Evolution and the Second Law of Thermodynamics


I am appalled by how many people (mostly fundamentalist Christians in my experience, though there are probably others) claim that the theory of evolution violates the Second Law of Thermodynamics. This can be clearly shown to be false. Such an immense mistake about the implications of thermodynamics would be excusable if admitted as simple ignorance, of course. But when the claim is positively insisted upon over the objection of those who are actually knowledgeable about the subject, then it is clearly a sign of blatant bias and outright hubris. Such extremely careless thinking deserves to be called out and reasonably answered for the sake of those who might otherwise find it persuasive.

The first thing we need to understand about the Second Law of Thermodynamics is that it is a very well-defined scientific statement, not a general philosophical idea. Things like the degradation of morals in society and the crumbling of civil empires are often compared to the Second Law, but no one who is knowledgeable and honest about the subject would call them true examples of the Second Law at work.

The only scientifically valid statements of the Second Law of Thermodynamics are those that can be proven to be equivalent to the idea that the thermodynamic entropy of a closed system cannot decrease. In classical thermodynamics, entropy is a quantity which is defined to be zero for a pure crystalline substance at absolute zero temperature, and is otherwise defined by the calculus equation dS = dQ/T, where dS is the change in entropy of the system, dQ is the change in heat of the system, and T is the temperature of the system.

In statistical thermodynamics and information theory, entropy is defined by a different equation, namely S = k·ln(W). Here S is the entropy of the system, k is a constant, and W is the number of microstates available to the system. The concept of microstates available to a system can be defined in various ways in information theory, but if we want to apply the Second Law of Thermodynamics, then we must define it in such a way as to make this definition of entropy equivalent to the definition in classical thermodynamics, provided that the right constant k is used. This is exactly what is done in statistical thermodynamics, where k is known as the Boltzmann constant.

This second definition of entropy is the one that more clearly demonstrates why entropy is a measure of the “disorder” of a system. If a system is highly organized into a rigid, crystalline structure, then the entropy is low because the molecular arrangement is relatively fixed, meaning there are not very many degrees of freedom, or microstates, that the system can be in. If a system is not so well-ordered, then there are many degrees of freedom on the molecular level, and the entropy is high.

Note that the concept of “disorder” is not necessarily the opposite of “complex design,” as often assumed. For example, the rigid structure with low entropy may actually be a very simple, repetitive design, whereas the other system with high entropy may have a far more complex design.

Since the universe is taken to be a closed system, the Second Law requires that the total entropy of the universe can never decrease, no matter what changes are taking place in the universe. For example, when heat flows naturally from a hot object to a cold object, the entropy of the two objects taken together always increases. From a classical thermodynamics viewpoint, this is because the T of the dQ/T expression is greater for the hot object than for the cold object, making the entropy lost by the hot object less than the entropy gained by the cold object. From the equivalent statistical thermodynamics viewpoint, it is because the number of microstates available to the combined system of the two resulting warm objects is greater than the number of microstates that were available to the combined system when one object was cold and the other was hot. The number of microstates available to the combined system is the product of the numbers of microstates available to each part, which means the entropy of the combined system is the sum of the entropies of each part, because of the logarithm in k·ln(W). Thus the entropy lost by the hot object is again seen to be less than the entropy gained by the cold object, and we can say that the composite system (as well as the entire universe) has become more “disordered” after the change.

Now let’s look at the application of the Second Law of Thermodynamics to the theory of evolution. The traditional creationist argument is that the evolution of lower life-forms to higher life-forms is not possible because it involves an increase in complexity, which they say amounts to a decrease in entropy. This argument is fundamentally flawed in two different ways.

First, we have no evidence that the complex life-form is actually less entropic than the simple life-form. Entropy is a measure of statistical disorder, not simplicity vs. complexity of design. There is no reason that I am aware of to believe that a man is less “disordered,” on a molecular level, than an equivalently-sized chimpanzee, for example. That might be the case, but if so, it has nothing to do with the complexity of our design. Consider the fact that, all else being equal (most notably mass and temperature), crystalline solids have lower entropy than either men or chimpanzees, and yet clearly their design is much more simple and repetitive, not more complex.

Second, even if higher life-forms do have lower entropy than lower life-forms, the Second Law does not say that they therefore can’t evolve. The burning up of the sun, with its corresponding energy transfer to the earth, along with other processes such as the geothermal activities within the earth itself, increase the entropy of the universe at a far higher rate than evolution could possibly decrease it. There are many processes on earth that result in a local entropy decrease, but this is permissible by the Second Law, since they are receiving energy from the sun. The entropy of the universe as a whole is still increasing.

Beyond these two fundamental flaws in the argument, there is a reason that we can be absolutely sure that evolution does not violate the Second Law. To see this, we must understand that entropy is an example of what is known as an “extensive property.” That means the entropy of an entire system is just the addition of the entropy of all of its parts. So, for example, the entropy of a collection of fifty identical 100-pound loads of organic material is fifty times the entropy a single 100-pound load of organic material. The entropy of a population of fifty 100-pound monkeys is fifty times the entropy of a single 100-pound monkey. And the entropy of a population of fifty 100-pound humans is fifty times the entropy of a single 100-pound human.

Now let’s assume (for the sake of argument) that the entropy of a monkey is less than the entropy of a 100-pound load of organic material, and the entropy of a human is least of all. For this example, we’ll arbitrarily assign an entropy of 1 unit to each human, 2 to each monkey, and 3 to each load of organic material. So, for the collections of 50 of each item we have an overall entropy of 1×50 + 2×50 + 3×50, which is 300.

Now what happens when a new human is born and grows up? The population of humans grows from 50 to 51, but where does the matter (which must be conserved) for the new human come from? It has to come from the surrounding environment. We can model this by saying that the collection of organic material decreases from 50 loads to 49. So if we again add up the overall entropy (of the items we’re discussing, not of the whole universe) we find it decreases by 2 to 298.

But here is the key concept. This change in entropy is completely independent of whether the new human was born to human parents or to monkey parents. It does not depend on whether or not there was already an initial presence of humans. We can start with zero humans and fifty monkeys, and postulate a sudden and bizarre evolution by assuming a human born to a pair of monkeys, and the entropy decrease is still 2, exactly the same amount of decrease as when the human baby is born to human parents.

Obviously, the birth of a human baby to human parents happens all the time and thus cannot be a violation of the Second Law of Thermodynamics. Therefore, it would not violate the Second Law of Thermodynamics if the human baby were born to monkey parents either. The same argument can be made for any evolutionary step, no matter how large or small.

The bottom line is, the Second Law of Thermodynamics does not argue against the theory of evolution in any way. Evolution doesn’t violate the Second Law of Thermodynamics any more than it violates Newton’s Law of Gravity. That doesn’t automatically mean evolution is true, of course, though I personally find the evidence for evolution to be quite convincing. But what it does clearly demonstrate is the gross intellectual crudeness of those who claim that evolution is proved false by this argument.


This page copyright © 2001-2014 Edward A. Morris.  Created May 6, 2001.  Last updated September 2, 2014.

Back to noble-minded.org home page