Science is not Truth

Let’s begin with a radical idea: science is not the truth/true, and does not give us the truth.

That is a mighty big claim, and so requires a lot of proof. The proof will come in 3 forms: theoretical, historical, and theoretical again. Some of these ideas have been taken from Thomas Kuhn’s The Structure of Scientific Revolutions, which ought to be mandatory reading by 8th grade.

So, let’s look at the proofs.

Proof 1: Theoretical

Let’s assume that someone tells us that science is true, and gives us truth. We can flip the statement around; science is not false, and it does not produce falsity. Yet, rather obviously, the latter position has been debunked to death. Between bad studies, bad application, and the apparent erosion of the peer review process, a lot of what gets published is simply false. Therefore, science does produce falsity. And then, because science is ultimately a collaborative effort, others go on to take false ideas and build their own scientific ideas atop them – ensuring that the resulting science is virtually incapable of being true.

But, you may object that the perversion of the scientific process (e.g. intentional publication of false or misleading ideas) is not science – it is a perversion of science. In a way, you would be correct. One should not blame the tool for its misuse. The point of ambulances and EMTs is to save lives. But if a deranged EMT decides to run people over with the ambulance, it becomes an instrument of death. Obviously, neither the profession nor the vehicle are to blame.

On the other hand, the question we posed was about the veracity of science. Science is what scientists do. Science is represented by its results, methodology, and common use. We cannot simply decide that, because we believe that science is true, any science that ends up being false is not to be considered science. That’s called data manipulation. And the fact of the matter is, a lot of scientific methods and results, as seen in the common use by scientists, are false. P-hacking alone is enough to negate the claim of science giving us the truth.

We cannot separate the field from the acts of its agents. Science in 2019 is represented by the scientific endeavors of the scientists working in 2019. Now, we can ask the question again, are the endeavors of the scientists working in 2019 such that the results are not false, and do not produce falsity? Besides the problems in veracity that arise from intentional corruption of the system, there are the problems of mistakes, insufficient data, biased data, bad interpretations, etc. All of these problems combine to create a system where the results are not always true, nor is the method capable of inherently avoiding all false claims. But, there’s more.

Let’s look at the idea that “science is true, and gives us the truth” from a different angle. If the statement is true, then science is incapable of being false. If it gives us the truth, then what it does not give us is falsity (the latter follows from the former).  Is it conceivable that any science, despite all the checks and tests we put it through, can arrive at conclusions that are simply not true? Of course it is. For one thing, science is being carried out by human beings – and human beings make mistakes. For another, human beings are limited in their knowledge, and are incapable of avoiding bias (even the selection of “relevant data” introduces a bias). And so, we see that, because science is capable of giving us false ideas (for any number of reasons), it cannot be true in such an absolute way.

Here, you might object that no human endeavor is capable of avoiding all possible pitfalls in advance, and so it is unfair to label science as not true and not giving us the truth. However, that notion would then limit the concept of truth in science to something like, science can, sometimes, be true and can sometimes give us the truth. But that statement is not only radically different than the initial one, but leaves us with a bit of a paradox: if we use science to get at the truth and know what is and is not true, then how do we distinguish between true and false science? Are we going to use science (which may or may not be true) to discover which claims of science are and are not true? That’s analogous to saying that, because Bob is colorblind, we will use Steve to check Bob’s color-coding. But, turns out that Steve is colorblind as well.

Proof 2: History

Whether science itself or by any particular claim, the history of science causes a massive problem for the notion of science as true and truth-giving.

Every science, and every claim in those sciences, assumed itself to be true. This should be obvious, because no one starts off from the assumption that their work and conclusions are false. Otherwise, there would be no science at all. Yet, from a historical perspective, just about every claim ever has been entirely debunked and discarded. Let’s take a walk through the graveyard of revolutionary science, shall we?

Aristotle’s explanation for gravity was true for about 1700 years, and then was entirely rejected. Newton gave us a new theory of gravity and physics, but that only lasted for some 200 years, when Einstein’s work completely debunked it. And now, we’re likely to debunk Einstein in the next 50-100 years. Each thought they were right; that they had the truth. And each of them was dead wrong.

Earth was at the center of the universe since long before Aristotle. Then, Copernicus, Galileo, and Kepler tossed that idea right out. But they replaced it with the sun in the center. A few centuries later, the Milky Way was the center, then the Big Bang is the center…

Phrenology was the cutting edge of science, until it was dismissed as nothing.

European scientists could demonstrate, with absolute certainty, that the people of African heritage were less intelligent than Europeans. And they did just that, all the way through the 19th century. Today, their demonstrations of infallible certainty are seen as a joke.

Darwin’s ideas were true, except that, like the Ship of Theseus, nearly every concept Darwin expressed has been debunked. Slow evolution? Nope! Comes and goes in bursts. Granted, the concepts got updated – but at this point, the original ideas are barely recognizable.

Pick any field of science you like, and you’ll find it filled with claims about absolute, unshakable truths – which were debunked and entirely rejected in a matter of centuries, if not decades. So, the notion that science is true/the truth is just not tenable. You may be tempted to protest, but turn the example around. Imagine if your money manager kept making claims that a stock was a great investment, and that you can – with no reservations or doubt – sink all your money into it. And now imagine that every time he said that, it turned out that the company went bust. How likely are you to keep him on? To keep your money in his care?

So far, we see that science can be false and produce false results; and that it has historically done just that – over and over again. In fact, you would be hard pressed to find any historical scientific claim that has not been debunked – or any science whose very existence is predicated on historical ideas that have been entirely dismissed.

Proof 3: Theoretical

By this point, you are likely to have thought up of the following two challenges:

  1. We might have been wrong, but we’re getting closer to the truth!
  2. If our science is wrong, how come things work in the way it predicts they will? How come my cellphone works?

We’ll take these on in reverse.

First, whether a science is true is unrelated to whether its predictions work. Sounds a bit funny, I know, but we have examples galore.

Malaria was understood to be an illness created by “bad air” – the literal meaning of the term. And so, the Roman doctors recommended that the military never stay in places that have bad air – i.e. swamps and other low-lying areas with stagnant water. While the diagnosis was wrong, the cure worked – and still does, because the carriers of Malaria are mosquitoes.

Newton was wrong about the very nature of reality, but still predicted where the cannonball will fall with extreme precision.

Ibn Sina worked on the assumption that there are 4 elements (earth, air, fire, water), and created a medical textbook that would serve the entire Islamic and European world very well – with great results in effective cures. Yet, he was dead wrong on the number of elements, etc.

The steam engine locomotive was designed by the rules of Newtonian physics; its description of the nature of reality is wrong – but works well enough, despite that problem.

What the “solutions” of science aim at is not the truth, but functionality. Being wrong about the nature of the world is irrelevant to science, as long as things behave the way we predict they will. If my explanation for gravity happens to be based on the love that the earth element feels for its fellow earth components, but the way I quantify that love tells me the rate of fall and force of impact in a predictable way, who cares if my explanation is wrong? That’s because my explanation is functional.

Further, my ability to be right in my predictions has nothing to do with being right about the truth. Every ancient civilization has had a whole lot of explanations for change in seasons. These are all incompatible. And yet, they all worked well enough to allow the people to correctly prepare for the coming of the next season. Who cares if my explanation depends on Hobbits, so long as the tribe survives?

Second, the idea of getting closer to the truth is one we base on the increasing functionality: things increasingly work the way we want them to. However, there are two distinct insurmountable problems here.

  1. To say that we are getting closer to the truth would require us to know what the truth is. That is to say, you cannot claim that you are getting further from or closer to something whose location is unknown. By analogy, if I do not know where I am going (what the destination is), simply driving about does not allow me to say that I am getting closer. That notion of closer is premised on a reference point – in this case the truth. So, if I do not know what the truth is, what my point of reference is, how can I claim to be getting closer?
    1. This is the equivalent of being placed in a pitch-black room, with a firearm, and told to hit the target – which cannot be seen. You take a shot, but get no feedback on what the bullet did or did not hit. In fact, you’re not even sure that there is a target in the room. So, how exactly do you adjust? And how can you say that, after such a blind “adjustment,” you are closer to the target?
  2. Given the historical pattern of every scientific claim arguing that it is the truth – only to be rejected down the line – there is no basis for arguing that we are now somehow closer than before. We may be in a different place, a new place, a place we think may be the destination. However, without knowing what the truth is, and with a long history of always being wrong, there are no grounds for making this claim.

You see, the story we have been sold (think: every science class you have ever taken) is that science builds on the earlier ideas – so that we stand on the shoulders of giants. That gives us a picture of the development and progress of science that we are approaching some magical point where truth lies. Something like this:

But, we now understand that this model does not work. Instead, as Kuhn notes, what we have is something like this:

Notice that we only have a function of time here, not of Truth. Perhaps we could replace the Truth axis with one based on functionality – i.e. problem-solving. However, as systems change, what we do is not solve the old problems. Instead, we generally throw away a lot (or all) of the old problems along with the old theory. And so, even a linear progress of functionality would not be quite right, as it would indicate we have the solutions to previous problems – problems that are no longer considered valid scientific questions.

Kuhn makes a crucial observation in his opening chapter:

The more carefully they [historians of science] study, say, Aristotelian dynamics, phlogistic chemistry, or caloric thermodynamics, the more certain they feel that those once current views of nature were, as a whole, neither less scientific nor more the product of human idiosyncrasy than those current today. If these out-of-date beliefs are to be called myths, then myths can be produced by the same sorts of methods and held for the same sorts of reasons that now lead to scientific knowledge. If, on the other hand, they are to be called science, then science has included bodies of belief quite incompatible with the ones we hold today.

The study of the history of science results in two options:

  1. If old science is to be understood as a myth (not true, based on crazy ideas), then the modern science is equally capable of creating such myths, and does so. Thus, in a few centuries, people will look back on all our work and consider us to have been some kind of insane.
  2. If old science is to be understood as science, then there is no truth in science as such – nor should we look for it. Instead, we should focus on functionality and problem-solving ideas generated by science.

This is Kuhn’s main thesis, and he goes for option 2. He does so by turning away from the truth, and towards functionality.

We kept nothing of phrenology. We kept nothing of Aristotelian explanation of gravity. Even the terminology itself – terms like Planet, which are Greek in origin – have been so redefined that they now encompass a range of things the Greeks most certainly did not intend, nor would they agree to.

Thus, the idea that science is the truth/true and gives us the truth takes yet another hit. Science is unrelated to truth. The latest and greatest of theories is, in fact, unconcerned with the truth. It cares only about functionality. So long as it gets results it predicted, the “truth” is irrelevant.

You may be thinking: but Copernicus was right – at least in the general sense! And you would be wrong. Copernicus claimed that we go around the sun – and that part may be true. However, the sun, for Copernicus, is the center of the universe – and so he is wrong. Entirely wrong. Also, as it turns out, the fact that the Sun is not the center of the universe, means that we’re not going around a stationary object. That sun is hauling ass at well over a million miles per hour; and we are chasing it. That’s a spiral, not a circle.

The issue with cells is the same. We’re not likely to discover that we’re not made of cells. But what that means is likely to be entirely overturned. See, when Greeks used the term “Planet” it meant a celestial body that wanders across the sky – i.e. is not in permanent position relative to the rest of the stars. But, the way you see a planet is by standing on Earth, and looking up at night. Sure, we use that term still, but now Earth is defined as a planet. Can you see earth wandering across the night sky by standing on Earth and looking up at night? Nope. The very meaning of the term was changed to the point that the original usage would reject it. Similarly, when we redefine the notion of a cell, as we inevitably will, we will still be made out of cells. Yet, what that will entail will be so radically different as to negate any previous truth claims.

Think of the idea of the atom. It means “indivisible” in Greek. The smallest unit out of which all things are made. We still use the term, but now it means something else entirely. For one thing, it no longer means indivisible – as atoms are divided into protons, neutrons, and electrons – into the nucleus and the electron fields. And then, those particles are additionally divisible into 20ish quarks. And quarks may be divisible…. Until you end up at string theory. You can see the incommensurate differences between the modern atom and the traditional one by the implications it has. The original concept of atom could never support the notion of string theory. The modern one made it all but inevitable.

Now, you may think that this position is somehow anti-science; that it denigrates the human efforts to understand the world. That is neither my intention nor the outcome of this idea. Pretending that science gives us something that it, rather obviously, does not, is a good way to harm science. We have plenty of historical examples of what happens when a field overpromises on what it can do, and then fails to deliver. Disillusionment, abandonment, suspicion regarding all its claims.

If we take science as what it is – a problem-solving method whose predictive and problem-solving powers seem to increase over time, yet which makes no claims about the actual nature of the truth – all we have done is ensured that everyone understands what science is and is not. No longer are bad claims somehow indicative of science as a whole; no longer is the act of being wrong (an inevitability) somehow detrimental to science as such. We are using these tools as best we can, to do the most we can. These tools are not magical; they sometimes work, sometimes don’t. Sometimes the tools fail, and sometimes it’s the user. But, they are the best we have to determine what is most likely to be a good solution to a whole host of problems we face.

The benefit of this perspective, besides the fact that the alternative cannot stand up to scrutiny, is that it avoids scientism. Scientism is the excessive belief in the power of scientific knowledge and techniques; the blind belief that either science knows, or that it will give us the truth. Scientism is a function of blind belief and fanaticism. It leads to a whole host of bad ideas, from trying to solve non-science problems with science (no different than trying to solve calculus problems by making waffles), to blind adherence for failed and harmful ideas. But it also stifles the development of science. The blind adherence ideologies become ossified and intractable. The problem-solving purpose of science loses all flexibility and dynamism and, like a Roomba with a few sensors missing, gets stuck in a corner, bumping into the wall.

The sooner we accept the idea that science is not true/the truth and does not give us the truth, the sooner we can start using science the way we should – as a great tool for solving all sorts of problems. But, when you’re working with tools, you know that if this tool won’t do the job, another one might. And you can more easily recognize which problems are beyond the use of your tools. If you have a broken A/C, tools are the way to go. If you’re feeling depressed, a screwdriver should not be your go-to solution.

As Maslow put it, it is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail. But once we recognize that we have more than a hammer, then we can introduce some nuance into our problem-solving. You may be wondering, who cares? Well, imagine that your brand-new laptop is having problems. You take it in for repairs, and the guy behind the counter picks up a hammer and proceeds to “fix” it. Suddenly you care. Wrong tool for the job. Ditto if you go to a therapist, and they proceed to beat you about the head with a hammer.

Knowing your tools, and the extent of their utility is crucial for meaningful function.