In the manga Medaka Box, the end of the last arc involves the moon dropping down onto the Earth, called down by the death of someone who should not have been killed. While such an occurrence would doubtless be immediately catastrophic in the real world, in the manga, the main character managed to destroy the heavenly body before the rest of the world even noticed. In the final chapter that shows the characters’ adult lives, we learn that she is planning a ‘Moonsault Plan’ to rebuild the satellite she destroyed years ago. This symbolically brought the story to a close, in a rather emotional finale. Speaking from a scientific point, though, not only is this an impossible plan but also makes one question whether what was built was even qualified to be called a ‘moon’.
The definition of what precisely makes a moon is surprisingly vague, mainly consisting of conventions and an I-know-it-when-I-see-it sense of things. The only concrete points everyone agrees on are that it has to be a natural satellite (thus what Medaka made would not have been a moon) with a relatively stable orbit; everything else is undecided because it is unsatisfactory. Let’s begin with the most apparent consideration: size.
If you keep up with the science section of the newspaper, you would doubtless have at some point seen an article like ‘New moon of Saturn discovered.’ By now, there are 82 outside the rings (if we include the ones within the rings, there will be several headaches), which will only increase as our technology gets better and better at spotting tinier things. Obviously, it doesn’t make sense to count every single dust particle orbiting a planet as a moon, but nobody knows precisely where the boundary lies. Some possible ideas I found on the internet are either too arbitrary or too complex, with one going the extra mile, involving Hill radii and Roche limits and some other terms I couldn’t even find on the web.
Just like there is a lower bound on size, there is an upper bound too. Pluto’s moon Charon is large enough relative to it that it doesn’t make much sense to say that it moves around it. Whenever two objects are in orbit, they really move around the whole system’s center of mass; it’s just that in most cases, one is big enough compared to the other that the barycenter can be approximated to its own center. For Pluto, however, this isn’t the case: the barycenter lies outside the planet, which makes for a confusing matter. As much as I want poor little Pluto to be a planet again, this point is rather hard to argue against: how do we justify a ‘planet’ which can’t even properly get its moon to revolve around it?
Perhaps size isn’t the way to go: how about visibility from the surface as a condition for moonship? It sounds sensible but then leads to certain questionable bodies like Phobos and Deimos qualifying as moons, even though they cannot even become spherical through their own gravity. But the biggest issue we run into is with the gas giants, who have no surface, being made primarily of hydrogen and helium.
There was one last criterion I found which could be used for moon qualification: roundness. Suppose a body’s mass and size are significant enough that it can achieve a spherical shape gravitationally. In that case, it is a moon, provided the barycenter of its orbit lies firmly within the host planet. This is actually somewhat better than the other definitions but tends to run into problems when lunar bodies like Hyperion are considered. Due to its porous nature, it is non-round yet is heavier than several moons. In true Universe fashion, things only get more confusing the more we try to make sense of them.
Despite this whole mess of what is and isn’t a moon, I find it strangely comforting that such a seemingly simple thing can have so much variety and possibility if you just look a little closer. It is a testament to the sheer complexity — and incomprehensibility — of the cosmos we live in, which, despite being full of baffling occurrences and unintelligible improbabilities, still has its own irreplaceable charm.
Proofreaders: Mokshit N.