The Day Pluto Got Demoted

A strange thing happened in 2006. At the beginning of the year we had 9 planets in our solar system. But at the end of the year we had only 8 planets in our solar system. What cataclysmic event could have caused us to lose a planet? The answer is that we made a small adjustment to the definition of the term “planet”, and as a result, Pluto no longer qualified. The solar system did not change — it was only our model of the solar system that changed. In other words, what changed was the way we think about the solar system, and the way that we structure that knowledge for teaching, learning, and discussion.

For many school kids around the world — and for many former kids as well — this was a catastrophe. After all, we had all been taught with absolute certainty that the solar system contains nine planets. To add insult to injury, many kids had considered Pluto to be their favorite planet. This incident provides a peek at a hidden truth: that much of our knowledge is based on useful models, but models are not the same things as facts — even though we learn them as facts. So when the models are updated, the so-called “facts” can change.

So how did this happen? Why was the definition of “planet” changed?

It has been known for centuries that our solar system contains several kinds of objects, in addition to planets. There’s the sun; there are moons; there are comets; there are asteroids. There are many other objects that don’t fit into any of these categories, for which we need additional classes. Furthermore, for some of these classes — such as comets — we know that we have not yet identified every object. In other words, we know that more comets will be discovered in the future. But in the case of planets, we have for thousands of years relied on a countable list of objects, even while the number of objects in the class has changed several times. The demotion of Pluto, and the reduction of the list from 9 planets to 8, was simply the most recent change to the class.

The original list of planets contained only 5 objects — Mercury, Venus, Mars, Jupiter, and Saturn. These are the 5 objects in the night sky that look a lot like stars to the naked eye, but which wander from one constellation to another. So that was the original definition of a planet — a wandering star. As our model of the solar system improved, and we realized that the planets orbit around the sun instead of the earth, we came to realize that our own Earth was a planet too. So now the list had grown to 6 objects. As telescopes became stronger, and the mathematics of planetary motion became more refined, we came to realize that there are additional planets in our solar system, not visible in the night sky to the naked eye. So we eventually added Uranus to the list, then Neptune, and finally Pluto. But with each addition, we made the assumption that the newly discovered object can be reasonably categorized as a planet, rather than some other type of body.

However, Pluto was always something of an odd planet. First, it is smaller than any of the other planets. Second, it has an eccentric orbit that is less circular than the orbits of the other planets, so that sometimes Pluto is closer to the sun than Neptune. (When this happens, we might ask whether Neptune or Pluto is the ninth planet.) Finally, the orbit of Pluto is tilted considerably compared to the orbits of the other planets.

None of these differences were very important when we thought that Pluto was unique. But in recent years we have come to realize that there are many Pluto-like objects out beyond Pluto. Most of these objects are smaller than Pluto, but at least one of these objects (named Eris) is more massive than Pluto. So this forced the question — should we consider some of these additional objects to be planets, and if so, then which ones?

To answer this question, we need a precise definition for the word “planet”. From that definition, we can derive a set of criteria — rules of thumb — that allow us to distinguish between planets and other objects in the solar system. In our attempt to formulate such a definition, it may be helpful to ask ourselves questions such as these: Why aren’t moons considered to be planets? Why aren’t comets considered to be planets? Why aren’t asteroids considered to be planets? Are there any other objects in the solar system — besides moons, asteroids, and comets — which need to be distinguished from planets?

Let’s start with moons. We exclude moons from the list of planets because moons do not revolve around the sun. Instead, moons orbit around planets. So our definition of a planet should mention that planets revolve around the sun.

Comets revolve around the sun, just as planets do. But their orbits are highly elliptical, rather than having the nearly circular orbits that planets have. Furthermore, the orbits often deviate significantly from the “ecliptic” — the plane that the planetary orbits occupy. In other words, the orbits of the planets constitute a set of concentric rings, all in the same plane (to within a few degrees). But the orbits of the comets are often not within that same plane. Therefore we might want our definition of “planet” to mention having a nearly circular orbit, and the definition could also mention that the plane of the orbit is nearly the same as that of Earth.

Asteroids revolve around the sun, just as planets do. Furthermore, most asteroids have nearly circular orbits, and the orbits are usually in the same plane as the earth (within a few degrees). Therefore it seems that asteroids would be good candidates to be classified as planets. However, we exclude asteroids from the list of planets primarily because they are too small to be considered planets. So our definition of a planet should mention a minimum size to qualify.

If we consider only the inner parts of the solar system — let’s say everything inside the orbit of Neptune — then our four traditional categories (planets, moons, comets, and asteroids) cover most of the objects of interest, other than the sun itself. But if we define the solar system as including all objects that orbit the sun — along with any moons of those objects — then the solar system extends a huge distance beyond the orbit of Neptune. There are countless objects in the outer solar system, and the vast majority of these objects have never been catalogued. But what are these objects? Are they planets, moons, comets, asteroids, or something else altogether? As we gradually discover these objects, what criteria should we apply in order to assign them to the proper categories?

If we were to travel beyond the orbit of Neptune, then the first region we would pass through — still within our solar system — is called the Kuiper Belt. Astronomers have already discovered more than 1000 individual objects within the belt, but far more objects remain to be discovered. One estimate is that the region contains more than 100,000 objects with diameters greater than 100 km (60 miles). The belt occupies a flat ring of space that is more or less in the same plane as the planetary orbits.

Beyond the Kuiper Belt — but still within our solar system — lies the Oort Cloud. The Oort Cloud is thought to be the source of all the long-period comets that enter the inner solar system, such as Hale-Bopp. Unlike the Kuiper Belt, which lies in the same plane as earth’s orbit, the Oort Cloud is a vast sphere that lies in all directions from the sun, extending as far into space as the sun’s gravitation field can have an effect. It is believed that the cloud may contain trillions of objects greater than 1 km in diameter. However, because of the great distance from the earth, no object residing in the Oort Cloud has ever been directly observed. The main evidence for the Oort Cloud is that the orbital paths of the long-period comets indicate that they all originated in this region.

Between the Kuiper Belt and the Oort Cloud lies a region called the Scattered Disc, although some astronomers consider it to be part of the Kuiper Belt. The periodic comets, such as Halley’s, appear to originate in this region.

As a general rule, the objects in the Kuiper Belt and the Oort Cloud are icy, not rocky or metallic like Earth and the other inner planets (Mercury, Venus, and Mars). Because of their volatile composition, objects from this part of the solar system become comets should they journey into the inner solar system. In other words, when such an object approaches the sun, some of the volatiles vaporize and stream away.

Given the above, how should we classify the various trans-Neptunian objects (TNOs) as they are discovered? Can we use the familiar categories of planets, comets, and asteroids for some of these objects? Do we need additional categories to describe some of these objects?

First let’s consider the term “comet” — which we can define as a small icy body that develops a fuzzy haze (and often a tail) as it approaches the sun. Such an object cannot have spent much time close to the sun, or else the volatiles would all have evaporated. Therefore such an object must have an extremely elliptical orbit around the sun, so that most of the orbit lies in the very cold regions of the solar system — but a small part of the orbit takes the object much closer to the sun. Given this definition, most of the TNOs cannot be classified as comets — their orbits never take them close enough to the sun. However, most of the TNOs are potential comets. If their orbits should ever be perturbed due to gravitational interactions with large objects (such as the gas giant planets), then they could become comets.

Next let’s consider the term “asteroid”. It used to be that any object orbiting the sun that is not a planet or a comet could be called an asteroid. But the definition of the term has narrowed over time, because such objects in the inner solar system are quite different from such objects in the outer solar system. The main difference is that the objects in the inner solar system are rocky or metallic, made of materials similar to the Earth and the other inner planets. Objects in the outer solar system are icy, consisting primarily of materials that turn into liquids or gases when exposed to temperatures like those found on Earth. The orbit of Jupiter is the approximate boundary between the rocky/metallic bodies and the icy bodies. Therefore the term asteroid is now limited to those small objects whose orbits are no farther from the sun than Jupiter. As a result, the trans-Neptunian objects are not asteroids.

So that brings us back to the term “planet”. Can any of the objects beyond the orbit of Neptune (including objects not yet discovered) be classified as planets? Given the earlier discussion, we might define the term “planet” as any object that orbits the sun, provided that the orbit is nearly circular, that the plane of the orbit lies within a few degrees of the ecliptic, and that the diameter of the object meets some minimum criterion. Keep in mind that this is not the only possible definition for the term planet, and that we will consider another definition before this discussion is concluded.

Let’s consider the 8 bodies that are unequivocally considered to be planets, along with Pluto. All 9 objects orbit the sun, so let’s see how they compare with regards to the other three criteria. As you can see from the table below, Mercury is the smallest of the 8 principal planets. It also has the most eccentric (non-circular) orbit, and the plane of the orbit deviates the greatest from the plane of earth’s orbit.

However, Pluto scores worse than Mercury on all 3 measures. It is less than half the diameter of tiny Mercury, its orbit is less circular than that of Mercury, and its orbit is far more tilted than that of Mercury. Therefore, depending upon what we choose as our thresholds, we might say that Pluto fails 3 of our 4 criteria to qualify as a planet. The only criterion that Pluto clearly meets is that it orbits the sun.

If we consider Mercury as setting the minimum standards that all planets must meet, then do any of the objects in the Kuiper Belt (and beyond) meet these standards? As it turns out, only one of the 1000+ known TNOs (Eris) is known to be larger than Pluto, but even Eris is smaller than Mercury. Therefore every known TNO fails our size test. We don’t even have to consider the other two criteria (orbit eccentricity and orbit inclination). If Mercury sets the standards, then our solar system contains only 8 known planets, and it does not seem likely that there are any undiscovered objects large enough to be considered planets.

If we can reject all the known TNO planetary candidates based on size alone, then do we really need to consider any other criteria? Can’t we just define a planet as any object that orbits the sun, provided that it meets a certain minimum size threshold? In fact, in 2006 the International Astronomical Union (IAU) did exactly that. In response to the 2005 discovery of Eris, the IAU revised the definition of the term “planet”. However, rather than choosing an arbitrary size threshold based on units of length (for example, a minimum diameter of 4000 km), the IAU took a different route. They chose a size-related criterion that would have a distinct physical effect. Thus the IAU’s definition of a planet is the following:

Planet — 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 neighborhood around its orbit.

Notice that this definition includes 3 criteria. The second criterion is clearly related to size — the object needs to be big enough that its own gravitational forces have shaped it into a sphere. The third criterion is also related to size. If the object is massive enough, then it will not share its orbit with other objects that are also orbiting the sun; it will have cleared its orbit of all these objects.

By this definition, both Pluto and Eris were excluded from the list of planets. Both objects orbit the sun, and they are large enough to have become spherical — but they are not large enough to have cleared their orbits of other objects. However, the IAU also created a new category, called a “dwarf planet”. A dwarf planet meets the first two criteria, but is not large enough to meet the third criterion. Pluto and Eris therefore fall into the category of dwarf planets. Several other objects in the solar system also quality as dwarf planets, including Ceres, the largest object in the asteroid belt. Unlike most of the other objects in the asteroid belt, Ceres is large enough to have rounded itself into a sphere. Two additional trans-Neptunian objects have also been designated as dwarf planets — Makemake and Haumea. There may be hundreds of additional dwarf planets in the outer solar system — either undiscovered, or not yet studied enough to know that they meet the criteria for a dwarf planet.

So back to the original question: How many planets do we have in our solar system? The standard answer (since August 2006), which we now want our kids to learn in school, is that there are 8 planets in the solar system — Mercury, Venus, Earth, Mars, Jupiter, Saturn, Uranus, and Neptune. However, when these kids grow up and have kids of their own, will this still be the correct answer? It is possible that new discoveries will have changed the count by then. But it is also possible that the definition of the term “planet” will have been tweaked again, resulting in a larger or smaller set of qualifying objects. The moral of this story is that we should indeed teach our kids the standard models of the time — but we should also make it clear that models, while extraordinarily useful, are not the same thing as facts, and that other models are possible!