"The whole problem with the world is that fools and fanatics are always so certain of themselves, and wiser people so full of doubts." -Bertrand Russell, 1872-1970
The philosophy of science is a rich discussion -- or in some cases a vigorous battle -- which carries (or rages) on to this very day. The question of "what is science?", or "what constitutes scientific thought?" has tormented thinkers for over a thousand years. The difficulty of these questions stems from the fact that science is not so vacuum sealed and perfect as so many would like to believe. Science exists in a context and in a culture of people -- people who have biases and who make mistakes. While it's nice to look to certain scientific ideals, such as the scientific method and peer review, in many places, those ideals do little justice to the messiness of the scientific world we live in. In light of these facts, approach science with the same attitude science would have you approach the world with: be curious, doubtful, and yet have an open mind.
For a more detailed discussion of this "messiness", a fun starting point is Conduct, Misconduct, and the Structure of Science, by Woodward and Goodstein (12 pages)
In part, I write this to warn the reader to take caution when assuming something about science. But more importantly, this is an excuse for skirting around a satisfying discussion of science in favor of a more operational approach. Moreover, below is not an explanation of what makes something science, but of what scientific inquiry (however that is defined) has accomplished, and through what mechanisms have those accomplishments been delivered.
What does science do?
From as early in our history as we have any credible record, people worldwide have found creative ways of explaining the rains, the seasons, the trees, the birds and the bees. However, most of this has been more to the tune of story-telling, having no root in viability (notice how I did not say "truth"). When science came into the scene, it very much stayed within the tradition of this sort of storytelling, but there is a distinction to be made about the results: scientific results were useful. To be more precise, the way in which scientists tell stories leads to concepts which seem to reflect some aspect of the world just closely enough.
When a scientist develops such a story it becomes a model. A model is able to take the complexities of our world and simplify some aspect of it. In principle, a model can be as close or far away from "reality" as a creation myth is, so long as its implications are useful for making predictions.
For example, the model of "Miasma" or "Bad Air", which posited that bad smells are responsible for the spread of disease, was actually a somewhat useful model because it was close enough to the truth at the time. In light of this model plague doctors wore long masks with perfumed beaks. This often protected the doctors, although not for the reasons they believed. Of course the model would be overturned by germ models two centuries later, but the story seemed to hold when tested, and offered useful ways of avoiding disease (although not as useful as the stories we tell today).
This is why it's important to question statements such as "Science has proven something", or more grievously "science shows us the truth". Science does not so much offer truth as useful results, it seems. The operational function of science is as an engine for producing models -- simplified explanations which our small minds can put to use.
The best stories of the day are called theories, which are models who's profundity-to-complexity ratio is pleasingly high. These models apply to a wide variety of phenomena, and have an uncanny ability to generate new lines of inquiry. Like poetry, a good theory is economical, saying no more than is necessary, but more than was expected. Later in this course you will learn about atomic theory, which simply asserts that macroscopic matter (that is, an object on a large scale) is composed of microscopic particles. This will probably seem very obvious to many already, but the results of this story are so profound and far-reaching that most theories in physics after its introduction rely on it in some way (thermodynamics, quantum mechanics, string theory, etc).
Another signal that a theory is great is that is unifies. For example, philosophers initially assumed there were different laws governing different "realms". It was thought the "perfect" movement of the heavenly bodies could not be reconciled with the jumpy, altogether badly behaving motion of objects on Earth. Newton's theory of gravity is rightly remembered as a great theory because it was able to unify the seemingly disparate phenomena, and with incredible accuracy. (Of course, Newton's theory of gravity has since been overturned by Einstein's theory of general relativity in 1915).
Overview of Models in Physics
The overall trend of physics since Newtonian Mechanics (1687) has been unification and generalization.
Newton's work became a foundation for how inquiry in physics should look, but was actually quite narrow in its scope. Scientists investigating light, heat, electricity, and chemical processes were forced to start from the ground and work up, as Newton's models were largely uninformative in those areas. However, starting around the 1800's, scientists began thinking of models which seemed to unify these topics.
For example, the atomic model of matter (around 1820) and the energy model of heat and temperature (1840s) would be merged with Newtonian Mechanics by Boltzman in what would be called statistical mechanics (1850s), which was able to explain heat and temperature in terms of the Newtonian interactions of trillions of atoms. The study of light, initiated by Newton in 1686, would be unified by Maxwell in 1873 with the study of electricity and magnetism, called electromagnetic field theory, showing that electric and magnetic fields are deeply connected, and that light is a combination of an electric and magnetic field which propagates through space. These theories, in conjunction with other specialized areas of physics such as fluid dynamics, form what we often refer to as the classical model of physics -- in contrast to what follows, which is usually regarded as modern physics.
Many physicists were fairly content with the classical model near the turn of the 20th century. In fact, Lord Kelvin is often attributed to having said in 1900 that
"There is nothing new to be discovered in physics now. All that remains is more and more precise measurement." - Lord Kelvin
This makes a funny story, seeing as how completely wrong that view was, but investigation shows he said no such thing (but why let the truth stop you from chuckling?). Kelvin's (fake) sentiment was one shared by many scientists of the period. Physics started running into a series of hairier problems in the decade following 1900. Namely, the behavior of light seemed to violate classical physics left and right. A revolution was around the corner. The entire foundation of physics would be turned upside down by the two new big theories in physics: general relativity (1905) and quantum mechanics (1926).
These theories were able to explain everything that classical physics could, but filled in the gaps classical physics couldn't. Now physics was split into these two theories, and physicists have been trying to get them to co-mingle for just shy of one-hundred years, and to some extent they have been successful. Early quantum mechanics and special relativity (relativity without gravity) were unified into quantum electrodynamics, which would later be extended to include the weak forces (what hold nucleuses together) and general relativity in relativistic quantum field theory. After a revolution in particle physics, the theory was extended to include the strong force (quark-quark interactions), and boiled down to a theory of fundamental particles known as the standard model, which is the currently accepted model of physics.
So are we done? Not even close. Many sources would have you believe that the standard model is complete, but the relativistic theories are only willing to snuggle up a little, rather than move in with quantum theories. In addition, bring up the graviton with a standard model physicist and she will likely create a diversion allowing her to sneak away, and never get back to you.
The takeaway is to stay curious, and not become the next unfortunate soul who is misquoted as saying we know everything.