The Illogic of Atheism Using Science to Defend Its Position

Science Cannot Do What Atheists are Purporting it Does

Ricky Gervais made a comment regarding why he believes atheism is a better world view than religion. His “in a nutshell” example is:

“one person says ‘there’s a God’, and an atheist says ‘can you prove that?’, and they say ‘no’, and the atheist says ‘I don’t believe in it’…that’s it, that’s all it is.”

He goes on to say that if you took every religious book and destroyed them, and you did the same thing with every science book, in a thousand years time those science books would be back, exactly the same, because the tests would all turn out the same. But those religious books will either never exist again, or they would look totally different, because there was no test.

To recast his statement in the form of an argument:

• science runs tests on the things it observes;
• running tests means we can get evidence of a thing’s existence;
• evidence is a direct observable example of a thing’s existence;
• getting a direct observable example of a thing’s existence is superior to belief because it confirms a thing’s existence.

therefore,

• atheism is a better (smarter, more rational, more realistic) worldview than religion.

This does not appear as a bad argument. It sounds to many as quite reasonable. In fact, it is logically consistent, as the premises do not contradict each other. It is also logically valid, since its logical structure guarantees the conclusion follows from the given premises.

But what the argument is not is sound. An argument is sound if it is both valid and all its premises are true. An unsound argument is considered a flawed or “bad” argument, because it cannot be relied upon to establish the truth of its conclusion.

What makes the above atheistic argument unsound? It is the 3rd premise that is false. It says evidence is a direct observable example of a thing’s existence.

This is not true. Science does not measure phenomena directly. Not even close. To understand why, we have to appreciate what a measurement is, how knowledge is validated, and the self-correcting process science uses.

Let us begin with an acceptable definition of a scientific theory. A scientific theory is a well-substantiated explanation of some aspect of the natural world that is based on a body of evidence, and has withstood repeated testing and observation.

Most people would agree with that definition, as do I.

Examples of scientific theories include the theory of evolution, the theory of general relativity and the germ theory of disease.

The properties of a scientific theory are:

• grounded in empirical evidence collected through observation/experimentation;
• allow scientists to make predictions about phenomena that have not yet been observed;
• testable, and capable of being shown as wrong through experimentation or new data (falsifiability).

The last property shows us that using the word “prove” does not belong in science. Scientific knowledge must always be open to revision and refinement. Science is an ongoing process. No theory is ever considered the final truth.

This is in contrast to fields like mathematics, where the word “proof” is appropriate, because mathematical statements can be proven with absolute certainty through logical deduction. This is because mathematics is establishing truths within mathematics itself, not natural phenomena.

But isn’t this just a matter of semantics? People toss the word “prove” around casually, as though there are degrees of proof. Thus, even if one’s use of the word violated its technical definition their argument isn’t necessarily invalid. If it turns out “prove” was being used in the context of confirming instances or copious demonstration then the atheistic argument stated above could still be a good one.

Correct.

But that is not what’s happening here. The atheistic argument is proclaiming that if an atheist asks a religious person “can you prove it” and the religious person cannot, then this shows us that atheism is a more valid worldview.

This assumes an inherent asymmetry. It assumes that what the religious person cannot do, the scientist can. This reasoning is what shows us that the argument in fact rests on the strict definition of “prove.” It is a yes or no question being asked by the atheist. One can either prove, or they cannot.

This is not what science is. In fact, this is precisely what science must not be in order to function. The dynamic and self-correcting nature of science demands that there is no such thing as an absolute confirmation.

And only through a proper comprehension of science can this be understood. This is because what science is, ultimately, is a systematic discipline that accumulates and organizes knowledge about the natural world. Because of its concern with accumulating knowledge, science is inescapably tied to epistemology, which is concerned with the nature, origin, and limits of human knowledge.

Epistemology stands outside science, and tells us how we know what we know, the methods used to acquire knowledge and the justification for claims. In short, if you do not understand the mechanism behind knowledge acquisition you do not understand science. Period.

Here is the key aspect of the epistemological mechanism that shows us why falsifiability is so crucial in science. There is an asymmetry in knowledge that shows us only negative observations are confirmatory, while positive observations never can be.

No amount of confirming instances that “swans are white” can ever prove a theory that claims “all swans are white” to be true. This is because there always exists the possibility that a black swan exists (the counterexample). But, this is not the case for disconfirming evidence. All it takes is a single counterexample (one black swan) to disprove this theory.

This is a critical aspect of how science does, and does not, work. A scientific theory must expose itself to potential failure, because there can be no absolute knowledge of a thing’s existence; there can always be a counterexample.

Scientists believe in a theory, not because it points directly to what is, but because of its survivability; when it is robust to attack, further testing, further observation. A theory gives us confidence, BUT, it is always a provisional confidence. To toss away the previsionary property of our confidence in a scientific theory is to destroy what makes the scientific process valid.

Without falsifiability, a theory could evade critique indefinitely, claiming that the absence of evidence against it is support for it. But as we know, absence of evidence is not evidence of absence.

Thus, the reason we can never “prove” a theory to be true in an absolute sense (which again, the original atheistic argument invokes this strict sense), is due to the asymmetry between confirming and refuting a theory.

To toss away the previsionary property of our confidence in a scientific theory is to destroy what makes the scientific process valid.

This also relates to our use of the heuristic known as Occam’s razor. The survivability mechanism, that science uses to accumulate knowledge about the natural world, is the reason we favor theories that admit the least amount of assumptions. This is not about using nice sounding, easy to explain theories, it’s about appreciating the epistemic barriers to knowledge acquisition. Only a “simple” theory can be destroyed, and it is through its destruction that we learn better descriptions of phenomena. Again, we cannot measure things directly, so we must look for a thing’s destruction or survivability to assess its validity. It is an indirect proxy to witnessing an aspect of nature. It is NOT, as per premise 3, a direct observable example of a thing’s existence.

A scientific theory is a placeholder that coaxes information from a complex setting (nature); it has no direct path to the actual phenomena of interest, no matter the measurement. This is the only way knowledge acquisition can function under real world complexity. Only a naive model’s survivability can lend credence to the theory that model rests on.

But what if we soften the word “prove” to mean something akin to increase our confidence in. This means that science provides us a framework whereby we witness a thing’s survivability, which gives us a glimpse at something real about the phenomenon of interest. Thus, while we cannot observe or measure a thing directly, we can at least accumulate some form of knowledge about its existence; albeit indirect. Would the atheist’s use of science be valid if we tweaked that 3rd premise to be softer?

But a scientific theory does not define a thing in any literal sense, it only puts forward an explanation that describes and predicts the properties of the things we observe. This is not a definition of existence, it is merely a description that helps us represent some aspect of our natural world. These descriptions are neither complete nor even “right” in the sense atheists are using it. In fact, there can be many ways to describe the same phenomena, using equally valid yet different representations.

A scientific theory is a tool, that we use as a placeholder to discuss and indirectly understand our world, and which we break in order to update our description.

Statements like “we have proven atoms exist” are meaningless. There is no such thing as proving the existence of an atom, there is only a scientific theory that helps us describe and predict the properties of what we call “atoms.” We used to think, as per the name, these were indivisible units of matter. And as for the electrons that surround the nucleus we have little idea of what an electron even is, and the descriptions we do have are some of the most indirect and ill-defined in all of science. Electrons take on countless paths, and seem to exist as what we call a wave function, whose collapse upon measurement reveals itself only because we chose to observe. There are many interpretations of what this might mean. Any intellectually honest account of what we call an atom would never proclaim we have proven its existence, as such a statement, epistemically, makes no sense.

All science can do is point to patterns in nature that appear invariant to our senses, and give those patterns labels. If those patterns have a repeatable consistency to them then we believe we are on to something. But this, at best, only confirms that the world continues to conform to how we perceive and reason about it. This is not the same as saying we have proven a thing’s existence. Again, that is an anti-intellectual, pseudoscientific statement.

In fact, whether you like it or not, the approach to defining patterns is not that unlike what religion does. Religions are not haphazard collections of naive convictions, they are typically systems of accumulated beliefs that have survived the test of time. Certain human behaviors have been shown to produce negative results over the long term, for reasons we cannot fully know, which compels societies to put in place conservative rules and rituals that govern social interactions. It is obtuse to suggest that these chosen patterns are random. These patterns are invariances amidst the vagaries of a lived life, that have been observed and codified into rules over time. There is interpretation, adjustment and refinement, but the deepest truths within these patterns reoccur with such frequency that they remain.

Again, these are merely descriptions of what humans experience. Whether it is a microscope or an emotional realization is irrelevant. Humans observe, perceive, and codify invariances. Both science and religion fit this human activity.

This is why, when Gervais says:

If you took every religious book and destroyed them, and you did the same thing with every science book, in a thousand years time those science books would be back, exactly the same, because the tests would all turn out the same. But those religious books will either never exist again, or they would look totally different, because there was no test.

…it makes little sense. Both science and religion are observing patterns and hanging onto the invariant parts. The scientist may choose to believe their invariant patterns are in fact reality, whatever that means, and the religious person may choose to believe their invariant patterns are in fact reality, whatever that means, but this step is where we move from the system of pattern recognition and description to the act of belief.

We should expect both science and religious books to return much the same, because both are codifying invariant aspects of nature (human experience is still nature). This is why a stoic book written almost 2000 years ago will still resonate with people today. This is why the Bible continues to do the same. Take almost any religious text and you will find invariant truths in there; not truth in the sense of some supernatural being existing, but in the sense that the life situations and patterns that rang true then still ring true today.

There was indeed a “test” when it comes to religious texts; that test was life and time. Only what continues to ring true for countless generations will survive as an invariant aspect of human nature. And again, I am not talking about anything supernatural, I am talking about the patterns in life that get codified.

The Irony

We can see that, ironically, a lack of falsifiability is precisely what the original argument claimed religious people are guilty of, without realizing it was committing the exact same transgression. Resting an argument on the premise that science proves things is unfalsifiable nonsense, because science can never prove things.

What Ricky is doing — what many/most atheists do — is invoking science to defend their disbelief in the existence of a God. It is essentially atheism because science.

But atheism because science is not a good argument. In fact it is logically unsound. It is unsound because it is using, as one of its premises, the claim that science can do something it cannot.

Science cannot be concerned with the existence of a supernatural entities because there is no way to observe/measure a supernatural being. This means whatever religious people believe in cannot, under any circumstances, be tested by science. In the entire arsenal of the scientific toolbox there is nothing to confirm or deny what religious people believe.

This is why science cannot be used to logically argue for the non-existence of God.

This is not about whether a God exists, this is about logic. It is logically unsound to argue that atheism is buttressed by science. To think otherwise is to not understand science.