Anti-Evolutionists Need to Stop Talking About Thermodynamics

The anti-evolutionists just never get tired of the second law thermodynamics! The latest bit of silliness comes from Barry Arrington, writing at Uncommon Descent. Here's the whole post:

I hope our materialist friends will help us with this one.

As I understand their argument, entropy is not an obstacle to blind watchmaker evolution, because entropy applies absolutely only in a “closed system,” and the earth is not a closed system because it receives electromagnetic radiation from space.

Fair enough. But it seems to me that under that definition of “closed system” only the universe as a whole is a closed system, because every particular place in the universe receives energy of some kind from some other place. And if that is so, it seems the materialists have painted themselves into a corner in which they must, to remain logically consistent, assert that entropy applies everywhere but no place in particular, which is absurd.

Now this seems like an obvious objection, and if it were valid the “closed system/open system” argument would have never gained any traction to begin with. So I hope someone will clue me in as to what I am missing.

I think Arrington is missing quite a lot, actually.

Let's start with the obvious. Many physical laws and theories only strictly apply to idealized scenarios, but that does not stop them from being very useful. There are no ideal gases in nature, but we have an ideal gas law that tells us how they behave. Physical objects never engage in perfectly elastic collisions, but classical mechanics tells us quite a lot about what would happen if they did. Heck, there are no triangles in nature, but trigonometry is still fantastically useful stuff.

So, yes, the only truly closed system is the universe as a whole, a fact pointed out in virtually every book on thermodynamics. But there are many systems that are close enough to closed for practical purposes, and that is enough to make the second law very useful indeed.

(Incidentally, for the purposes of this post I won't belabor the distinction between a closed system and an isolated system. The former refers to one where no mass is crossing the system's boundary, while the latter requires that neither matter nor energy is crossing the boundary. If you are making the statement, “Entropy cannot spontaneously decrease,” then you had better be talking about an isolated system. While we're at it, for the purposes of this post I will be discussing everything in the context of classical thermodynamics. I will not discuss statistical mechanics or anything like that.)

The bigger thing that Arrington is missing, however, is that there is so much more to the second law than the statement that entropy cannot decrease in an isolated system.

One frustration in learning about thermodynamics is that you can consult a multitude of textbooks and popularizations and never find the second law stated the same way twice. Sometimes it is boiled down to the simple statement that heat always travels from a hot body to a cooler body. Sometimes it is expressed in terms of heat engines. Sometimes it is presented with an impenetrable amount of mathematics. Making things worse is that it is very hard to pin down what, precisely, entropy is. That's why you get a lot of talk about complexity, or randomness, or useful energy, in popularizations of this topic. These ideas capture some of the spirit of the concept, but they also fool a lot of people into thinking they know what they are talking about.

When creationists first noticed that the second law could be used to rhetorical advantage, they tended to do so in a shockingly naïve way. For example, here's Henry Morris, from his book The Troubled Waters of Evolution:

Evolutionists have fostered the strange belief that everything is involved in a process of progress, from chaotic particles billions of years ago all the way up to complex people today. The fact is, the most certain laws of science state that the real processes of nature do not make things go uphill, but downhill. Evolution is impossible!

And later:

There is ... firm evidence that evolution never could take place. The law of increasing entropy is an impenetrable barrier which no evolutionary mechanism yet suggested has ever been able to overcome. Evolution and entropy are opposing and mutually exclusive concepts. If the entropy principle is really a universal law, then evolution must be impossible.

Now, when creationists are saying things like that, it is perfectly reasonable to emphasize in reply that the second law only precludes spontaneous decreases in entropy in isolated systems, which the Earth certainly is not. But that statement is hardly the entirety of what physicists know about entropy.

To fully understand the magnitude of what Arrington is missing, we should consider what the second law was accomplishes. The principles of thermodynamics make certain claims about what sorts of processes are physically possible. The first law asserts that energy must be conserved in a thermodynamical process. Energy can be inventoried, you see, and if the tally at the end differs from the tally at the start, then your hypothetical process is not possible. In practical situations, if your tallies do not match then you have probably overlooked some source of energy. It was from such reasoning that neutrinos were discovered, but that's a different post.

The trouble is that there are many hypothetical processes that, empirically, never happen, but which are not ruled out by the first law. For example, heat is never seen to flow spontaneously from a cool body to a warmer body, but such a thing is not precluded by the first law. This is where entropy comes into it.

As I said, it's very hard to say what, exactly, entropy is. What matters, though, is that, like energy, it can be inventoried. The second law then makes a definite statement about the manner in which entropy can change as the result of a thermodynamical process. Here is that statement:

$latex \Delta S \geq \int \dfrac{dQ}{T}$

On the left we find the change in entropy as a result of the process in question. Fully understanding the integral on the right takes some doing, but there are really only two things we need to know for the purposes of this post. The first is that this inequality applies to any sort of system: open, closed, isolated, whatever. The second is that in an isolated system the integral is quickly seen to evaluate to zero. In this special case, the second law reduces to the statement that the change in entropy in an isolated system cannot be negative, which is to say it cannot decrease.

And for an open system? In that case the integral might very well evaluate to something negative, in which case the change in entropy can be negative as well. Which is to say that in an open system the entropy can, indeed, decrease.

So it's pretty clear that Arrington has missed rather a lot. His view of the second law is so impoverished that he has missed most of what the law actually says.

Can we save the anti-evolutionist argument by showing that evolution would violate that inequality? Good luck with that. If you figure out how to determine the entropy of an elephant, much less an entire biosphere, you will have accomplished something very impressive indeed. In practical situations, that integral can be very hard, or impossible, to evaluate. As it happens, some physicists have attempted to estimate the change in entropy on Earth as a result of evolution, bending over backwards to make extravagant assumptions that are harmful to the evolution side. Their consistent finding is that, even given those assumptions, the change in entropy is many orders of magnitude smaller than what the second law precludes. Feel free to challenge their details, but that is what you are up against if you want to revive the creationist argument from the second law.

Confronted with these basic facts about thermodynamics, anti-evolutionists respond in a number of ways. Some, like Arrington's blogmate Granville Sewell, try mockery. Sewell likes to talk about the “illogical compensation argument” that evolutionists invoke to avoid conflicts with the second law. By this he means the idea that spontaneous decreases in entropy are possible so long as they are offset by greater increases elsewhere. Here on Planet Earth, what Sewell describes as illogical is understood simply as a straightforward consequence of what the second law says.

Others prefer something like this: “So what if sunlight is entering the system? How does that explain anything? If the parts of a 747 are lying disassembled in a junk yard, shining a light on them won't cause them to make a functioning jet! Hahahahaha!” This is deeply silly, of course. There's a big difference between showing that an hypothesized natural process does not violate the second law, and showing that it is reasonable to believe that it happened. Consistency with the second law is a very low hurdle. In principle there might be all sorts of reasons for challenging theories of evolution or the origin of life. The point for now, though, is that those reasons will not have anything to do with the second law.

The reason anti-evolutionists seem so impervious to the elementary points I have made here is that, for all their blather about the second law and entropy and whatnot, their arguments generally have nothing to do with thermodynamics. They almost never employ any of the machinery that you see laid out in the textbooks. They are really just saying that they find it hard to believe that evolution can cause complexity to increase. The thermo-jargon is just a way to make the argument from personal incredulity sound scientific. So, when a scientist explains the facts of thermodynamical life to him, the anti-evolutionist does not hear a reply to the argument he is really making. And that is why he will usually retort with a non sequitur.

Anti-evolutionists really need to stop talking about thermodynamics. Every time they do they just show the world why scientists are so contemptuous of them.

More like this

Come on JR, you know better than

"The reason anti-evolutionists seem so impervious to the elementary points I have made here is that, for all their blather about the second law and entropy and whatnot, their arguments generally have nothing to do with thermodynamics."
Just replace thermodynamics with "science". Hence your header should be

"Anti-evolutionists need to stop talking about science."
Indeed The Sensuous Curmudgeon has pointed out that the IDiots from Seattle move in that direction. This year they have more or less admitted that they talk about religion.

Any working scientists here, feel free to tear the following to shreds if I've screwed up somewhere:

The definitions of entropy I'm familiar with are:
a) a measure of dissipation of energy from a source to a sink.
b) a measure of potential configurations of bits.
c) a measure of increase of disorder.

As for organisms & evolution, that part's "easy."

Organisms are open systems that gain complexity by harvesting ambient energy. Sunlight is absorbed by plants, converted via photosynthesis to plant matter and energy to drive plants' metabolic processes. Plants are eaten by animals: stored energy from plant matter (e.g. carbohydrates) is converted to energy for use driving animal metabolism. Animals eaten by animals: same with another conversion step or two along the way.

Another term for these systems is "dissipative structures," which include nonliving chemical reactions that gain complexity by utilizing ambient energy. The term "free energy" is another term for "ambient energy."

The complexity gain in evolution is nothing more than increase in adaptive mechanisms.

The interesting puzzle is, at what point does complex chemistry cross over to biology? But to my mind, the answer to that doesn't require a deity or whatever supernatural "snap of the fingers." It's just a logical progression from complex chemistry to self-reproducing chemistry. One of these days we'll find the points at which the reactions cross over and start reproducing.

Wild speculation department: I suspect the crossover from chemistry to biology will be a kind of gray zone of intermediate steps rather than a sharp clear demarcation. Prions are an example of part of what's in that border zone. They appear to "reproduce" by causing other proteins to change shape from "normal" to "prion" configuration. That appears to be nothing more than a change in topology as it were, configuration of bits, for which there's no inherent thermodynamic penalty, thus no need for an increase in the energy conversion capability of the proteins to get from non-reproducing molecules to molecules that reproduce that way.

OK, what have I gotten wrong, and how badly;-)?

And why don't the creationists get the point about organisms being open systems? Seems to me that's obvious. If organisms are individually open systems, then evolution is a reflection of that fact, taken to the level of the collectivity of all organisms. Personally I suspect it has something to do with disdain for being compared to "dirty little monkeys."

A nice post, except for this historical blunder:

The trouble is that there are many hypothetical processes that, empirically, never happen, but which are not ruled out by the first law. For example, heat is never seen to flow spontaneously from a cool body to a warmer body, but such a thing is not precluded by the first law. This led people to suspect there was another principle that would cover such things, and this is where entropy comes into it.

As it happens, the 2nd law of thermodynamics was discovered before the 1st law. In fact, Sadi Carnot's Reflections on the motive power of fire, the founding document of thermodynamics, got a few things wrong because Carnot didn't have the 1st law, but instead used the caloric theory of heat (later discarded). The later work of Clausius and William Thomson showed how the adapt Carnot's work to be compatible with the 1st law.

Carnot's essay starts off with an extensive discussion of the importance of steam engines to Britain's economy, and the hope that France can achieve parity. This episode in the history of science is the poster child for thoroughly practical concerns leading to first-rate theory.

By Michael Weiss (not verified) on 18 Jul 2014 #permalink

Jason, if you're responding to drivel like Arrington's, I worry that your SIWOTI syndrome is returning. ;-)

By Richard Wein (not verified) on 18 Jul 2014 #permalink

Gedanken-experiment: It is proposed in a given thermodynamic system, e.g. a Rainforest, there have to be at least 3 or an odd-number of species competing with enough energy resources in a continually-habitable region for optimal, e.g. regular evolutionary opportunities for development. However even in a non-dynamic thermodynamic system there is no reason for or against 1 or 2 species mutating to advance species development or competition over a certain metric of Time. However they may also remain stagnant having optimized the energy usage of their environment and not "evolve". Therefore a pro- or anti-evolutionary developmental polemical viewpoint is not necessarily relevant for any rate or nonrate of species development given a perpetuated thermodynamic system. How any given climactic event species or machine utilize energy and transfer it into evolving power systems is the more exact question to be answered.

By Jonah Lissner (not verified) on 18 Jul 2014 #permalink

Anti-evolutionists really need to stop talking about thermodynamics.

But they won't. They are drawn to that argument. And they demonstrate their ignorance every time.

The one who surprises me is Granville Sewell. He is smart enough to be able to understand the second law, if he would only try. But he seems unable to resist making a fool of himself.

By Neil Rickert (not verified) on 18 Jul 2014 #permalink

Systems on Earth are neither closed nor isolated: they are powered by a gigantic fusion reactor located circa 93 million miles away.

Invoking thermodynamics to explain anything much other than clustering on Earth, or in the entire universe, will remain an invalid argument until all of the stars have ceased fusing matter into energy. Even then, Brownian motion will still play a huge role in clustering during the increasing entropy of the universe.

Isn't the currently popular model of the universe that it started at a very high temperature, is expanding and adiabatic (closed), and is, therefore, "running down". Thus entropy for the entire universe is increasing.


Yes, the universe seems to be increasing in entropy when the evidence is integrated over billions of years. But, long-term integrals are just averages of the plethora of ongoing short-term differentials. E.g. The winner of a marathon is determined by the total distance covered divided by the time it took to complete it; it is not determined by the maximum short-term metres-per-second that the runner achieved during the race.

The fundamental message from thermodynamics, which is so frequently misunderstood (usually for the purpose of obscurantism), is that entropy definitely does not increase monotonically at any specific location in space-time.

Is it possible to calculate the entropy change of a single elephant, as it grows from an egg to an adult? This would seem to involve a large local decrease in entropy, and it could be interesting to understand what happens with the second law here. This is an easier system to understand that the evolution over time of the elephant lineage, one would think. Why is not life itself impossible due to the 2nd law??

By George Bell (not verified) on 18 Jul 2014 #permalink

Michael Weiss--

Thanks for pointing out my historical error, which I have now corrected. Since my scientific point was entirely separate from my historical point, it was a matter of just changing a few words.

RIchard Wein---

Ha! I fear you're right. During my recent blog break, I didn't just stop writing my own blog, but I also mostly stopped reading other blogs. Then I made the mistake of checking UD, and I was immediately sucked back in! I see Arrington has a new post up in which he explains why, for some reason, a consistent materialist should behave like a total psychopath, so I guess I'll have plenty of blog fodder for a while.

@ anyone interested enough to answer -

I see the 'increase in complexity' line trotted out a lot (Dembski of course, if he's still about; but even Jason here alludes to it). Is there actually a definition of 'complexity' in this sense?


By Pan Outeast (not verified) on 18 Jul 2014 #permalink

You are, of course, correct; but, considering that the 2LoT is near the top of the Answers in Genesis list of arguments to avoid, I don't think the creationists still using this argument will pay any more attention to you than they have paid to AiG.

It's actually the ID folks who have revived the second law argument recently, so maybe this is one place where the young-Earthers are ahead of the ID crowd. As it happens though, the argument AiG is referring to is the claim that the second law began with the fall of Adam and Eve. At the end of their discussion they explicitly say that it is appropriate to use the second law to challenge evolution, though they are now savvy enough to acknowledge the open system/closed system distinction.

The cooling flow from spiral galaxies is a loss of energy by matter that is too hot for the elliptical galaxies. The formation of suns and the infall nucleosynthesis serves the same purpose in spiral galaxies. The development of life requires more energy than lack of life development. The inflow of matter into spiral galaxies causes the development of suns and of life. This is more time efficient than cooling flows.
Similarly, life serves the purpose of dissipating energy, also. A developing model of life proposes life is more efficient at eating energy and dissipating energy as heat \citep{england,crooks}. This process is constrained by the laws of thermodynamics. The increase in entropy is higher for life and the complex organisms than for the mineral components. The evolution of life is toward a greater rate of entropy increase. This idea balances the natural selection of evolution to include the rate of entropy increase alongside the efficiency requirement of survival of the fittest

Crooks,~G.~E., 1999. {\it{Entropy production fluctuation theorem and the nonequilibrium work relation for free energy differences}}, Phys. Rev. E, {{60}}, 2721.
England,~J.~L., 2013. {\it{Statistical physics of self-replication}}, J. Chem. Phys., 139, 121923.

So, Barry Arrington,
“As I understand their argument, entropy is not an obstacle to blind watchmaker evolution, because entropy applies absolutely only in a “closed system,” and the earth is not a closed system because it receives electromagnetic radiation from space.
Fair enough. But it seems to me that under that definition of “closed system” only the universe as a whole is a closed system, because every particular place in the universe receives energy of some kind from some other place”
Seems correct.

Yet I accept evolution. The IDers are getting smarter.

Is it the wrong place to present my theory of Designed Snowflakes? Since the second law of thermodynamics says we can't have entropy decrease, it must be that Snowflakes are not from a natural source, but instead from some ID style source.

By MobiusKlein (not verified) on 19 Jul 2014 #permalink

"Since the second law of thermodynamics says we can’t have entropy decrease"

No, it does not state that -- even in closed or isolated systems.

RIchard Wein --

I thought the point was not to persuade the unpersuadable, but to provide another resource for those in the middle.

What I like about this post is the carefully calibrated level of technicality. Plenty of websites mention the deal about closed systems, and leave it at that.

Incidentally, talking about the "entropy of the whole universe" is problematical (see the Wikipedia entry for "Heat death of the universe"); indeed, it's not clear what is meant by the "energy of the whole universe" (see the Physics FAQ entry.)

By Michael Weiss (not verified) on 20 Jul 2014 #permalink

Sorry Pete A., should have used my /sarcasm tag on my post. I blame Darwin.

By MobiusKlein (not verified) on 20 Jul 2014 #permalink

Apologies for my error. I'm usually good at spotting sarcasm, but earlier that day I'd been accosted by a pair of evangelical indoctrinators.

Creationist and ID invocations of SLoT irresistibly remind me of the poor deluded creationist who almost discovered the sun:

One of the most basic laws in the universe is the Second Law of Thermodynamics. This states that as time goes by, entropy in an environment will increase. Evolution argues differently against a law that is accepted EVERYWHERE BY EVERYONE. Evolution says that we started out simple, and over time became more complex. That just isn't possible: UNLESS there is a giant outside source of energy supplying the Earth with huge amounts of energy. If there were such a source, scientists would certainly know about it.

Another hindrance to the 2LoT objection is that even if it held weight, it would only rule out evolution when combined with an assumption that theistic evolution is necessarily false.

After all, we already have tons of evidence that evolution of some kind occurred (such as transitional fossils), and cdesign proponentsists obviously believe in a being who can perform miracles and who often works "through" nature. A theist insisting that 2LoT disproves evolution altogether is like a ufologist declaring "Ancient people could never have built the pyramids — therefore, the pyramids don't exist."

One possible counterpoint is that those who make the 2LoT argument aren't really trying to rule out evolution altogether, but simply to disprove pure materialism. In that case, there are still plenty of hurdles, such as the apparent ability of humans to "break" the creationists' re-envisioned version of the law.

In response, some say "intelligence can overcome" the law, at which point we have wandered into the outright surreal. Intelligence doesn't "overcome" natural laws, especially not thermodynamics, arguably the most fortified area of science period. (Especially thanks to the work of scientists and engineers who spent decades trying to break those laws!)

Another hindrance to the 2LoT objection is that even if it held weight, it would only rule out evolution when combined with an assumption that theistic evolution is necessarily false.

If the earth was a closed system, it still wouldn't rule out evolution of life on earth, because there are plenty of stored energy sinks and nonequilibrium systems on the earth that will take billions of years to run down. Plenty of battery left even if we were disconnected from the wall power, so to speak. Really the only way the 2LOT would prevent evolution, ever, is if it is (wrongly) interpreted in the strongest way possible to never allow even local decrease in entropy. 2LOT objections can't "hold weight" unless they deny the naturalness of reproduction and development, since evolution is really nothing more than differential, imperfect reproduction and development.

The 2LoT has circa 6 preconditions that must be fully met before it can begin to be applied. It does not apply to our morning mug of tea/coffee and it certainly won't apply to life on Earth until our planet has reached the limiting condition of thermal equilibrium within the entire universe, at which time every last particle has cooled to within a tiny fraction of a degree kelvin above absolute zero temperature.

It fascinates me that even in this 21st Century the anti-science evangelicals continue to rely on the most simplistic summary of the 2LoT and deliberately hide all of its preconditions and caveats. Of course, the only means to indoctrinate one's audience is by gathering an audience that is scientifically illiterate.