Does the Second Law of Thermodynamics (Entropy) Contradict Evolution?

A friend of mine recently asked me if I knew how entropy and evolution could be reconciled.  This is not a theological question, nor is it a biological one – it is primarily rooted in physics.  Now, as a biologist I can talk for hours on the experiential evidence for evolution without ever once having to think or refer to the Second Law of Thermodynamics.  I have been told that the Second Law of Thermodynamics, which deals with entropy, is the one solid empirical fact we have about our universe, that it will hold true even if everything else we know turns out to be false.  I cannot speak to the truth of that, but I do know that evolution is a real phenomenon, and if entropy is as real as physicists claim it is, then the two truths cannot contradict one another.

Simply reading the Second Law article on Wikipedia is enough to make my head spin.  I need you to know that before I start in on this subject – I don’t want you to think that I am pretending to be an expert.  I am not.  I am basing much of what I am about to say on an article by Daniel Styer, published in 2008 in the American Journal of Physics, entitled ‘Entropy and evolution’.

From what I can gather, the Second Law of Thermodynamics states that the entropy of the universe is always increasing (at least, this is how Arthur Eddington defined it in 1928).  The universe can never decrease in entropy.  In 1854 the German physicist Hermann von Helmholtz came to the unnerving conclusion that this law of entropy means that the entire universe will one day die out in what is termed ‘heat death.’  You can see how this works simply by putting your hand on your belly.  Do you feel that heat?  We humans are able to maintain a constant internal temperature in part because of this warm-bloodedness.  But where did this heat come from?  Well, it was stored as chemical energy in the food we eat.  Our bodies took that energy and stored it in a chemical form our bodies could use (called ATP), and then used that ATP energy to drive the motion of our bodies, our thoughts, our heart beat, the production of enzymes and cell membranes and all sorts of things.  But each time chemical energy is used to do work, at least part of that energy gets lost as heat.  That heat which we feel radiating from our skin primarily came about as this by-product of metabolism.

This is why dead people are so cold.

But what happens to this heat once it radiates from our bodies?  Not very much.  If there’s enough people together in a room, it will raise the temperature of the room enough that it is noticeable.  This heat could possibly melt some snow that people tracked into the room.  But there is not much work that it could do.  Once it radiates from our bodies, that energy, that had at one time been sourced in the sun, that was stored in sugars by a plant, and which was unintentionally released from our bodies as heat, comes to a sad end as it causes a few gas molecules in the atmosphere to vibrate a little faster than normal for a brief moment of time.

And this is the way it is for all things.  The sun is pumping out energy that life is converting into useful workable energy, but eventually all of it will dissipate as heat.  One day the sun will run out of fuel and life on earth will end.  But work will continue throughout the universe as other stars are formed and die out.  Eventually all of their energy will also dissipate as heat, and all that will be left is a cold, lifeless universe.  Heat equilibrium means death.

This is what the second law of thermodynamics demands.

Entropy, then, has something to do with heat.  Its equation is essentially this: a change in entropy = Q/T, where Q = the heat that is gained, and T = temperature.  We often think of entropy as an increase in disorder (if you gain entropy, you are more disordered, if you lose entropy, you gain order).  The main Creationist quibble with entropy is that, an increase in complexity is a decrease in disorder.  Thus in order for complexity to appear, entropy must be decreasing, which according to the Second Law is impossible.  Therefore, you need supernatural intervention to account for complexity.

The complexity the Creationist is talking about comes in three major forms:

1. An increase in complexity from the start of the Big Bang to the rise of the stars and planets and galaxies, and

2. An increase in complexity from non-life to life, and

3. An increase in complexity, from single-celled life to multi-celled life, to consciousness and self-consciousness, society, etc.  For the rest of this blog, I will only focus on this third form of complexity, as the first two lie well outside my field.

You should be able to immediately see an intuitive problem with the Creationist position: reproduction.  We, each of us, began as two single cells which fused together to form a zygote, and from those initial conditions we developed into multicellular beings, each with a unique personality and the potential for reproduction ourselves.  And yet no physicist says that our development from a zygote caused a decline in the universe’s entropy.  And no Creationist argues that our development was only made possible by direct intervention from the Creator.  Yet surely a zygote is less complex and less ordered than a human.

So why, then, is evolution even considered a problem?

Less intuitively, you may have seen another problem with the Creationist’s quibble.  It presumes that entropy does, indeed, have something to do with disorder.

But this simply isn’t true.

As you can see from the entropy equation discussed above, there is no measure for disorder in the equation.  There is only temperature and heat gain.  Disorder, according to Styer, is only a useful metaphor for entropy, but is not an exact image of entropy.  Thus there can be numerous instances in which entropy increases, but disorder does not.

For example, a hot coffee, left to its own devices, will cool down.  As it loses heat, its molecules go from a highly excited state in which they are vibrating around, to a less excited state.  The coffee’s heat dissipates into the surrounding environment; the environment has gained entropy.  But the coffee, despite doing absolutely nothing at all, has in fact lost entropy.  It has gone from a high entropy state to a low entropy state.  But wait a minute, doesn’t the Second Law say that’s impossible?  Not at all.  It says it is impossible for the universe as a whole to lose entropy, but not for individual things within the universe.  The lesson here is that anything in the universe can lose entropy, so long as somewhere else in the universe there is an even larger gain in entropy.

As another example, Styer asks us to take an icecube out of the freezer and smash it with a hammer.  Go ahead, I’ll wait.  How disordered do those shards of ice look?  Now let them melt into a bowl.  Is it not intuitive that the liquid water is more ordered than the shards of ice, since the water is uniform in appearance?  But, in fact, the entropy of the water is higher than the entropy of the ice.  The lesson here is that disorder is not a perfect metaphor for entropy.

To answer the Creationist question, then, we could provide a very simple answer:

Evolution and entropy are compatible because we live in an open system: we constantly get energy from the sun.  So long as the universe is gaining more entropy from the sun than the entropy that is lost on earth during evolution, there is no conflict.

A More Detailed Answer, from Styer’s Article

Before we can get into the details of this answer (and it will be detailed), we first need to keep in mind what the question actually is.  The question is: does the second law of thermodynamics invalidate evolution?  It is important to understand the question being asked; a good number of articles on this topic do not.  For example, the answer I gave above is often derided by Creationists because all it does is throw energy at the problem.  Energy alone, they say, will not drive evolution.  That may or may not be true, but that is not the question they had originally asked.  They were not asking if entropy proves evolution; such a question is absurd.  They were asking, as are we, if entropy and evolution are compatible. 

To do that, we need an estimate of how much entropy is lost during evolution, and how much entropy is gained by the universe.

But first, what powers evolution?  Like all processes on earth, evolution gets its energy from the earth’s core and from the sun.  For simplicity, we will only consider the sun.

The following information is from Styer’s paper.

The sun has an average surface temperature of 5778 K (Kelvins, a unit of measurement for temperature often used in science, where 0 K = absolute zero, the temperature at which molecules stop moving, the absolute lowest temperature that can be reached, -273.15 degrees C).  The earth has an average surface temperature of 288 K.  The earth’s temperature is relatively constant; so, as the sun shines down heat, nearly all of that heat is reflected back into outer space, which has a temperature of 2.728 K.  Here we have a system in which energy is sent to the earth, and the earth reflects nearly all of that energy back into the surrounding universe.

Every second, the sun, by generating energy, has an entropy decrease of 20 900 000 000 000 Joules/Kelvin.  The earth receives this entropy and reflects it back to the universe, with an entropy throughput of 420 000 000 000 000 Joules/Kelvin each second, such that the earth’s entropy remains relatively constant.  The universe gains 44 400 000 000 000 000 Joules/Kelvin in entropy each second simply by this sun to earth to universe pathway.  So, the sun declines in entropy while the universe experiences a major increase in entropy.

The entropy throughput on earth is the entropy that evolution has to work with.  The sun is providing a huge amount of energy.  But how much entropy does evolution lose?  Is the sun sufficient to account for it?

To figure this out, we need to find some estimate for complexity.  Now, if we are to define complexity as ‘improbability’, we can give a ballpark guess that every one hundred years, each organism on the planet is 1000 times more improbable than their ancestor from one hundred years before.  This is a pretty unrealistic assumption, in that it is way too fast given normal evolution, but that is not a bad thing.

We can then use a statistical equation for entropy (which is in the paper and which I will not explain here, mostly because it is above me) to calculate that the change in entropy between an organism from one hundred years ago, and its new and improved progenitor, is -0.0000000000000000000000923 Joules/Kelvin.  It is negative because entropy is decreasing.

This is for 100 years of evolution that leads to a single living organism.  To know the total amount of entropy decline over 100 years, we can multiply this by the number of living things.  It turns out that there are probably over 100 000 000 000 000 000 000 000 000 000 000 organisms on the planet, including bacteria.  Multiplying everything out, we get the change of entropy that occurs each second due to evolution as -302 Joules/Kelvin.

Yet compare this to the amount of entropy that comes to the earth from the sun every second, which we saw was 420 000 000 000 000 Joules/Kelvin.  Styer writes, ‘In other words, at a minimum the Earth is bathed in about one trillion times the amount of entropy flux required to support the rate of evolution assumed here.’  And that is with a minimum estimate for entropy throughput for the earth, and a maximum estimate for entropy loss through evolution.

What this means is that, if we had two identical planet earth’s, and one was dead and one was full of evolving life, and they both received equal amounts of energy from the sun and reflected equal amounts back, we would be unable to measure any entropy differences between the two planets.  The amount of entropy throughput on earth is so great, that life barely makes a dent in it.  And that little tiny dent that it does make, that dent that is so small we could never actually detect it, is more than compensated by the huge increase in entropy that occurs in the surrounding universe.  In other words, even if evolution, powered by the sun, does cause a decline in entropy on the surface of the earth, this debt is paid, and then some, by the total entropy gain, also caused by the sun, to the universe.  The Second Law, which states that entropy can only increase in the universe over time, holds true.

Whew.

So, that’s that.  Again, I am not a physicist.  For me, evolution does not rise or fall based on how good Styer’s estimates are (and they are just estimates).  But, by making an estimate, Styer has at least shown that it is totally possible for the complexity of life to increase on the planet over time, and the second law of thermodynamics to still hold true.

What this does not address is how complex things come from simpler things.  The second law of thermodynamics does not address that.  All we have learned is that entropy and evolution are completely consistent with one another, and thus we have answered our question.  As to what drives increasing biological complexity?  That is a separate matter entirely, and has nothing to do with physics.

But you wouldn’t know that from reading the Creationist literature.

Additional Note: I just discovered another paper after posting this, by Sharma and Annila, published in Biophysical Chemistry, entitled ‘Natural process – Natural selection’ which apparently shows (I have not read the whole thing yet) that evolution in fact leads to an increase in entropy.  Styer’s paper only looked at evolution in general and was based on energy input from the sun, and showed that the loss of entropy is surprisingly small.  Sharma and Annila, instead, focus on the chemical processes that occur in living things, and demonstrate an overall increase in entropy.  Although some aspects of evolution lose entropy, other aspects gain entropy, with the gain being greater.  These two papers are not in conflict with one another, they just approach the question from a different angle.  Both of them provide interesting solutions to the entropy problem, and both demonstrate the sort of mathematical and biological acumen you need to properly understand and answer the entropy/evolution question.