Alternate title: “The Problem With Supernatural Explanations”
People often tell me “science can’t prove that God does not exist, because you can’t prove a negative”. How often?
Ok, it’s really not that often.
It’s said that you can’t prove a negative because you can’t exhaustively examine and eliminate every place in the universe, so you can’t be 100% certain that something does not exist. Interestingly enough, however, the claim that “you can’t prove a negative” is itself an unrestricted negative. It effectively says “there are no proofs of unrestricted negatives”. If so, then no one can prove that no one can prove an unrestricted negative. And if no one can prove that no one can prove an unrestricted negative, then it must be logically possible to prove an unrestricted negative. The claim is self-refuting.
It should not only be possible to prove a negative, but I intend to show that a number of them have already been proven.
How would you go about proving a negative? 2,500 years ago, Parmenides realized a fundamental rule of logic–the law of non-contradiction. Anything that involves a logical contradiction cannot exist. We know for certain that there are no married bachelors, highest numbers, or triangles with other than 180 degrees (in Euclidian space, at least). These things are self-contradictory, because nothing can both have and lack some property at the same time. So, one way to prove a negative is to show that something contradicts itself.
God is traditionally defined as a supreme being–as St. Anselm put it, “a being than which none greater can be conceived”. As an analogy, however, imagine a supreme number. We know this can’t exist, because every number can have 1 added to it, creating a larger number. A supreme number would be able to have 1 added to it, and would not be able to have 1 added to it, and that’s impossible. The notion of a supreme being could be just as incoherent as a supreme number. (Sidenote: there’s also no such thing as a smallest number)
Other contradictions exist in the traditional definition of God. He is said to be all-good, making him both perfectly merciful and perfectly just. If he is perfectly just, he always makes sure everyone gets what’s coming to them, and that the punishment fits the crime. If he is perfectly merciful, he’s always lenient and forgiving. He can’t possibly be both.
These are just a few inconsistencies in the traditional conception of God. Here are others. Theists will invariably reply that, properly understood, the properties of God have no contradictions. They could be right, and it could be logically possible for God to exist (I don’t think the traditional definition can be rescued, though). Does this mean we cannot prove he doesn’t? No, actually. It’s not necessary to show that something cannot exist because it’s logically impossible. It’s sufficient to show that it is epistemically unnecessary–that it is not required to explain anything. Science has proven the non-existence of many things in this way, such as phlogiston, the luminiferous ether, and the planet Vulcan (not the Star Trek one–that one is still possible). Let me be very clear before we move on–scientific proofs do not establish perfect certainty, unlike logical or mathematical proofs. There is still the possibility of doubt. But they are proofs nonetheless, for they establish their conclusions beyond a reasonable doubt, and that is all that is needed to justify them.
Some quick science history–phlogiston, the luminiferous ether, and Vulcan were theoretical entities that were postulated to explain various observations and phenomena. Phlogiston attempted to explain heat, the ether was thought to be the medium of light propagation through vacuum, and Vulcan was hypothesized to explain perturbations in the orbit of Mercury. Science has since shown that these phenomena can be explained without them. Heat is caused by the movement of atoms, light is propagated by photons (weird-ass wave/particle things), and Einstein’s theory of relativity explains wobbles in Mercury’s orbit. By demonstrating that phlogiston, the ether, and Vulcan aren’t needed to explain anything, science has shown that they do not exist.
God is a theoretical entity postulated by theists to explain various phenomenon–the origin of the universe, the apparent design inherent in the universe, and the origin of life. Modern science can explain all of these things without resorting to God(1). In the words of Laplace, science has no need of that hypothesis (2). Phenomenon which remain currently unexplained, such as human consciousness, quantum entanglement or the exact steps of abiogenesis (see my last post) will very likely be illuminated with further work. By showing that God is not needed to explain anything, science has proven that there is no more reason to believe in the existence of God than to believe in the existence of phlogiston, the luminiferous ether, or Vulcan. This may help explain why more than 90% of the world’s top scientists disbelieve or doubt the existence of God.
Now, at this point, I’m reminded of another thing I often hear/read: “science is close-minded and arrogant, ignoring the possibility that there are things it doesn’t understand”, or the charge that science is “reductionistic”, as if that’s a weakness instead of a strength. This is because science, and scientists, prefers natural explanations over supernatural ones.
They do so not because of any metaphysical bias or closemindednes, but because of one simple reason. Natural explanations produce more understanding than supernatural ones. To say “God did it”, or an equivalent supernatural statement, is to simply offer an excuse for having no explanation.
How are explanations determined to be “good”? The quality of an explanation is determined by how much understanding it produces (how much ground the explanation covers), and how much it unifies and systematizes our existing knowledge. The extensiveness of this unity is measured by various criteria, including simplicity (the number of assumptions made), scope (the types of phenomena explained), conservatism (fit with existing theory), and fruitfulness (ability to make successful novel predictions and direct further questions).
Supernatural explanations are inherently inferior to natural ones because they do not fit this criteria well. They are usually less simple because they assume the existence of at least one additional type of entity (or an entire unseen realm of entities). They have less scope because they don’t offer mechanisms of how the phenomena in question are produced and thus they raise more questions than they answer. They are less conservative because they imply that certain natural laws have been violated or heavily contradict existing knowledge. And they are usually less fruitful because they don’t make any novel predictions. That is why scientists avoid them.
The realization that the traditional God of theism is not needed to explain anything–that there is nothing for him to do–has led a number of theologians to call for the rejection of this notion of god. In Why Believe in God? Michael Donald Goulder argues that the only intellectually respectable position on the god question is atheism. In Why Christianity Must Change or Die, Reverend Spong, former Episcopal Bishop of New Jersey, argues that the traditional theistic conception of God must be replaced by one grounded in human relationships and concerns. Both agree that religion should not be in the business of trying to explain the world.
What if we don’t yet understand a certain phenomenon, and have no explanation? Does that immediately mean it has a supernatural explanation? No, of course not. It’s far more likely that it’s only our ignorance of the natural forces in play that currently prevent us from having an explanation. Many things–earthquakes, eclipses, disease–were once attributed to supernatural causes or beings but can now be explained in purely natural terms. Apparent miracles are not contrary to nature but contrary to our knowledge of nature.
Given the inferior explanatory power of supernatural claims and the incompleteness of our knowledge, theists are only justified in offering God as an explanation for anything if they can prove that not only do we not currently have a natural explanation, but that no natural explanation is possible. And I believe that that is an unrestricted negative that cannot be proven.
(1) See, for example, Richard Dawkins, Climbing Mount Improbable; Stephen Hawking, A Brief History of Time; Lee Smolin, The Life of the Cosmos.
(2) When the French physicist Pierre Simon de Laplace explained his theory of the universe to Napoleon, Napoleon is said to have asked, “Where does God fit into your theory?” to which Laplace replied, “I have no need of that hypothesis.”