While I really like Virmaior's answer, I would like to add a few points, mainly regarding the ideas of verifiability and falsifiability as they relate to scientific theories.
Prior to the turn of the last century, it was nearly universally believed that scientific theories could be proven in the same way as theorems in mathematics could. The surprising collapse of Newtonian mechanics, Newtonian gravity, and classical electrodynamics caused philosophers to realize that no amount of verification can lead to a guarantee that the next experiment won't produce a null result.
Karl Popper recognized that theories were indeed not provable, but maintained that they (or at least the good ones) were falsifiable. This idea is still deeply entrenched in popular opinion and is still commonly offered as a solution to the problem of demarcation. However, the claim that scientific theories are falsifiable does not hold up under scrutiny.
As Thomas Kuhn points out, in his book 'The Structure of Scientific Revolutions,' the scientific community will go to great lengths to prevent an accepted theory from being falsified. Take the famous example of the observation of perturbations in planet Uranus' orbit, which were not predicted by Newtonian mechanics. These discrepancies between observation and theory were known for nearly 70 years. As Kuhn points out, the scientific community does not reject a theory the first time there is a null result. In the case of the orbit of Uranus, the scientific community, rather than rejecting Newtonian mechanics and Newton's law of gravity, instead hypothesized the existence of a yet undiscovered planet: Neptune. They even calculated the exact location where the new planet would have to exist in order to explain the inconsistencies, which lead to the discovery of Neptune—a great achievement for Newtonian physics. However, never during that 70 year period (between the discovery of the discrepancy and the discovery of Neptune) did the scientific community ever consider rejecting Newtonian mechanics.
A few years after that, the existence of the planet Vulcan was hypothesized to explain persistent irregularities in the orbit of Mercury (irregularities that had first been observed a century earlier). However, unlike Neptune, the planet Vulcan was never discovered, and Mercury's orbit was only explained after Einstein published his theory of general relativity in 1916.
A third example, from antiquity, was the inclusion of epicycles by Ptolemy in the Aristotelian system of astronomy to account for irregularities in the observed orbit of the planets. All three of these examples highlight the scientific communities ability to defend a theory in the face of inconsistent experimental data.
The point of these three examples is to show that the scientific community can always tweak the theory or tweak auxiliary hypotheses (i.e posit a yet undiscovered planet) in the face of incongruent data.
Moreover, even when a theory is 'falsified,' there is no guarantee that it won't come back to life a century later. Take, for example, the particle (corpuscular) theory of light, which was supposedly falsified in 1819 when the French physicist Dominique-François-Jean Arago observed a bright spot at the center of the shadow of a circular disk (a bizarre prediction of Fresnel's wave-theory put forth to discredit the theory). Resistance to the wave theory collapsed and by all accounts, the particle theory of light was completely and utterly destroyed. However, fast-forward to 1905, and the particle theory is resurrected by Albert Einstein to explaining the photoelectric effect and ultraviolet catastrophe.
The point is that the scientific community can always explain away a null result by challenging one or more of the auxiliary hypotheses, and/or by adjusting the theory to account for the results. And even if the scientific community agrees that a theory is falsified, it still might be resurrected at some point in the future new and unforeseen reasons.
- The only conclusion you can draw is that just as you can not prove scientific theories, you also cannot disprove or falsify them either. A 'proven' theory might turn out to be wrong just as a 'falsified' theory might turn out to be correct in some way we could never imagine at the time.