Are scientific theories ever proven? Or does the evidence for them just become stronger?
If there’s enough scientific evidence, a theory might be proved, thus no longer being a theory. But if it’s only consistent 9 out of 10 times, then it’s a theory with limitations.
Nope, never. Proof is for math and alcohol.
Adman I have to object - could you cite an example of a scientific theory that has been 100% proven? I have to side with Joe Random here - as time goes by, we just accumulate more evidence convincing us a theory is true - but we’re always open to refining theories in the light of new evidence.
Now, theories in math can be proven.
Hi Major Kong.
The empirical sciences progress by way of gathering data and formulating theories which account for those data. In these sciences, a theory is an attempt to describe why something is the way it is, such that we can make successful predictions based on past experience. The theory enables us to extrapolate from what we have observed in the past to what we expect to observe in the future.
A theory is regarded as successful only to the extent that the predictions which follow from it tend to correspond with what we observe in real life. The longer the period of time over which a theory seems to be successful, and the more people who tend to agree that it’s a useful way to describe experience, the stronger the theory is deemed to be.
Sometimes, fresh evidence comes along which is in conflict with a given theory. This provokes attempts to either show that the theory is, in fact, correct, or the devise a better theory which can account for all the existing observations just as successfully as the old theory did, but also works for the new data as well. This is what the empirical sciences are all about.
If a theory seems to be consistent with all the evidence over a very long period of time, to the extent that there is universal agreement (throughout the scientific commnity) that it never falls down and can be trusted, then it tends to be regarded as an ‘established fact’. However, in science, it is always recognised that if fresh evidence comes to light which contradicts the theory, then the theory will have to be revised and improved.
More informally, and away from the strictly accurate terminology of preofessional science, if a theory has held good for a long time we tend to say “It’s true” or “It’s been proved”. But it’s always worth bearing in mind that throughout history, people have considered things to be “true” which turned out to need revising - in other words, we found out more about the way the world works and we needed to revise a long-held theory to improve it and make it stronger, more successful.
Within a given, isolated experiment, you can run a test and say that the result ‘proves’ that a given theory works, but there is always the possibility that one day someone will conduct a different experiment and show that the theory has a limitation or a flaw we didn’t know about before.
It’s all about consensus over time. ‘Truth’ is a label, a term we give to the correspondence or agreement between experiences. If you want to check out whether mercury or lead melts at a lower temperature, you can check various sources, devise an experiment to find out, read about it, ask people… and everywhere you turn you will find that everyone says mercury melts at a lower temperature than lead. It makes sense to regard this as “true” because all sources of information agree or correspond. In fact, no-one has ever observed any evidence to the contrary. The greater the degree of correspondence between different people’s experience, over time and throughout the world, the more safe it is to regard something as “true” or “proved”.
Rather interestingly, the school of thought championed by a man called Karl Popper says that a theory is only worth anything if it is falsifiable - in other words, if it is at least possible that some facts or some evidence could come along which would show the theory is flawed. According to this school of thought, a theory which is non-falsifiable simply has no place within the empirical sciences.
When you move away from the empirical science, words like “proof” and “proven” are used in different ways. They might no longer have anything to do with experiments, observations and figuring out the way he world works, and is likely to work in the future. For example, in logic and mathematics, if you can show that a particular conclusion follows from a given set of premises, then you can say you have ‘proved’ that conclusion is valid i.e. that if the premises are true, then the conclusion is also true. But this doesn’t have to have anything to do with the real world, and things we can detect and experience.
For example:
Premise 1. All people called Cecil wear glasses.
Premise 2. This person is called Cecil.
Conclusion: This person wears glasses.
This is a form of reasoning called a syllogism. The conclusion follows correctly from the premises. However, a syllogism doesn’t necessarily have anything to do with real life (in real life, there are people called Cecil who don’t wear glasses).
In mathematics, sometimes we try to define and describe the mathematical relationships between real-life data. For example, Pythagoras’s theorem tells us something about every right-angled triangle we ever meet or encounter in real life, and it can be useful in things like construction and engineering. But some mathematicians explore the types of mathematics which have nothing to do with real life so far as we know. For example, in geometry there is Euclidean geometry, which is basically ‘real life’ geometry, and non-Euclidean geometry, which is based on what things would be like if different basic rules applied.
You might think it’s rather useless for maths geeks to explore areas that have nothing to do with real life. However, there is always the chance that by exploring these other mathematical possibioities and frontiers, we will achieve new breakthroughs in terms of our understanding and mathematical capability that will eventually prove to be useful in real life.
The words do not have universal, precise, meanings and need to be taken in context. For example:
For instance: the law of gravitation says bodies attract each other with a force directly proportional to etc. And yet we have no theory which explains why that law works like that.
Oops. hit submit too soon.
In other words, a law describes what happens and allows us to predict what will happen. A theory explains why it works that way.
There is no point in getting too hung up on words though and these words are used differently by different people.
An example and an instance! 
One of the reasons that we have no theory which explains why that law works is because we now have a new theory (and observations) that show that it doesn’t really work that way.
Just to help out Adman, here.
Crick and Watson’s theory about the double-helix structure of DNA was later proven by Kornberg.
Would it be correct to to say theories are often said to be proved if evidence if suffecient that they are almost certainly true?
For that matter, Newton’s (or Einstein)'s theories of gravity are known to be not 100% correct, but are a simplification of whatever really happens, that’s very nearly true for almost everything we can observe (or technically, might be true in some limit, eg. speeds ‘almost’ zero). These are accepted in some sense, despite not being true.
Well, Newton’s theory is no longer used in serious work since it’s sufficiently flawed to make a difference in real-world calculations, i.e. the orbit of Mercury. But Einstein’s theory of General Relativity is so accurate it can be used with great precision for any normal calculations in the universe. The only place it breaks down is where the very small meets the very large, that is to say, when quantum dimensions meet macroscopic forces, such as in a black hole, where the dimensions involved are extraordinarily tiny, yet the force of gravitation is huge. Here, a new theory is taking over, and in fact supplanting General Relativity: M-Theory, which is based on the proven work of supersting theory. An excellenyt read on this subject is Brian Greene’s The Elegant Universe.
Proven? When?
Poor choice of words. Rephrasing, it has been shown experimentally to agree with reality to many decimal places, and has made predictions that have subsequently been shown to be correct. For example, string theory gets rid of the pesky infinities that pop up when we try to apply Einstein’s equations to black holes.
I have to nitpik a bit here. We sent a man to the moon and all the trajectory calculations used were Newtonian. I’d say that’s a real world calculation. Ditto for the orbits of the hundreds of satellites cirlcing the earth. No building or bridge that I’m aware of needed anything more that Newtonian Phsysics for it’s construction. Trajectories of missiles fired (eg, the ones we shot in Iraq) use Newtonian Physics. I could go on…
Watson and Crick proposed a hypothesis to be tested; a model to be used to guide future investigations. Indeed, DNA has since been shown to take forms other than the Watson-Crick model (B form), including the bizarre, left-handed Z form.
No, that’s revisionist, Terminus Est. What was then called a theory, and later shown as fact, you relabel as hypothesis.
And the existence of other forms doesn’t negate Kornberg’s proof.
I guess that it just has to be accepted that there’s no real consistent use over the centuries of the terms “law” and “theory”.
No, sorry, Terminus Est has it exactly right. Scientific work is always to prove a hypothesis. A theory should be reserved for the overarching body of explanations that uses these facts. The structure and function of DNA is one tiny piece in the huge intellectual structure that is evolutionary theory.
It’s true that the word theory is misused constantly by the general public and even by scientists at time, especially when speaking to the public. Obviously this causes confusion. But the confused view is still a wrong view.
Watson and Crick never had a theory; they had a hypothesis. There is no revisionism involved.
Evaluate Watson and Crick’s original paper for yourself. It’s just one page long and contains the understatement of the century:
Is this the same as methods, algorithms, and rules?
I’m sensing a little nerdtivitiy here… 
What we know to be facts today have to start out somewhere before today, right? If what was thought to be a theory turns out to be the actual outcome of an activity, it has been a theory that has been proven.