The Definition of a Planet

In 2006, the International Astronomical Union (IAU) voted that Pluto was not a planet, just months after the New Horizons spacecraft was launched with a mission to…visit the last planet that had never been visited by a spacecraft.

Well…that was awkward.

The IAU vote on the definition of a planet was a weird and highly politicized affair in which planet-hunting scientist Mike Brown, the person with the most to gain from keeping Pluto a planet, was one of the loudest voices for demoting it (though he was not a member himself). Dr. Brown wrote a whole book about it: How I Killed Pluto, and Why It Had It Coming. You can read an excerpt online about how part of the circus at the IAU went down. Leading the charge for Pluto ever since has been, Alan Stern, the scientist in charge of the New Horizons spacecraft, who (rightly!) says that the IAU doesn’t speak for all planetary scientists and rejects their definition.

But for me personally, I have to come down on Dr. Brown’s side and say that Pluto is not a planet. This is not based on the IAU definition, which I believe has serious problems, but on a more basic conceptual understanding of the structure of the Solar system. This essay is based on several posts I wrote after the New Horizons flyby of Pluto in 2015 to explain my position.

What’s Wrong with the IAU Definition?

The IAU definition of a planet says that a planet must orbit the Sun, be large enough to become round under its own gravity (larger than a few hundred kilometers across) and must have “cleared the neighborhood” around its orbit. Pluto orbits the Sun and is round, but it sits in the Kuiper belt with a bunch of other similar objects, some of which are nearly as large, so you can’t say that it has “cleared its neighborhood.” Therefore Pluto is not a planet. Instead, the IAU created the new category of “dwarf planets” to describe these kinds of objects, and I think the IAU did the right thing for the wrong reasons. It really doesn’t make sense for Pluto to be a planet, but the IAU definition is clunky, unclear, and incomplete.

What’s the problem? Let’s look at the three parts of the definition and see.

  1. A planet must orbit the Sun.

Yes, that means our Sun. Yes, the definition specifically says our Sun. And yes, that excludes more than 99% of all known planets because the definition doesn’t apply at all to extrasolar planets. From start to finish in the debate, the IAU never even tried to address the problem of extrasolar planets (and this was 2006, when we already knew about 100 or so of them). With out the full picture of the universe of planets that we get from exoplanets (and we’re still trying to understand that today), you can’t construct a definition intelligently. A big part of the IAU’s mistake was trying to make a definition based on this limited information.

  1. A planet must be round.

This part of the definition is not bad in itself. I agree that planets should be large enough to compress into a round shape under their own gravity. The problem is that it’s very hard to tell whether an object is large enough to be round in the Outer Solar System. You can make a good guess based on how bright it is, but you can’t be sure. There’s a large range in how reflective the surfaces of distant objects are. Worse, the IAU has appointed itself as the gatekeeper of which objects are declared to be “dwarf planets” (large enough to be round). Not only is this not at all normal scientific practice, but they have also been very, very cautious about it, not certifying objects that are so large they are dead sure to be round because the telescope technology doesn’t exist to see it directly. Objects that are clear dwarf planets are not listed as such. Mike Brown made the same complaint here. In short, the IAU wrote a definition that is too difficult to actually apply.

  1. A planet must “clear the neighborhood” around its orbit.

This statement is both unclear and misleading–misleading because no part of the Solar System is truly “cleared” and unclear because there is no accepted definition for what that means. Basically, what this rule really says is that a planet must be gravitationally dominant over all the other objects around it. This actually isn’t so bad. Asteroids and comets can be found all throughout the Solar System, but in most parts of it, they are either penned in a particular area by the gravity of a planet, or on an unstable orbit that will see them kicked out of the area after a few million years or so. However, the IAU definition fails to capture that nuance. (Granted, it is thought that planetary systems evolve in such a way that they are always just barely stable over the age of the system, but that’s another story.) It wrote an unclear definition that opened them up to further controversy.

Honorable Mention: Small objects that are neither planets nor dwarf planets are not called something simple like “asteroids” and “comets” by the IAU. Instead, they are given the clunky moniker, “small Solar System bodies.”

At the end of the day, the IAU adopted a definition of a planet in a rush and for unscientific reasons, and they got something that doesn’t quite work as a result. Unfortunately, it probably won’t be revisited unless we discover something that directly challenges our concept of a planet again, like a Mars-sized object in the Outer Solar System (which is possible), which would not be called a planet under the current definition. (Note that this would not apply to the hypothesized Planet Nine because it should be big enough to be gravitationally dominant.)

So What Is a Planet?

The IAU definition needs a lot of work, if we should even have it at all, but that doesn’t mean the situation is hopeless. We can still ask how the word “planet” should actually be defined? However, it’s not simple.

The basic problem here, which may not be obvious to many people, is that “planet” isn’t a scientific term, and it never has been. Scientists often compare it to defining the word “continent.” How many continents are there? Seven…except that Europe and Asia are only continents by convention, not because they’re separate land masses. What’s more, geologists don’t care what people mean when they say “continent” because they mostly don’t use the word in that way, and if they did, they could reasonably declare that Madagascar is a continent because it sits on its own chunk of continental crust. All this means that it’s hard to define a continent in scientific terms. If you try to, you would probably say there are six of them (though that’s not totally crazy because people do talk about Eurasia as a continent).

Similarly, most astronomers don’t particularly care (scientifically) about the definition of a planet, and there are plenty of scientific papers that call objects planets that really aren’t really. For example, the object DENIS-P J082303.1-491201b is described by the NASA Exoplanet Archive as a planet, but at 28.5 times the mass of Jupiter, it must be a brown dwarf, which most people agree is not a planet. The reason it’s called a planet there is probably that, as seen in a telescope, it looks very similar to one, and you use the same techniques to study it.

So in practice, astronomers don’t have a stake in this, but the public very much does, and astronomers do care quite a bit about social outreach, so it is important to establish a clear and defensible concept of what a planet is, for lack of a better phrase, for PR purposes. Notice that I wrote “concept” there. Mike Brown says that scientists work by “concepts” rather than “definitions.” His meaning is a little vague, but essentially, a concept is a general set of properties that a group of objects have in common, and it doesn’t have to be as precise as a formal definition.

So what is our concept of a planet? Here, we can pin down a few things.

  1. A planet should be big.

A planet should be smaller than a star, of course, but it should be bigger than all of the asteroids and comets and such in the solar system. This includes being big enough to be round, which is part of the IAU definition.

  1. A planet should not be a moon.

This even supersedes the “big” qualifier because there are some moons that are bigger than some planets. But it’s pretty well agreed that a planet shouldn’t orbit a bigger planet. (This is ignoring the concept of a double planet, where the two planets are close to the same size because that’s a whole other can of worms.) Note that I did not say a planet must orbit a star, even though most astronomers say it should. This is because the cultural concept of a planet  includes rogue planets, which do not orbit a star.

  1. Planets should be few enough in our Solar System that any five-year-old can memorize them with ease.

This is not a scientific concept. Not by a long shot. But it has been the reality from time immemorial. In all of recorded history, there have never been fewer than five planets and (except for a few years when we were figuring out what’s up with asteroids) never more than eleven. This was the problem with the earlier proposed definition of a planet that would have included the dwarf planets—everything large enough to be round. This definition would eventually lead to hundreds of planets in our Solar System, completely wrecking our cultural concept of the word.

With this concept in mind, I don’t think that we can rigorously define the word “planet,” even for our own Solar System, because we don’t know what’s still out there for us to find, but I think we can address Pluto, and we can also divide our known Solar System into planets and non-planets, although there are still some pitfalls to watch out for.

A Better Classification

If you were a space alien visiting our Solar System for the first time, you would see that there are five main zones characterized by different kinds of objects. Starting from the Sun, first there would be the realm of the four inner planets, Mercury, Venus, Earth, and Mars. These objects are significantly larger than most of the other objects in the Solar System, except the gas giants and a couple of moons, and all the asteroids that come near them are small and in unstable orbits and thus aren’t important to the dynamics of the Solar System as a whole.

Second, you have the asteroid belt, which is filled with thousands of small objects in mostly stable orbits, none of which are dominant over the others. The largest of these objects, Ceres, is much smaller than Mercury (and Pluto) and, while significantly larger than the others, doesn’t exert gravitational control over the zone.

Third, there are the four outer planets, Jupiter, Saturn, Uranus, and Neptune, which are by far the largest objects the Solar System. All other objects in this region are small and in unstable orbits (besides moons) and thus aren’t important to the dynamics of the Solar System as a whole.

Fourth, you have the Kuiper belt, which contains many small objects in mostly stable orbits. There are several large objects that are about the same size, including Pluto, but they are still only half the size of Mercury and don’t exert gravitational control over the zone.

Fifth, you have the region beyond the Kuiper belt. We don’t know much about this region, so we can’t describe it in detail, and it may be possible to divide it into further zones such as the scattered disk, the realm of detached objects, the inner Oort cloud or Hills cloud, and the outer Oort cloud. This region is also composed of a bunch of objects, some of which may be bigger than Pluto. If Planet Nine exists, it would be gravitationally dominant over this zone, but otherwise, nothing out there is.

Now, from this description, it’s pretty clear that there is something special about the eight large objects in our Solar System, so it’s reasonable to call them planets and everything else not planets. That is why I think Pluto shouldn’t be a planet, in a nutshell, not just because it doesn’t fit with the other planets, but because it fits much better with its fellow Kuiper belt objects.

But there is one big caveat that says this shouldn’t be a hard and fast definition, and that is that we don’t know what we’ll find out in the Oort cloud. Our concept of a planet is that it should be big, but what if we find an object in the Oort cloud that is bigger than Mars, or even bigger than Earth, but smaller than Planet Nine. Astronomers think is very possible, but such an object would not be gravitationally dominant, so under this “definition” and the IAU’s definition, that object would not be a planet.

That just feels wrong. It feels wrong for Earth to be a planet and something bigger than Earth to not be a planet. That’s not at all scientific, and many scientists will say that’s a bad reason to make a definition, but it violates our concept of what a planet should be. So what should we call that object? If you make it a planet, the case for Pluto not to be a planet becomes much weaker. If you make it not a planet, it’ll be a much harder sell than with Pluto. I would say you can’t fairly call it a planet, but you definitely can’t call it is a “dwarf planet,” either. I’ve never liked that term because it’s completely misleading. It’s not a planet, and if it’s that big, it’s not a dwarf either.

The bottom line is that I don’t think it’s possible to create a scientific definition of “planet” that perfectly fits with our concept of it. Something has to give, and I think that thing is that an Earth-sized object that doesn’t clear its orbit should instead be called a “planetoid”–something like a planet, but not quite. If we actually had an Earth-sized object that we could point to and say, “That’s a planetoid,” I think that would solve a lot of the cognitive dissonance of Pluto and other objects out there not being planets, and it would show how silly the term “dwarf planet” is. Alas, that’s not likely to happen at least until the LSST turns on in 2023.