[lixtra]

From the suicide note:

> Considering that this will have an impact on my career in the future and my reputation in the area of Computer Architecture, my future life will be worse than death and I will be totally in a dilemma.

Maybe we also need to change academic culture to allow people to confess academic fraud, receive punishment and than be forgiven and continue academic careers.

Currently once you started, there is no exit and fraudsters are incentivized to continue to sustain their building of lies.

With financial debt we allow people to go bankrupt and be reborn free (after some time and a honest effort; not including American student debt).

Why not with your past academic fraud?

[tticvs]

This is similar to the advantage to amnesty for corrupt dictators proposed by Mesquita & White in The Dictators Handbook.

Basically giving authoritarian regimes an alternative to fighting to the death might increase overall utility.

[chii]

but it also gives the moral hazard of someone wanting to start such a regime, because they know they can profit/garner personal wealth during their reign, then beg for amnesty after a while (when it looks like their grip on power is lowered), and possibly, hiding their wealth gained away somewhere secret.

Where as, in a world where such amnesty doesn't exist, they may have to face the possibility of being executed when they lose power (which, if history is to be believed, they eventually do). Therefore, some who are more cowards but cunning won't want to be a dictator, and thus, you end up with a more democratic system.

[marcosdumay]

There must always be punishment. The comments above are too summarized to get into it, but when that kind of amnesty happens, it is never complete.

But if the punishment is so crass that the people prefer to lose their lives instead, that's a clear signal that your society has a problem, and your new regime isn't all that good either.

[tlringer]

Students under abusive advisors like this should not be punished and in fact are currently not punished. The power structures are too fundamental.

The problem is that if you cross your advisor, good luck getting a job later.

[zaik]

I think the authors are Bruce Bueno De Mesquita and Alastair Smith.

[MaxBarraclough]

I'm reminded of this recent comment on the consequences of zero-tolerance policies in the military.

https://news.ycombinator.com/item?id=25506316

[boublepop]

> Maybe we also need to change academic culture to allow people to confess academic fraud, receive punishment and than be forgiven and continue academic careers.

Are you suggesting we forgive the professor who’s pressure and influence seems to have cause the phd students suicide?

If anything I think we need to stand firm on the harsh one direction only nature of academic misconduct, and simply provide better training and whistleblower options for students. There should never be any doubt in a students or professors mind that academic misconduct when discovered will be an end to their academic carrier.

[s1artibartfast]

> There should never be any doubt in a students or professors mind that academic misconduct when discovered will be an end to their academic carrier.

This is a big part of the reason for the suicide. Reread the quote with this in mind

[pm90]

I don’t think that’s a good idea.

The analogy with financial debt is misleading: people go bankrupt even if they don’t do anything “wrong”, things don’t pan out, or disasters hit. For people that do commit financial crimes, they are prosecuted and face criminal punishment and often lose their ability to continue to be involved in finance.

Academia doesn’t have that kind of strict laws, likely because academic fraud doesn’t affect the general population and is limited in the damage it can do, or hasn’t received similar attention as financial fraud. So making it even more tempting to commit academic fraud isn’t something I would recommend.

The really straightforward solution here seems to be to make these institutions recognize the professors (even the prestigious ones) are human and can commit fraud, and thus to create ways for students to give feedback on such kinds of frauds to the institution, based on which the institution must take action.

[yowlingcat]

> For people that do commit financial crimes, they are prosecuted and face criminal punishment and often lose their ability to continue to be involved in finance.

Wealthy white-collar criminals often only face minor fines and are able to keep operating. The industry perception of people who get caught (who are known to still be only a small percentage of those who commit them) is almost universally "they were stupid enough to get caught."

I wonder whether this is yet another tendril of the student loan debt bubble and the privatization of a public good (education) contributing to deep, structural rot in American society. Most of the quality academic talent gets picked off into the industry, where there are real opportunity costs for mis-execution, and a much more healthy and liquid market for labor supply and demand. Does that mean that maybe the only plausible end-state for academia is a market for lemons?

[pm90]

The low rate of prosecution doesn’t mean that the laws are useless. It has coerced the financial industry into behaving in more pro-public ways than it would have without them. In this case it would likely ensure that academic frauds are not engaged in lightly.

I don’t think Research fraud specifically has much connection to student loan debts. Most PhD students are fully funded by their programs. Most of them are international students anyway and would likely not have access to the sweetest US based loans.

[murphyslab]

Forgiveness definitely should have its place, particularly for minor offenses, however the consequences are often multifaceted and tangled. For example, as research often uses public (government) funds, there are sometimes criminal consequences attached. [0][1]

A secondary issue is trying to determine what penalty is fair and trying to determine whether an error was intentional or unintentional. The boundaries are admitted by most to be fuzzy. Moreover there is a desire to exact punishment:

> A recent survey of roughly 1,800 people in the United States found that 90 percent of those who responded believe that fabricating data is morally repugnant. Preferred punishments include being blacklisted from university positions and being banned from government funding for future research, but don’t end there.

> “Most respondents who support criminalization prefer a sentence of incarceration, rather than a fine and/or probation,” according to the researchers, Justin Pickett and Sean Patrick Roche, both of the University at Albany. “The results indicate that slightly over half of all Americans would prefer both to criminalize data fraud and to sentence fraudsters to a period of incarceration.” [2]

... and then there's China, where the penalties sought seem even higher. [1][3] However, given the investment in most researchers and their specialized skills, it seems a huge waste to lock up that person in a way that prevents society from benefitting.

Finally, you have the problem of coordinating the action. With criminal behaviour, society coordinates action (for the most part) in order to achieve reasonable justice: Otherwise every person willing to act independently (as a vigilante) to extract an additional pound of flesh. We would need everyone to agree, otherwise you have situations where there is no punishment because someone breaks an embargo or you have too much punishment because multiple agents inflict punishment separately.

Your typical person breaking the law is stuck within one country. It's a bit more complicated for researchers, since most are highly mobile. A PhD in a STEM topic means visas are almost automatically granted, so it's relatively easy to get a position in a different country.

(It would be really interesting to see a follow-up on all of the researchers featured in [1], since that was written back in 2007.)

[0] http://navier.engr.colostate.edu/CH693/prot/Nature_445_244_2... [1] https://journals.sagepub.com/doi/10.1177/1747016119898400 [2] https://www.statnews.com/2016/08/04/fraud-science-jail-time/ [3] https://www.statnews.com/2017/06/23/china-death-penalty-rese...