Philosophy: Occam's Razor and Evolution

TORONTO, ONTARIO - It occurred to me recently that one reason that Americans have such a tough time accepting the concept of evolution is that it doesn't seem to meet Occam's Razor. A simplistic but incorrect view of Occam's Razor, that it calls for applying the simplest explanation for any given phenomenon, seems to imply that evolution is too complicated. What's missing is an understanding--not scientific per se, just intellectual--that simpler explanations can explain a broad view of a topic, but the details often require more nuanced explanations.

One of the arguments used more often by creationists of late is that the basic concept of evolution as explained by Darwin and others does not adequately explain the fossil records that are found. The original idea of evolution was that random mutation was constantly occurring, and that beneficial traits would be selected for and survive, and undesirable traits would be selected against and die out. As the process continued, enough mutations would occur that the original and mutated organisms would no longer be able to reproduce together, thus creating a new species. If this were actually happening at a constant rate, then the fossil record would show a steady introduction of new species over time and a steady extinction of less adapted species. That's not what is found. The fossil record instead shows species that exist together for millenniums, then in a short period of time a large number of new species appear and a mass extinction occurs, and this happens repeatedly. The most famous such episode was the extinction of the dinosaurs.

The argument goes that since the theory doesn't really fit the data, it must be wrong. Scientists concur, and Niles Eldredge and Stephen Jay Gould came up with the idea of punctuated equilibrium. In simple terms, it states that speciation is more likely to occur on the boundary of a population or in an isolated population, where the mutation is not diluted by the larger population. Otherwise, the sheer size of the population of the original species serves to keep the trait in one individual from spreading and the species stable--only when a boundary population undergoes a significantly advantageous mutation does the new species have a chance to displace the existing species. Thus, they concluded that the fossil record is actually what would be expected--long periods of comparative stability combined with rapid displacements.

There are additional nuances to the theory, which likely has already lost some readers. Thus, the creationists argue, the theory is too complicated. It is actually a simpler explanation to say that God created the various species and the fossil record. It's an explanation that everyone can understand, and thus is superior to punctuated equilibrium based on Occam's Razor.

The fallacy here is that Occam's Razor doesn't really just call for the "simplest" explanation. Occam's Razor is actually about not including elements in an explanation that are unnecessary. The original, continuous evolution theory didn't explain the data, so it was inadequate. The ideas of punctuated evolution extended the theory of evolution in a way that explained the data that had been found. In reality, punctuated evolution actually stands up to Occam's Razor, in the absence of a better theory that explains the same data. Creationism actually fails Occam's Razor, since it introduces an actor--an omnipotent being--that isn't necessary to explain the fossil record, and introduces all kinds of implications that have nothing to do with the data, which in this case is the fossil record.

To go back to my favorite medical simplification of Occam's Razor, "when you hear hooves, think horses and not zebras," imagine for a moment that one has heard hooves. The simplest explanation is that a hoofed animal--not necessarily a horse, even--is passing by. Only if the sound is accompanied by the sight of a horse-like animal with black and white stripes would the existence of a zebra need to be theorized (but if such a sight was seen, it would need to be theorized; a generic hoofed animal would not be enough to explain the data). The equivalent of creationism in this situation would be that God was directing hoofed animals to pass by one's location to teach one some sort of moral message. That doesn't pass Occam's Razor. Moral messages may or may not be interpreted in the situation, but to explain the observations, all one needs to do is theorize the existence of the zebra.

The concept of more complicated explanations being required to explain more complicated phenomena should not be a hard idea to grasp. Most people learn Newton's Laws of Motions in high school, and find these theories perfectly adequate to explain the physical phenomena around them. However, it doesn't take understanding of quantum mechanics to recognize that Newton's Laws aren't adequate to explain quantum mechanics. That scientists have had to extend those laws with additional things that aren't significant in everyday physics (and reduce to Newton's Law in everyday situations) should be easy to accept, even if one has no clue those extensions work.

Getting away from "pure" science, in politics, when one candidate has clearly more experience and more logical ideas than another candidate, it would normally be easy to predict that the first candidate would win the election. Yet, if that candidate is a member of a party that has just made a colossal mistake that angered the public, predicting the result of the election might be much more difficult. The more complicated and nuanced the political environment, the harder it is to predict the election result, or how each candidate may behave during the campaign.

For engineers, this kind of thinking is second-nature. They are always taking rigorous theories and finding simplified rules-of-thumb that give an equally accurate result assuming certain conditions are met, and then use that simpler theory to make their work easier. For non-engineers, this concept is more foreign, but it should still be understandable.

An explanation that is "more understandable" is not necessarily "simpler" if it contains elements that have nothing to do with the phenomena that needs to be explained. In the case of the fossil record, modern views of evolution (that are more complicated even than punctuated equilibrium) have come into existence precisely to try to offer the simplest possible explanation of fundamentally complicated and nuanced data. That they happen to be complicated does not make them fail Occam's Razor. As long as they exist and explain the data, the introduction of an omnipotent actor is unnecessary, and thus it is actually creationism that fails Occam's Razor.

Philosophy: Glitch's Razor

TORONTO, ONTARIO - I violated one of my strongest personal principles in my Friday post on the Fiscal Update Crisis in Canada. I tried to create a case for a conspiracy within the Conservative Party. This was a violation of what for lack of a better name I'll call Glitch's Razor: Never blame on a conspiracy what can be blamed on simple incompetence.

Pretty much everyone has heard of Occam's Razor. While more formal treatments and even Wikipedia explain how the modern interpretation of it is not strictly accurate, in modern North American culture it can be summarized as "the simplest explanation is usually the best." Sometimes it is even equated with the KISS principle--Keep It Simple, Stupid. I personally always liked the wording that is apparently common in medical school, "When you hear hooves, think horses, not zebras."

Glitch's Razor is a corollary of Occam's Razor, postulating that incompetence is a simpler explanation for a low-probability event than a conspiracy, and thus in applying Occam's Razor, the explanation involving incompetence would be expected to be correct. Even under more rigorous views of Occam's Razor, Glitch's Razor would still follow, as it is about ignoring superfluous portions of an explanation (the conspiracy, which by definition involves at least a small group people if not a large one) when the core explanation (incompetence, often of only one person or perhaps a small group of people) is all that is needed.

It seems such an obvious corollary that I am surprised it does not appear to have ever been named. Clearly, thousands of people have had the same thought, and the similar wording "never suspect conspiracy when incompetence can achieve the same result" appears in many counter-arguments to conspiracy theories. If anyone knows of a citation of the origin of this thought, I will be happy to quit calling it Glitch's Razor, as it is far from my original thought.

As a principle, Glitch's Razor applies throughout one's life. In a number of work situations, I have used it to accurately predict the behavior of remote management. Once, my immediate supervisor was convinced that he was to be fired because of a string of events that he took as directed at him personally. Applying Glitch's Razor, I predicted that the events just showed a lack of understanding by the remote management team--basically, incompetence in managing our site, not a conspiracy against the site or my boss in particular. While I was right on that point, the same incompetence detected by Glitch's Razor eventually caused the whole site to be closed in an ill-advised manner.

Do you think your bank is out to get you? More likely, their programmers didn't test their own software adequately. Do you think the government is trying to bankrupt you? More likely, they can't understand and administer their own taxation laws. Was there a conspiracy to make your train late? More likely, a maintenance or operations employee didn't perform some task, and the train was delayed enroute as a result.

Obviously, Glitch's Razor applies in politics, even in Machiavellian politics. Even people that are trying to put something past their rivals or the populace can mess things up. Perhaps the best recent example occurred in Washington state, where professional initiative organizer Tim Eyman simply wrote Initiative 985 poorly. There was no big, successful conspiracy ganging up on him--people just didn't like the initiative.

So that brings us back to what has happened since last Thursday in Canadian politics. It is entirely possible that the Conservative over-reaching in the Financial Update was strictly incompetence by Prime Minister Stephen Harper. He could have misjudged the reaction that would come from the opposition parties, in particular the Liberal Party. By Glitch's Razor, the possibility of incompetence is more likely than a conspiracy within the Conservative Party.

I still have a hard time believing that Stephen Harper is actually that incompetent. But, in the absence of evidence for an alternate narrative, it is an explanation that probably should be accepted, at least if I really believe in my own Razor.