The Suicide PhD Candidate Huixiang Chen (2019)

There is this odd tension between academics and society where there was this system that was never designed for it is being treated like some sort of vocational certification system. It'll be a pity if academic culture gets destroyed on the way through.

Anyway, TLDR; I agree that the academic system is archaic and paternalistic but if that is a problem people should expect to go build a new system. Preserving a culture is going to get archaic in less than a generation.

The more central problem is the incentive structure of the academy. Genuine training outcomes cannot be measured by the bureaucracy and therefore don’t matter.

Without solving the incentive problem nothing can really be improved. In aggregate the most successful PIs will be the most ruthless blood sucking fraudsters. A few exceptional empathetic geniuses will sneak through on principle, but mostly we’ll continue hobbling along.

I had to do this and even though everyone involved were exceptionally kind people with the utmost integrity, it was made pretty clear to me that if my old supervisor would have launched an even halfhearted campaign against me (they did not), nobody at the university would have stuck their neck out for some junior grad student.

And this was primarily b/c there is no established process. One should be able to:

1. Confidentially state to an ombudsman that you wish to change supervisor

2. Have the ombudsman obtain the consent of the would-be new supervisor to take on the student

3. Have the ombudsman inform the old supervisor that a decision had been made to change supervisor

4. The ombudsman, old, and new supervisors sit down and discuss what projects the student can keep persuing

If anyone here is faculty at a university and cares about their grad students, please promote something like this.

The combination of complete control over their students (mainly due to being able to decide when/whether they can graduate) and the pressure of tenure and the publishing race makes it no surprise that so many professors slide into abuse. Reporting abuse while still a student is obviously extremely risky, and even reporting it after graduating can cause a career setback by giving up the recommendation letter.

I wonder if it would be possible to expand the role of the PhD committee (which is currently pretty much relevant only for the thesis defense) to become a replacement for a single advisor.

Blaming the system is wildly off base.

Thanks but no thank

They had no way out of their predicaments - it seems their mentor had forced them into a corner, where either they had to produce fraudulent work, or abandon their entire career.

I think there is also a lot of cultural pressure at play also. Being caught after producing fraudulent work is also an easy way to have the same effect of abandoning your whole career.

And without a fruitful result of your life's work, it would feel that you have wasted your whole life.

And in some cultures, that would mean being terribly judged by your family and friends, and worse.

So they did really think that they had no other options out and everything else was "worse than death".

This is why it's so important to have some "spiritual" life of some sort as you cultivate an academic career, where you can practice appreciating the truly valuable aspects of life, and be grateful and peaceful when you need to be. For many people on this path you don't learn these things until much later in life. This is reinforced by your entire life being dictated by grades and exam scores until the moment you start your research.

While killing oneself over an academic pursuit seems extreme, getting a PhD is a pretty extreme thing to do and one of those things where it seems like one's life to that point has been entirely dedicated to the pursuit.

Reading Philip Guo's "The PhD Grind" several years back ended up giving me one of those multi-year Kafkaesque nightmares that caused me to wake up in a cold sweat. I get it.

I only found this petition demanding the disclosure of investigation - https://www.change.org/p/demand-that-uf-release-the-investig...

I actually still have nightmares about not finishing my thesis for some reason, or having my topics rejected (although for some reason, the mental stress didn’t start until I got past that point).

Design topics are the worst: you can’t measure how fast they are or how effective they are. My advice to anyone who wants to do a PhD is to get into a field where the metrics are obvious, and then lean into Goodhart’s law for all its worth (but be super honest with numbers and careful with your methodology).

I'm a poor experimentalist (so poor I later switched to computational chemistry), but these procedures were absolutely impossible to follow once you got down to the details.

My advisor told me "yeah. we don't try to replicate anything that comes out of that lab." and left it at that.

Sobered me up real quick to the realities of how widespread these bullshit cheating practices are.

People who have been up to shifty things should be very nervous, and those entertaining doing shifty things should consider that it's very likely it will be exposed in the next decade.

Warren Buffett is pretty open that some kind of fraud is probably occurring at every one of his companies.

I worked in one of the most prestigious labs in the world and seen it first hand.

And people wonder why 85%+ of papers in natural sciences are not reproducible. The incentives are all wrong in academia and science, it's just farming papers, farming citations, money and politics.

The two major wrong incentives are the pressure to publish on high impact papers for PhDs and postdocs, as their career depends on it (this gives them limited timeframes to fail and really explore), and lab funding being tied to the same incentive.

The pressure to publish is also a major factor behind tweaking data. As your PhD is ending you need those publications or you can forget an academic career. And it's a pyramid, so you need to get ahead of others in your situations. You tweak data, you pull your strings, you leverage your or your labs connections.

On top of that, add the fact that as soon as you move in any direction you are effectively entering a niche where the number of people qualified to review your paper is small, and everybody knows each other.

If you're working on helicoidal peptides as energy transmitters on metal surfaces it's a niche. Advanced perovskite synthesis? Niche. Hormonal regulation of some gland? Niche.

There can't be that many experts on pretty much any topic in science, those are all small clubs, even when you get to popular topics.

It's a perverse system.

Maybe take a step back and consider where your understanding is sourced from. I don't think I'd take the claims or ravings of a suicidal person as completely true representations of facts you need to account for the fact that this is all just the representations of someone that we know holds particularly extreme views.

1. His future doctoral program would be hard.

2. It would be impossible to graduate.

3. He would be killed by the supervisor.

It's hard to say how much of these is warranted. He said 3 was concluded from a conversation with his supervisor. However, tragically he thought these are real enough for him that there was no other way than to suicide.

You may believe that they simply lack the desire to use their intelligence over a longer timespan, but there is effectively no difference between never wanting to do it and not being able to do it.

There may occasionally be cheaters who cheat because they get a kick out of it, and happen to be intelligent, but that isn't the norm.

You are completely wrong. Cheating is an art of its own, and mastering the skill of cheating is similar to mastering any other skill, like, for example, playing an instrument. Every time you cheat you take a risk because someone might get to know, and this risk might be thrilling.

It's not surprising at all. In Computer Science at least, intelligent people quickly realise that CS university curriculum is not particularly useful or relevant for their further career or life. The degree (and high GPA in case of US) is just a hoop to jump through, required for getting a good job. So, they focus on what's important - degree and high GPA, while treating knowledge acquisition as a secondary priority.