One of the most tediously common creationist claims is the Second Law of Thermodynamics argument. It is wrong, and was wrong from the beginning; but despite the fact that it has been refuted countless times, it continues to circulate in anti-evolutionist organizations, inevitably accompanied by the claim that it presents an insurmountable problem for the theory of evolution. Nothing could be further from the truth, as this essay will demonstrate.
The Second Law of Thermodynamics is a fundamental principle of physics. It deals with entropy, a term that has a precise mathematical definition. For this essay, it is sufficient to say that entropy corresponds to the amount of energy in a system available to do work. The more energy available for work, the less entropy there is, and vice versa. The Second Law of Thermodynamics (or 2LOT for short) says that the net entropy of a closed system must always increase.
To better understand the topic, a further explanation is worthwhile. The First Law of Thermodynamics, another well-established physical principle, says that the total amount of energy in the universe is constant; energy can never be created or destroyed. However, it can take many forms – kinetic, potential, electromagnetic, thermal, mechanical, and others. When you throw a ball, you are converting chemical energy stored in your muscles, in the form of the cellular fuel ATP, into mechanical energy to move your arm, which is in turn transferred into kinetic energy in the ball. The leaves of green plants collect electromagnetic energy – light – and convert it into chemical energy in the form of sugar. Virtually every interaction involves an energy conversion of some kind.
The 2LOT states that no energy conversion is ever 100% efficient. Whenever energy is converted from one form to another, some of it is lost, in the form of waste heat. When electric current travels through wires, some of the energy is dissipated as heat by the resistance of the wire; when you move your body, some energy is dissipated as heat by friction in your joints, and so on.
Energy cannot be destroyed, but this waste heat is effectively lost. It spreads out into its surroundings, becoming diluted among all the available molecules until no temperature differential exists and it is stretched too thin to ever be recovered. Entropy has just increased. This is the heart of the 2LOT: in every reaction, usable energy decreases and entropy increases, even if only by a small amount. The most common phrasing of the 2LOT is that in a closed system, net entropy always increases. A system is any collection of interactions; a closed system is one that exchanges neither matter nor energy with its surroundings.
Equivalently, entropy can be described in terms of disorder. In a system containing atoms at different temperatures, the low-entropy state would consist of those atoms segregated by temperature, for example, one reservoir of hot atoms and one reservoir of cold atoms. This temperature differential is what makes it possible for the system to do work. The high-entropy state would consist of atoms of all different temperatures randomly mixed together, so that the temperature of any part of the system is on average the same as the temperature of the whole system. The high-entropy state is more disordered than the low-entropy state, in the sense that rearrangements of the system are less likely to change it importantly. When atoms are separated into hot and cold pools, any random rearrangement will probably mix the two pools together, increasing entropy; but when the atoms are mixed together in a homogeneous gas, any random rearrangement is far more likely to leave them as a homogeneous gas, producing no change in entropy. (This can be analogized by pouring ink into a glass of water. If one randomly stirs the glass immediately after pouring in the ink, the result will be to diffuse the ink throughout the water’s volume, but if one stirs the glass after the ink and water are well mixed, it is extraordinarily unlikely that the ink and water will unmix themselves and separate out; rather, they will remain mixed.)
It is important to note that an increase in entropy is not necessarily irreversible. If the system is left to itself, random chance makes it overwhelmingly likely that entropy will increase to the maximum and then stay the same, just as it is overwhelmingly likely that the ink will diffuse throughout the water and never spontaneously re-separate. However, if the system is not “left to itself” – if work is done on it – the entropy of that particular system can decrease, although only at the cost of increasing the entropy of the system that did the work – and thus the total amount of entropy in the universe – by an even greater amount. Many kinds of human-built technology are designed to change a high-entropy state into a low-entropy state; one common example is a refrigerator.
We now turn to the creationist argument. Typically, it goes something like this: Evolution predicts that organisms (biological systems) evolve from less complex to more complex (more highly ordered) states. The 2LOT predicts that all systems, including biological ones, inevitably degrade and break down into less ordered states. Therefore, evolution contradicts the 2LOT and must be wrong.
With the above explanation in mind, several things can be said about this. First, it is a common mistake to equate entropy with everyday notions of chaos or disorder, but it is a mistake nevertheless. Strictly speaking, entropy is equivalent to disorder only on the atomic level, in the technical sense discussed above. If the terms “order” and “disorder” are instead used in the way most people would understand them – the way creationists use them – then entropy can increase without there being any corresponding increase in disorder, and some processes that increase entropy actually increase order. For an example of the first condition, if you set up a heater in your bedroom and leave it running for several hours, you will have significantly increased the room’s entropy, regardless of whether the arrangement of items in the room has changed. The temperature of your bedroom, rather than whether your clothes are folded in the bureau or strewn about the floor, is the overwhelmingly important factor when determining the entropy of the room (see this Talk.Origins Post of the Month for an explanation); and this applies to every other macroscopic system as well.
As for the second condition, there are many physical processes that spontaneously increase order without in any way violating the 2LOT. For example, the process of crystallization produces highly ordered, complex structures such as geodes and snowflakes. Random weather interactions give rise to highly ordered systems such as tornadoes and hurricanes. Evaporation can separate a mixture of water and salt; wave action on a beach can sort pebbles by size. Random processes of freezing and thawing can produce eerily regular geometric patterns of stones (Self-Organization of Sorted Patterned Ground; see also Stone circles explained). And finally, highly organized stars can form through the gravitational collapse of random clouds of hydrogen gas. None of these processes violate the 2LOT in any way.
When the true definition of entropy is used, rather than the creationists’ straw man, it is obvious that evolution in no way violates the 2LOT, regardless of whether it produces an increase in some vaguely-defined notion of order. For thermodynamics to be satisfied, the only requirement is that entropy increase, and it does. What energy is dissipated to cause this entropy increase? The answer is fairly obvious: the energy of the Sun. A small fraction of the energy the Sun emits reaches the Earth, where it is absorbed by living things and used for photosynthesis. The rest of the Sun’s energy is radiated into space and dissipated, causing a huge increase in entropy.
In other words, the Earth is an open system – it is not isolated, but rather is constantly receiving energy from the Sun. Therefore, there is no 2LOT constraint on what its net entropy gain or loss must be. (Living organisms, for that matter, are also open systems, since they continually take in energy.) Though the 2LOT, in varying formulations, does apply to all systems, it does not say that the net entropy of an open system must increase. On the contrary, the net entropy of an open system can either increase or decrease. This is because an open system can import energy, as the Earth does, and export waste heat, so that the universe’s total entropy does increase “somewhere else”. Such is the case if one steps back and considers the larger system of the Earth and the Sun. Any small decreases in the entropy of the Earth are more than compensated for by the huge increases in entropy produced by the Sun each moment.
If evolution violates the 2LOT, there must be some specific step in the process of evolution that violates the 2LOT. Which step is this, according to creationists? It is not mutation; the process of replicating the genome uses up energy just like any other chemical reaction. It is not natural selection; for every one organism that emerges victorious and prospers, many more must die, dissipating their stored biochemical energy. It is not reproduction; all living things are dependent on the usable energy of the Sun, whether used directly for photosynthesis or passed along the food chain. There is no step in evolution that violates the 2LOT, which is why creationists cannot point to one. And if no single step in the process violates this law, then the process as a whole does not, either.
When confronted with this argument, creationists will usually speak vaguely about “the law of conservation of information” or “energy conversion mechanisms”, or say that only intelligence can “overcome” the 2LOT to produce an increase in order. These arguments are pure fabrication. No such concepts exist in the scientific formulation of the 2LOT or in the field of thermodynamics in general. The equations of the 2LOT are called path independent, meaning that it is irrelevant for their purposes how an energy conversion is brought about. Intelligence has no privileged position in the equations; a human producing an ordered system no more “overcomes” the 2LOT than a human who throws a ball up into the air overcomes the law of gravity. Nor do the equations of thermodynamics mention “information”, a term which creationists deliberately leave ill-defined and vague so that they can employ it in a manner convenient to them. Before such arguments can be entertained, creationists must first define exactly what “information” means in this context and provide a precise way to measure it.
There is one final point to consider: If the creationists’ arguments were correct and the 2LOT did indeed forbid evolution, then there should be a massive conflict between biologists and physicists. There is none. Even someone who has no understanding of entropy or thermodynamics can appreciate this point. If the clash were as obvious as it is made out to be, then every physicist in the world should be an anti-evolutionist, aware that the theory is a gross violation of one of their most basic laws. But since the community with the most comprehensive understanding of thermodynamics seems to see no conflict between it and evolution, the obvious conclusion is that the creationists have invented a problem where none exists.