On Pluto and planethood: or, how science isn’t very good at classification

As I write, New Horizons is within celestial spitting distance of Pluto. When the craft was launched in 2006, Pluto was a planet. It isn’t any more.

Later in 2006, the International Astronomical Union (IAU), decided to finally get their shit together and define what a planet actually was. You see, in the run-up to 2006, improvements in observation methods had led to the discovery of what is scientifically known as a fucking fuckload of objects, at least one of which was more massive than Pluto. This would not do, because with each new discovery, it had been possible to say it wasn’t a planet because it was smaller than Pluto (pleasingly, the object more massive than Pluto was named Eris, after the goddess of fucking shit up and ruining everyone’s day). And so the IAU decided to finally figure out what the hell comprised a planet, and couldn’t manage a definition that left you with My Very Easy Method Just Shows Us Nine Planets.

And so, Pluto was stripped of its planetary status, because it only met two out of the three criteria they’d settled on for planethood: yes, Pluto orbited the Sun and was massive enough to be approximately round, but unfortunately for poor old Pluto, it didn’t meet the third criteria–clearing the neighbourhood of its orbit (i.e. there’s other stuff around where Pluto orbits).

This isn’t the first time a planet has been demoted. A little over 200 years ago, it was hypothesised that there was a small planet between Mars and Jupiter, to account for the relatively large gap between the two planets. They found the hypothesised planet in 1801 (then they lost it, then they found it again, but that’s a different story entirely). They named it after the goddess of agriculture, gave it a symbol and everything. Just a year later, another object in the vicinity was found… and then another, then another, and then basically they had a fucking fuckload of the fucking things. It was eventually decided that we would call these small objects orbiting between Mars and Jupiter the asteroids, and they were a different kettle of fish to the planets.

Funnily enough, with the 2006 definition of planets and dwarf planets, Ceres has been sort-of-promoted, to occupy the same dwarf planet class as Pluto. Good for it, I suppose.

The 2006 definition of planethood has been criticised, because of the “clearing the neighbourhood of its orbit” criteria, and not just because it means Pluto is no longer a planet, which kind of makes everything we learned at school about the nine planets a fiction (or something). The thing is, it also means that Earth isn’t a planet, because we have Cruithne, and Jupiter isn’t because of its trojans, and Neptune isn’t a planet either because of its army of plutinos… turns out that this “clearing the neighbourhood” gig is not particularly well-defined either, so if I fancied, I could demote that oversized shitlord Jupiter, with its more moons than it has any business having and its smug red spot.

It’s worth noting at this juncture that while the science upon which the “clearing the neighbourhood of its orbit” is based doesn’t deplanet Earth, Jupiter and Neptune, the wording of the definition adopted by the IAU does. Stern and Levinson wrote a discussion paper, laying out a proposed new scheme which contained size criteria for a planet–small enough to have never ignited a fusion reaction, but big enough to be roundish. As well as this, they suggested dividing “uberplanets” from “unterplanets” by their ability to be gravitationally dominant. This is not simply clearing the neighbourhood of everything: a gravitational interaction is acceptable. The paper is an interesting, and reasonably accessible one, and well worth a read as it outlines a lot of the problems with approaches to defining a planet. It’s assumed that the IAU’s neighbourhood clearance criterion was based on Stern and Levinson’s work, although Alan Stern himself points out that it’s a bit of a shitty criterion and excludes literally half the planets.

So, in short, what happened was we started out with no definition of a planet: it was something we just knew when we saw one. And now, we have a definition of a planet, and it’s pretty much just a manifestation of what feelings we have about planets, which Alan Stern described as “sloppy science and it would never pass peer review”.

That’s a thing with the science of categorisation, though. For the most part, it simply reflects what prejudices and constructions we have about whatever is being categorised, because it’s humans doing the categorising. It is something which needs to be undertaken with a degree of reflexivity. A lot of the time, we can pootle on for years lumping and splitting things into various categories, only for that to be thrown into disarray when new information emerges (take, for example, the advent of DNA analysis, and how that has moved a lot of animals around the tree of life). With new information, we tend to redefine based on what we already think.

Some areas are more behind than others. For example, if we defined planets in the same way we considered biological sex, we’d only count the five or six planets we can see with the naked eye.

I expect that within my lifetime, I will see the definition of “planet” changed once or twice more, to incorporate new discoveries, but, more than that, to maintain a comfortable number of planets that falls within Miller’s Magic Number: I suspect this is the ultimate motivation for creating a reductive approach to categorisation.

So, is Pluto a planet, or is it not a planet? Does it really matter? I consider it so, not due to any scientific criteria–although I am sure I could rustle up a definition. All right, fine, fuck it. A planet should have an atmosphere which isn’t predominatly just particles flying at it from the Sun. This makes Pluto a planet (it has an atmosphere and literally as I write this a spaceship is studying it!), but it makes Mercury not a planet since it doesn’t fit this criterion; which is fine, because fuck Geminis, they don’t need a ruling planet, the two-faced bastards.

This, then, is the challenge in classification, categorisation, and building definitions, and similar problems pop up everywhere. Remember this, when you see definitions presented, and categories built by scientists. Remember it, and question it–what do they stand to gain, or preserve?

__

A couple of years after Pluto got demoted, I was doing my PhD (I never finished). My subject matter involved classification and categorisation, and I turned to the natural sciences to have a look at how they did stuff. I suppose, in its own way, this post is the reinterpretation of the findings of the literature review I published; this is something which lurked in the back of my mind, but I didn’t really have the courage to say at the time, because it would put my work on a pretty shaky footing from the off. I needed to believe that classification systems had some sort of “objectivity” to them. I might have nodded in Borges’s direction then, but now I think I’m basically with him.