Planets go missing

Even before the discovery of Eris, an icy body of the outer solar system larger than Pluto, a heated debate was raging over how we should define a planet. Something had to be done. Either Eris was a planet, or Pluto wasn’t.

So the International Astronomical Union set up a committee to consider the issue, to report back to the IAU conference.

The committee came back with the new definition that a planet had to be orbiting a star, big enough to have collapsed into a spherical shape under its own gravity, and not be the satellite of a planet.

I had to think about it for a while, but in the end it was clear that this was a good definition, not because it saved Pluto, but because it was a robust, physical definition. It could be applied to new discoveries without the need for further arguments and conferences. This definition included the nine existing planets and added Eris, Ceres and Charon making a total of 12 planets.

Alas, the IAU didn’t ratify the recommendation. Instead of thinking about what a planet really is, physically, it decided instead to draw an arbitrary line in order to preserve an arbitrary and by no means broadly-held perception of what a planet should be.

The change to the committee’s recommendation was to invent a new category called “Dwarf Planets” to apply to Pluto, Eris and other similar bodies.

Even the name doesn’t make sense. Dwarf planets, the ruling goes, are not determined by size. In fact, they’re the same as normal planets, except that they have not “cleared out the neighbourhood around their orbit” of other bodies.

The orbital neighbourhoods of all planets are cluttered with other bodies. What exactly is an orbital neighbourhood? Here again we have an arbitrary decision and description, not a physical one.

Even Jupiter has a whole retinue of asteroids which share its orbit, the Trojans. We now need to insert a special “secondary” set of definitions - again arbitrary - to exclude the Trojans, just so that Jupiter can still be called a planet.

Space in the outer solar system is huge. If we finally find out, in years to come, that there is nothing else sharing Pluto’s orbit, will we then call it a planet after all? How close to it does another body have to be before it is deemed to have not been “cleared out”?

And what about ellipticity? Just because the orbits of the planets out to Neptune have low ellipticities and obliquities, what logic is there in deciding that all planets have to have this? How small an orbital ellipticity and obliquity does a body need to have before it is a planet? Here is yet another arbitrary, non-physical, line which has been drawn.

What if we find a body the size of Earth out in the Kuiper Belt? Is this still a “dwarf planet”? What if we find a body the size of Neptune out there? Will we suddenly insert some post hoc reason to call it a planet after all? Maybe its ellipticity will be low, so we could just “tweak” the definition to include it that way, without including all that other icy riff-raff out past Neptune.