Occam’s Razor or the law of parsimony states that in investigating a theory that has multiple competing hypotheses that are equivalent in their conclusions, one should select the hypothesis with the fewest assumptions. I have often heard this popularly reduced to “the correct explanation is the simplest” or the like, but that’s not really what it’s about. To me, Occam’s razor is a tool, a guide, a knife used to pare away special pleading or post-hoc reasoning.
Special pleading is a fallacy where arbitrary new elements are injected into an argument in order to prop it up. “There’s tons of evidence that points towards evolution my natural selection.” “But God just made everything look that it had evolved!” These arguments often cannot be falsified by their very construction. In essence, it’s a problem of creep, where new arguments are added on top of old creaky ones to try to fix the structure when the problem is that its foundations are weak. Occam’s razor is the notion that these spurious should be cast aside and the simplest hypothesis with the fewest number of assumptions should be given priority. “The evidence directly shows that evolution occurred, so we’re going with that hypothesis.”
This is not to say that a more complicated explanation for events is necessarily untrue. It simply states that when operating with limited information and knowledge the most valid avenue of further investigation and research is that with the fewest a priori unnecessary justifications. To allow special pleading as valid means that essentially any claim can be subverted arbitrarily. “Well, wouldn’t this group of people have known about and reported on the conspiracy?” “Well they’re just part of the conspiracy, too.”
If a hypothesis is shown in light of new evidence to be unsatisfactory, then by all means it may be necessary to introduce new assumptions. But as few as possible should be admitted with each step, and they should be treated critically to see if special pleading has entered the picture. Also, Occam’s razor can be misused by attempting to reduce a hypothesis to such simplicity that evidence begins to be rejected so that the model does not get too complicated, possibly for aesthetic reasons. This caution is sometimes called Einstein’s razor, and I’ll end off with the relevant quote from the man himself: “It can scarcely be denied that the supreme goal of all theory is to make the irreducible basic elements as simple and as few as possible without having to surrender the adequate representation of a single datum of experience.”