![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Obvious only in retrospect
Dec. 5th, 2004 10:32 amOver the past 15 years or so, a large number of extrasolar planetary systems have been discovered, mostly through spectroscopic observations of wobble in their primaries. The big surprise is that there are so many that astronomers could detect this way as early as they did, because they have huge Jupiter-sized planets in tiny orbits around the star (often much smaller than Mercury's orbit).
It's hard to say what fraction of solar systems are like this; there's a huge sampling bias, because these are the ones that are easiest to detect. Astronomers can't detect terrestrial-sized planets yet, except for those oddball planets orbiting pulsars. But it's fair to say that almost nobody expected that there would be so many solar systems with hot Jupiters.
Why not, though? We already knew that most star systems aren't anything like ours. It's well-known that multiple star systems are common, and many of these are "spectroscopic binaries" with extremely-close-in, fast-orbiting stars. It's not much of a stretch to conclude that if these systems are all over the place, and stars seem to become more numerous the smaller they get, then spectroscopic multiples where some partners are not stars at all, but substellar hot Jupiters, ought to be as common as lint.
What we think of as commonplace is conditioned by our mental categories. We popularly categorize celestial objects into stars, planets, comets and asteroids; but these categories are based on observations from the pre-telescopic and early telescopic era with an extremely limited data set. It leads to silly arguments over the precise division between a planet and a substellar "brown dwarf", or over whether Pluto is a planet. I remember that when Chiron was discovered in the 1970s, people argued over whether it was a tenth planet (resulting in continuing woo-woo interest), a comet or an asteroid; it didn't fit well into any of those categories, and it ended up with both a cometary and an asteroidal designation. Now there's a whole known class of these things, called Centaurs and thought of as probable former Kuiper Belt objects. They're not really comets or asteroids; they are what they are.
So when people thought about extrasolar planetary systems, they typically extrapolated from the data they had on planetary systems, of which there was only one known, and not from the relatively large amount they knew about multiple star systems. But the latter, in retrospect, seems more relevant.
It's hard to say what fraction of solar systems are like this; there's a huge sampling bias, because these are the ones that are easiest to detect. Astronomers can't detect terrestrial-sized planets yet, except for those oddball planets orbiting pulsars. But it's fair to say that almost nobody expected that there would be so many solar systems with hot Jupiters.
Why not, though? We already knew that most star systems aren't anything like ours. It's well-known that multiple star systems are common, and many of these are "spectroscopic binaries" with extremely-close-in, fast-orbiting stars. It's not much of a stretch to conclude that if these systems are all over the place, and stars seem to become more numerous the smaller they get, then spectroscopic multiples where some partners are not stars at all, but substellar hot Jupiters, ought to be as common as lint.
What we think of as commonplace is conditioned by our mental categories. We popularly categorize celestial objects into stars, planets, comets and asteroids; but these categories are based on observations from the pre-telescopic and early telescopic era with an extremely limited data set. It leads to silly arguments over the precise division between a planet and a substellar "brown dwarf", or over whether Pluto is a planet. I remember that when Chiron was discovered in the 1970s, people argued over whether it was a tenth planet (resulting in continuing woo-woo interest), a comet or an asteroid; it didn't fit well into any of those categories, and it ended up with both a cometary and an asteroidal designation. Now there's a whole known class of these things, called Centaurs and thought of as probable former Kuiper Belt objects. They're not really comets or asteroids; they are what they are.
So when people thought about extrasolar planetary systems, they typically extrapolated from the data they had on planetary systems, of which there was only one known, and not from the relatively large amount they knew about multiple star systems. But the latter, in retrospect, seems more relevant.