Posted by I’m sure we’re all familiar with the creationist chestnut that evolution violates the second law of thermodynamics. For people with a basic science education it is something of a litmus test. As soon as you hear someone make this argument you can be certain that you are dealing with a crank. You see, if someone says that he has considered the evidence for evolution and finds it unconvincing, we might shake our heads in disbelief or wonder how seriously he has studied the matter, but in the end it is a matter of opinion whether the evidence is compelling. But if someone says that evolution contradicts the second law then that person is simply wrong about a question of fact.
Most people know that the second law has something to do with entropy and that entropy has something to do with randomness and disorder. Creationist uses of the second law seldom go beyond this rudimentary understanding. Usually there is little more to the argument than the claim that the second law says natural forces always lead to increases in entropy, while evolution implies that the entropy of the Earth has decreased.
It s important to realize, however that the second law is ultimately a mathematical statement. In its classical form it says this:
\[
\Delta S \geq \int \frac{dQ}{T}.
\]
Giving precise definitions for all these variables, and understanding exactly what they represent in concrete cases, is a daunting task, which is why thermodynamics textbooks do not make for light reading.
But a quick overview will be sufficient for the point I want to make. The basic scenario is that we are imagining some well-defined physical system that has gone from an initial state to a final state via some transfer of heat.
The term on the left is the change in entropy of the system that resulted from the change in state. The T represents something called the “equilibrium temperature” of the system. The term dQ represents (very roughly) the quantity of heat that has been transferred during the process.
The inequality above applies to any sort of thermodynamical system, regardless of whether or not it is open to its surroundings. But if we imagine that our system is completely isolated, so that neither matter nor energy is crossing the boundary, then dQ will be 0. In this case, elementary calculus tells us that he integral will be 0 as well. We then have the statement that the change in entropy for an isolated system must be positive or zero, which is equivalent to saying that it cannot decrease. Moreover, it will only be zero in the special case of a “reversible” process, which no actual physical process is. This is where we get the notion that the second law says that a spontaneous natural process in an isolated system always causes entropy to increase.
Don’t sweat it if you have long forgotten your freshman calculus class. The point is simply that the second law is ultimately a mathematical statement. That means that if someone claims that evolution, or anything else, violates the second law, they must back up the claim with an actual calculation. If they don’t, if they’re arguing solely at the level of generalities about order and disorder, then the second law is not really playing any role in their argument at all. One hardly needs fancy principles of physics to understand that things tend to break down unless energy is expended to prevent them from doing so, but creationist presentations of the second law are rarely any more sophisticated than that.
Now, anti-creationist literature tends to argue at a popular, qualitative level. In responding to the creationists the point is usually made that the Earth is not isolated, since it receives energy from the Sun. The decrease in entropy on the Earth is more than compensated for by the increase in entropy of the universe generally as a result of the heat radiated by the Sun and Earth into space. These are cogent points, and are sufficient for seeing the flaws in most creationist versions of the argument. (Sometimes creationists will retort that entropy can only decrease if some sort of “energy conversion mechanism” works to overcome the second law, but this is just a lot of made-up gibberish. You will not find such language in any serious textbook on thermodynamics.)
Entropy calculations tend to be very difficult in practice, and no one knows how to calculate the change in entropy of the Earth in the course of evolution. However, certain crude estimates can be made, with the intent of determining whether evolution needs to be nervous about a possible thermodynamics problem. One such calculation was carried out by physicist Daniel Styer in a paper published in The American Journal of Physics (“Entropy and Evolution,” Vol. 76, No. 11, Nov. 2008, pp. 1031-1033.) His conclusion was that, “the Earth is bathed in about one trillion times the amount of entropy flux required to support the rate of evolution” assumed in the paper.
In a subsequent paper for the same journal, physicist Emory Bunn argued that Styer’s calculation contained an unrealistic assumption. (“Entropy and the Second Law of Thermodynamics,” Vol. 77, No. 10, Oct. 2009, pp. 922-925.) He redid the calculation, bending over backwards to make assumptions that would overestimate the entropy reduction on the Earth in the course of evolution. His conclusion was that “we find that he second law … is satisfied as long as the time required for life to evolve on Earth is at least 107 seconds, or less than a year. Life on Earth took 4 billion years to evolve, so the second law of thermodynamics is safe.”
As it happens, I discussed these points in more detail in this essay. I was specifically responding to an attempt by mathematician Granville Sewell to revive the second law argument against evolution. Sewell has now popped up again, in this blog post.
Sewell has been peddling this argument for quite some time, having first presented it in two opinion pieces for The Mathematical Intelligencer in 2000 and 2001 (the second being a reply to critics of the first). Recently he presented the same argument
in a paper which somehow made it through peer review at the journal Applied Mathematics Letters. As I shall make clear, this represents an astonishing breakdown in the peer review process.
As the article was in press, the editor had it called to his attention that it wasn’t very good. The editor then suddenly refused to publish the paper. This led to a kerfuffle culminating with the journal paying Sewell ten thousand dollars, nuisance money to avoid a lawsuit one suspects.
It is close to incomprehensible that a competent reviewer could have signed off on Sewell’s paper. For one thing, it contains very little that was not in his Mathematical Intelligencer pieces, meaning this new paper consists mostly of previously published material. Even worse, here’s the first sentence of the abstract:
It is commonly argued that the spectacular increase in order which has occurred on Earth does not violate the second law of thermodynamics because the Earth is an open system, and anything can happen in an open system as long as the entropy increases outside the system compensate the entropy decrease inside the system.
The reviewer could have stopped right there, since that is not at all what’s commonly argued. I would be interested to see a quotation from a reputable person who has claimed that anything can happen in an open system. The actual argument is much simpler. It is that the second law only precludes processes that violate the inequality presented earlier, and it is perfectly clear that evolution does not violate it.
Later in the paper Sewell writes:
Of course the whole idea of compensation, whether by distant or nearby events, makes no sense logically: an extremely improbable event is not rendered less improbable simply by the occurrence of “compensating” events elsewhere.
But this is just bizarre. That local decreases in entropy are possible so long as they are offset by global increases is not some excrescence tacked on to the second law by people desperate to protect evolution. It’s just an immediate consequence of what the second law says. Moreover, the assertion that evolution does not contradict the second law has nothing to do with whether evolution is a credible or reasonable theory. There might, in principle, be a hundred good reasons for rejecting evolution. It’s just that “It contradicts the second law” is not among them.
(As an aside, the reference to probability here involves the interpretation of the second law the derives from statistical mechanics. The details of this interpretation is not so important for my present purposes, but I explain the basics in my essay, linked to earlier.)
We move now to Sewell’s current blog post. With standard creationist arrogance, it opens as follows:
The American Journal of Physics article by Daniel Styer which was offered as a “concise refutation” to my Applied Mathematics Letters article by the blogger whose letter apparently triggered the withdrawal of my AML article is possibly the dumbest work ever published by a major physics journal. To demonstrate how absurd the logic in this article is, I wrote a little satire ( here ) which extends Styer’s attempts to quantitatively demonstrate that the decrease in entropy of the universe due to biological evolution is easily “compensated” by the increase in the “cosmic microwave background&rdquo, to the game of poker.
Go to the original for links.
Harsh words, but whenever I read such things I recall the remark made by Richard Dawkins when responding to an especially vicious, and badly misinformed, criticism of his book The Selfish Gene. We are in danger of thinking that no one would dare to be so rude without first taking the elementary precaution of being right. Alas, Sewell’s little satire bears the uninspiring title, “Poker Entropy and the Theory of Compensation.” But, as we have noted, there is no “theory of compensation.” There’s just what the second law says, what evolution says, and the self-evident fact that these statements are not contradictory.
In discussing certain probabilities derived by Styer and Bunn, Sewell writes:
Since about five million centuries have passed since the beginning of the Cambrian era, if organisms are, on average, 1000 times more improbable every century, that would mean that today’s organisms are, on average, about 1015000000 times more improbable than those at the beginning of the Cambrian … And since nothing can have probability more than 1, this would mean today’s organisms have a probability of less than 10-15000000 … of having arisen.
The details of where those numbers came from is not important. It’s that final statement, about the probability of today’s organisms “having arisen” that I find remarkable. You see, neither Styer nor Bunn made any pretense whatsoever of calculating the probability of modern organisms having arisen.
Styer and Bunn were treating organisms as ensembles of atoms and molecules. In keeping with the statistical mechanical interpretation of the second law, they were asking about the number of microstates (roughly, arrangements of atoms) that correspond to the organism’s macrostate (roughly, the gross physical properties of the organism). Styer’s estimate that organisms are, on average 1000 times more improbable every 100 years means essentially that for a given organism, the number of microstates describing the modern organism is smaller, by a factor of 1000, then the number of microstates describing the corresponding organism a century earlier. (It was precisely this assumption that Bunn challenged, replacing it with a much smaller, and better justified, factor.)
My apologies for the jargon. The point is simply that these numbers have nothing to do with the probability of modern organisms “having arisen,” unless you think evolution assembles organisms by trying out random configurations of atoms. I suspect, though, that Styer and Bunn would be among the first to point out that natural selection is a great probability shifter. Indeed, the whole point of evolutionary theory is that structures that almost certainly would never evolve by chance become vastly more probable when the role of selection is considered.
In the end this whole issue is really very simple. Things that are thermodynamically impossible do not occur. But evolution by natural selection plainly does occur, and there is no reason in principle why it cannot account for the growth in complexity in organisms. Scientists routinely apply evolutionary thinking to their work and are rewarded with tangible progress in their research. Rough order of magnitude calculations regarding the decrease in entropy of the Earth show that evolution is not even close to violating the lower bound given by the second law.
Against this Sewell offers nothing but the usual hand waving.
I’d say the presumption should be that evolution does not violate the second law, wouldn’t you?
Now we come to the big finish of this whole saga. Sewell sent his little poker essay to the American Journal of Physics (AJP). It was immediately rejected. At his blog post, Sewell presents the reason he was given:
I do not see any educational value in your manuscript. Because it is well established in the physics community that there is no conflict between the second law of thermodynamics and evolution, we can consider manuscripts which help students understand why. However, papers that promote views that are contrary to accepted understanding in physics should be sent to research journals not to AJP.
Sewell was outraged, of course:
In other words, we will print anything that supports the accepted view, no matter how stupid, and won’t consider anything that challenges it, no matter how logical.
Yes, that’s clearly a perfect paraphrase of what the editor said. But if we read carefully and piece together the clues found in the editor’s reply, we might discern that AJP is actually an educational journal directed at students and teachers, and not a research journal directed at professional scientists. If we jaunt on over to the “About” section of the journal’s website, we find this:
AJP publishes papers that meet the needs and intellectual interests of college and university physics teachers and students. Articles provide a deeper understanding of physics topics taught at the undergraduate and graduate level, insight into current research in physics and related areas, suggestions for instructional laboratory equipment and demonstrations, insight into and proven suggestions for better teaching methodologies, insight into how college students learn physics, information on historical, philosophical and cultural aspects of physics, annotated lists of resources for different areas of physics, and book reviews.
Given the aims of the journal, I’d say the editor’s reply was entirely correct. A paper claiming that the physics community is totally confused on a question as basic as whether evolution contradicts the second law belongs in a research journal, not an educational journal.
Sewell continues:
Any suggestions as to which “research journals” might consider papers which “promote views that are contrary to accepted understanding?”
I’m really more familiar with the mathematical literature than I am with physics. But I will conservatively go out on a limb and suggest that it’s actually every research journal in physics that is interested in papers that challenge accepted understandings. In fact, I am guessing that it’s actually pretty hard to publish a paper that just says, “The conventional view is entirely correct.” It’s just that research journals usually have this silly requirement that the challenge be backed up with strong evidence and cogent argumentation, both of which are lacking in Sewell’s papers.
All in all, it’s pretty standard stuff. An ID proponent presents a bad argument, then pompously struts around as though it is everyone else who is confused. Nothing to see here, everyone move along now…
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 23, 2011
June 24, 2011
June 24, 2011
June 24, 2011
June 24, 2011
June 24, 2011
June 24, 2011
June 26, 2011
June 26, 2011
June 27, 2011
June 27, 2011
June 27, 2011
June 27, 2011
June 27, 2011
June 28, 2011
June 29, 2011
July 5, 2011
May 7, 2012
May 7, 2012
May 7, 2012
May 9, 2012
May 9, 2012
