All Astronautical Evolution posts in 2015:

Britain Takes the Wrong Approach to Manned Spaceflight (Dec.) (Comments)

“Drowning in Process” (Nov.)

Does Intergalactic SETI Make Any Sense? (Oct.)

SETI and Sanity (Oct.)

SpaceX, SpaceY, SpaceZ (Sept.)

A Letter to Britain’s New Space Minister (June)

Mars: 25 Years After Mars Direct – Discussion (May)

The Astronist Mars Strategy (May)

Mars: Still So Distant, 25 Years After Mars Direct (May)

The Mariner Anniversary Calendar for Mars (April)

Mars: An Awful or an Awesome Place? (April)

Should We Phone ET? (March)

More Pluto Controversy (Feb.)

The Pluto Controversy (Jan.)

New in 2015:

Short story The Marchioness

AE posts:

2017: Mars…

2016: Stragegic goal for manned spaceflight…

2015: The Pluto Controversy, Mars, SETI…

2014: Skylon, the Great Space Debate, exponential growth, the Fermi “paradox”…

2013: Manned spaceflight, sustainability, the Singularity, Voyager 1, philosophy, ET…

2012: Bulgakov vs. Clarke, starships, the Doomsday Argument…

2011: Manned spaceflight, evolution, worldships, battle for the future…

2010: Views on progress, the Great Sociology Dust-Up…

Chronological index

Subject index

General essays:

Index to essaysincluding:

Talk presented to students at the International Space University, May 2016

Basic concepts of Astronautical Evolution

Options for Growth and Sustainability

Mars on the Interstellar Roadmap (2015)

The Great Sociology Debate (2011)

Building Selenopolis (2008)


Issue 109, 1 January 2015 – 46th Apollo Anniversary Year

Site home Chronological index About AE

The Pluto Controversy: When is a Planet not a Planet?

In 2006 the 26th General Assembly of the International Astronomical Union attempted to settle the status of the Solar System’s smaller worlds. They made two things absolutely clear:

  1. Pluto, Ceres and similar bodies are planets – specifically, dwarf members of this class;
  2. Pluto, Ceres and similar bodies are not planets – because “planets” and “dwarf planets” are two fundamentally different categories of body.

This is called: trying to please everybody, but pleasing nobody.

Artist's impression of Pluto

This year the controversy will inevitably be revived, with both Pluto and Ceres due for their first spacecraft visits. In April 2015 Dawn is scheduled to enter orbit around Ceres, and in mid-July New Horizons will make a flyby of Pluto and its moons. Two worlds will be revealed in close-up. Or two planets?

The reason why the IAU’s ludicrously inconsistent definitions were adopted goes back to the existence of two factions within the IAU: orbital dynamicists and planetary geophysicists, as described by Alan Boyle in chapter 9 of his book The Case for Pluto [1]. The dynamicists focus on the orbital dynamics of a body: Pluto is not gravitationally dominant in its region of the Solar System, but rather a member of the population known as the Edgeworth-Kuiper Belt. The geophysicists focus on the nature of a body in itself: Pluto is large enough to have a spherical shape, a complex internal structure and a surface shaped by geological processes, in contrast to an asteroid, which is merely an irregular mountain-sized lump of undifferentiated rock, metal or ice. In order to keep the peace and attempt to please both groups, the IAU was therefore driven to define Pluto as being both a planet and not a planet.

In its haste, the IAU defined planets as necessarily being in orbit round our Sun. This was obviously out of date even in 2006, as most planets known then, as now, orbit stars other than the Sun. Further, the makeshift nature of the definition is demonstrated by the fact that dwarf planets were singled out for special treatment, while terrestrial and giant planets were not. Important questions were left hanging:

For all these reasons the IAU definition remains unsatisfactory, and will certainly be changed in the near future. But how? What form of words can resolve the conflict between dynamicists and geophysicists?

Defining planets and related bodies

(1) We start with a common-sense definition: a planet is a substantial body orbiting a star, larger than an asteroid, but smaller than a star.

Note that everyone seems to be already agreed that Pluto and Ceres are indeed planets – the controversy hinges on the adjective to be applied to that noun, and whether the noun may be used without the adjective.

(2) How does it orbit? Three broad cases:

(a) It is gravitationally dominant in its region of the planetary system: call this a major planet.

(b) It is not dominant, but a member of a population of similarly sized bodies which have similar orbits or are subordinate to a much larger body in the neighbourhood: call this a minor planet. (Unless it actually orbits the larger body, in which case it is of course a satellite.)

(c) It is not in orbit round a star at all, but an interstellar wanderer. I find the term nomad attractive: it can be used to qualify planet or as a noun on its own, and unlike orphan does not imply origin in a planetary system.

Note that the major/minor planet distinction is already well established in the case of the main belt asteroids, equivalently long known as minor planets.

(3) How substantial is it? Three broad cases:

(a) A giant planet, like Jupiter.

(b) A terran planet, like Earth.

(c) A dwarf planet, like Pluto.

This gives us a binomial system capable of describing a wide range of planets with two independent terms which specify its orbital state and its size separately. Thus Jupiter is a giant major planet; Pluto is a dwarf minor planet.

(4) The system may thus be extended to cases found in exosolar systems which are not present in the Solar System. For example, if a planetary system is found in which one terran or giant world orbits at the L4 or L5 point of another, or if they have overlapping orbits stabilised by a period resonance as in the case of the Neptune–Pluto 3:2 resonance, then the smaller would be a terran minor planet or even a giant minor planet.

Note that under the current IAU definition, an Earth-sized to Neptune-sized body which was dynamically controlled by a planet of say several Jupiter masses in the same orbital neighbourhood would be called a dwarf planet, which would be an absurd use of language. For full versatility, the orbital characteristics need to be described separately from the mass.

(5) But there is a problem. Orbit dynamicists are happy that they can distinguish a dominant body from a subordinate one, so the major/minor categories seem to be clear enough. (Is this, however, still the case in the remote outer Solar System, hundreds to thousands of AU from the Sun, where the time taken for a planet-sized body to clear its orbital neighbourhood, however that may be defined, of rivals may be on the order of the age of the Solar System?) But we still need to draw dividing lines between a brown dwarf (intermediate in size between a giant planet and a dwarf star) and a giant planet, between giant and terran and between terran and dwarf planets, and between dwarf planets and mere asteroids.

These boundaries are in practice fuzzy ones, depending upon both mass and composition. At the top end we are asking whether a giant planet has ever experienced the nuclear fusion of deuterium in its core, which would make it a brown dwarf star. At the bottom end we are asking whether a dwarf planet is too flimsy to have compressed itself into a properly spherical shape, which would make it an asteroid. But deciding these questions would require detailed observations, which in the case of a distant world may not be possible, leaving its status in limbo.

We therefore set the niceties of theory aside and adopt the following rough and ready practical solution.

The most fundamental property of any celestial body is its mass. This is one of its first properties to be measured. If we know anything about it at all, we have at least an estimate of its mass. We define the giant planet/brown dwarf star dividing line to be at 12.6 Jupiter masses, equivalent to 4000 Earth masses. In doing so we must accept that some bodies a bit heavier than this have not experienced any nuclear fusion, and some a bit lighter have done so, because whether fusion of deuterium takes place depends not only on mass but on composition as well. But we need to draw an arbitrary dividing line somewhere.

Starting with this mass and working downwards, the terms giant, terran and dwarf are therefore defined purely as mass classes, thus:

Giant: 4000 to 4 Earth masses;

Terran: 4 to 0.004 Earth masses (or: 4000 to 4 milliEarth masses);

Dwarf: 0.004 to 0.000004 Earth masses (or: 4000 to 4 microEarth masses).

Each mass class thus covers a range of exactly a factor of a thousand in mass. Within each mass class, subcategories of super-, mid- and sub- may be used to cover a range of a factor of ten in an obvious way.

This simple scheme maps surprisingly well onto the known Solar System bodies, as follows. (Only named bodies are included, thus excluding some recently discovered trans-Neptunian objects. Bodies are given in decreasing order of mass.)

A lot of the smaller bodies listed are of course satellites. The mass classes therefore provide a uniform set of descriptive terms to cover all major and minor planets, satellites and nomads.

The terran/giant boundary at 4 Earth masses comes reasonably closely to the point in mass where Earthlike surface conditions give way to those on Uranus and Neptune, but again this will clearly prove to be a fuzzy boundary, and again we may know the mass quite precisely for a world whose surface conditions remain unknown.

At the bottom of the range, the IAU’s stated criterion for a planet that it should be in hydrostatic equilibrium (having a closely spherical shape) was rather weakly linked to mass, and depended upon observations of how rounded a body was. The solution shown here is more inclusive than that would have been: it allows planethood to subdwarf and even some middwarf bodies such as Vesta which would have been too misshapen to make it under the IAU ruling. These were however already traditionally known as minor planets, which they remain under this new system.

How many planets are there in the Solar System?

The upshot is that a scheme such as this provides rigorous definitions to satisfy the view already being put forward some years ago:

John Davies (p.208): “What the whole debate may be telling us is that there are at least three types of planets; rocky terrestrial planets like the Earth and Mars, giant planets like Jupiter and Neptune and a recently recognised class of ice dwarfs which encompasses Pluto, Charon, some of the large icy satellites and the large trans-Neptunian objects” [2].

Alan Boyle (p.197-198): “Even before Pluto was discovered, the solar system was divided into two classes of planets: the rocky worlds like Earth, and the gas giants beyond. Pluto has pointed the way to the solar system’s third great class of planets, no less important than the other two.” [1]

And the last word should go to Alan Stern (PI of New Horizons), as reported by Alan Boyle: “the new view is four terrestrial planets, four gas giants, and hundreds of Plutos” [1].

In more detail

About a year ago I wrote all this up in a paper intended for publication. The journal which was recommended to me and which I submitted it to was not interested, and at present I’ve not got the energy to try submitting to another journal.

If anyone wants to take the trouble to read the paper, it is online here (PDF) and here (Word). But I think I’ve managed to summarise all the key points on this page.

If you can see any publication possibilities for this scheme, please let me know!


[1] Alan Boyle, The Case for Pluto: How a Little Planet Made a Big Difference (John Wiley, 2010).

[2] John Davies, Beyond Pluto: Exploring the outer limits of the solar system (Cambridge University Press, 2001).


Please send in comments by e-mail.

Jerry Stone (1 Jan. 2015):

Hi Stephen, Happy New Year to you!

I’m glad that you share my view that the IAU definition of a planet is an absolute mess! What I really don’t understand is why it wasn’t amended at the 2009 conference, or 2012. So far I have seen no indication that it will be discussed at this year’s conference either! Unfortunately it’s such a closed shop that only IAU members (who have to be professional astronomers) may submit items for discussion, which rules out people like us.

As you know, I present space workshops in schools all over the UK, and when talking about the solar system I discuss the status of Pluto and just what the term “dwarf planet” actually means. I ask for a show of hands from everyone who has seen any film featuring the actor Kenny Baker, who plays the robot RD-D2. I then explain that Kenny is 3' 8" tall, and so he might be referred to as a dwarf. “But,” I then ask, “what exactly is Kenny Baker?” and it doesn’t take long before we all agree that he is still a human. So if a dwarf human is still a human, then surely a dwarf planet is still a planet – it’s just a small one. In fact, that’s what the term “dwarf planet” means – “small planet”, not “small rock”, “small piece of broccoli” or small anything else. According to the IAU, Pluto is a small planet.

I then add that if the IAU didn’t want Pluto and the other dwarf planets to be known as planets then they shouldn’t have called them planets! They could have labelled them “small solar system objects” or come up with a new term, such as “hooglyflif” – a name I invented. That’s what happened when William Herschel came up with the term “asteroid” when Ceres, Vesta, Pallas and Juno were downgraded from being planets.

I have developed a full presentation called “Is Pluto a Planet?” which covers the thorny question of the status of Pluto. It looks in detail at the IAU definition of a planet and some of its unexpected – and unintended – consequences. For example, I can show that according to the definition, we may only have 3 planets in the solar system – and that does not include the Earth. Also, Jupiter – the biggest object in the Solar System other than the Sun – is actually a dwarf planet! I can also show that it isn’t a planet at all…

The presentation covers far more than in the workshops. For example, I mention that our Sun is a dwarf star, as you have in your article, and I then ask if there is a single person who is going to say that because of that label, the Sun is not actually a star? Even ignoring the case of Kenny Baker, I point out that if the astronomical objects known as dwarf stars are recognised as being stars, and the astronomical objects known as dwarf galaxies are recognised as being galaxies, then to be logical and consistent, the astronomical objects known as dwarf planets must be recognised as being planets!

Last year I gave this presentation for the first time, at the Astronomical Society of Haringey, and it was very well received. With New Horizons reaching Pluto this year, I already have a number of bookings for this, including at the International Astronomy Show. I did suggest it for AstroFest but it was rejected as they apparently have a significant speaker on Pluto, though there is nothing on the advertised programme about this. I have wondered about proposing it to the Oxford Union…

Regards, Jerry

Jerry Stone FBIS
Freelance Presenter on Astronomy and Space Exploration
Contact me about space workshops for schools and public lectures on space
Author of One Small Step, commemorating the first men on the Moon
Founder, The Sir Arthur Clarke Awards
President, The Mars Society UK
Leader of The SPACE Project
STEM Science Ambassador
Interplanetary Poet

Stephen Ashworth (2 Jan. 2015):

Jerry, thanks, and Happy New Year to you and the Space Colony Project!

The obvious term for a sub-planetary object would be a planetoid. Not sure how ready people would be to accept the word “hooglyflif”.

I actually think the “planet” definition based on orbital dominance makes sense within the Solar System out to Neptune. Here’s why. Earth is by far the largest object within the orbital zone that can be strongly gravitationally affected by Earth over the lifetime of the Solar System. The next largest object is of course the Moon, and the remaining material consists of near-Earth asteroids whose mass is trivial in comparison. Thus Earth has embodied within itself more than 98% of the material in its orbital zone, and captured almost all the remaining mass into orbit around itself. Similar situation with Jupiter.

Looking at Ceres and the asteroid belt, however, Ceres contains only one third of the mass in that orbital zone, and the next largest bodies are half as massive as Ceres itself. The orbital dynamicists’ perfectly reasonable point here is that in the case of Earth, Jupiter and the other six major planets, the accretion of planetesimals has run its course and the overwhelming majority of the mass collected into one single body, far larger than anything else in the neighbourhood. The concept of the “neighbourhood” seems to be a bit vague to me, especially in the outer Solar System beyond Neptune, but the essential point remains that Earth, Jupiter and so on are the product of runaway accretion, with only crumbs left over relative to the mass of the dominant body. Ceres, on the other hand, is clearly a different case: the accretion process has stalled and Ceres is a member of a population of similar bodies orbiting with similar semi-major axis and period. The largest member, to be sure, but one that contains less than 50% of the mass remaining in that zone.

Pluto, again: the whole debate started when it was becoming apparent that Pluto was not in any sense unique in its locality, but rather merely the largest member of the Kuiper belt population. (Eris is mostly much further out, but its perihelion is within the aphelion of Pluto, and such eccentric orbits throw the whole concept of orbital neighbourhood into question, I would say.)

So I think the orbital dynamicists’ point of view should be respected so far as it goes.

It works currently within the Solar System because there is a clear dividing line in mass between the smallest major planet and the largest minor planet, using the historical terms major and minor to reflect the dynamical difference between Mercury and Ceres/Pluto. The basic problem with the “dwarf planet” term at present is that (apart from obviously being a kind of planet, as you so rightly point out) it uses a size term (dwarf) when it’s actually intended to reflect a dynamical property (unique in its orbital zone versus a member of a population occupying an orbital belt). But this will not necessarily be the case in exosolar systems, where, say, an Earth-sized planet might be found at a trojan point or crossing the orbit of a supergiant planet.

Even here, if a body greater than the mass of Mercury is discovered in the Kuiper belt or the scattered disk, which is perfectly possible, then we would have a so-called dwarf planet that was more massive than a regular planet, and again the terminology would be shown to be stupid.

Hope these thoughts help.


Site home Chronological index About AE