Defining The Head Of A Pin is a series looking into the definitions of various words and concepts, discussing the reasoning behind them, and exploring the limits and limitations of both the definitions and the way we use them.
Definitions are tricky things. Quite a few discussions have decayed because the participants are working with differing definitions for the same word. In many circles, the question of what art is is one of great interest, whether that is artists exploring it in their work or politicians attempting to defund endowments for the arts on the grounds that what they are funding is not true art. But actually discussing what definitions are is…somewhat unpopular. When you poke at many definitions, you tend to find more definitions lurking underneath, like a fractal argument. Those that don’t are usually either artificially created, like well agreed upon country borders (when they are not well agreed upon, we get some of the worst arguments about definitions), or specific, well separated natural states, which are far rarer than one might expect.
So then, one might ask…who wants to spend ages discussing definitions? Who would want to make a whole series just poking at this?
Me! I do! I love poking at definitions, from working out who created them and for what purpose, to finding weird counter examples at the fuzzy edges of them. This is something I do for fun, even for definitions that don’t really have a practical purpose; ones where you may as well ask “How Many Angels Can Dance Upon The Head Of A Pin”* for all the good it does.
* Oh look, we have the series title!
For now though, let’s look at a definition that would seem to be nice and clear, based in the world we live in. In fact, it is directly related to the world we live on.
What is a Planet?
Classical Gas (Giants)
For most of history, the term planet, from the greek for wanderer, has been both well and fuzzily defined. There are five planets that can be seen with the naked eye, other than the Earth: Mercury, Venus, Mars, Jupiter and Saturn. All of these were well known throughout cultures and the world throughout the lifespan of humanity. In a world without light pollution, the night sky is a vast shared connection around the Earth. They were called wanders because they were bodies that didn’t share the motion of the other observable objects in the sky. The stars are so far away that their own motion is not visible on human timescales, and their apparent path through the sky is an illusion, controlled by the rotation and orbit of the Earth. The planets meanwhile were unmoored by these rules, wandering across the night sky by their own logic. Geocentric models of the universe had to produce fascinating models to predict these movements; circles within circles, wheels within wheels. Indeed, early astronomers referred to the sun and the moon as planets, albeit rather inconsistently. This group is sometimes called the Classical Planets. Earth was added to the list after the move from geocentric to heliocentric models of the universe, with the sun becoming the star that the planets orbited around and the moon being relegated to a satellite of the Earth, helped along by the discovery of other moons around Jupiter to provide examples of a body orbiting a planet.
The history of astronomy is fascinating and long, but for our purposes we are going leap forward to 1801 with the discovery of Ceres. When first discovered, Ceres was claimed as the next planet (indeed, the initial basis for the search that found Ceras was a now outdated idea there was a pattern in the orbits of the then known planets, which included the recently discovered Uranus, and this pattern was only broken between Mars and Jupiter). This was quickly followed by a number of other discoveries of planet-like objects orbiting in that area, including Vesta and Juno, and, by 1850, there were 15 known in total! Now we have a new circumstance to deal with: the other planets, with the exception of Pluto, which wanders into Neptune’s orbit (and, of course, neither had been discovered by this point) all hold unique and distinct elliptical (oval) orbits around the sun. But here we have a number of wanderers across the sky, and all in roughly the same place!
Eventually it was decided that these bodies would not be classed as planets, being too small and too clustered to count. William Herschel, the discoverer of Uranus, suggested that they be separated out into their own category, called the asteroids, but this was not widely accepted till after his death. By the latter half of the 1800s, the second major removal of planets from the category had occurred.
This introduces two new ideas for the definition of a planet; one, it must be above a certain size, and two, it must stand roughly alone in its orbit. “Two stars keep not their motion in one sphere”, to quote Shakespeare. Still, this wasn’t too difficult to handle. The other planets were all much larger than the newly classified asteroids, and none of them at this point crossed over each other’s orbits, so the definitions of planet and asteroid could be rather wooly and informal without the risk of creating a grey area between them.
Around this time and into the 20th century, we have the discovery of the outer planets; the two ice giants, Uranus and Neptune, and, beyond them, Pluto. Pluto was unusual even then. It has a far more irregular orbit around the sun, which drops it for some of its orbit within the orbit of Neptune. It was also far smaller; it was originally discovered due to variations in the orbits of Neptune and Uranus, the same that were once thought to be due to a massive body beyond them, sometimes called Planet X. But as more accurate measurements of the ice giants’ weight were made by space probes like Voyager 2, these orbital variations disappeared when the new masses were used. Similarly, when the largest of Pluto’s moons, Charon, was discovered, that allowed Pluto’s mass to be explored, and it was found to be far smaller than first thought. Still, it was accepted into the group.
Properties of Planets
If you are my age or older, these are the planets that you learnt, the familiar list of bodies. All of them share two characteristics; each is nearly spherical, and are orbiting around the sun, with their own spinning rotational movement as well. Beyond that, it is difficult to classify them all together; between them they cover a huge range of sizes, temperatures, rotational behaviours and material compositions, from the huge quickly rotating gas giant Jupiter to the tiny rocky Mercury zipping around the sun. Even the orientation of their rotations differ; while most point at least roughly perpendicularly to the orbital plane and rotate in the same direction as each other, but Uranus rotates at a nearly 90 degree angle to the rest of them, while Venus rotates in the opposite direction to the rest.
One of the traits that most of them share is that they are the only object of their size or larger within their orbital paths; all other objects have either been captured or removed by the larger planets, either turned into moons or banished from the orbit. The one exception to this is Pluto, the smallest of the nine, whose orbit crosses over that of Neptune at its closest to the sun, and flings itself out to the far edges of the solar system at the it’s furthest point.
But despite all these differences, their similarities are enough to give us a perfectly useful definition of a planet; it’s a body orbiting the sun. Heck, we can even throw in the idea of it being spherical (which we will get back to later).So there we have it; nine bodies that we call planets.
However, as a pretty good illustration of a problem with, or perhaps an important trait of definitions; we generally want them to be able to absorb new examples of themselves into the group. If a celestial body is orbiting a star, is not a star itself, and is a spherical body, then that should count as a planet, right? Nothing in our definitions preclude adding new members to our list of nine, and with Ceres we even have an example of a body already known that could be added.
And indeed, the solar system provided plenty of other potential planets. Most are out beyond Pluto, even more of those tiny ice spheres orbiting so far out the sun is just a slightly brighter speck on the sky. In 1992, a small body known as Albion was discovered outside of Neptune’s orbit, and has been joined by then by thousands of other tiny objects in what is called the Kuiper Belt; a ring of icy debris that is the source of many of the comets that fly into the inner solar system before returning to the depths of space. While Pluto’s extreme orbit is unusual for a planet, it is actually fairly standard for bodies in the Kuiper Belt, who’s orbits tend to be extreme and often at an angle to the plane that the rest of the solar system lies in (called the ecliptic). In 2005, Eris was discovered, named after the Greek goddess of strife and discord. Aptly so, since while Eris is smaller than Pluto in terms of volume, it actually has a larger mass than it, which means that if we accept Pluto as a planet, we should probably add Eris.
Now we have a problem: we have a whole array of bodies that should fit the category of being a planet, with a notable difference between those bodies large enough to clear their orbit (that is, they are the only bodies in their orbit of comparable size, other than any moons they have which are trapped in the planet’s gravitational pull), and Pluto, Eris and Ceres, all of which have very much not managed that. It’s got three potential solutions. Either we tack on all these new planetary bodies, or we create a new definition that potentially removes some that we already have, or we give up on having a clear cut definition of a planet and just say “a planet is what people accept to be a planet”. Some astronomers had at this point already started not calling Pluto a planet, but there was no consensus throughout the scientific community. But perhaps one could be made.
Enter the IAU
The International Astronomical Union (IAU for short) is the worldwide professional body made up of astronomers from PhD level and up which are active in studying and educating on the subject. Founded in 1919, it meets once every three years. As the largest international body in the field of astronomy, it is considered to be the authority on naming and defining astronomical objects and phenomena. However, discovering new planets was still rather rare, so it wasn’t until 2005 that a committee was set up to discuss potential new definitions of a planet. Three initial proposals were made: cultural (do enough people think it is a planet?), structural (is it large enough to form a sphere?) and (orbital) dynamical (is it large enough to remove other objects from its orbit?).
Eventually, a second committee built a preliminary definition on the second of those three, the structural idea. To quote the initial proposed definition in full,
A planet is a celestial body that (a) has sufficient mass for its self-gravity to overcome rigid body forces so that it assumes a hydrostatic equilibrium (nearly round) shape, and (b) is in orbit around a star, and is neither a star nor a satellite of a planet.
This would have resulted in an expansion to the definition of planets such that Eris and Ceres would be included, raising the number to 11, or potentially 12 if we include Pluto’s largest moon, Charon. See, the size difference between Pluto and Charon is so low that Charon does not actually orbit Pluto, as a normal moon does, but rather both orbit a point outside of the two bodies. It was therefore proposed to list Pluto and Charon together as a double planet.
There are two major advantages to this definition. First of it, defining a planet based on it being spherical is a lot clearer than setting an arbitrary size distinction on the size of a planet. Objects become spherical when they are large enough that their gravity overcomes the internal forces keeping them in a particular shape, deforming a less regular shape into one where all of it is as close to the centre of mass as it can be. This point is known as hydrostatic equilibrium. (Note that this is talking about the planets as they would be without their rotation, which causes them to bulge out at the centre and squash down at their poles).The exact size required does depend on the composition of the planet and how resilient the internal forces are to gravity, but overall it’s a pretty useful definition. Also, this is a currently observable property of a body, so it will hopefully be more resilient to time as new scientific knowledge is discovered, which definitions based around more specific, less currently well understood processes like planetary formation.
The other advantage is that it was thought people would be more willing to add new planets to the list rather than removing Pluto, which, as the most recent planet discovered, had a special place in a lot of people’s hearts.
The major disadvantage is that it doesn’t have any notes on the orbital effects of a planet. Over millions of years, the dominant gravitational body in an area will either force the surrounding bodies into a particular orbit or it will eject them from its orbit, and as Pluto shows, just because you are large enough to reach hydrostatic equilibrium doesn’t mean you are large enough to do that. (In fact Pluto is locked in its orbit by Neptune, in a particular type of orbit that is beyond the scope of this post but means that they will never actually collide despite their orbits crossing over). The definition also discards so called rogue planets; spherical bodies that have been banished from their stars, or perhaps never had one, that are now drifting through the universe unmoored from a parental light giver. On the other hand, these bodies can easily be classified by themselves, and can be linked to planets without being technically classified as them.
One possibly surprising problem is that while, as noted above, defining planets as needing to be spheres is better than defining them as, for example, 2,000 km across or more, what I put in that paragraph above was…I want to say deliberately wrong, or perhaps that I was just giving the elevator pitch. While it is true that whether or not an object is a sphere is less arbitrary than the size, which has a very wide range of values (compare Mercury, 4880km in diameter, to Jupiter, ~70,000km), nature rarely gives us a perfect sphere. There is still an arbitrary definition hiding here, it is just in terms of “how far off do we allow an object to be from a sphere to count”.
There were several steps between the preliminary definition and the final one announced in 2006, but by the end, the final proposed definition, in the form of IAU Resolution B5, read thusly.
The IAU therefore resolves that planets and other bodies, except satellites, in our Solar System be defined into three distinct categories in the following way:
(1) A planet  is a celestial body that (a) is in orbit around the Sun, (b) has sufficient mass for its self-gravity to overcome rigid body forces so that it assumes a hydrostatic equilibrium (nearly round) shape, and (c) has cleared the neighbourhood around its orbit.
(2) A “dwarf planet” is a celestial body that (a) is in orbit around the Sun, (b) has sufficient mass for its self-gravity to overcome rigid body forces so that it assumes a hydrostatic equilibrium (nearly round) shape , (c) has not cleared the neighbourhood around its orbit, and (d) is not a satellite.
(3) All other objects ,except satellites, orbiting the Sun shall be referred to collectively as “Small Solar System Bodies”.
 The eight planets are: Mercury, Venus, Earth, Mars, Jupiter, Saturn, Uranus, and Neptune.
 An IAU process will be established to assign borderline objects to the dwarf planet or to another category.
 These currently include most of the Solar System asteroids, most Trans-Neptunian Objects (TNOs),comets, and other small bodies.
A short follow up was also proposed, Resolution B6, which simply read
The IAU further resolves: Pluto is a “dwarf planet” by the above definition and is recognized as the prototype of a new category of Trans-Neptunian Objects
With a note to set up a new committee to come up with a better name for the new category. They ended up just keeping the name. So. I wanted to just write *shrugs* here, using the asterisks to indicate I was performing an action since I didn’t really have any sentence that sums up my response better than *shrugs*, but I didn’t want to confuse any readers who aren’t used to that rhetorical technique. Luckily I have defined it now, so I can end this section with it. *shrugs*
On a practical level, the main thing that was introduced was the concept of “clearing the neighbourhood”, or orbital dominance, or dynamical dominance, which we have discussed above. Some astronomers argued that the concept isn’t actually that well defined, since there is still a fairly fuzzy boundary with no clear transition point between having cleared the neighbourhood and being surrounded by debris, particularly for objects that might be large enough to eventually clear the neighbourhood around them but haven’t had the time to do so due to the size of their orbit or other factors. Some conditional boundaries have been suggested, with the criteria varying with the mass of the star and how far the planet is away from it, but the mathematics behind this are I think a bit too beyond the scope of this piece. Still, for the solar system, there’s no bodies that really fall into that grey area; the dwarf planets (now defined and including Pluto, Ceres and Eris) clearly are not large enough to dominate their regions, with Ceres surrounded by the rest of the asteroid belt and the outer two bodies being located in the Kuiper Belt and heavily influenced by Neptune.
However, that solar system centricity is another issue with the definition. The field of exoplanets is huge and again, is beyond the scope of this particular article, but just in what we can find from Earth we see a terrifying array of different planets, including gas giants larger than Jupiter orbiting closer to their star than Mercury. Some of these gas giants are so big that, while they can’t fuse normal hydrogen (that is, hydrogen with only one proton in its central nucleus), they would be able to fuse deuterium (hydrogen with a proton and a neutron). This is a lot rarer, and would mean that the gas giants would only been able to act as a star for a short while before the core runs out of deuterium…but does that make these a type of star, known as a brown dwarf, or do they return to being a planet when they run out of fuel?
There is also a semantic argument to be made about the definition. Pluto and its fellows were defined as a dwarf planet, but that still involves the word “planet”, which creates a potential confusion by implying that dwarf planets are a subset of planets, rather than a seperate class of celestial body.
The vote was held at the end of the IAU conference in Prague in 2006. There were 424 astronomers at the vote, and the session notes states “Resolution 5A: “Definition of Planet” was not counted but was passed with a great majority.”, while the removal of Pluto was more contentious, but “was passed with 237 votes in favour, 157 against and 17 abstentions.” This was, in and of itself, a controversial move; not every member of the IAU attended the conference, and even among those that did, as it was one the last day of the conference many attendees had already left. Still, 424 out of 9000 members (at the time of the vote: at the time of writing the individual membership list is over 13000) is actually enough to get a representative sample…if we assume that the members who stayed to the end of the conference would be a random sampling of members of IAU. I’ll leave it up to the reader to make their minds up about how acceptable they find that.
The discussion had revealed a number of splits in the astronomical community, discussed in an article by Robert Roy Britt for Ad Astra (although I found it on Space.com) in 2006. In the article, David Morrison, a member of the IAU and NASA researcher who was at the final vote said he saw two splits in the community; firstly, there was one between planetary astronomers who focused on their orbital dynamics, who introduced the third criteria, and those planetary astronomers who focused on the make up and formational development of a planet. Secondly, there was a perceived split between the American delegates and the rest of the world; Pluto was the only planet discovered by an American and was very popular in the country, and so Morrison argued that some members from other countries saw attempts to keep Pluto as a planet to be somewhat nationalistic.
After the vote was made, there was a popular backlash against Pluto being downgraded from a planet, indicating the initial proposal was indeed correct that Pluto’s position was very popular. Astronomical institutions had letters of varying degrees of politeness arrive, people created catchy signs and protested, and there have even been songs written defending Pluto’s definition of a planet. Former mayor of Seattle Mike McGunn wrote an article arguing that Pluto represents an under-dog, and that it isn’t right to banish it just because it isn’t large enough. Curiously however (and I realise I have not been able to do a fully representative analysis) but few of these passionate defences of Pluto argue for Ceres or Eris being added to the definition, leaving us back where we started…to a degree. Despite claims that it would be overturned, the IAU definition has remained for over thirteen years at the time of writing. However, we still have three camps on the issue (There are still more potential definitions, including ones that would class bodies we would usually call moons planets, since they do orbit the sun as well as their parent planet but those are not really popular enough for me to really want to get into here):
Do we add new planets, potentially up to 200 of them? (Imagine the mnemonic rhymes needed for that.)
Do we remove the Dwarf Planets and just keep the Big 8, as the IAU did?
Or do we follow the route of many who argue that the definition should be cultural, and thus we just keep the old nine planets many are familiar with? Mike Brown, the discoverer of Eris, made this argument in an online post prior to the IAU decision, that there was no way for the astronomical community to win; either they removed a planet, or told the public they needed to learn multiple new planets at once. He proposes that instead they stay out of it, using a metaphor that we will come back to in the next installment in this series.
If I may editorialise slightly, from my perspective I do see the point of separating the bodies in the solar system by their orbital dynamical effects. While I do understand the idea that a planet should be based entirely on their makeup, whether it is able to achieve hydrostatic equilibrium and become a sphere, removing the orbital criteria removes an important aspect of the nature of a planet. Perhaps the issue was just the terminology; “planet” is a term that has been around for centuries, if not millennia. Maybe the bodies now called simply planets could be called major planets, or a similar term that acts as a sub-category of planets, joined by dwarf planets. Still, I suspect that terminology would have disagreements as well. Eh. *shrugs*
Defining a pincushion
I wanted to start off this series by looking at the definition of a planet because it throws up a whole load of issues with any definition, even one as seemingly fixed by reality as “what is a planet”.
Who gets to make the definition? Why are they doing so? In this case, the obvious example is the IAU setting the definition of a planet, even above some of the protests of its own members.
How do we handle those grey areas between the clear cut examples that our definitions carve out? What definitions do we use everyday that are not actually particularly sharp definitions, but it doesn’t matter because we don’t have any examples of the edge cases? A good example of this here would be the separation of planet and asteroids after the initial wave of discoveries in the 1800s
How do we handle definitions that vary depending on who is saying them, even if the context remains the same?
And most importantly, how useful are our definitions, and what power might they hold?
All of these are worth exploring, and there’s no shortage of potential examples for us to examine. But for now, let’s drop down from the astronomical scales onto a planetary scale and ask:
“What is a continent?”
Britt. R. R. 2006, Why Planets Will Never Be Defined, accessed from: https://www.space.com/3142-planets-defined.html
Brown M. 2005, The discovery of
2003 UB313 Eris, the 10th planet largest known dwarf planet, accessed from: http://web.gps.caltech.edu/~mbrown/planetlila/
Brown M. 2006, The Eight Planets, accessed from: http://web.gps.caltech.edu/~mbrown/eightplanets/
McGinn, M. 2016, In Trump era, a case for underdogs like Pluto, accessed from: https://kuow.org/stories/trump-era-case-underdogs-pluto/
IAU, 2006a, Resolution B5, accessed from https://www.iau.org/static/resolutions/Resolution_GA26-5-6.pdf
IAU, 2006b, IAU 2006 General Assembly: Result of the IAU Resolution votes, accessed from: https://www.iau.org/news/pressreleases/detail/iau0603/
NASA, 2019, What is a Planet?, accessed from: https://solarsystem.nasa.gov/planets/in-depth/
NASA Solar System Exploration, Overview | Ceres, accessed from : https://solarsystem.nasa.gov/planets/dwarf-planets/ceres/overview/
(A note on referencing: I still haven’t quite decided how I want to reference these articles yet. I want all the references to be online so they can be easily accessed by anyone reading these, but whether to use full academic style references or just add links in the text I haven’t decided yet, and it was getting to the point where I just decided to post it with a list of references and see what people prefer. Let me know in the comments. I will probably come back to this with more references in a later edit.)