Let's review what we have seen about how science "works."
We initially described the goals and methods of science as follows:
The goal of science is to render our universe understandable -- to discover the underlying principles that govern how things (ultimately everything, we hope) behave. It is much more important to understand the concepts and principles than just the facts to a scientist. The reward for gaining this understanding is the ability to predict things in completely new circumstances. In some cases we can gain a deep enough understanding to distinguish what happens inevitably (e.g., if we could find another Universe, these things would be the same) and what has an element of chance.
The essentials of the scientific method:
1) Examine something or observe some process.
2) Develop a hypothesis to explain what is seen.
3) Apply the hypothesis to new data or a new situation and
see if it continues to explain what happens -- better yet,
make a prediction of what should be observed, and then go
make the observation as a test.
But actually, the process from Ptolemy to Newton seems a bit more complex than we originally indicated. Philosophers of science describe it by saying that scientists tend to work the process listed above within a certain intellectual framework, or "paradigm". When this paradigm starts to lead to many failures in the testing they conduct, there is a period of broader questioning that usually leads to a new paradigm and a new direction for that part of science. Thus,
Ptolemy established a paradigm based on pure circular motions of the planets, around the earth
For centuries, astronomers worked within this paradigm, refining it by adding new features so it could reproduce the motions of the planets better.
Because these efforts produced an excessively complex system that still was not very accurate, Copernicus broke with part of the paradigm and had the planets (including Earth, which had previously not even been considered a planet!) orbit the sun.
However, he did not abandon circular orbits, so the accuracy of his model was not improved. Tycho proposed a model with geometry identical to that of Copernicus but with the earth still the center of motion. The predictions were still no better because he retained circular motion. Thus, we ended up with three different theories, none of which was making good predictions for the positions of the planets.
Kepler realized that he could achieve great simplification and at the same time much greater accuracy with planets orbiting the sun on elliptical orbits. He completed the paradigm shift that had started hesitantly with Copernicus. From Kepler on, all astronomers worked under the basic assumptions of his theory.
Meanwhile, Galileo had initiated a paradigm shift in physics. Previously, physicists had worked to interpret the writings of Aristotle, and were satisfied if the behavior of nature was only approximately explained. The term "to save the appearances" was widely used at the time to describe science that conveniently described the observations but had no deeper significance in terms of general application to new situations. Galileo made detailed, quantitative observations and insisted that physics should explain them in a general way that allowed predictions in new situations, and his approach opened up scientific progress in this field.
An implication of Galileo's new paradigm was that physics had to find a way to account for Kepler's Laws. Newton succeeded in doing so, and in terms that applied generally to many more aspects of physics. Thus, he established a new paradigm for physics that guided much of the field until an accumulation of inconsistencies led to a new class of theory and new paradigm, established by Einstein (which we will discuss later).
But is this process guaranteed to make progress? Frequently, when a new paradigm comes forward, it is not obvious that it is an improvement. We saw the issue with Copernicus' concept for the solar system. More interesting is the fate of Aristarchus' proposal that the sun lies at the center of the system. It was rejected in favor of the paradigm of constant speed around perfect circles, with the earth at the center of the system. Thus, an inferior paradigm was retained, and in this case for more than 1500 years. However, eventually it was rejected and progress began again.
--William Gaddis, "JR," 25
We have also tried to distinguish science from pseudo-science, which looks the same but isn't. The easiest test is to ask if the hypotheses being advanced can be tested and if they fail the test, would those proposing them either withdraw them or modify them to be consistent. In general, only scientific hypotheses pass this test.
W. Sharrock and R. Read, "Kuhn, Philosopher of Scientific Revolution," Blackwell Publishers, 2002 (first recommendation)
T. Kuhn, "The Structure of Scientific Revolutions,", University of Chicago Press, 1962
K. R. Popper, "The Myth of Framework," Routledge Press, 1994