The following is a version of the article published in Mercury
in the Nov./Dec. issue of 2003.
Even before civilization, people looked into the sky and recognized different celestial objects. The Sun defined daytime, and the stars provided a fixed background of faint, twinkling lights at night. Among them moved the Moon, and a few special steadier lights. The Greeks called those which moved "planets" (it is worth noting that the Sun and Moon were originally included, since motion against the stars was the defining characteristic). Most cultures have an analogous word for these "wanderers". Both the stars and the planets were thought to revolve around the Earth.
After the Copernican Revolution, we recognize the Moon as the only body that orbits the Earth. The Sun is a very nearby example of a star, and the visible planets are other large bodies that orbit the Sun. We see them by reflected sunlight, while stars produce their own visible light. This understanding yields the dictionary (lay public) definition of the word "planet": a large heavenly body that shines by reflected light and orbits the Sun. In the past century we gained much understanding of our Solar System, and even visited most of the planets robotically. Yet today, professional astronomers find themselves unable to agree upon a succinct definition of "planet". Replacing "the Sun" with "a star" is obviously necessary now that many extrasolar planets have been discovered, but the problem goes well beyond that.
Two recent controversies that found their way to the popular press illustrate further difficulties. One is the "Pluto controversy". This arose because of the discovery of a large belt of icy objects beyond Neptune. They are the outer remains of the original protoplanetary disk. This "Kuiper Belt" is a natural outcome of incomplete planet formation in the outer Solar System, and is the source of some of the comets we see. As Kuiper Belt objects (KBOs) were discovered in increasing numbers in the 1990s, including a population of "Plutinos" which share Pluto's orbital characteristics (somewhat different from the other planets), some astronomers began to suggest that Pluto itself (which shares many properties with, but is the largest KBO known so far) does not qualify as a planet. The recent discoveries of Varuna and Quaoar (which are KBOs half the size of Pluto, like its moon Charon) may presage the time when we find another Pluto-sized KBO.
The current situation is much like that in the early 1800s, when the first asteroids were discovered. Ceres was originally hailed as the fifth planet, particularly since one in its position was expected from "Bode's Law" of planetary spacings. It lost its status within a few years, when other members of the asteroid belt began turning up. Herschel, who had been the only person to have discovered a new planet before then, aided the effort to demote Ceres. The arguments against its planetary status were 1) that it is much smaller than the other known planets, and 2) since there are many other such objects in similar orbits, they should get a special name and not be called planets. Otherwise we might suffer a planetary embarrassment of riches whose names would be difficult to remember.
It is worth noting that Pluto was originally thought to be close to the size of Mars, but as its icy nature became known, its inferred size shrank, since the inference relied on how reflective the surface was rather than a direct size measurement. With the discovery of Pluto's moon Charon (and its subsequent eclipses) it became apparent that Pluto is substantially smaller and less massive than Mercury (or, for that matter, several of the larger moons of Jupiter and Saturn). Now that many other KBOs are turning up, Pluto has exactly the same problems that caused Ceres to lose planetary status. The main difference is in the length of time (several decades) before these problems became apparent.
The second controversy arose in the late 1990s. Objects so low in mass that they cannot shine by nuclear fusion were found in isolation in nearby star-forming regions. Some of their discoverers called them "free-floating planets". These objects are too small to be brown dwarfs, and seem to have the same mass as some of the newly discovered extrasolar planets (but do not orbit a central star, and are much younger). Some astronomers felt that the use of the word "planet" should be restricted to objects orbiting stars. Again, opinions as to the validity of this classification were offered and debated in newspapers. It became clear that the International Astronomical Union (the only body empowered to settle both matters) has no formal definition of "planet", and valid opinions either way could be expressed. The IAU Working Group on Extrasolar Planets offered a couple of instances which they felt disqualify planetary status: 1) objects which manage core nuclear fusion, and 2) objects not orbiting stars. The first constraint establishes an upper mass limit to planets (which therefore also constitutes the lower mass limit to brown dwarfs). The second deals with the environment of planets, and I'll return to both later.
One might fairly ask whether the issue of what a "planet" is has any real importance or scientific utility, or is "just semantics". Some astronomers opt for the latter choice, noting that Nature produces objects with a continuum of masses, found in a variety of circumstances. However, we humans like to classify and name objects. Since everyone now agrees that we live on a planet, it is hard to maintain that it doesn't really matter exactly what a planet is. In particular, the discovery of other "planets" tends to garner much more attention from the media and the public (and astronomers) than the discovery of, for example, merely substellar objects (which include both planets and brown dwarfs).
Why is "planet" hard to define? Part of the problem is that there are three logically separate arenas in which the definition might be constructed. I alliteratively list them as 1) characteristics (physical properties of the body itself), 2) circumstances (primarily questions regarding orbital properties), and 3) cosmogony (formation mechanisms and histories). Much of the debate is really about the relative weight each of these arenas should carry (and to a lesser extent, what the defining properties should be in each arena). For example, the dictionary definition contains a bit about characteristics (a planet doesn't shine by its own light), and specifies a circumstance (a planet must orbit a star). Among the problems with this definition are: it doesn't specify a lower mass limit to what qualifies as a planet, and it doesn't cover cases like the asteroids or KBOs (which technically satisfy the dictionary definition). On closer inspection, it doesn't really specify the upper mass limit either, nor does it distinguish between various ways in which a body may "shine" (neither the source of the luminosity nor the spectral region in which it occurs). Finally, it turns out that astronomers (much more than the public) care about how planets and brown dwarfs form, which is not covered in the definition at all. The loudest (and least well-informed, both theoretically and observationally) arguments tend to occur among professionals in the formation arena.
In the arena of characteristics, one property of interest is luminosity. Given that the Sun is powered by fusion, and the Earth is not, there is general acceptance that one reasonable dividing line for the upper mass limit on planets could be placed where fusion becomes possible. This happens when the self-gravity of the object renders the core hot and dense enough to ignite deuterium (which is the most easily fused nucleus, but rather rare). This occurs at about 13 times the mass of Jupiter. Above this limit, objects fuse other elements when they are sufficiently massive to manage it (and after they have exhausted easier fuels). The next dividing line is when brown dwarfs can manage lithium, then hydrogen fusion (at around 60 jupiters). We don't call objects truly stellar until 75 jupiters, when contraction (which has continually supplied substantial luminosity since birth for all objects, though slowed by fusion) is fully halted for an extended period of time by hydrogen fusion - the "main sequence". At successively higher masses helium, then heavier elements, can eventually be fused (but we don't give separate names to these various levels of stars). To simplify these technical concerns, I'll call any object capable of fusion at some time during its life a "fusor".
Another possible dividing line could be chosen based on the means the object uses to support itself against gravity. Very low mass objects (like the Earth), use Coulomb forces between their atoms (solid or liquid forces), to hold themselves up. Gas giants use gas pressure (the interior of Jupiter reaches 20000K to generate enough pressure, though one might argue about whether the state of matter there is "gaseous"). The luminosity of such objects comes from the fact that as they cool off, they shrink, releasing gravitational potential energy. Above about twice the mass of Jupiter, a new form of pressure takes over. Called "electron degeneracy", it is the means of support for objects all the way up to true stars (which revert to thermal gas pressure generated by the intense heat of their fusion). We needn't worry about the physics behind this here, except to note that if you add mass to a degenerate object, it actually compresses (to smaller volume and higher density). Adding mass to stars and objects under 2 jupiters makes them larger. Thus, we could place the upper mass limit for planets at 2 jupiters, finding another name for degenerate non-fusors (like "superplanets"; degenerate fusors are brown dwarfs). This is not really under consideration, because some already accepted extrasolar planets would violate this limit (particularly in view of the fact that the Doppler method yields only lower observed limits on mass).
At the low mass end, there is one easy way to draw a line. All the accepted or putative planets are round. This is because they have enough mass to allow their self-gravity to overcome any material forces (the sources of pressure support listed above) that might produce asymmetric shapes. One can calculate the lower limit for this mass (which depends a bit on whether the object is rocky or icy). It is a below the mass of Ceres (reaching objects less than about 500 km in diameter). This seems like a good place to draw the bottom line for planets. Almost all asteroids and comets are too small to make the grade. It is hard to come up with another physically motivated criterion. One might quibble over what constitutes a violation of "roundness" (technically roundness means conformity to the equipotential surface, for you lawyers out there who note that rotating planets are actually oblate), but the basic limit is clear enough.
As regards circumstances, the first question is whether orbiting a fusor is a required property of a planet. Both the dictionary and the IAU come in with a "yes" to this question. Any disagreement, therefore, stems from the judgment of some astronomers that circumstances (and cosmogony) do not belong in a basic definition of an astrophysical object - only characteristics should count (this is true in most other cases). I began from that position myself, but have concluded that the cultural tradition is too strong for this reasoning to hold sway. In the end, we need a definition that both the public and most astronomers are comfortable with.
Another issue is "orbital dominance" - is the object sufficiently massive to remove all comparable objects from similar orbits, or is it accompanied by a number of such objects? This depends on the mass of the object, the mass and orbits of possible companions, and the distance and mass of the star. A reasonable calculation of dominance in our Solar System finds that Jupiter and Saturn are bullies, Uranus, Neptune, Earth, and Venus are all similar, and the low tier has Mars, Mercury (and Luna, if it were by itself). Pluto and Ceres are quite substantially below them, and similar to each other. For extrasolar planets (all Jovian so far) this is not an issue yet. The main reason for bringing it up here is to note that one cannot distinguish Pluto and Ceres in this manner. I suggest that orbital dominance is not needed as part of an simple, enduring definition.
Finally, we come to the arena of cosmogony. There is a "standard" model of planet formation, based on what we see in our own Solar System. In a nutshell, it builds planets up from "planetesimals" - objects in the kilometer size range that can be rocky, icy, or both. Planetesimals are responsible for the extensive cratering of solid surfaces in the Solar System, and some current asteroids and comets are representative planetesimals left over from the era of formation. There is almost no question that the terrestrial planets were built up this way. Since many more icy than rocky planetesimals can be made in the outer protoplanetary disk, one might expect bigger planets there. Once planetesimals (or planetary embryos, which might be Mars-sized) have merged to make an object with 10-15 Earth masses, if the hydrogen (and helium) gas which constitutes 99 percent of the original protoplanetary disk mass is still around, it will begin to be rapidly accreted onto the massive core. The predominantly rock/ice cores of Uranus and Neptune with 10-15 Earth masses are highly suggestive. Saturn was more successful in gathering gas, for reasons that are not fully understood. Jupiter is yet more gaseous (but for Jupiter the evidence of a rock/ice core has proven elusive). A substantial (and vocal) contingent of professional astronomers has the position that a planet is defined by formation from planetesimals (which automatically includes the requirement of orbiting a star).
The problem is that we are not sure this is the only way to make non-fusors in orbit around stars. The incidence of stars orbiting stars is quite high (almost half of stars have stellar companions), so it is clear that massive objects can form in orbit around stars without needing planetesimals. The question is whether non-fusors can do so, and whether some of the current crop of extrasolar planets (or brown dwarf companions) formed in a "direct" way. This would involve gravitational instabilities in the protoplanetary disk, in which the gas (and any accompanying solid material) in a region of the disk collapses relatively quickly under its own gravity to form a massive planet (which could be augmented by further gas or planetesimals afterwards). Some theoretical models have managed this, but it is not clear whether the conditions they were started under can realistically occur in Nature. There is even theoretical work underway suggesting that planetesimals may form by gravitational instability (but in the thin dust-rich midplane of the disk).
A related question is whether non-fusors can form at the center of their own disk, with no star involved at all. This would be the likely mode of formation for the "free-floating planets". Indeed, some observations now appear to show very low mass objects forming in their own disks. Again, some calculations have also managed this, but we are not sure how realistic they are. There is another way to make "free-floating planets", which is more circumstantial. One might make several giant planets around one star with the traditional method, but then they may interact gravitationally with each other. This can result in the ejection of one or more of them, which will be seen later floating between the stars. This probably is not the dominant source of isolated non-fusors; they are numerous enough that this scenario would have to be surprisingly common.
In any case, I assert that using cosmogony to actually define "planet" is a bad idea. We not only don't know enough about planet formation, it is also very difficult to distinguish the mode of formation observationally (witness our problems with Jupiter and its core). We should have a definition that can be applied through observations in a relatively straightforward way, and which doesn't rely on inadequately tested theories. The case of free-floating non-fusors is an example where one cannot usually tell whether the object's history qualifies it as a planet (if defined by cosmogony) from any current observations. It is true that there is some evidence for separate modes of formation for brown dwarfs and extrasolar planets. The distribution of masses for bodies in few AU orbits around solar-type stars is greatest at the lowest detectable masses (one jupiter or less), and tails off rapidly at higher masses, with almost none above the fusion limit (this is the so-called "brown dwarf desert"). On the other hand, many brown dwarfs are found in other circumstances. That suggests that the currently observed non-fusor companions have a different formation channel from fusors under similar circumstances. By itself, however, that hardly makes the case for cosmogony as a defining characteristic.
How can this be resolved? A consensus is slowly developing (I believe) for the following solution. We can first define what we mean by "planetary mass", and base this only on physical characteristics. Then we can include circumstance into the definition of "planet". I propose the following three definitions:
FUSOR - an object that achieves core fusion during its lifetime.
PLANEMO - a round non-fusor.
PLANET - a planemo orbiting a fusor.
These definitions are succinct and observationally testable. I'm using "planemo" as shorthand for "planetary-mass object" (time will tell whether any of these terms or definitions are adopted). Note that "planemo" is more general (and astrophysical) than "planet", because circumstance doesn't count. These definitions resolve all the outstanding controversies, and are unlikely to be upset by future discoveries. Non-fusors found by themselves can be called free-floating planemos (or "isolated planetary mass objects"). It does not matter what their origin was. Any "extrasolar planet" whose true mass is above 13 jupiters is actually a brown dwarf companion. Stern & Levinson have already proposed the "roundness" limit to the IAU, and I have forwarded the above definitions as well. It is not clear the IAU must act; there is no formal definition of "star" either (and as one looks closely at the upper and lower mass limits of stars, a lot of potential discussion looms there too).
A number of moons (including our own) in the Solar System are planemos, but they are not planets because they orbit non-fusors. Pluto is indeed a planet, but so is Ceres. We should make liberal use of adjectives, which can be argued over with less heat. There are 8 "major" planets in the Solar System, and that is that. I would call Pluto, Ceres, and their ilk "minor" planets (as has been done in the past). To borrow a joke from American politics, perhaps planets without orbital dominance could also be called "beltway" planets. It has been pointed out that there is no fundamental problem with calling Pluto both a planet and a KBO; they refer to different classifications. Although the formal number of planets would be increased, schoolchildren need not learn all the names of the minor planets (but could share in the excitement of new discoveries). The total number of planets in our Solar System is probably fewer than 25. One could still refer to the 9 "historical" planets.
We already use compositional and size-based adjectives like "terrestrial", "gas giant", "ice dwarf", and others. "Degenerate" or "super" could be put in use for planets above 2 jupiters. Those who feel the roundness limit includes objects smaller than their preconceptions of planets allow can apply a suitably dismissive term. We should also deal with cosmogony by using adjectives -- such as "agglomerated" planets, "runaway gas-accretion" planets, "direct collapse" planets, and possible future suggestions. It is currently unproductive to define "planet" itself based on unconfirmed (and difficult to test) prejudices about formation modes. The point of adjectives is to add information, and perhaps focus attention on particular aspects of study. Let's keep the concept of "planet" itself simple and easy to understand.
I'm sure there will be further debate on this topic, and I would be delighted
to hear from you regarding it (email@example.com). As with any such topic,
a consensus will only emerge after protracted and vigorous debate, along with
quiet thoughtfulness, further experience, and future discovery. Eventually the
IAU may decide to codify things, or they may choose to leave it to the lay and
professional cultures to continue evolving a definition of "planet". My main
hope here is to have helped you add your (more informed) voice to the process.