 If you want to make a biologist friend feel special, just send them a cladogram. What kind? Um... In February of this year, the research journal Cladistics published an editorial which instigated one of the nerdiest of nerdy flame wards the internet has ever seen. Cladistics is a journal for scientific papers about taxonomy, the classification of living things in the different categories to better understand how life evolved here on earth. Specifically it's about figuring out the branch structure of the evolutionary tree, the order in which new species split off from supposed common ancestors to do their own thing. A tree like this one that shows which species split off from which and when is called a cladogram, and figuring out how to arrange those species inside it can be a really big deal. Like putting this split here instead of here tells an entirely different story of the development of this creature. Taxonomists have to put in a lot of careful thought in deciding between two cladograms like these because they can inform our understanding of how other creatures might have evolved the same traits in different ways. There are several different analytical methods for figuring out how to judge if a particular tree is likely to be the correct one, each with its own benefits and drawbacks, and there are certain scientists that have a preference for one over the other, which is where this editorial comes in. In this short piece, the lead editor of Cladistics pretty snoodily declared that one of these methods, maximum parsimony, was philosophically superior to any other, that all submissions to the journal should use it and that if anyone opted to use something else, they should be prepared to defend that decision on philosophical grounds. Now, I'm a big fan of thinking philosophically about things, even in science, but experimental methods are generally justified practically, statistically, or best of all, scientifically. A scientific journal going all in on one particular methodology because it's supposedly philosophically superior is weird. It would kind of be like a doctor prescribing one medication over another on poetic grounds because the name sounded better. However, the editor's link with philosophy wasn't entirely misguided. To understand why a particular method of cladogram analysis would prompt such an audacious editorial and an ensuing flame war, it would probably help to understand the underlying rationale of its operation, the law of parsimony. You might know it by a different name, Occam's razor. Occam's razor is definitely a cool name, but it's not a particularly descriptive one. First, Occam isn't a person. It's a small rural town in the south of England with a population of around 400 and one pup. Just one. Second, the person who the razor is attributed to, William of Occam, a 14th century Franciscan friar and philosopher, didn't invent it, not by a long shot. He just really liked using it. It's fairly common, even in ancient philosophy, with written records of razor-like arguments going as far back as Aristotle, and it's still an important philosophical tool. Unfortunately, the way that it's most commonly stated in popular culture, the simplest explanation is usually the correct one, is pretty much wrong in almost every way. Some people use this misunderstanding of the razor to dismiss more nuanced viewpoints of complicated issues, saying that answers that are simpler and easier to understand are more likely to be right. The way that William posited the razor when he used it was, plurality must never be posited without necessity. Basically, if you can explain the same phenomenon in two separate ways, one which assumes the existence of many things and another which only assumes a few, so long as both work equally well to explain all known aspects of that phenomena, then you should lean towards the one that has fewer moving parts. In the context of clodistics, Occam's razor, also known as Maximum Parsimony, is actually a formal algorithmic method for comparing two trees of gene sequences to see which one is best, in this case the one which requires the fewest steps. Let's say that we have a sequence of genetic code, like G-A-T-T-A-C-A, that we found in a creature that evolved a long time ago. Let's say that we want to find out its relationship to two more recently evolved creatures, G-A-T-T-A-C-C and G-A-T-T-A-A-A. Now, it's technically possible that this final letter in the sequence mutated into a C and then both final letters mutated again into A-A. That would mean that we'd have a tree that looks kind of like this, with three total mutations, but there's an easier way to get from here to there. If we were to say instead that the C-A sequence mutated once into A-A and once into C-C, then we'd be looking at a tree more like this, with only two mutations required. Fewer branches, fewer mutations, same outcome, according to Maximum Parsimony, this is a better theory. There are good statistical justifications for using Occam's razor in a vacuum of any other meaningful information, both in cladistics and in general problem solving. If you have to make fewer assumptions to achieve the same explanatory power with a theory, then you're less likely to make an error in evaluating one of those assumptions, and so your conclusion is less likely to be wrong. But ultimately, it's a heuristic, a rule of thumb for getting the right answer more often than not. It's not a logical necessity, and sometimes razor-like thinking can lead us to the wrong conclusion. In cladistics, one of these razor happy problems is something called long branch attraction. Occasionally, through pure random chance, two totally unrelated species will just happen to get the same nucleotide in the same spot in their DNA. The longer a pair of DNA sequences get, the more likely these sorts of random matches are to occur. And unfortunately, when a Maximum Parsimony algorithm sees these sorts of things, it's prone to spit out an answer like, oh yeah, these things are totally related. Now, every cladogram analysis method that we know of has some sort of theoretical Achilles' heel. Some weird situation, which would cause it to report two unrelated species as being closely linked, or vice versa. This is why, in many cases, good taxonomists will cross-reference a couple different methods to verify that their hypothesized trees more or less check out in several different ways. Which was one of the reasons the cladistics editorial blew up so quickly. Taxonomists are all too aware of the potential shortcomings of Maximum Parsimony, the sort of errors that it's prone to. It was bad enough that the editorial came off a little bit haughty, but to behave as though, even in situations where Parsimony would dependably produce worse results than other methods, it was still somehow superior, that was crossing the line. Twitter exploded with the hashtag Parsimonygate. Biologists the world over expressed their opinions on the issue. Nobody got too badly burned, but for a couple of weeks there, biology Twitter was just crazy. In retrospect, I could sort of see how a misguided editor might get a little hung up on the law of Parsimony and its philosophical importance. Occam's razor is absolutely a crucial part of rational scientific inquiry. It takes an awful lot of convincing evidence to say something like, no guys, we really need another force to explain how atoms hang together. But Occam's razor isn't foolproof, either in philosophy or in taxonomy, and ignoring other methods of analyzing information might help keep our answers simple, but it's not going to make them better. What do you think of hashtag Parsimonygate? Please leave a comment below and let me know what you think. And a quick reminder, thunk episode 100 is coming up, so if you have any questions that you'd like me to answer, please either leave them below or send me an email. Thank you very much for watching. Don't forget to follow us, subscribe, and share. And don't stop thunking.