The Sad Story of Heisenberg's Doctoral Oral Exam (1998) | Hacker News Reader

The Sad Story of Heisenberg's Doctoral Oral Exam (1998)

aps.org

262 pointsbladecatcher65 comments

This highlights what was a huge misconception for me about a PhD. I thought a PhD was about making an original contribution to my field. I thought Academia was a place where you took risks in pursuit of knowledge, and a PhD was your first real go at it.

In retrospect (I'm a PhD dropout) the PhD is really more about training in the fundamentals of scholarship. It's about building up background knowledge, and learning the mechanics of research and publishing.

The actual scholarly contribution matters almost not at all. This is why faculty will pressure you to pick a conservative project... the results are besides the point. The point is demonstrating that you can do all the steps. Because lots of great people can only do half the steps. A PhD means you can do all.

Once you have your PhD, then its your career on the line and you can do whatever you want. Before that point, you're really working on borrowed (from your advisor) time, and as much as it might seem like you are supposed to blaze a path, they really just want you to show that you can walk in a straight line.

I understand the part about building up the background knowledge (of the particular field) is going to vary wildly, but it seems like the part about learning the mechanics of research and publishing should be something so well documented it shouldn't be that big of a deal. Do you have any good resources that summarize that knowledge and/or process?

No amount of documentation can teach that. The entire discipline of research, collaboration, iterating, grant writing, etc was an amazing experience (and I didn't stay in academia).

> but it seems like the part about learning the mechanics of research and publishing should be something so well documented it shouldn't be that big of a deal.

No.

I think you're thinking: "how to format/structure a paper/grant, how to conduct an experiment, how to write a proof, how to write a bibliography, how to conduct a literature search, how to submit a paper to a conference, ..."

Those are all very easy and formulaic by comparison.

What your parent actually probably meant: "how to pick a research problem, how to develop a multi-year research program that addresses that problem along with other related problems, how to find collaborators and succeed in working with other researchers, how to write a convincing paper, how to write a winning grant proposal, how to find funding, how to present your work in a compelling way at meetings/conferences, ..."

Those things all have a million particularities and can't be taught from a text book or lecture. You have to just do them and experience the millions of little things that come up and see a pro handle each situation.

It's like the difference between knowing how to incorporate a Delaware C and put together a passable pitch deck vs. actually running a particular type of business through every stage. Like a start-up focused on X, where X is your research field and there are maybe only a few other labs (companies) doing X in the world.

I'm getting a negative tone from this post and I don't really see why. Training is the point of graduate school. That is why we still call them graduate students.

> I thought a PhD was about making an original contribution to my field

> the PhD is really more about training in the fundamentals of scholarship

You realize that these two concepts are not mutually incompatible? And furthermore, that the former is contingent upon capability in the latter?

> as much as it might seem like you are supposed to blaze a path, they really just want you to show that you can walk in a straight line

The vast majority of graduate students are not capable of conducting independent research when they start (I certainly was not!). Often, it takes years of training to get them into a state in which they are sufficiently knowledgeable to take a real leadership role. Expecting someone to be capable of independent research would be wasteful: they need more guidance! This is why they are called 'advisers', not bosses. Without them most students would zig off into a direction that is fruitless.

However, to suggest that the work you do at this time is meaningless is also wrong. Your adviser almost certainly has funding sources that support the work, and this is likely directed towards productive and meaningful pursuits. The fact that students are not leading these effort is not a detriment, and in my experience even smaller research projects quickly provide amply opportunities for taking initiative. Furthermore, very few students finish a doctorate without writing papers, so I find the notion that there is no actual scholarly contribution inaccurate.

Hm, that's an interesting reaction. I also don't really see why that you're getting a negative tone. I don't think the post is suggesting that the work is meaningless (as your comment interprets) but rather that the meaning of a PhD is mostly not in the thesis itself, but in the ability to conduct independent research, like you said.

In science original research is only original if you don't know the outcome. There is a huge temptation to assume the outcome matters, but as long as you can validate why you looked into something then the outcome should be meaningless. Thus the thesis matter's, but only up to a point.

Otherwise you push people to either make things up, or only reward the lucky.

PS: Some non science fields have other criteria, but they are not experimenting.

> Your adviser almost certainly has funding sources that support the work,

This depends a lot on the advisor and the field. I just wrapped up an engineering Ph.D., which I came within inches of bailing on. While I had funding for the first four years, it was for unrelated work. For the last two years, I got a job outside of the university. All I got from my advisor in terms of support for my doctoral research was access to the computing resources I needed.

No negative tone intended. I genuinely wish I had understood this about Academia so I could've maybe made different decisions.

I'm also a grad school dropout, so I have a bit of perspective on this as well.

You're correct, IMO that the PhD is largely about training to become a junior researcher, and less about making an original contribution to science. But, based on the dissertation defenses I've attended and theses I've seen, I would say they are less "conservative" than "scope limited." Many people I know have gotten two years worth of papers out of their dissertation topics.

This may be field-specific, however (I was in math).

There's a mechanics of research? Why is my supervisor hiding it from me?? It must be his plan to weaken and kill me...

My advisor put it like this: Any person with a doctorate should be able to, on their own, begin a department in the undergraduate basics of their subject if hired by a university that doesn't have one.

I doubt that 80% of the doctorates in biology could do that.

I dropped out of academia, but I effectively did that myself, by starting a nonprofit research organization and acquiring the resources to conduct science from almost zero.

A reassuring story for those of us who feel we are strong in many areas but fear we have deep, dangerous holes in certain fundamentals.

This vignette explains part of something I hadn't understood about the emergence of Heisenberg's work (https://en.wikipedia.org/wiki/Werner_Heisenberg#Matrix_mecha... ): he seemed to work out the core theory of QM without really developing a sensible, general approach. Compare this to Newton, who did develop calculus to explain mechanics (even if we these days use Leibniz's contemporaneous work). In Heisenberg's case, Born was the one who realized that we should use matrices.

It's still weird that Born didn't get the nobel for this work and had to wait 20 more years to get one.

Wasn't it Schroedinger who approached QM using matrix algebra?

no, Schroedinger approached QM via differential equations

Thanks, I keep mixing them up...

Nothing wrong with that -- I find Schrödinger very hard to follow!

> A reassuring story for those of us who feel we are strong in many areas but fear we have deep, dangerous holes in certain fundamentals.

Then you will likely enjoy The Man Who Knew Infinity. Hardy and his peers were often shocked at the gaps Ramanujan had on basic fundamentals given his brilliant findings--based on this lack, his discoveries seemed like magic. Wolfram reviewed the movie here: http://blog.stephenwolfram.com/2016/04/who-was-ramanujan/

"But that fall Heisenberg's worried father wrote to the famed Gottingen experimentalist James Franck, asking Franck to teach his boy some experimental physics. Franck did his best, but could not overcome Heisenberg's complete lack of interest and gave up the effort. If Heisenberg was going to survive at all in physics it would be purely as a theorist."

I had not heard this part; and might shed some additional light onto why the German atomic project was significantly behind. When the lead of your project is a famous physicist, but who isn't strongly grounded in experiment, but who nevertheless feels like he can't simply be a manager and must have input, you're likely going to have problems.

Maybe not though; be an interesting line of investigation though. Anyone know if this was a documented issue? I know that Heisenburg had seriously overestimated the amount of necessary fissile material needed for a bomb.

http://blog.nuclearsecrecy.com/2013/09/13/what-did-the-nazis...

One of the reasons why the German atomic project was so much behind because there was no concerted effort, but instead several competing organizations (Wehrmacht, SS, Post ministry, possibly others) that didn't have nearly enough fissible material in the first place.

Heisenberg's greatest contribution to humanity may be his (accidental or intentional) scuttling of the Nazi nuclear program.

I had heard that one of the issues was that the Germans preferred to do the fission calculations in the context of a plane rather than a sphere because the math was easier. I read that at one point, but I have absolutely no idea where it came from. Maybe that was why they overestimated the required amount of fissile material?

As I understand it, the overestimation was (at least partly) about a failure to understand how cascades worked. In particular, to develop a proper model of a 3D cascade in a sphere.

The result was a consistent underestimation of how quickly fission would advance, implying the need for a larger reacting body than is actually required to sustain a chain reaction.

Jerry Lettvin used to tell a story, to support his contention that students were no longer being taught how to think. He said there was a grad student at whose orals Jerry was one of the professors. Before the student arrived, Jerry noticed a beverage bottle sitting in a window, with the sun shining on it. He went to the window and rotated the bottle 180°, and sat down. The student came in, and Jerry asked him to examine the bottle. "Which side is warmer?" Jerry asked. The student replied that the side facing away from the window was warmer. Jerry said, "Explain this."

To my recollection, Jerry didn't relate the student's response, but from the way he told the story I gather it was unsatisfactory.

The way you think when suspecting someone may be misleading you is different from the way you usually think. To find the truth when someone is actively trying to fool you is just fundamentally harder, and if we all went round in the mindset of being prepared for that we'd never get anywhere.

That is hilariously cruel.

"Find the height of a building with a barometer."

"Lower it on a string"

I love this story! This isn't a quip: read the article, then finish reading my comment.

"Was Heisenberg a good physicist" - well, he was and he wasn't.

This works at the macro and micro level.

If you ask Schrödinger, he would comment "which depends on whether you ask him or not."

It's always encouraging to be reminded of how the giants in the history of science were also just people.

"Accustomed to being always at the top of his class, Heisenberg found it hard to accept the lowest of three passing grades for his doctorate."

The only sad thing I can see is that, according to the story, receiving such a low grade at his final oral exam in experimental physics undermined his confidence in his own skills in experimental physics.

I would hate to sound blunt, but receiving a low grade for being unable to answer basic questions should not be a surprise to such a theoretical genius. He got his doctorate anyway.

Indeed. It should never surprise anyone when mediocrity is valued above all else and ass kissing trumps brilliance. Never think that you are allowed to focus your efforts where you can actually make a difference. Not while in school anyway.

Where are you imputing surprise?

It is somehow reassuring that even a mind as brilliant as Heisenberg's had its limits.

Mandatory lame joke - "I love driving my Heisenbergmobile, but every time I look at the speedometer I get lost."

I don't know what's supposed to be sad about this story. It's interesting, reassuring even about the high level of scrutiny physicists are in if geniuses like Heisenberg even have such troubles. But sad?

I agree. "Sad" is the wrong adjective.

I've seen similar stories take place when I was in grad school.

In one instance, a professor wanted to fail a student (in his qualifying exam) b/c he didn't seem to take one of his courses seriously (and showed limited knowledge of this subject area).

The prof was willing to let the student re-take the qualifying exam, but he wanted to prevent what he perceived as a student "skating through" the program.

I think that's what happened to Heisenberg here. The only reason it's remotely interesting is b/c Heisenberg is a legend and so we all feel like he deserved to be forgiven for his distaste for experimental work.

It shouldn't be surprising for anyone who takes an exam unprepared that you may walk out with only an average grade. I mean, it's not like he didn't know he was going to get questions on experimental physics, nor the impact they would have on the final grade...

I was lucky to hear Heisenberg speak at MIT shortly before his passing. But I forgot what he talked about, something historical I recall. His son was a MIT professor. Dirac gave a talk that year too. But it was his silly numerology topic on large numbers.

Some important scientists are smart about everything- polymaths- and other smart only in their field of interest. Dyson and schrodinger were more polymaths, while Heisenberg and Einstein more specialisrs.

Well, if he did more interest/better skills in practical experimental/engineering type applications, that might have had some, well, unfortunate political consequences later on in life...

Sort of a more high stakes version of the Grothendieck "prime"... makes you wonder how many more folks haven't made it through the gauntlet.

Luckily now we have this useful FAQ on dealing with doctoral exams:

http://www.mcsweeneys.net/articles/faq-the-snake-fight-porti...

What does the snake represent?

Benzene.

Thesis defense, I think.

It represents the type of A/V cable used to connect your laptop to the projector

Oh, so THAT's why my projector that I used for my PhD defense didn't display red.

Why, did the red elements mean something special?

</dijkstra>

That is only one of many possible snakes. It was not known to exist in thesis defenses before circa 1995. Since that time it has become a regular foe.

FTFA:

Q: This whole snake thing is just a metaphor, right?

A: I assure you, the snakes are very real.

Yeah, that's great, but what does it actually represent? The writer is doing the usual "wink wink, nudge nudge, know what I mean, get it" thing, but I never did postgrad and I don't get it.

I believe they're alluding to your PhD defense, due to questions like "does my advisor pick the snake?". Your advisor doesn't pick whom you do your defense to, but its not a random choice either.

I'm assuming it's the exam team member that wants to torpedo a student to show how smart s/he is by asking something tricky or tangential.

I was lucky. My supervisor asked for recommendations for who to ask from the Math department and I replied "a push-over". He thought for a second then said "Ok, I know who to ask."

It depends on which department you ask.

My advisor occasionally kept snakes.

He asked the question "What is a random variable?" in an EE qualifying exam (not mine). The correct incantation ("A measurable function into the reals") was not heard from the candidate. A lot of other stuff was heard, at length.

The cure for this particular case of snake bite was a remedial class in measure theory.

Although if you get a poisonous snake, it often means that there was a problem with the formatting of your bibliography.

That's genius.

Q: Do I have to kill the snake?

A: University guidelines state that you have to “defeat” the snake. There are many ways to accomplish this.

I'm pretty sure that I just waited the snake out until it got tired.

I never actually went in for the curriculum that included advanced study with the non-optional arena fight at the end, but I can appreciate biting[1] satire done so well.

1: That's not a pun. A pun has no place near something this well constructed.

Me, too. But then...

"Q: Could the snake kill me?

"A: That almost never happens. But if you’re worried, just make sure that you write a good thesis."

Follow-up Q: How do I know whether my thesis is good?

A: Only by the size of the snake you are presented with.

If he hadn't died of boredom, I would have had a very large snake.

I got the committee to argue among themselves about the snake, and then another class needed the classroom (2 hrs after start).

... which reminds me of this:

http://www.ccs.neu.edu/home/shivers/autoweapons.html

I'd almost bet that if you posted that to a public list/newsgroup today, you'd get A. a call/visit from the FBI, ATF, etc., B. suspended or fired by your organization, and/or C. harassed to no end by idiots on the Internet, probably including being doxed.

I hope I'm wrong...

Today i learnt haha

Reply on news.ycombinator.com