Neil deGrasse Tyson recently tweeted something that, for most scientists is uncontroversial.
The good thing about Science is that it’s true, whether or not you believe in it.
But is this statement accurate? With all due respect to Dr. Tyson, I object to his use of the word “true”.
I do so not because I disagree with it in spirit but because I feel like for years such statements have been misunderstood as statements of philosophical truth and, ironically, used against science, as if science’s claim to truth is in contradiction with its tendency to be self-revising. But, it is not. Rather it is a shorthand way of saying that scientific theories have the weight of evidence to back them up and therefore they are independent of personal feelings, beliefs, or convictions. Yet, they are not “true” and that is actually a good thing for science.
The whole point of the scientific method, according to many scientists, is to determine truth from falsehood by observation, hypothesis, experiment, and revision. This is how theories are formed that we generally regard as “true”. Unlike theological or philosophical statements about God, free will, ethical behavior, and so on, scientific propositions are backed up with data that rule out alternative propositions.
But are those propositions “true”?
Truth, for my purposes here, refers to logical truth, meaning that a proposition or statement follows from premises to conclusion.
For example, I have a statement such as “If it is raining, then the grass is wet.” Then if I say “It is raining”, the statement “The grass is wet” logically follows. Therefore, it is true.
If I just say, “the grass is wet”, however, I cannot conclude that “it is raining” because the “if” statement only goes one way. I may “infer” that it is raining, but I can’t say that it is raining is a true proposition. The grass could be wet because of a sprinkler.
Looking at scientific statements, then, such as Newton’s law of universal gravitation, these are largely inferences rather than deductions. The scientific theory or hypothesis acts as the premise not the conclusion. For example, if Newton’s law of universal gravitation is true, then each member of a two body system (Sun and planet) will obey an elliptical orbit. We observe that two body systems do obey elliptical orbits, therefore we conclude that Newton’s theory is true. But logically, that is an erroneous deduction. We can only infer that Newton’s theory is true, but, if another theory also implied elliptical orbits, then that theory could equally be true.
There are many premises that can imply the same conclusions, especially for scientific propositions that are intended to be universal. The goal of science is to find the best ones.
That is why I prefer, when I’m wearing my philosopher’s cap at least, to use the word “best”, “better”, and “worse” for scientific theories rather than “true” and “false”. Getting used to this language can help you to understand how a scientific theory can be replaced but still be useful.
Indeed, we know that Einstein’s theory of gravity explains phenomena that Newton’s theory does not. For example, it explains the precession of the orbit of the planet Mercury, light bending, gravitational redshift, and many other measured phenomena that Newton’s theory does not. Therefore, the premise that Newton’s theory is true does not lead to the conclusion that we would observe those things. Since we do observe them, it implies that the premise is logically false. Newton’s theory is false.
This is how implications work. If the premise is true, the conclusion must be true. If the conclusion is false, the premise must be false. It cannot be true.
That does not mean, however, that Einstein’s is true nor does it imply that Newton’s theory is useless. After all, Newton’s theory is used extensively in rocket science and ballistics. For most purposes, it is the go-to theory for modeling the motion of every day objects. Einstein’s certainly is not.
Therefore, it is better to say that Einstein’s theory is “better” than Newton’s as an overall explanation of gravity. Meanwhile, Newton’s theory, while false, is still useful.
Somewhere out there is probably a theory that is better than Einstein’s at explaining gravity, a quantum theory perhaps, and yet that does not make Einstein’s useless either. This new theory may be more complicated than Einstein’s and only useful in special cases. Even if we feel certain that Einstein’s theory must be false just like Newton’s, we cannot dismiss it as no better. It is better. And at least for now, it is the best. And both theories are certainly better than our own intuitions or desires which easily crumble under the slightest evidence.
Science does not chase after truth for it cannot determine what premises are true, unlike mathematics where premises are simply defined. Science only chases after improvement.
And it gets it.
The objective nature of scientific evidence all but guarantees that science improves its models over time. Any time you gather evidence that a scientific model fails in a particular regime, such as the failure of Newton’s law for black holes, and another model succeeds in that regime as well as all the others that the other model succeeds at, like Einstein’s, then you can objectively say that the second model is a “better” explanation of the given phenomenon, like gravity.
Likewise, you can say that the old model is false, not just worse, but logically false.
There are times, however, when you can have more than one hypothesis explain the same data with no real way to distinguish between them. In this case neither is false, so how do you determine which is better?
In this case we have a tool that can provide a ranking of theories from best to worst: Ockham’s razor.
Ockham’s razor is a rule of thumb or self-evident axiom. It says that the best theory is the one that explains all the data and is also the simplest. You can’t prove Okham’s razor is true but it doesn’t have to be. Rather it is a way of ranking hypotheses so as to provide the most useful and therefore best answers to scientific questions. It also doesn’t tell you which theory is true. It only tells you which one is the best. And in the end that is all that matters.
Many of us implicitly use Okham’s razor as a shortcut for finding the best explanation for what we experience everyday. If you come home and find your shoes chewed up, you will easily conclude your dog did it not some random stranger. It is only if you didn’t have a dog that your mind would have to stretch to more complex explanations. When this habit goes awry, however, and we start believing complex explanations instead of simple ones, we seem odd and out of touch to others. Just think of how conspiracy theorists look as they espouse complex theories involving numerous secretive actors to explain mundane events.
Then again sometimes complex phenomena demand complex explanations. When out of ignorance we ignore complexity in favor of simple explanations, we get another kind of crackpot, not the conspiracy theorist but the extremist who would like to boil everything down to a single principle.
Science is a balancing act between these two extremes.
One question is whether science will one day discover a theory of everything that cannot be improved upon. Will that theory finally be true?
Unfortunately, logic just doesn’t work that way. The only way you would know a scientific theory is true is if you defined it yourself. In which case it would be tautologically (automatically) true.
You would have to be God.
As long as we are part of a universe and not creating them ourselves, we will have to live with the uncertainty of never knowing what is true, only what is false, and believing not the truth but the best theories we have given the evidence. Yet, to do otherwise would be illogical.