Those who divide the world into two kinds of people, and those who don't. If you wish to understand the scientific worldview, you have to divide beliefs into scientific and non-scientific.
What exactly do scientists believe in?
First, let's get something very clear. There are at least two broad domains of belief, which we can call is and ought. Tremendous miscommunication, waste of oxygen, and rancor results from people getting them mixed up. Sometimes Ozzie is talking about is while Harriet is talking about ought. If they had only noticed it in time, they wouldn't be divorced. Sometimes a person notices a fact about nature -- be it human biology or the behavior of waterfowl -- and concludes that it proves how people ought to behave, or how society ought to be organized. Sometimes people reason from what they believe ought to be, to a conclusion about what is. All of these are common, and grave errors.
Got that? Because when I say that scientists don't believe in anything, that does not imply that they have no ethics, find life pointless or meaningless, or can't be passionate advocates for causes, including truth, justice and the Albanian way. I'm talking about is, not ought. (Also, the statement isn't actually true, although it perhaps it ought to be. As we shall see momentarily, scientists cling fiercely to ideas. But then again, maybe that's a good thing.)
So, with the exposition out of the way, this plot concerns how we distinguish between science and pseudoscience or unscientific beliefs. This is called, repulsively, the "demarcation problem." I have mentioned previously the ideas of intersubjectivity -- the notion that the truth is "out there," as it were, subject to observation, and that we accept observations when we share them. Cold fusion went down in flames because nobody but Pons and Fleischmann -- okay, almost nobody -- could see the same thing, not because their explanation for their observations was necessarily implausible.
But that is just the beginning of the story. New observations extend our knowledge -- yup, there's an icy body we hadn't noticed before, orbiting beyond Pluto -- but by themselves, they don't extend understanding. That takes a theory. A theory is a general model of the relationships among constituents of nature, and to be satisfying, it usually has to include one or more causal statements. (Yeah, heavenly bodies go around each other, but why?)
Newton thought that he had proved his theories from facts but of course, he was wrong, as Einstein later showed. His theory was only approximate (whatever that means), not true. And as people have thought about this since Newton's day they have realized that it is impossible to prove any theory, at least one that is sufficiently interesting. (Again, whatever that means.) There could always be another explanation. Lots of very smart people are working very hard right now to try to find a deeper explanation for gravity than Einstein's.
You may have heard about Karl Popper's idea that a theory is "scientific" if its proponents specify, in advance, observations which would prove it false. (Popper was an Austrian-born philosopher who became a British subject. He died in 1994.) Just because a theory is falsifiable doesn't mean it is true, and obviously, specifying falsifiability criteria means we think it might not be. But, once we prove our theory wrong, we can go on to find a better one -- one that accounts for all the obserations the old theory explained, plus the ones it cannot. And when that theory is falsified, we go out and get an even better one. And so on.
It is now considered the mark of a rube to think this "falsificationist" doctrine is correct. Supposedly it has been falsified. Actually, I think the critics are partly just quibbling and that their objections are not nearly as fundamental or profound as they seem to think. So I guess I'm a rube. But anyway, the arguments, which can get awfully technical, basically come down to the idea that if an observation seems to falsify a theory, it is often possible to explain it away. The measurement might have been inaccurate. The sub-theory you used to argue that the observation would falsify the main theory is actually the one that's wrong. (E.g., it's not my date for Etruscan civilization that's wrong, it's the method of Carbon-14 dating.) My theory is right, it just needs a little something extra to account for the observation. (Ptolemaic astronmers kept adding "epicycles" to explain retrograde planetary motions.)
Scientists do indeed have a very hard time giving up their precious theories. As Thomas Kuhn famously obseved in The Structure of Scientific Revolutions, observations that might be considered to falsify a widely accepted theory can pile up for a long time, and they just serve as puzzles to be solved by explaining them within the confines of the theory. Then quite dramatically, a new scientific consensus emerges: the Copernican Revolution; the immutability of the elements and the periodic table; etc. Kuhn makes this seem like a purely sociological event, not really explained by the world "out there," but by the social hierarchies and conventions of the scientific enterprise.
I say, fiddlesticks. It may have taken longer than it should, but no sane person can deny that the Ptolemaic universe has been falsified. So have the phlogiston theory of combustion, the miasma theory of disease, and, oh yeah, the theory that the world was created by divine fiat 10,000 (or less) years ago. These theories are false, and they have been proved false. There is no conceivable combination of arguments about faulty measurements, faulty experimental assumptions, or missing epicycles that can rescue them. We have sent robots to Mars. We can see microbes under the microscope. We can count the layers in the antarctic ice. Case closed. The argument that single observations don't falsify theories is also largely a quibble. If you have to add an epicycle, you've been forced to change your theory. It's just a matter of degree.
However, just because theories can indeed be falsified doesn't mean that they can be proved. (Some philosophers claim that this is illogical because by upholding falsification, I am claiming that the theory that the universe does not revolve around the earth is provable. It's true, I consider it proved, but I don't consider it a "theory" in the technical sense we are using here, since it doesn't explain anything, it just states something we don't know.)
But if we can't prove anything, and all theories are likely false, what makes a theory scientific? I am satisfied by the formulation of the mathemtician and philosopher Imre Lakatos, whose very accessible discussion I excerpt thus:
[A]ll the research programmes I admire have one characteristic in common. They all predict novel facts, facts which had been either undreamt of, or have indeed been contradicted by previous or rival programmes. . . .Halley, working in Newton's programme, calculated on the basis of observing a brief stretch of a comet's path that it would return in seventy-two year's time; he calculated to the minute when it would be seen again at a well-defined point of the sky. This was incredible. But seventy-two years later, [when both Newton and Halley were long dead,] Halley's comet returned exactly as Halley predicted. . . .Thus, in a progressive research programme, theory leads to the discovery of hitherto unknown novel facts. . . .
The hallmark of empirical progress is not trivial verifications: Popper is right that there are millions of them. It is no success for Newtonian theory that stones, when dropped, fall towards the earth, no matter how often this is repeated. But, so-called 'refutations' are not the hallmark of empirical failure, as Popper has preached, since all programmes grow in a permanent ocean of anomalies. What really counts are dramatic, unexpected, stunning predictions: a few of them are enough to tilt the balance; where theory lags behind the facts, we are dealing with miserable degenerating research programmes.
So, what of the theory that God made the world and all the species in it? This predicts nothing. It's proponents claim that no-one can know the mind of God, so god might do anything at all, without notice, rhyme or reason. Its proponents spend all their time trying to explain away observations, in ever more desperate flailings to somehow cram observations of reality into their procrustean program of dead and degenerating ideas. Their sought after destination is human helplessness and impotence, and ignorance is inscribed on their banner.