A Theory on Flat-Earthers

The modern-day flat-Earther movement is one of the strangest phenomena in modern times. It’s an organization that’s exactly what it sounds like- a group of people who legitimately believe that the Earth is a flat disk. One of the more well-known doctrines of this system is that the Earth is a disk with the North Pole at the center and a wall of ice around where Antarctica should be. The account for gravity apparently by stating that the world is perpetually traveling up at 9.8 meters per seconds squared. Of course, the specifics of they’re theory aren’t particularly captivating; the real question of their movement is “why?”.

It’s probably easy to dismiss these people as lunatics, simple gullible fools who got tricked into believing this rancid hogwash. Perhaps we can argue that mental illness or egotism plays some part. Unfortunately, these explanations work only for cartoon villains and shallow caricatures: and flat-Earthers are real people with lives and families. No, in my opinion, it’s neither satisfying nor respectful to simply label anyone as stupid and close the book. There has to be a compelling reason for holding this set of beliefs; something that justifies a person holding these ideas (from a human perspective, not a scientific one).

A while ago I read this nice article about a brave individual who dove headfirst into this movement. One paragraph, in particular, intrigued me:

For millions, technological advances meant not freedom, but utter misery—and just as it declares that everything can be known, instrumental reason abstracts that knowledge beyond immediate experience. “Enlightenment,” Horkheimer writes, “has always aimed at liberating human beings from fear and installing them as masters. Yet the wholly enlightened Earth is radiant with triumphant calamity.” Faced with a reality that could no longer be intuitively understood, whose secrets had become the property of a small class of scientists and administrators, the early flat earthers tried to claw back some of their autonomy.

Now, this is an interesting point. These days, it almost seems like people aren’t allowed to have opinions anymore. So many issues are topics with exactly one correct answer, and if you ask for justification, you’re told, “because science says so.” What was in the past reserved for simple truths about the basic physics and chemistry is now being applied to nearly everything. Science now says that carbs are bad for you and coffee is good for you. It’s even dictating the most personal aspects of people’s lives. Science will even tell you the right way to raise your kids. Also, atheists are less tolerant than religious people, so if you’re an atheist you should probably stop because your atheism is causing you to be closed minded. Just don’t become religious either because then that’ll cause you to be dumber.

These aren’t “suggestions” either, this is science, after all. These are absolute truths. Therefore, you must follow them exactly or be content with being the moron who questions the literal experts and the unchanging nature of reality. Every day, there are fewer and fewer things about which the average person is even allowed to have an opinion on. As science discovers more and more information about the world the average person must sacrifice more and more of their autonomy. There is a right side and a wrong one and you need to be with the former. No debate, no arguing. This shouldn’t be a problem, because people should do this with information if that information is the truth. It’s pure mental pathology to try and argue with facts.

That’s true, but none of those articles I linked were from peer-reviewed journals. They were from random new sites and similar sources. Those ones are known for misrepresenting scientific data. This is important because this is generally all that non-scientists read. Heck, even the primary source material has issuesCoffee, in particular, is known for being a case study for this weird phenomena.

The point is, we have people being told how to live more and more aspects of their daily lives by information which, as far as Joe Schmo can tell, doesn’t stand up to the scrutiny that science is supposed to stand up to. What this boils down to is that people can quite justifiably start thinking that they’re not only being controlled in an increasingly totalitarian way but by sources which cannot be trusted. This is a recipe for rebellion. Science is about as close to absolute truth as humans can get. So much so that critical thinking is almost discouraged in certain situations. If a large group of scientists in a field agrees to proposition X, and you’re not an expert in the field, you should defer to their opinion. This is not a bad thing. Scientific information compels authority because it is undeniable. However, if you are fed bad information, then it is deniable, and it’s not scientific. It’s an opinion, perhaps even an unfounded one. And all that critical thinking and skepticism you were expected to put aside becomes essential again.

Maybe these flat-Earthers are the end result of a system has allowed enough untruth into itself so that it has instead become more like forcing opinions on people. If the newspaper says that scientific experts now believe X and you find out that X must not be true, then why trust the person who told you to believe X? Clearly, the scientific experts don’t know what they’re talking about! Remember, this example person hasn’t read the peer-reviewed journals, he just reads newspapers and watches TV. At that point, why not start following some guy on youtube? Sure, you’re questioning science and looking at a guy whose only qualification is that he has a camera and can upload videos, but when you can’t trust a source of information, it is your duty as a scientific mind to find the truth elsewhere. Watch this clip of Joe Rogan arguing with a flat-Earther; listen to Eddy Bravo’s constant refrains:

“It looks fake to me!”

“you can’t trust that!”

“The pictures are terrible!”

He doesn’t trust the way the data was collected; he doesn’t trust the scientists to tell the truth, and if he had spent years reading the above examples, it’s understandable how someone could think that way. Granted, it’s still not justifiable for him to think that way. The media’s inability to get the stories right all the time doesn’t mean that NASA is concocting a grand conspiracy, and trouble within medical journals doesn’t change the fact that geologists all believe the earth to be round, but it makes a leap to such a bizarre belief much more explainable.

What if these flat-Earthers are just a group of people who feel alienated and ostracized from the pursuit of truth? We live in a world where more and more facts about life are dictated by a small handful of experts in a way that can barely be understood by the layman. A world where it is expected that everyone believes these facts unquestioningly or risk social excommunication. A world where, even so, the “science” that the average person sees is often subject to correction unbecoming of an epistemology that demands total submission. Is it any wonder then that some would latch onto a belief system that at least gave them back their autonomy? Their humanity? In the flat-earth community, you’re free to do your own research and your own critical thinking, no Ph.D. required.

Estimating Political Spectrum with Machine Learning

With the primary elections well under way, it’s looking like the political season is already starting to heat up. I don’t know about you guys, but my personal online echo chamber facebook feed has become saturated with blogs urging me to like candidate X and hate candidate Y. Oddly enough, while I hear the mantra of liking candidate X because he’s “against the partisan system” and “fairly moderate” I hear this applied to everyone and everything; including clearly non-moderate candidates like Bernie Sanders and Donald Trump. This begs the question: what exactly defines political moderation?

On a related note, what defines the political spectrum period? It’s not a well defined problem (and probably isn’t something where consensus can be reached period). Noam Chomsky claims to be a “socialist libertarian” despite the fact that those two philosophies are, at face value, total opposites. but it is still a question in my mind. The Wikipedia page on the issue has no less than four charts espousing different theories. Take the following, for example:


In this chart we can plot any political philosophy as having a score in two dimensions- authoritarian-ness and liberalism…-ness. Anyway you can plot a point on this like you would on any XY axis.

This chart actually shows a common theme amongst political spectra (that I’ve seen anyway): the idea that, in two dimensions, the political spectra range from liberal to conservative and authoritarian to libertarian. This makes sense intuitively, but it isn’t quite comprehensive. I feel that a more empirically justified spectrum is in order.

The Plan

Essentially, anyone’s political designation is some combination of their opinions of agreement or disagreement on various issues (gun control, taxes, etc…). If we can assign a numerical value to this, it means that we can represent it as a vector. If we get multiple people’s opinions together, we can make a matrix, and if we have a matrix, we can do PCA.

PCA (principal component analysis) is a way of reducing data to fewer dimensions. Essentially, it takes multidimensional data points and reduces the dimensionality from m dimensions to n dimensions, and it chooses the dimensions that best preserve the information given. Long story short, if I have data points in 300 dimensions, I can use PCA to find the best one or two dimensions that represent the data. It doesn’t have to be any of the original dimensions though. PCA can just find dimensions that it likes!

There are already websites that gives us all us presidential candidates positions on a set number of issues (abortion, gun control, etc) all ranked from “strongly disagree” to “strongly agree.” I took this data and vectorized it (-2 for “strongly disagree”, 2 for “strongly agree,” etc…). With this, I can do PCA. The website I’m using to get my data is here.

Also, for this experiment, I’m using the following not-too comprehensive list of candidates. I realize that a lot have dropped out by now, but more data is always better…

  • Donald “make American’s hair great again” Trump
  • Christ Christie
  • Hillary Clinton
  • Jebcito Bush: la tercera parte
  • Mike Huckabee
  • Rand Paul
  • Rick Santorum
  • Ernie “cutting the B.S.” Anders
  • Ted Cruz
  • Marco Rubio

For political issues, I was able to find all candidates issues on the following:

  • abortion
  • same sex marriage
  • affirmative action
  • presence of God in the public sphere
  • limiting EPA restrictions
  • easing voter registration
  • strict punishment for crime
  • second amendment rights
  • expansion of Obamacare
  • school voucher promotion
  • prioritization of green energy
  • marijuana as a gateway drug
  • support of the 2008 stimulus
  • higher taxes on the wealthy
  • path to citizenship for illegal immigrants
  • privatization of social security
  • promotion of free trade
  • expansion of the military
  • belief in American exceptionalism
  • aversion to foreign conflicts

The Results

To start off simply, we take the data and reduce it to one dimension. This, in principle, should place all the candidates on a single line. The results are as follows:

pca1.pngInterestingly enough it seems like this almost perfectly matches the popular notion of the “liberal-conservative” spectrum with more conservative candidates, such as Santorum and Cruz, on one end and more liberal ones, like Clinton and Sanders, on the other. Note, however, that more conservative values were given lower values, and liberal ones were given higher ones. Because of this, the “right wing” is on the left, and vice versa.

The real surprises, however, come with a reduction to two dimensions. pca2.png

As before, we see conservative/liberal appear along the horizontal axis. The vertical one is a bit more perplexing. The fact that Rand Paul appears so distant from the others hints that the y axis may be authoritarian/libertarian (he certainly isn’t shy about how much he values personal freedom). If this was the case, it would make our spectrum the same as other graphs (which also have liberal/conservative and authoritarian/libertarian dimensions). It’s very possible that this representation doesn’t correspond to anything easily intuitive.

Whatever the interpretation may be, we can still gauge candidates against one another. We can measure the distance between candidates like we would measure the distance between any two points. Although in what way the two points are close or far from each other isn’t entirely clear, it still provides a way to measure who thinks like who in this election.