Knowing Right From Wrong

On October 21, 2004, I gave a talk at the Warren Astronomical Society. It was based on a book written in 1909 called Curiosities of the Sky, written by Garrett Putnam Serviss. Since it is out of copyright, it is available for free at Project Gutenberg. Download it and read it. That’s what I did.

The book consists of a series of descriptions of the current problems in 1909 astronomy. For each issue, Serviss presents a brief overview of the available evidence, and then a series of current ideas that had been proposed to explain it. Finally, he picks his favorite idea. It struck me that he was wrong a large fraction of the time.

Would I have been wrong that often? I doubt it. Maybe it’s my century-later perspective. But to me, evidence is evidence. Maybe I have higher confidence in spectroscopy than was common a hundred years ago. In any case, many of these concepts were cutting edge at the time, so it shouldn’t be a big surprise if a mistake was made here and there.

My talk wasn’t a book review. Nor was the talk about 1909 cutting edge science. My angle was to attempt to engage the club to use the past to try to see the future. My big idea was that with hindsight, we can see which (if any) of the explanations of 100 years ago was most correct. So, could this be a guide, of sorts, to learn how to tell if a modern idea is likely correct or not? Is there a way to learn thought processes that can teach us how to interpret modern discoveries? With that in mind, let’s go over just a couple examples, after the break.

The Andromeda Nebula

I should point out that the nomenclature of the time had The Milky Way, and The Universe as basically synonymous. The Milky Way was thought to be a flat disk of stars. But it was not known if there was anything else. In particular, the Andromeda Galaxy was called the Great Andromeda Nebula. Measuring distances to even nearby stars is extraordinarily difficult. It was first achieved for a handful of stars around 1800. These were stars that were at most a few tens of light years away. We now estimate the distance to the Andromeda Galaxy as 2.5 million light years away. As galaxies go, it’s nearby. And that could be inferred by it’s being the biggest and brightest of these “spiral nebulae”.

In 1885, astronomers were surprised to see a sixth-magnitude star glimmering in the midst of the hazy cloud of the great Andromeda Nebula. It soon absolutely disappeared. Its spectrum was remarkable for being continuous, like that of the nebula itself. A continuous spectrum is supposed to represent a body, or a mass, which is either solid or liquid, or composed of gas under great pressure. That is, both the “nebula” and the “new star” or nova, had a stellar spectrum, not a gaseous spectrum like the Orion nebula.

In other words, they thought that M31 was a nebula, despite spectrographic evidence to the contrary. Then, a faint supernova appears there, at least 9 magnitudes fainter than usual, indicating a factor of 4,000 reduced brightness, suggesting that the object is at least 60 times further away than usual. Perhaps, since they have really no idea how far away the bright ones are, this has no significance for them. If the bright supernova are at least 10,000 light years away (in the band of the Milky Way), then M31 ought to be at least 600,000 light years away, quite outside of the Milky Way. But this deduction didn’t seem to happen. Note that they thought that nova are stellar explosions. Further, they were known to brighten by 7 magnitudes or more. This should indicate that these nova are pretty far away. Perhaps this is what Hubble was thinking some 15 years later, but wanted better proof.

An ancient Greek did better. One guy drilled a series of various sized tiny holes in a plate. He viewed the Sun through them. Then, at night, he attempted to remember which hole was closest to the brightness of the brighter stars. With the hypothesis that the Sun was a star, only closer, he wanted to figure out how far away the stars must be. And the estimate wasn’t terrible, despite the inaccurate assumptions (like stars are the same brightness, and so on). For that matter, what was the state of arithmetic 2000 years ago. No zero. I’d hate to do the math with Roman Numerals. Given evidence that Andromeda is star-like, then the apparent brightness gives you a distance. This distance is well outside the Milky Way. I want to tell them, “get over it”. But this is assuming that they could resolve any stars at all. We may have needed to wait for the 100 inch Hooker telescope for that.

Anyway, Serviss thought that Andromeda, whatever it turns out to be, must be smaller, closer, and inside the Milky Way.

There’s another problem. It had been only a hundred years since the distance to the nearest stars had been measured. 20 trillion miles or more. People couldn’t wrap their heads around such huge distances. They sort of gave up, calling such huge numbers Astronomical Distance. They didn’t seem to be prepared for numbers a million or more times that far.


In 1901, a pre-maximum sighting of a nova was reported. Pictures days before showed that no star down to 12th magnitude existed. Then, the nova appears, and for two days brightens to maximum. Then, in jerks, it fades over time. Finally, pictures were taken and the nova was surrounded by a nebula – which they did not see before. The spectrum transforms from stellar to nebular over 6 months. The speculation is that two stars collided.

During the collision: the photospheric envelope of the star is destroyed, the internal incandescent mass gushes out.

The theories of what causes nova include collision of two stars, collision of a star and a nebula, chemical change in a star (including oxygen/hydrogen), and atomic explosion.

One theory is that most or all stars nova, but the ones we see are short lived stars. Most stars take “millions of years” to get to this point.

OK, so a stellar collision is fairly reasonable. I was amazed that there was any idea of an atomic explosion. I don’t think anything was known about fusion. At the time, it was a total mystery how geologists could claim that evolution must have taken hundreds of millions or even billions of years, and the Sun must have been shining all that time. So the idea that only large stars with short lives explode was not known. These days, the idea is that stars generally don’t collide, except in crowded globular clusters, or maybe the crowded center of the galaxy. Stars are just too far apart to have much chance at collision. And this might have been known at the time.

A Modern Example

The Hubble Space Telescope was launched, in part, to determine the age of the Universe. The idea was to get a better calibration for the Cephied Variable Star brightness to pulsation rate phenomenon. It would do this by measuring the distances to nearby examples and get better values for their brightness. Then, distances to nearby galaxies would be measured. Type Ia supernova in nearby galaxies would be calibrated, and finally distances to more distant galaxies would be obtained, along with their cosmological red shift values. All very good. It was called the Hubble Key Project.

But the team released an early report that the Universe must be a bit over 11 billion years old. I immediately thought that this value was not old enough. There are stars thought to be older than that. There are clusters thought to be older. And a correction was issued a few years later. And now, we have multiple lines of evidence, including a totally independant measurement that comes from the study of the cosmic microwave background. But i’ve no idea why i came to my conclusion so quickly. I turned out to be right. And, in all modesty, i usually am right. But if this is something that i learned, is it something that could be taught?


longer version of this article is up on my before-blogs-were-invented web site. In my opinion, the original book, Curiosities of the Sky, is more entertaining. Really, go read it.

Science is an endeavor that attempts to discover truth. And, it’s been extraordinarily successful. It is tempting to many to use it as a religion. After all, religions essentially all claim to either have the truth, or be in the process of seeking it. Further, just as most religions have a creation myth story, so does science. It’s not our first creation myth based on evidence. So the question is, why do we feel that it’s so much more correct than other attempts? The lay person doesn’t have the time to personally investigate the current cutting edge claims coming out of science. So, it’s easy to fall into pseudoscience, like the 2012 end-of-the-world predictions. How does one weed out the nonsense?

…the educated person is not the person who can answer the questions, but the person who can question the answers. -Theodore Schick Jr.

And yet, the show has an “Ask the Astronomer” segment, not “Answer the Astronomer”.

Previous Post
Leave a comment

Leave a Reply