Wednesday, August 13, 2014

You've Been Punked! Occam's Razor

In science, just like in every other human endeavor, there are some underlying assumptions, which are ultimately arbitrary - value choices or random procedures. In every other human endeavor, this is quite clear. In science, however, the most fundamental assumptions are often presented as facts or fact-based, automatic or the obviously most reasonable. The problem is these are only arbitrary assumptions, which shouldn't be uncritically accepted, just like you shouldn't uncritically accept anything other people choose to believe or to live by. Since each of those is used as a way of criticizing or attacking astrology, I think it is a good idea to examine them closely. The first one is probably the most popular, the one you will hear mentioned most often - Occam's razor. According to this guiding principle, astrology shouldn't be pursued, because it is an unnecessarily complicated theory. Let's analyze that.

As formulated originally by William of Ockham, "Entities should not be multiplied beyond necessity". In this view, the simpler a theory is, the (supposedly) more likely it is to be true. How to determine the level of simplicity of a theory is already a problem, though. You can define the simplest explanation as the one with the least assumptions, but that doesn't tell you very much about the quality of those assumptions. You could explain everything (in theory) by assuming that particles move because they want to, but they just don't feel like changing the way in which they're doing it, so it appears to be guided by laws. This theory has only two assumptions, but they are a bad kind of simple - the kind that defines simpletons. Measurably, the particle behavior is ultimately chaotic, or in other words random, meaning there is no obvious pattern to it, and that is, according to scientific consensus, the right kind of simple. Therefore the theory of the randomness of the universe is currently the best one, according to the Occam's razor, and that's why science is rolling with it.

Since there is no proof that there isn't any fundamental pattern explaining away the perceived randomness, it is important to understand that it is only an assumption of scientists (that the world is fundamentally random). Scientists do not actually know that to be true for a fact. According to Karl Popper for example, you can't even prove anything to be true scientifically, you can only disprove it. Which would be fine, if scientists really meant it, when they say that they don't know, and constantly tried to disprove what they believe at the moment. The problem is, in regards to Occam's razor, they do not act like they don't know - they act like they do know for a fact. Equating what's likely (in their opinion) with what is a fact, is a choice that scientists make every time when they authoritatively state things like there is nothing after death, there are no mythical monsters, long lost cities, aliens on Earth, etc. Even if you grant to scientists that some claims or theories are indeed unlikely, you need to realize they can still be true. There is a fundamental difference between unlikely and impossible, which some skeptics tend to overlook on a regular basis.

Now you may ask, like most scientists and skeptics often do, why should we study unlikely-to-be-true pseudosciences like astrology, parapsychology, ufology or cryptozoology at all? Wouldn't that be just a colossal waste of time? My counter would be - but how do you know that BEFORE you try to prove their hypotheses, exhaustively? If any of them were proven to be true in any respect, it would disprove the current consensus, which, let me remind you, is largely an assumption, which we as scientists should be constantly trying to falsify in all possible ways to ensure that we are right about it. You can try to support it statistically, but it would only be a thin masquerade for what it really is - a "common sense" gut feeling. How likely did quantum physics look before their assumptions were put to a test? Even after they were proven many times, some of the top physicists still couldn't accept them, including Albert Einstein. Scientists were wrong countless times, since what they thought was likely, was often just a result of their ignorance - planets couldn't move in elliptical orbits, airplanes were impossible, oh and the city of Troy was just a myth, along with a couple of species that have actually survived to this day from prehistory, etc.

The scientists and skeptics are also very inconsistent when dealing with such "pseudosciences" - criticizing people who pursue unlikely hypotheses for not being scientific enough about it, but no qualified respectable scientist could even attempt to do proper research in those fields without being immediately ridiculed and ostracized by the rest of the academia - if they publish positive results. I am not saying that there definitely have been aliens visiting Earth, that all the myths are true or that everyone claiming to be a psychic has an actual superhuman mental ability. Unlike many scientific skeptics, I am also not declaring all gods to be definitely non-existent. I am only trying to keep in mind that what seems likely to some people, is not a proof of fact, and that research of unlikely hypotheses is no less necessary than the research of likely hypotheses. On the level of historical precedent, some apparently "likely" theories were later completely discarded (such as eugenics), while some apparently "unlikely" theories have been later shown to be true. On the level of statistical reasoning, as expert statistician Nassim Taleb writes in his bestseller book Black Swan, the best risk-reward strategy is to bet majority of assets on smaller, but likely returns, but also distribute the remaining minority of assets among several high-risk, high-reward options.

Just as it is the case with quantum mechanics, proving something going entirely against the common sense is extremely valuable, and it is precisely that which moves science forward the most. As Thomas Kuhn puts it, through a revolutionary change of a paradigm, instead of a slow series of incremental evolutionary steps. Even if only one in a hundred wacky theories is shown to be true, it is still worth pursuing all hundred of them, in my opinion. Imagine that we discover that aliens do exist and have visited Earth, or that psychic abilities or ghosts are real, or that there was an even more advanced ancient civilization than we could imagine right now, or the missing link in the evolution of man, or that astrology works, and so on. Each of those discoveries would be paradigm changing. If we follow Occam's razor and reject them beforehand, we will never find out. And if the current way of pursuing these hypotheses is not scientific enough, let's make it more scientific. Who knows, even if the expectations will not be fulfilled, maybe we will discover a newly evolved species, or not miss the aliens' future visit, or better understand something about the human brain we wouldn't otherwise even think to look for. As it is with conspiracy theories - just because you're paranoid, it doesn't mean that nobody is after you. As long as we keep our imaginations in check and methods straight, we can only gain from looking, even for the equivalent of teapots orbiting the Sun in deep space.

No comments:

Post a Comment