There is an interesting article in the January 2007 issue of Scientific American entitled "What Is a Planet?" by Steven Soter.

It doesn't take long to see what conclusion he's going to reach, which suggests that an examination of the axioms he adopts would be more rewarding than to have him posit his axioms and then spend three pages grinding through the same-old obvious consequences of those axioms.

"Most of us grew up with the conventional definition of a planet as a body that orbits a star, shines by reflecting the star's light and is larger than an asteroid."

Actually, I think most of us grew up with an extensional definition: a list of the planets. Or a definition that, imprecise though it may be, elided the "larger than an asteroid" part. For example, on a K-12 site, I found:

<<A planet is a body in the solar system that orbits the sun.>>

Soter thus loads the dice before he takes the first roll.

"In the 1990s, however, a remarkable series of discoveries made it untenable."

So we should not try to define planet at all, right? Or will you just argue past that point, assuming that it should be defined formally?

Soter offers an answer to that, but a totally unsatisfying one:

"This is not just a debate about words. The question is an important one scientifically. The new definition of a planet reflects advances in our understanding of the architecture of our solar system and others."

But wouldn't abandoning the attempt to define the term scientifically reflect the advances even better?

We once thought there was a significant size gap between Ceres and Mercury with no objects inbetween. We've learned otherwise. Ceasing to pretend there's a gap would reflect the advances rather well.

"[accretion] produces a small number of massive bodies--the planets--and a large number of much smaller bodies--the asteroids and comets, which represent the debris left over from planet formation."

That's one of many ways to posit subcategories. Another would be to call the giant planets one sort of thing and to lump Earth with other flotsom. It would certainly disrespect the depth of the issue to assume that planet-versus-asteroid is the classification that reality has handed us when it was actually history that handed it to us. Reality has make it clear that the distinction has vanished.


"By 1851 their number had increased to 15, and it was becoming unwieldy to consider them all planets."

Really? What made it unwieldy? Not enough paper to print the 15 names on? Is the number of states in the USA unwieldy? Is the number of countries in Europe unwieldy? Are the rivers of North America too numerous for us to call them all rivers? Are there too many species of insect?

Say, I thought this was a scientifically-motivated definition, not an aesthetic one.

Page one, and Soter has talked past all of the meaty issues so he can spend three pages grinding forward with the conclusions his axioms force one to agree with -- if you agree with the axioms:

1) There is a clear gap between the planets and asteroids+comets.
2) We cannot afford to have too many planets.

Those axioms don't yet decree anything about Pluto/Eris, but they already abandon the premise of a scientific discussion. There WAS BELIEVED TO BE a clear gap, and discoveries have eroded it away. And there is nothing scientific about deciding how big a category should be. That's blinkering yourself to reality if anything ever was. God help us if we find fifty-six bodies larger than Earth orbiting 0.1 light-years out. Then what? Would that soften up those "scientific" axioms any?

A third axiom comes later:

3)
"the large gap in dynamical power provides a clear way to distinguish the planets from other bodies."

So does mass. When did the anti-Occam flu make the rounds?

Why exactly was it we were HUNTING for a clear way to distinguish the planets from other bodies? Oh, right, axiom #1, which I didn't buy into because Soter didn't argue it, but assumed it.



The terrible thing about this issue is that the zeal for arguing to a conclusion has stifled the real and interesting debate around what are being treated as axioms. Both sides sense that if they don't over-reach, they'll "lose" to the side that over-reaches past them.

Here are the real questions:

IS there a clear divide between the planets and the asteroids/comets? Or is there any such divide (between Earth and Neptune, between Saturn and Uranus, between Mars and Venus, between Juno and Vesta), come where it may?

Are we trying to base definitions only on objective reality, or are we OK in principle with history playing a role?

Do we need to avoid having too many planets? That sort of constraint also rejects objectivity. Which is fine, just be HONEST about it instead of putting on the scientist hat while bandying about aesthetic and pedagogical considerations.

Anybody talking past these questions is talking junk. if you have some different answers to them, fine: ARGUE them, don't assume them.

Let me point out the fact that mass directly determines the dynamical power for an object to be able to effectively clear or dominate its orbit. The critical factor in determining a scientific definition of "Planet" must ultimately be mass...no matter how it is expressed. The problem with the new IAU definition comes from the fact that the farther an object is from the Sun, the larger the object must be to effectively dominate or clear its orbital path. At very large orbital distances absurd situations could easily arise.

Certainly if found, each and every one of those 56 hypothetical objects larger than Earth near 0.1 ly from the Sun should be called planets upon discovery. According to the new definition, they cannot be called planets. Of course, you pointed out that such an exclusion is absurd on its face. Therefore, there are obvious huge potential flaws within this new dynamical definition of planet. But it is only a hypothetical situation. You cannot viably create a working scientific definition solely on a hypothetical situation--that in reality may not even be possible(we simply don't know yet).

Although I may not agree with it, the new definition for planet is workable for the immediate future. I only view it as a transitional definition. It is pretty obvious to me that this definition must eventually be reworked to conform to more objective criteria.

I disagree with you that Soter assumes axiom #1 as worded. In my view, he seems to be arguing that the disappearance of a PREVIOUSLY assumed gap necessitates the more precise definition of planet now. I just don't think the IAU is nearly as precise as he suggested. And Soter uses the IAU definition to define a new type of GAP as measured by the variable of 'µ'. Perhaps this is the GAP you refer to, but it is not clear.

You are, however, correct that Soter does assume axiom #2. Axiom #2 does limit objective reasoning when it comes to a more scientific definition of a planet.

The biggest argument for not simply using mass as a critical limit for the definition is that the line drawn would necessarily be arbitrary. (1) Is this true? (2) Even if this is relatively true, why argue an equally arbitrary standard using orbital populations?

Despite the argument given by Soter, he seems to contradict those arbitrariness objections by inserting an equally arbitrary standard using 'µ', when he says:

"I have suggested setting the cutoff at a µ value of 100. That is, a body in our solar system is a planet if it accounts for more than 99 percent of the mass in its orbital zone. But the exact value of this cutoff is not critical. Any value between about 10 and 1,000 would have the same effect. "

This arbitrary number is in addition to the changing relationship that 'µ' represents over the vast differences between the dynamic relationships relative to clearing ability that is unique to each of the various orbits! Large objects can potentially be excluded, while small objects could potentially be included as planets.

Remember, the farther that an object is from its star, the larger it must be to be able to effectively clear its orbit. In other words, the closer to the star, the smaller an object can be and still be considered a planet. So Soter's, assertion that the roundness criteria of the IAU definition being unnecessary is itself incorrect. Without the roundness criteria, the IAU definition would be potentially unworkable for other star systems--in situations that although not known, could easily exist. Use of the GAP created by 'µ' for a planetary boundary within our solar system works in a relative manner within obserable parameters. The GAP created by 'µ' really exists within the limits of our current observations. However, adjustments clearly need to be made, especially when considering other star systems. Soter indirectly addresses this problem by adjusting the model himself by defining the problem away through the claim that large objects not completely clearing their orbits according to his criteria would be called planetary ‘embryos’. So he himself must be aware of problems.

The orbital clearing ability of an object criteria is only a start at a workable definition. The problem with it is the use of orbit to describe neighborhood. Clearly, obits beyond Eris' (or maybe even as close as Pluto) are not accurate in describing an object's neighborhood. Objects within similar orbits at vast distances from the Sun can be separated so far as to never interact with one another--both because of distances involved and resonances. In fact, the farther the object is from the Sun, the more resonance comes into play as to describing its neighborhood. If resonance were included into the definition as a possible factor for inclusion (instead of exclusion), then the IAU definition could become workable over a much longer period, if not for good. How that would specifically be done...I don't know. But variations on the idea should be considered, so that those 56 hypothetical larger than Earth objects at 0.1 ly from the Sun would be included as planets. If we ever do find 2 or more Earth sized objects sharing a similar orbit, should we simply called them embryos? Can mature and embryonic planets exist in the same system over a long period of time?

Granted the stability of such a system is in question--but there was speculation before the space age about the possibility of a twin Earth planet sharing our orbit. They even created science fiction movies about the idea in the 1950’s. At that time, there was no question whether that hypothetical object would be called a planet. Before the 1990's, there would have been no question about it at all. If there ever could possibly be such a stable situation because of resonance in a different system, would we now call these potential objects embryos? Observation has proven that there are many different stable orbital resonances.

The truth is that despite claims to the contrary, orbital clearance has never been a criteria for calling an object a planet until recently. The problems associated with the asteroids in the 1800’s were primarily growing numbers and the small size of the objects. Somehow over the 1990’s, somebody (it may not have been a single person, but an evolution of thought processes by a number of people) construed numbers as being the same thing as orbital clearance because of convenience while arguing planetary size limitation. Many people ‘bought’ the equivalence of the concept without much thought of true historical context, even if they initially did not ‘buy’ the argument for getting rid of Pluto. At least, that is how I see it.

This book might be of interest to those who want to know how the definition of the word 'planet' has changed over history.

Is Pluto a Planet? A Historical Journey through the Solar System

Bill

I just want to point something out about the whole Pluto furor. For those of us who question why it matters whether or not Pluto is a planet or some other thing. It's a bit of a sideways look at it, but it says something important.

Here in Minneapolis, there is a comedy troupe called the Brave New Workshop. Their latest production is entitled "Pluto, and Other Lies My Teachers Told Me."

Sort of puts the whole mess into a new perspective, doesn't it?

-the other Doug

But that's what I meant by axiom #1: the idea that a definition is needed at all.

We have no terms to distinguish between birds that are smaller than a foot and birds that are bigger than a foot, because there's nothing interesting about that dividing line. If there had been a large gap in the size distribution of birds, then there probably would be some reflection of that in the terminology. Then, suppose such a gap existed only in the Eastern Hemisphere but not the Western. Let us say that pre-1492, every bird in Europe and Asia was either smaller than 12 inches or longer than 24, and the terms BigBird and LittleBird had been coined. When the Americas were discovered, how much sense would it make to try to come up with a more precise definition of BigBird and LittleBird to apply a razor-sharp delineation to the Western Hemisphere's birds? It's when you discover the gap isn't there that you realize that the old terminology didn't make sense, and not because you weren't precise enough, but because the distinction was nonsensical with ANY threshold as the dividing line. It wouldn't matter if you shift from the length of the bird to the mass to counting feathers. The discovery of the new birds tells you that the terms themselves are pointless. Arguing, as you say Soter does, that a more precise definition is needed is assuming something. Something wrong.



That's interesting background. I, however, see the orbital clearance shift as equivalent to taking the bird categorization from length to feather-count in an attempt to find a new gap. Because the old gap went away, but you forgot that the gap was an assumption based on casual observation in the first place.

Tnew definition, besides being based on errant and undefended axioms from the get-go, is a Las Vegas -style bet that nothing exists out past Neptune that will embarrass it empirically the way it is already embarrassed theoretically. There are many scenarios that could do so, many of them involving a putative icy body 2x the mass of Ganymede (surpassing Mercury) or 4x the mass of Ganymede (surpassing Mars). If such a discovery happens, it will be a travesty if anyone still thinks, "OK, we still know that there has to be a gap between planets and nonplanets -- we just have to find out in what dimension of measurement it lies."

Perhaps. Personally I like the idea that scientists should look for order in the universe as we observe it, and I think it's right and proper that they should revise their taxonomies when new data comes in. Studying those patterns is typically how we find deep, underlying principles.

The arguments against that all seem to operate under a single, undisclosed axiom: "Nothing may ever be removed from the list of planets I learned in grade school." I personally find that uncompelling -- no matter how many words are used to dress it up -- but lots of people seem to go for it.

--Greg

I disagree with your statement that the arguments for Pluto retaining its planet status stem from the sentiment that "nothing may ever be removed from the list of planets I learned in grade school." This seems very much like a way of discrediting supporters of Pluto's planethood with the equivalent of an ad hominem attack. The reality is that the new data that has come in does not tell us with any degree of certainty what is and is not a planet. Arguments for demoting Pluto are based on interpretation of that new data, not on the data itself. In his book Is Pluto A Planet? David Weintraub illustrates that defining the term "planet" is more difficult than ever. He raises the issue of "pulsar planets," "orphan planets," etc. and clearly shows that we still know far too little to establish a firm definition of "planet" with clear specifications. Weintraub argues that Pluto is a planet, but so are Quaoar, Orcus, Eris, etc.--a proposition that clearly makes sense and is not driven by sentiment related to what one learned in grade school. I still find it hard to comprehend why people have a problem with KBOs large enough to achieve hydrostatic equilibrium being counted as planets. If the answer is simply that it's inconvenient for the solar system to have 50-100 planets, then we're back to making a determination based on what is convenient for us rather than on the reality out there.

How about this:

Any body that's half the size of Earth's Moon or larger, and that orbits directly around the Sun and not around another body, is a planet. Smaller non-satellite bodies in the inner system are asteroids. Smaller non-satellite bodies outside of the inner system are iceteroids.

Think we could get that past the IAU?

-the other Doug

Ok, iceteroids are made out of ice, asteroids are made out of ...

(ducks and runs)

Steroids are made from stereos?

And Polaroids are found in the Polar regions.

Olive oil is made by squeezing olives;
Peanut oil is made by squeezing peanuts;



Baby oil is made by squeezing ........

And, of course, from the bodies known as hemerhs, we derive...



-the other Doug

I quite agree. In a case like this, "revising taxonomies" must extend to utterly doing away with terms that have been shown not to have scientific meaning. We can see that the scientific use has been done away with by Soter's need to include Axiom #2, bringing in the subjectivity.

We're not looking at a malfunctioning machine wondering how to fix it. We're looking at a distinction that we've discovered not to be a distinction.



Actually, I find the "for" argument here to be: "No term may ever be deleted from the list of scientific terms that I was taught in grade school."

It's not that we lost track of the distinction between asteroid and planets and need to look under every piece of furniture until we find it again. It's that it wasn't there in the first place. Our discoveries have shown that. Maybe it's time to pay attention to the discoveries.

The term retains a popular usage, and render unto Caesar what is Caesar's. Scientifically, it's dead. And has been since the last time anyone paid serious attention to Bode's Law.

My objection has nothing to do with what is and isn't a planet under the definition....

well, actually, it does.

As it stands, the status of Jupiter, Mars and Earth as planets is frankly, shakey.

So - as long as you have no problem with a rule that excludes Pluto....and Earth....then we're all good


Doug

Well, that could be spun into an April Fool joke. IAU proclaims that Earth is a minor planet now because, as it turns out, it failed to clear its neighborhood. Earth gets its provisional MPC number - now it's 100000 Earth. NASA assures everybody that it's not curtailing its Earth observation programs yet, as long as financing allows, but it will gradually assign more resources to the real planets, such as Mars. In the meantime, protests against the IAC decision are planned worldwide.

I want to argue strenuously for a definition that will distinguish objects of Neptune-size or larger from the rest of the rubble. Maybe "large enough not to have a clearly defined solid surface"? I realize there's a need to distinguish small bodies like Neptune and Uranus from real planets like Jupiter and Saturn, but I don't believe that the difference is so great that we can't call them planets anymore, even if they weren't discovered until after the American Revolution. If necessary, I'll settle for a compromise and call Neptune and Uranus "dwarf planets".

Funny stuff. And yet, as you look down the list of all the sun-orbiting bodies, ranked by size, the biggest gaps (ratio of an object's diameter to that of the next-biggest object) are Earth/Neptune (by FAR the biggest gap: 0.258) and Uranus/Saturn.

There is a gap of about 2:1 between Mercury and Eris, although there's a good chance that future discoveries will wipe that gap out entirely. And that is the gap that the IAU's planet definition has chosen to make hay of, although they chose different variables that correlate with size to try to make the gap seem bigger. Because they are clinging to the assumptions that "planet" is a meaningful category, that it shouldn't have too many members (how scientific!) and that Mercury has to remain in the set.

If it takes more discoveries to drive this illogic out of people, then so be it -- the trend suggests that we may keep discovering yet-bigger objects out there for some time. But some more thought should do the same thing without more discoveries.

Okay, I would like to step back from this debate and ask: What is a scientifically defensible classification system that includes all objects that could be included under the term 'planet'? In fact, the system I came up with includes all stars, and all sub-planetary bodies as well.

Basically, this is a two-axis system. One axis classifies objects by mass, with 'bins' spanning a factor of two in mass, while the second axis classifies objects by their position in an orbital hierarchy.

MASS

The largest stars appear to be somewhere between 100 and 150 times the mass of the Sun. Occasionally, a larger-mass candidate is announced, but they inevitably turn out to be binary systems, or otherwise below the limit. The Sun is a bit over 330 thousand Earths in mass, so the most massive known stars would have masses of around 50 million Earths.

Class : Mass range (in Earths) : Examples

Class 1 : >32,768,000
Class 2 : 16,384,000 - 32,768,000
Class 3 : 8,192,000 - 16,384,000
Class 4 : 4,096,000 - 8,192,000
Class 5 : 2,048,000 - 4,096,000
Class 6 : 1,024,000 - 2,048,000
Class 7 : 512,000 - 1,024,000
Class 8 : 256,000 - 512,000 : Sun (332,830)
Class 9 : 128,000 - 256,000
Class 10 : 64,000 - 128,000
Class 11 : 32,000 - 64,000
Class 12 : 16,000 - 32,000 : lower mass limit for hydrogen burning (~24,000)
Class 13 : 8,000 - 16,000
Class 14 : 4,000 - 8,000 : HD41004B b (>5,850 - largest known 'planet' by radial velocity - might be a brown dwarf)
Class 15 : 2,000 - 4,000
Class 16 : 1,000 - 2,000
Class 17 : 500 - 1,000
Class 18 : 250 - 500 : Jupiter (317.9)
Class 19 : 125 - 250
Class 20 : 62.5 - 125 : Saturn (95.18)
Class 21 : 31.25 - 62.5
Class 22 : 15.625 - 31.25 : Neptune (17.14)
Class 23 : 7.812 - 15.625 : Uranus (14.54)
Class 24 : 3.906 - 7.812 : Gliese 876 d (>7.3117 - smallest known planet by radial velocity)
Class 25 : 1.953 - 3.906
Class 26 : 0.976 - 1.953 : Earth (1)
Class 27 : 0.488 - 0.976 : Venus (0.815)
Class 28 : 0.244 - 0.488
Class 29 : 0.122 - 0.244
Class 30 : 0.061 - 0.122 : Mars (0.1075)
Class 31 : 0.0305 - 0.061 : Mercury (0.0553)
Class 32 : 0.0152 - 0.0305 : Ganymede (0.025) Titan (0.0225) Callisto (0.018) PSR 1257+12 b (0.0222 - smallest known pulsar planet)
Class 33 : 0.0076 - 0.0152 : Io (0.015) Luna (0.012) Europa (0.008)
Class 34 : 0.0038 - 0.0076
Class 35 : 0.0019 - 0.0038 : Triton (0.00359) Eris (0.00268) Pluto (0.00219)
Class 36 : 0.000905 - 0.0019
Class 37 : 0.000452 - 0.000905
Class 38 : 0.000226 - 0.000452
Class 39 : 0.000113 - 0.000226
Class 40 : 0.000056 - 0.000113
Class 41 : 0.000028 - 0.000056
Class 42 : 0.000014 - 0.000028
Class 43 : 0.000007 - 0.000014
.
.
.

In theory, this could be extended to individual 'dust grains'

ORBITAL HIERARCHY

a : free intergalactic wanderer
b : bound to a galaxy, but no smaller object (not counting open clusters, which are dynamically unstable over billions of years)
c : bound to a globular cluster, but no smaller object
d : bound to a single object which is itself class a, b or c
e : bound to a class d object
f : bound to a class e object - i.e. a moon of a planet orbiting the less massive star of a binary star system

This axis could be sub-divided further, with designations for Trojan-type orbits, 'double planets', etc.

EXAMPLE DESIGNATIONS

Sun : Class 8b
Jupiter : Class 18d
Earth : Class 26d
Ganymede : Class 32e
Pluto : Class 35d

If it were up to me, I might junk the existing system of nomenclature and go with something like this. Does this sound like a reasonable system? (not that it has any chance of being adopted)

Bill

Argh...

I'm about ready to junk the term "planet" in favor of "stuff orbiting around a star" (STOAAS)! After all, the entire Solar System is just a minor fraction of the mass of the whole when the Sun is included, and most of it exists in discrete packages along a continuum with local variables based on mass and the amount of solar radiation received.

Planet is a subjective term...let's just call 'em as we feel 'em. Unscientific as hell, of course, but realistic...

I don't think you'll ever get rid of the term "planet", if for no other reason than a century and more of science fiction has entrenched the term solidly within the human psyche.

When humans imagine going out into space and exploring the Cosmos, they think about living on chunks of rock with roughly the same size, shape and general characteristics as the Earth. We imagine exploring, but not taking up residence on, other chunks that are less hospitable.

It doesn't matter if you're talking about David Brin or Arthur C. Clarke or Gene Roddenberry -- when you tell stories about traveling to distant worlds, you have to have a generic name for the bodies you visit, and that name, for better or for worse, is "planet".

So, "planet" will continue to describe those chunks of rock that humans imagine visiting. Otherwise, billions of words of science fiction will instantly become "quaintly out-of-date," and that just ain't gonna happen...

-the other Doug