in thermodynamics, does not agree with you. Atkins does not agree with you, as you yourself have admitted. So you must not try to portray Atkins as supporting your view, because he does not.