I'm halfway through Graham Farmelo's Dirac biography at the moment, and enjoying it quite a bit. Farmelo cites Dirac as one of the first physicists to evaluate theories on their mathematical beauty, rather than waiting for experiments. This is in stark contrast to his Cambridge colleague Rutherford, who was highly skeptical of abstract theory, and preferred to deal in concrete experiments.
This is one of the great chicken-and-egg problems in science: Should data come before theory, or should theory come before data? Thus, this seems like a good topic for a poll:
So, whose approach do you prefer, Dirac's or Rutherford's?
- Log in to post comments
For really good science, you need both approaches. You need the theories which tell the experimentalists what to look for, and you need the occasional "who-ordered-that" experimental result to drive theoreticians. If you have only one of the two, science is likely to stagnate. String theory is a classic example of theory running amok without experiments to guide it. OTOH, when experimental results are all you have, as with much of pre-1950s medicine, you tend to get bogged down in empiricism and phenomenology, and it becomes easy to lose sight of the big picture.
What is best in science? To crush your theories, see them driven before you, and to hear the lamentations of their predictions.
Science works best when the two hold hands. Sometimes one is ahead, sometimes the other, but the important part is not to let them get too far apart.
@cisko: Excellent!
I agree with Tony @ 3. Experimental data and new theories go hand in hand, sometimes one leads, sometimes the other, but if you exclude one of the two, there is no new science.
I'm also here for the Steelykid pictures, and Emmy's discussions of physics with you.
How can science be "best" when ...? Either you're doing science or you're not.
Also, where're the options for "Both" and "Neither"?
Also, where're the options for "Both" and "Neither"?
"Both" and "Neither" are cheap, lazy, cop-out options. I want people to take a stand, dammit.
Science is a dish that is best served cold.
That said, it's obvious to me that you need the data first. Once you get going, the theory and the facts kind of whipsaw each other. But, until you have at least one fact, you don't even know where to *start* a theory.
I picked option 5, because I really wanted to pick a version 4: I do actually like science, but I'm just here for the cute baby pictures....
Historically, data comes first. You have to know something before you can start wondering about what it all means or how it works or why it works.
As an experimentalist, I'm deeply ashamed to side with Dirac over Rutherford, but his mathematically beautiful theory ostensibly predicted the existence of the (insert obscenity here for emphasis) positron.
So since I'm anonymous, I'd say it works best when the theories come first and - as cisko (and Conan) say - the experiments crush the weak among them.
Both. Gotta have both. I like BOTH the doggy pics and the baby pics; they're all pretty cool. I've taken your Appa idea to my own grandson. His pics are pretty cool, too.
It's just way cooler when we have a theory for that looks beautiful on paper....then maby a couple decades later we find proof.
Paul Adrien Maurice Dirac:
"Physical Laws should have mathematical beauty."
This statement was Dirac's response to the question of his Philosophy of Physics, put to him in Moscow in 1955. He wrote it on a blackboard that is still preserved today.
Quite rightly so!