Doublethink, Nondisprovables, and the Scientific Method

The following is an edited and much expanded version of a TEDx talk I gave at the Baylor School in Chattanooga, Tennessee on September 16th, 2017. A video of the original can be accessed here.

This isn’t going to be some tirade against the relationship between politics and truth, although there is plenty to talk about there.

Rather, I’d like to discuss the broader, macro shifts in the relationship between society, truth, and the scientific method. I feel it’s being misunderstood and misapplied. This has many downstream effects; politics is just one of them.

The Scientific Method

The Scientific Method is arguably the most fundamental and powerful mental device ever conceived by man. It’s what allowed us to master nearly every field of knowledge, from construction to circuitry to chemistry to physics and beyond. It’s what allows me to type this on a computer, it’s what allows you to wear the clothes you’re wearing, through the glasses you’re reading them through. It is also a thought process unique to humans.

It can be used to understand nature, to solve a problem, or to invent. Let me briefly lay it out for you.

It usually begins with an observation. We see something about the world, and it intrigues us. We want to learn more. Why ask the all important questions: why or how.

Then, we develop a hypothesis. We apply our existing understanding with the subject to make a best guess as to the answer of our interesting question.

We then formulate a testable prediction. If my answer is correct, then this other thing will also be true. We know that in the situation we observed that our hypothesis fits, but we change the conditions and question the assumptions until we can determine if the hypothesis is still correct.

Then, we conduct the experiment, which either rejects or accepts our hypothesis. We then iterate on this discovery. We take it further, we ask deeper questions, we think of different tests. Most importantly: We see to push things, our hypotheses, assumptions, and understandings, until they break. Only then do we discover; only then do we learn.


Most of the above is most likely not new to you. Every middle schooler in the U.S. is introduced to the above concept, but even if the then you understood it, your teachers didn’t take the time to explain the consequences and long range effects of this type of thinking. I mean these words not in a negative way, but completely literally. At least, that’s what current events seem to show.

1: Certainty

First, it’s important to understand that nothing is for certain with the scientific method. Remember from the above—we simply keep iterating on our hypotheses, coming up with new tests until they break. The only time we can be certain is when the hypothesis is disproven. Until then, the hypothesis is not proven, but simply not disproven.

Scientists can say that they believe something to be more or less likely, or more or less certain, but you can never ask a scientist if they are “sure.” Because if they’re following the scientific method, they never will be.

Take a hyperbolic example: that the Earth orbits the Sun. Before the heliocentric theory (note my emphasis) there were various hypotheses for how the planetary bodies functioned. Most prominent was Geocentrism, meaning that the Earth was the center of the universe. This was also a theory. We made observations, and created models that supported it. Think about it—from our perspective, the Earth doesn’t move. Why should we assume that we’re flying around the small ball in the sky? When our measurements become more precise, and the supposition that the other planets orbited us were questioned mathematically, we created more and more complicated amendments to the geocentric theory to make it still work.

Nicolaus Copernicus, however, figured it all out. He published  De revolutionibus orbium coelestium (On the Revolutions of the Celestial Spheres), but it was mostly a mathematical model that was more elegant, simple, and altogether more sensical than geocentrism. There was no irreconcilable assertion, at least until Galileo.

Galileo, among his many contributions, made an incredible physical discovery. He witnessed the moons of Jupiter. He proved, with undeniable physical proof, that there were bodies orbiting other bodies. This throws a whole wrench in geocentrism, yes fits completely with the theory of gravitation and heliocentrism.

We kept believing in geocentrism until we absolutely broke it, beyond deniability. Then, we adopted a new model of the world.

Every observation, every space mission, every piece of data so far collected supports heliocentrism as the correct model of the solar system. We’re extremely unlikely to find evidence that disproves this hypothesis.

That the Earth orbits the Sun (or more precisely the center of gravity of the Sun-Earth system) is a fact. But it’s also a hypothesis.

We will hold this view until we can disprove it. We accept it as fact and base our calculations, and lives, on it. But there’s always some probability, however small, that we are wrong. That we aren’t orbiting the Sun. And ultimately the foundation of the scientific method is holding it both as a fact and as a hypotheses at the same time. Simultaneously believing it, and building our calculations and lives around it, yet being completely open to being proven wrong.

This is a counterintuitive mental state—it’s no wonder people don’t understand the nuance. It was never explained to them this way! It’s almost doublethink—holding two opposite ideas to be true at once. Except this is science, and fundamental for the expansion of knowledge. If we can’t hold electromagnetic theory to be true, while still being open to it being false, we couldn’t invent computers and sensors and electron colliders to disprove our existing understanding of the fundamental nature of the universe.

Both have to be true, and false, at once.

Therefore, we need a better understanding of certainty. Or at least a better appreciation of how scientists articulate it. When a scientist says something is “extremely likely,” that means, in normal parlance, certain. When a scientist says something is “likely,” that means “we’re pretty confident.” When a scientists says “the data doesn’t support your statement,” that means you’re wrong. Period.

2: Nondisprovable Things

Science says we can prove things, generally not disprove things.

Here’s an example. We can all agree that the Easter Bunny, or Santa Claus, are false myths or stories. They don’t exist. Any rational adult would agree.

But can you disprove the existence of either of those two stories? Sure, you can prove to me that Santa Claus exists, by catching him red handed in the act, studying his magical reindeer, finding his lair in the North Pole. Those are all things that can be proven to exist. But can you disprove him? How can you prove to me that he is a figment of a child’s imagination?

You can’t.

Instead, we have to look at all available evidence and use Occam’s Razor—what the most simple and logical explanation for the observations we are making?

What’s more likely—that a man living in the icecaps has animals that defy the laws of physics and carries enough presents in his small sleigh that would normally occupy the volume of a small mountain—or that loving parents create, in a societal tradition, the idea of such a man, while at the same time leaving gifts for their children under the tree?

What’s more likely, that a human sized and sentient rabbit exists, or that the Easter Bunny is, again, the creation of loving adults?

You can expand this idea to a lot of topics, including religion. You can’t disprove the existence of (for example) the christian God any more than you can disprove Santa Claus—and is it really likely that in all of the universe, the creator made this planet, and our species, special? Hardly. But more on this in a later post.


Our Current World and Why This is Important

The above my seem, perhaps rightfully so, like semantics, the definition of which being “the meaning of a word, phrase, sentence, or text”, but semantics, in this context, is extremely important. As our society advances, science, and the general public’s understanding of it, become increasingly important.

Our world, quite simply, is becoming more important, more complex, and more nuanced, all due to science. The significant changes in the last millennium, and of the next millennium, will all be due to science. But in order for all of us to adapt to these changes, we have to understand the scientific method and its implications.

If we don’t truly comprehend and appreciate the finite distinctions in articulation scientists use, we’re doomed to misunderstanding. These misunderstandings are not trivial—they can be fatal.

Vaccines

In the world of today, the “Vaccines cause autism” (sigh) movement is a great example of both of the above, applied to a topic with significant ramifications.

In colloquial terms, we’ve proven that vaccines don’t cause autism. Yet, every year, thousands of parents neglect their duties because they don’t appreciate this fact.

I’d like to believe it’s just a misunderstanding.

Scientists say things like “no link has been found between vaccines and autism,” but that doesn’t seem to impress or impact people. They want to hear “we’ve proven that vaccines don’t cause autism,” but that isn’t how science works. We can’t disprove the cause and effect relationship—instead, we look at large samples and see of people who were vaccinated and see if they are more or less likely to have autism. When they aren’t more or less likely, there is quite literally no link. We can’t disprove that there is a cause (a nondisprovable). We can’t say with certainty that vaccines don’t cause autism (there is never complete certainty). We can only say that no link has been found.

But people have to realize, that in scientific terms, scientists are screaming “Vaccines don’t cause autism” at the top of their lungs!!

Climate Change

Fortunately/unfortunately, everything I write has a tendency to come back to climate change. This one bugs the hell out of me; it has to do with certainty.

Fact: we cause climate change. Through a myriad of actions we take, we’re causing an overall increase in the amount of Energy in the Earth biosphere system, which results in higher temperatures and higher kinetic energies (storms).

Here’s a great graphic from the IPCC that describes all the various ways we impact the climate (includes natural changes in solar irradiance).

But when you put scientists in front of congressional committees, they have to answer truthfully. They have to answer precisely. They have to answer scientifically.

They can’t be hyperbolic. In my opinion, they can’t, really, be direct.
When scientists are asked if humans cause climate change, they say “the evidence supports the assertion that human activity is the primary driver of climate change.”

This doesn’t make sizzlingly interesting television. It’s not quotable.

It pains me to complain about precise language, but the reality is that those qualities are needed to propel action. Without directness, or hyperbole, or emotive words, there is no action.

The Galileo ‘Defense’

One of my favorite (read: least favorite) defenses often used by Anti-vaccinators or climate “skeptics” is that: “hey, we’re contrarians! Galileo was a contrarian and persecuted for his beliefs. We’re the victims here, let us talk! We’re just as worthy of screen time as the other guys!”

This is such a contortion of reality that it makes my blood boil.

Here’s the thing: any true scientists is able to accomplish the doublethink mentioned above. They are able to look at the evidence that vaccines don’t cause autism and hold that conclusion, while still being open to new evidence that proves them wrong. When there’s new data, they listen. There isn’t suppression of new thought or data. Quite the opposite.

What is being mistaken for suppression is, in fact, healthy skepticism. Scientists are more skeptical of something contrarian the more substantive the existing evidence is. The more confident their conclusions, the more comprehensive data proving an alternate conclusion needs to be. Here’s two conclusions we’re reached: 1) red cars are the most likely cars to get speeding tickets 2) the Earth orbits the Sun.

Which conclusion would need more data to change?

Exactly. That’s why most scientists are extremely skeptical of research that disproves anthropogenic climate change: because the existing data is so comprehensive. That doesn’t mean they ignore it—not at all. Papers are peer reviewed. But that does mean they’re more skeptical. This relationship is integral for science to function; otherwise we’d always be changing our opinions every time a new piece of shinier research came out. That’s now how you find truth.

Just because you’re contrarian doesn’t make you worthy of attention. Substantive evidence makes you worthy of attention. And modern day Galileo-wannabes are most usually extremely lacking of evidence. I have a feeling he would turn in his grave if he knew his life was being contorted in this way.

Emotion > Evidence

Again, this is hardly a novel observation, but humans are often persuaded much more by emotion than by evidence. Why?

I don’t know. Honestly, I don’t think anybody knows. But let’s turn to psychologists; if anybody would know, it would be them. It seems to boil down to three basic principles.

I’m quoting from an article by Joachim I Krueger, Ph.D in Psychology Today [here].

1: Extrapolation From Limited Data

Clearly, one of the functions of emotions is to guide us towards pleasure and away from pain. To succeed in gaining what is good and avoiding what is bad is difficult in an uncertain environment.

In essence, emotions are our way of extrapolating conclusions from limited data sets. Makes sense. I see why we need this.

2: They Help More Than They Hurt

The lesson from these studies is that the experience of negative emotions can help and hurt decision-making; it all depends on the context. Considered in isolation, emotions are rather arational (neither rational nor irrational)…it may be the case, however, the type of context in which emotions help is more common in our world than the type of context in which they hurt.

Maybe, but that would only be the case if evolution actively weeded out emotion in decision making. Perhaps when natural selection was most active the environment did indeed result in emotion driven decisions being a net positive, but times have changed.

Before, we didn’t have collective, comprehensive data and theories spanning time and individuals. Now we do. But, unfortunately, the timescale that natural selection operates on is very different from the timescale in which the environment of knowledge and reason has emerged.

3: Quick & Decisive

The final reason not to discard emotions remains the fact that they make us act quickly and decisively.

I can understand why evolution would favor this. In the days when natural selection was highly forceful, the predominant dangers we as individual and tribal units faced were immediate, direct, and physical. Questions such as: do we run or fight, which cave should we explore, and is that tiger going to eat us, were the most important questions of their time. Those are the types of questions that separate the living of the dead. Paralysis, in these instances, is death by default.

But the kinds of threats we face have changed. Yes, of course, we still need to make decisions about when to cross the street.

But the important, existential-threat questions of our day work on very different timescales. They work on decades, not seconds. They are extremely nuanced, requiring considerations of a multitude of input and output variables. The outcomes are not binary either.

Emotions are extremely poorly suited to anything but straightforward, binary decisions where deliberation must last only seconds. Although they may have been at one point, emotions are not equipped to solve the problems we now face.

Because, emotions, it seems, take into account only a very limited set of data. They value direct sensory data above all else, and are heavily weighted toward recent experiences. When making emotional decisions, numbers are of little consequence. What does it matter what the global temperature averages are when it’s snowing outside?

Emotions, therefore, when applied to the most important questions we face today, are a direct contradiction to wise and logical long-term thinking. Emotions, quite simply, aren’t built to make these kinds of decisions.

This, instead, is where our logical prefrontal cortex must take over. Where we must trust numbers over direct sensory data. Where we must intellectually appreciate waterfall effects, and comprehend the timescales in which we are making these decisions.

The Fundamental Disagreement

And herein lies the problem. We, as voters and citizens of a democracy, are being fed these facts pertaining to long-term decisions in short-term news cycles.

It’s not that the language we’re using to describe these problems is wrong—it’s that we’re being given these words in the wrong context.

Scientists have two different information streams: normal news outlets being for short-term decisions filled with bombastic emotional language; scientific papers for evidence-driven, carefully articulated and precise language pertaining to long-term conclusions.

But the general public doesn’t have this choice. They only have the news cycle, and right now we’re trying to fit the latter into the former. And, clearly, we’re failing miserably.

We have a few options

  1. Create a different news cycle for long term conclusions, and train the public to use it, with the public evaluating data and reaching their own conclusions.
  2. Create better forums for experts to articulate their conclusions to the public, but still allow the public to make these decisions.
  3. Have the public delegate long-term decision making to experts: the public establishes goals and empowers experts to take action, as best supported by evidence, to fulfill those goals.

Unfortunately, #2 seems to be the only logical option. #1 seems extremely unlikely to be practical—it’s impossible to educate the public on the knowledge and thought processes required to make intelligent and nuanced decisions on a broad range of subjects. #3, although very tempting, seems like a very short road to centralized, dispassionate rule, eventually leading to monarchy, etc.

Therefore, we’re left with one real option: create a better forum for experts to educate the public on making better choices.

Sigh. Needless to say, when I started to write this post I didn’t think that I’d end up writing about how the media needs to change… yet here we are. This is perhaps the question of our time, and I’m no more qualified to give the solution to the problem than our best journalists.

I may go into more detail of my proposal in another post, but I think it may be helpful to have a few requirements of such a forum.

  • Only science it discussed. This is a forum for climate change discussion, not immigration policy.
  • Falsehoods are stopped in their tracks, period. Each side is allowed to fact check the other and stop them midsentence, live. If one side wishes to protest their assertion being called a falsehood, there is an appeals process within public view where the flaw in logic, either in the null or alternative hypothesis, is found. Get to the base truth.
  • Scientific peer review is validated and held in proportion. If 1% are protesting, the viewer should be made extremely aware that it’s not a 50/50 “disagreement.” But most importantly, the flaw in their logic must be made extremely evident.
  • No commercials. No time constraints. Yes, there can be 3, 10, or 20-minute summaries for the casual viewer. But you can’t cover something as complex as climate change in a couple 7-minute segments. If it takes 5 hours of debate over a subset of a subject, so be it.
  • The language should be simple. Complicated words only obscure truths and are used for pseudo-intellectual credibility. Start from fundamentals up and teach. No need for anything but simple language, even to convey complex ideas.

At the end of the day, there are a few takeaways.

  • The scientific method is a specific process that has enabled, and will continue to enable, humanity’s most important advancements.
  • It has several implications that must be understood be the general public for society as a whole to comprehend the results of science.
    • 1) Nothing is certain
    • 2) Many things are possible to prove, but many things are impossible to disprove. What’s actually important is their relative likelihood. We must use Occam’s Razor.
  • Scientific doublethink is necessary for science: hold something to be true, but be open to it being disproven, given evidence.
  • Shouts of contrarianism are not worthy of attention, as they are rarely accompanied by substantive evidence. Substantive contrarian evidence is respected in the scientific community, as it should be.