"Everything we thought we knew was flawed" scenario

Has there ever been an instance of a student pointing out a error during a lecture that changed a fundamental concept of the discipline? (My question is a bit awkwardly put, sorry.)

In other words, Professor Sneedle is giving his talk up at the board, then the student raises his hand and says, “Um, Professor? Wouldn’t that make X behave like Y?” The Professor studies the current slide up on the whiteboard and exclaims, “Great Scott! You’re right! This changes everything we know about the subject!”

Every terrible disaster movie ever.

I doubt that it could really happen - it’s based on viewing science the way pulp science fiction did (and a lot of later sf copied), that science is pretty much just one guy having a great idea and no one else really gets it. The reality is that by the time anything is making it to an undergrad lecture, lots of really smart and really dumb people have been thinking about it and poking holes in it, and that generally all the little bits of science intertwine with a lot of other bits. The closest thing to what you’re thinking of is probably a student accidentally solving a major unsolved problem in the field thinking it’s homework, the RL case of it is is George Dantzig (from http://www.snopes.com/college/homework/unsolvable.asp)

yes. For example, when Einstein developed Relativity, it was already well known that something odd was going on with the speed of light. No matter how they measured light it was always the same speed. How was that possible? Einstein came up with the not-obvious answer, but everyone was thinking about it.

Similarly, “everyone knew” there was a chemical genetic component - it’s just that the “discovery” of the double helix was an aha! moment of Crick and Watson, and it explained a lot of things beautifully - how the molecule stayed together, how it could reproduce, etc.

But then, these were puzzles that people had thought about long and hard - not revolutionary overturning of established though.

Perhaps a similar more modern overturn of scientific thought would be he acceptance of plate tectonics. Several people had the idea that chunks of the earth moved, based on assorted evidence (continent shapes, mountain characteristics, earthquakes). Simply, as more evidence became stronger, such as undersea rift zones and subduction zones, the theory took over as the accepted science.

But only after they call the POTUS

Even if it had ever happened, the response wouldn’t be “Great Scott, you’re right!”. It would be more like “Hm, that’s a good point, I’ll have to get back to you on that.”.

Indeed. “Gadzooks!” would be far more appropriate in such a situation.

Feynman describes his revelation about beta decay and the tau-theta puzzle:

Note that in the Feynman revelation, he wasn’t a student who spotted an unknown problem which overthrew an established theory as the professor was lecturing on it. He was already a professional in the field, and the other people he was talking to in the field were aware that there was a problem with results matching the established theory. He didn’t actually spot the problem himself, and it the problem was found by delving into new experimental results - not from something that would be taught at a lecture. So while it was an impressively huge insight on his part, it’s nowhere near the disaster movie trope in the OP.

From my days as a prof in Computer Science.

Once in a great while, some student would think he/she had a brilliant insight and all I had been lecturing about was wrong. It ranged from knowing “for a fact” that the binary representation of 1/10 could be represented exactly in X bits to overturning the Halting Problem.

Most of the time it quickly became clear they had the “true believer” mentality and there was no point in finding the error in their thinking.

There have been cases where someone would point out during an advanced seminar or such a minor to huge problem. I did a lot of that as a student and it helped me get set up with my advisor (and a Re$earch Assistantship). A few things led to papers, but nothing field changing in any one case. (My big stuff didn’t arise this way.) But mostly it lead to me getting an acknowledgement in the speaker’s paper later on.

I think that sort of stuff is common to a lot of stronger researchers. (Let me tell you, having one of the top people in the field sitting in when you give your latest and greatest stuff is quite scary. None of them ever found anything wrong with my stuff, but it’s always on your mind.)

Yes, but it has to be a black man…or sometimes a woman

As for the actual student thing…as a student I DID see a few times where a bright/lucky student DID ask a question that DID make the prof stop dead in his/her tracks and have to get back to us later about it. And about something fundamental too.

And these were not freshly minted profs either.

I will say once when a student asked me a question in class, the answer made me realize there was a extension to a model which in hindsight is obvious, but had never been mentioned or utilized before. It was one particular “aha” moment in my career, and resulted in one of the fastest-start-to-finish papers I ever wrote. (Well sort of. It was actually combined with something else related.)

There’s the supposed story that Ludwig Wittgenstein was explaining some simple concept to a class and started out by saying “Suppose X is the number of sheep in a flock…” and one of the students raised his hand and asked “But, Professor, what if X isn’t the number of sheep in the flock?”

Wittgenstein supposedly later said that this was either the dumbest or the most profound question he had ever heard in his career and he couldn’t decide which it was.

I had a moment like that once. We were rehearsing our presentations for an upcoming conference, and one of the other students mentioned that he was assuming for simplicity that a certain parameter was very small. I pointed out that that parameter couldn’t be small, as it would require an angular momentum smaller than hbar. The other student then went on to find a looser but still adequate bound for that parameter that allowed his work to proceed.

Of course, this was only one small aspect of one piece of research on one very limited topic, and far from being something “everyone knew”, it showed up only in that one presentation that, at the time, only a half-dozen people in the world had ever seen.

The OP isn’t describing the way science works. You never say “Wow, I’ve got this great idea and I know it’s right!” And no scientist is ever going to say “Your idea is amazing! This changes science!” Science isn’t a set of beliefs and opinions. If a student ever said “Um, Professor? Wouldn’t that make X behave like Y?” the best response he or she could hope for is “Great Scott! That’s a possibility! We need to conduct experiments to see if it’s true!”

More likely than that would be, “Interesting, how would you like to be my understudy?”

Well, there’s sort of something like that going right now, mostly in the social sciences. After some checking, it turns out that a great many studies and published findings can’t actually be reprocessed reliably, which makes their validity in the first place suspect. This is problematic because it includes studies which have been accepted as fact for many years. But it’s definitely not a single question or a lone student pointing out the problem.

After reading the OP, I’m thinking something like: What if, someday, a beginning algebra student were to discover two real numbers, a and b having the property that
a + bb + a
and calling this to the attention of the teacher. This would certainly be a moment where:


Which studies? Sounds very interesting.

It’s more true than people realize:

That’s just one example.

You should be able to easily prove this to yourself.’
How many headlines have you seen about a vitamin or supplement that was supposed to help with X?

I’m guessing I’ve seen over 100.

How many are currently accepted?

Virtually none. Vitamin D, some stuff for pregnant women, and maybe a couple others.

Remember all that hype on anti oxidants. Walter Willet - the second most published author of nutrition studies - admits that there is only ONE study that convincingly proves any benefit. And it took - I think 18 years for it to show up.

But I believe the poster was probably referring to this:

Oh and don’t get me started on statistics - can’t remember the percentage, but well over 25% of studies that use ‘statistically significant’ - don’t calculate them correctly or give them meaning that isn’t accurate.

I get to see the data that doctors and scientists use and it is atrocious in many cases - as is what they think is appropriate to do with it. Everything you read in published journals should be taken with a huge grain of salt - until you see someone else replicate it.