How Do We Know What's True?A Story by Rick PuetterEssay concerning objective reality. A companion piece to "Why Do We Believe?"This image is licensed by R M Media Lid under a Creative Commons Attributions-ShareAlike license which permits the use for any purpose, private or commercial, with or without modification. How Do We Know What’s True? Facts vs alternative facts. Belief vs knowledge. How things work vs why things work. What we can know vs what we can’t.
How do we know what’s true? That’s a good question, isn’t it? And it’s probably the most important question of all, since all of our other beliefs and actions are based on what we believe to be true. But before this question can be addressed, we have to address a different question, namely what is truth? This has different answers depending on context and, in the end, we have to admit that this has different answers depending on the person. After all, in the largest context, the person asking the question is, in fact, the ultimate context.
Belief versus Objective Truth
Now if there is nothing that can be agreed upon by a group of individuals as to what is actually true, then we’re done and there is nothing more to write. Then there is no objective truth, i.e., truth outside of the individual, and each individual has his own truth. However, there are areas of human endeavor where large groups of people agree on things, and this lends itself to the idea of objective truth. So, in this essay, we will not be discussing matters of individual truth, personal religious beliefs, etc. Such “truths” may be very real for an individual, but demonstrations of these truths rarely extend beyond the individual. We generally describe these truths as “beliefs”.
So how does one wrap their arms around “objective truth”, and derive confidence that any particular truth is on “solid ground”? This is a challenging question, and we shall see that there are limits and qualifications to these truths. We shall also see that in a final analysis that such truths also have subjective qualities, but we’ll postpone that discussion until the end.
So here is the basic problem of “objective truth”. I believe this and you believe that. Prove to me beyond a shadow of a doubt that I am wrong, and that you are right. And remember, I will try with all my might and wit to uphold my belief, and to disprove yours. So, to win this debate, you must utterly shatter all my arguments, and even more, you must make me believe as you do.
This is not a lazy person’s definition of truth or proof. This takes lots of work. What’s more, demanding this type of proof really requires training. For if one is not trained in the spotting of faulty arguments, and getting to the bottom of truths, this program can at worst be foiled, and more typically proved to be non-productive.
Revealing Objective Truth
If people haven’t caught on, I am talking about science. Because in its broadest definition, I have just described the tenants of science and the scientific method. To more formally define this discipline, to convince me of the objective truth of your statements, you must show me how to demonstrate that you are right and that I am wrong. And to avoid being flimflammed by you, since you are a clever deceiver, you have to describe how to do this demonstration so that anyone can do it, and more importantly that I, myself, can do it. Hence the demonstration must be honest and must not depend on trickery or things you’re not disclosing to me or to anyone else that performs your demonstration.
Can a better procedure for discovering truth be made? Well, people have been trying to find one, but so far this is the best procedure of which we’re aware. Now in the abstract, this “scientific method” sounds pretty good, but is it the ultimate method? Are there any pitfalls to the method? Well, yes there are. The scientific method only guarantees that if I or anyone else performs the demonstration of a given truth (performs an experiment), that we are almost certain to get the results described by the inventor of the demonstration. I say “almost certain”, because the inventor has almost certainly performed this experiment before, always getting the same results, and others who have performed the experiment have also always gotten the same results. And a betting man would wager large sums that if I performed the experiment that I would also get the same result. Is this certainty? Well, no, but it is tremendous confidence. What could go wrong? Well, the experiment inventor’s “theory” of how objective reality works can be only a partial and incomplete theory. It could be that there are conditions under which the experiment’s results turn out differently. And this is how science advances and why scientists continue to perform different experiments under very different conditions.
Facts versus Alternative Facts
Now what are the various aspects of “discovering objective truth”? Well, most scientists would say there are two parts. The first is establishing the facts. The second is interpreting the facts. Now in truth, the only really secure aspect of the scientific method is establishing the facts. And despite claims to the contrary, there is only one set of facts. The facts are the facts. The purpose of experiment is to establish the facts, and if this is done honestly, and by a variety of people, the facts should be relatively secure. The facts (experimental results) are simply a list that says: when this was done, this happened. And that could be the end of the goals of science. However, scientists usually go on to develop a theory, i.e., “an explanation” of the experimental results. The “explanation” is mostly to satisfy the human desire for a reason that things are the way they are. The secondary purpose for a theory is that it might predict the results of experiments that have not yet been performed. Indeed, this is often used as the measure of how good a particular theory is. If it can predict new results, scientists say that the theory “has legs”, i.e., probably encompasses truths beyond just the current experimental results, or facts that are already known. Still, all the purposes of science would be fulfilled by a complete list of experimental results without an “explanation” of why this should be. And we’ll get into a deeper discussion of the limits of human understanding toward the end of this essay.
Problems with the Scientific Method
So, what does the scientific method teach us? Well, it teaches that if we have what we think is an objective truth, or a “theory” about how things are, that if we can’t find a demonstration of this truth that everyone can perform, then that “truth” is only a belief and that other people (“scientists”) will withhold judgment until someone can come up with a “fool-proof” demonstration. Further, other scientists will be very interested in understanding the details of any “fool-proof” demonstration that is discovered, but they will always withhold judgment about whether any demonstration is “fool-proof”. In fact they will try to find situations in which this new demonstration fails. They will continually test the new theory. They know that scientific knowledge is not infallible, but that it only hasn’t been proven to be false so far.
Now how is this different than other beliefs held by individuals? Well, typically other beliefs, such as religious beliefs, don’t have demonstrations that other people can perform to demonstrate their truth. Also, it is not typical for the holders of a particular religious belief to welcome doubters into their fold, and welcome them to challenge their beliefs, and hoping, and helping them to prove that their current beliefs are wrong. Now this is a perfectly human trait, and to be honest, most scientists are disappointed when their cherished theory is proven wrong. But ideally, this is how the scientific method is supposed to be practiced, i.e., emotional attachments to beliefs are to be purged, and one hopes that scientists follow the facts and only the facts.
Critical Thinking and Modern Society
Often scientists equate the scientific method to “critical thinking”. Why might this be? Well certainly scientists are trained to be highly critical of their beliefs. Their goal is to discover “objective truth”, and even to upset current beliefs and ideas. There is nothing that advances one’s scientific career more than successfully advancing new results, new ideas, and upsetting current theories.
Now it is important to understand that new experimental results would never (should never) undo the results of previous experiments. If carefully performed, the results of previous experiments are facts, i.e., if you do this, then this happens. But the new results might be of the type: well you never did this before, and then something unexpected happens. But such results simply add to the list of facts. They do not displace the old facts. As always, facts are facts. These new results, however, can contradict a current theory of why (how) things happen. (Science only tells the “how” of things, never the “why”.) The theory might predict that although we’ve never tried it, if you do this, then this should happen. And the experiment might turn out that, no, that doesn’t happen at all. Something else happens. At this point the current theory must be thrown out or revised in some manner to explain both the old list of facts and the new facts (results of the new experiments). An example is the Newtonian theory of gravity and Einstein’s theory of General Relativity. For a long time, Newtonian gravity explained all of the known facts about the motion of stars, planets, and galaxies. But eventually more modern results from new experiments demonstrated that Newtonian gravity got some things wrong. These are small, but still observable, effects such as with the motion of planets in our solar system, or effects that occur on scales of the size of the Universe, and/or when gravity gets very strong, like near black holes. Under such conditions, the Newtonian theory fails and produces the wrong results. And I should point out, that recent experiments with LIGO have detected extremely powerful events such as the orbital collapse and merging of blackholes, which are totally beyond Newtonian mechanics, but which remarkably agree with the predictions of Einstein's General Relativity.
What is a Good Scientific Theory?
Now so far, we have been discussing two aspects of science: (1) facts, and (2) theories. It is time to get even more critical about our efforts to discover objective truth. Now as has been said so many times already, the facts are the facts. They are not subjective. But what about scientific theories? We have discovered that sometimes they are wrong. So, clearly, they are not a fact. And indeed, the whole edifice of building a theory is to satisfy our subjective need to understand why the facts are the way they are. So, theories are really model building, and there’s nothing fancy or elevated to that. And maybe there are other theories (models) that give the same prediction, or that, in fact, work better. So how do we know if we have a good theory? Well certainly our theory must reproduce the known facts, but generally there are lots of theories that do that. One obvious way to test a theory is to perform new experiments and see if the predictions of the various theories are correct, and then the one with the best predictive power is the better theory.
Now thinking further, it soon becomes apparent that predictive power alone is an unsatisfying criterion for selecting the best theory. We want something more. We want something that makes us feel that we “understand”. We want something that gives us a warm and comfortable feeling in our stomach. We want something that makes us say: “Of course. It’s so simple. Why didn’t I see that before?” But you know what? These are all subjective things. They are not objective at all and have nothing to do with objective truth. In truth, the goal of science is to produce a list of all the facts. Any interpretation or theory is secondary and really not needed. The Universe doesn’t care if you have a warm feeling in your tummy. And different people would need different things to have this feeling. And different alien life-forms would almost certainly need different things if they had a “tummy” at all. So even though we hate to admit it, a theory of how things work is unnecessary. Only a list of how things work is needed. So really, once the facts are in, there is really no such thing as a “good” scientific theory. Good scientific theories were only beneficial when there were still things that were unknown. Good scientific theories might suggest new experiments and speed the collection of scientific facts, but once these facts are known, scientific theory really offers nothing but subjective comfort.
Ockham’s Razor
Before moving on, we should also mention that many scientists judge the merit of a theory by its ability to predict and its simplicity. They would say that to find the best theory, line up all the theories that accurately predict the results of experiments and then pick the simplest one. That is the most likely to be correct. This idea was first introduced by William of Ockham (also Occam), who was an English Franciscan monk (c. 1287 - 1347), and this principle often goes by the title: Ockham’s razor. Now while Ockham had philosophical reasons to believe in his principle of simplicity, modern scientists generally think of this in terms of Bayesian statistics, named after Thomas Bayes (c. 1701 - 7 April 1761), an English statistician, philosopher, and Presbyterian minister, who developed a simple, but powerful, formula for the joint probabilities of two things happening at the same time. Now we’re going to get a bit technical, but I think most people can follow this. Bayes’ powerful formula is p(A,B) = p(A|B)p(B) = p(B|A)p(A). Okay, let’s not get excited about the symbols and mathematics, because it’s really pretty simple. Spoken in English, this says the probability that both A and B happen at the same time, i.e., p(A,B), is the probability that A will happen given that we’re certain that B will happen, i.e., p(A|B), times the probability that B will happen, i.e., p(B). And, of course, this is symmetric and we can reverse A and B which gives the second expression. While that is a simple idea (I hope my readers think so and I haven’t lost them), it is surprisingly powerful, and consequently has developed in to a whole discipline in statistics. But it also “explains” Ockham’s razor. How does it do that? Well, let’s suppose we want to evaluate the probability that a certain scientific theory is correct, given that we’ve done a bunch of experiments and have known results. We could then let A stand for our experimental results being true and let B stand for our theory and just plug that into Bayes’ formula: p(experiments,theory) = p(experiments|theory)p(theory). Again, in English, the probability that both the experiments and the theory are correct is equal to the probability we would get our experimental results if the theory was true, times the probability that the theory is true. So, these are the two terms we were talking about at the beginning of this section. The first term, p(experiments|theory) is how well the theory predicts the measured experimental results, but it is the second term p(theory) that speaks to Ockham’s razor. This second term is how likely the theory is in the first place. Simple theories are more likely than complicated ones. Now at first that may seem surprising, but here is how you can see that. Let’s suppose theory 1, let’s call it t1, has two important parameters, and the theory 2, t2, has a million important parameters. Let’s suppose both t1 and t2 explain our experimental results to the same accuracy. But now ask the question, a priori how likely were we to guess the perfect values of t1’s two parameters compared to having guessed the perfect values of t2’s one million parameters. It’s much easier to get two things right than to get a million things right. Hence as both Ockham and Bayes would point out, the simple theory is more likely to be correct. So, Bayes’ theorem is an important insight. But now that I have enamored you of the power of Bayesian statistics, let me argue that it is all a ruse. When it gets right down to it, the simplicity of a theory is all a matter of language. We are the ones that picked the parameters of t1 and t2 in the above example. We made t1 simple because we had two parameters. We made t2 complicated because we used one million parameters. Let’s use an example of trying to predict what dogs look like. Let theory 1, t1, be that dogs only come in certain breeds, and let t2 be that dogs only come in certain arrangements of atoms. Both t1 and t2 can predict dogs, but t2 is more complicated because there are more possible arrangements of atoms than there are breeds of dogs. But just because describing dogs in terms of the arrangements of atoms is more complicated doesn’t make the dogs more complicated. We simply chose a complicated way of describing a dog. And, in fact, when describing dogs, we do not need to consider all possible arrangements of atoms that make up the dog. Most arrangements would not yield a dog at all. So, we picked a stupid language to identify dogs. Why is it stupid? Well it’s stupid because it is overly complicated. Using breed to describe dogs is much better. Why? Because it’s simple. So, what is happening here? We never try to describe a dog to our friends by listing the exact arrangement of its atoms. When a friend asks what kind of dog do you have, we say we have a black and tan Doberman pincher. That is much simpler. But this also requires something we’ve been sweeping under the rug. It requires that my friend knows what a black and tan Doberman looks like compared to other breeds. So, there must be shared, unspoken information, in this case about dog breeds. Hence the trick and power of Ockham’s razor and Bayesian statistics is you need to pick out an appropriate language to talk about any particular subject. If you don’t pick a good language, there is no way to assess the a priori probability of the theory and its complexity. Among information theorists, they would say there is no information without context. Using our terms we would say there is no complexity without picking a language. And our language is honed by experience. We talk about different types of dogs in terms of dog breeds, but what that means is learned and shared. We talk about physics in terms of mass, velocities, and forces. But that is all learned. And when I describe the elliptical orbits of planets to a physicist by telling him about gravity, he’ll say (she’ll say), wow, how simple. That’s so much simpler than the orbital cycles and epicycles we had before Kepler and Newton. But it’s only simpler because we’re using a learned language that includes gravitational laws and a sun-centered solar system. The Earth-centered cycles and epicycles are still a correct description of planetary motion, it’s just more complicated because we hadn’t learned the right language. And when I describe planetary orbits to another physicist, I just say, “yeah, it’s just gravity”. And that’s all I need to say because being physicists, we have shared information of what that means. So, where are we? This was a long section, but I wanted to make clear that as scientists we can’t find the truth by finding the simplest explanation for things. Rather, it is the other way around. We find the truth and then that truth becomes part of our language and shared knowledge, and from that point on, the explanation sounds simple. Well it sounds simple, at least, to scientists that have that shared information. Higgs bosons, the strong nuclear force and quarks, etc., rarely seem simple to non-physicists. Well, they don’t have all that shared information. They’re unable to speak the right language. They don’t know what the words mean. They’re not physicists. And when I try to explain this to them, I have to describe each term, etc. So effectively they need to get a PhD in physics, and then it becomes very complicated. Once they have that PhD, I’ll describe it to them and they’ll say: “Of course. It’s so simple!”
The Inherent Limitations of Scientific Theories
And now we come to the ultimate question. Can the human species answer all the questions we might pose about the Universe? Unfortunately, the answer is no. We are most likely limited by the nature of our minds, i.e., some important understandings might be simply beyond our grasp. We have evolved here on Earth in a tiny corner of the Universe. Our minds have not evolved to answer many of our fundamental questions. Evolution of our bodies and minds depend on improving survival. But our survival is surely not influenced by an accurate and encompassing knowledge of number theory or the properties of gravity at the event horizon of a black hole. We did not advance at the dawn of man because we had a better understanding of particle physics or the nature of dark matter. So, Mankind is really unprepared and poorly conditioned to grasp the full nature of reality at its most fundamental levels. And while there are fundamental levels at which we can’t address or “understand” the nature of reality, there is also the fact that we are finite creatures. So, we cannot even learn all the things we are capable of learning. There simply is not enough time. We learn slowly. There is much to learn, and we now believe that the Universe is only around for a finite amount of time. Again, the Universe doesn’t respect our aspirations. We live. We die. We learn, then die and forget. And this is true of the whole human endeavor. Civilizations rise. They fall. The comet comes and destroys all life on Earth. Or the sun becomes a red giant, enveloping the Earth. So we need realistic expectations. We don’t want to be naive, but we should realize that there is only so much we can achieve.
So, back to practicality. We are thinking creatures. At least we try. There are methods that are more conducive for discovering “objective” truth. Such a goal is noble and wise. This, I believe, is the path we should follow. It leads to a more fulfilling life. Be we should not get too haughty. And with that, I’ll end this essay on how we know what is true. I hope this gave some food for thought.
Copyright 2020, Richard Puetter. All rights reserved.
© 2021 Rick PuetterFeatured Review
Reviews
|
Stats
506 Views
5 Reviews Added on October 15, 2020 Last Updated on January 5, 2021 Tags: truth, science, knowledge, facts, understanding AuthorRick PuetterSan Diego, CAAboutSo what's the most important thing to say about myself? I guess the overarching aspect of my personality is that I am a scientist, an astrophysicist to be precise. Not that I am touting science.. more..Writing
|