Is understanding the significance of calculations important in physics research?

[spaghetti3451]

A few of my lecturers have told me that most working physicists, half of the time, don't even understand what they're doing. They work through the calculations but have no idea of the significance of what they're doing or why it's important.

Do you think this is true?

 

[DEvens]

Heh.

I am recalling what Feynman said about quantum mechanics. I don't have the exact quote at hand, but it was something along the lines of nobody understands quantum mechanics.

Seriously though, the situation you describe can exist, even be desirable, in some contexts. But in other contexts it cannot be tolerated.

Example: If you are performing a calibration procedure it is important that somebody, someplace, understands all the details of the process. They need to know how to tell when the process is being done correctly. And they need to know how to be aware that something is wrong with the process. Otherwise, when the process does not go as expected, then you get real problems. The example is the Hubble telescope. The calibration process had a problem. There was a corrective action taken, but this only masked the problem. The result was, the Hubble went to orbit with its mirrors badly out of calibration. One of the findings when this was investigated was that nobody knew how the entire process was supposed to be done and how each result was supposed to look. So warning signs were misinterpreted, and additional corrective action was taken that did not help and obscured the problem even more.

On the other hand, the person "tightening the bolts" does not always have to be the person who understands the big picture. Usually it is not an effective and efficient use of talent to have the theoretician in the lab tuning the detector. Probably Carlo can make better use of his time in other ways than by being the guy who installs the new vacuum pumps.

https://en.wikipedia.org/wiki/Carlo_Rubbia

Similarly, there are probably grad students doing data manipulations who don't fully understand the physics being tested. They know how to write a good computer program that extracts data from the lab. He does need to be given, and to understand, instructions on his task. He needs to know what to look for, what indicates a problem, what indicates incorrect results, and so on. So that grad student needs to know how to tell if the detector is out of calibration, for example.

 

[JorisL]

A few of my lecturers have told me that most working physicists, half of the time, don't even understand what they're doing. They work through the calculations but have no idea of the significance of what they're doing or why it's important.

Do you think this is true?

I think the language is a bit harsh. Usually a working physicist should have a goal in mind.
Of course this means that you'll do quite a bit of wrong calculations or calculations that don't lead to useful/new results.
For example last year me and another student investigated whether we could see non-equilibrium effects on a lattice gas as suggested by our professor.
It turned out not to work (simulation and first approximation theoretical derivation) much different.
So even if you think there might something (non-trivial/interesting) happening in a system you'll have to check.