[NeedMoreTea]

> Do we now just throw away the knowledge that was gained?

Simply? Yes. Most of it was scientifically poor research, of no value, anyway. Only the Luftwaffe hypothermia experiments had any value.

Following the doctor's trial at the Nuremburg Military Tribunals the Nuremburg Code[1] established a minimum set of ethics for human experimentation. It's subsequently shaped many international and national laws and treaties.

Whilst some of that research has been used and cited, there's been many times when referencing it was explicitly ruled out[2].

[1] https://en.wikipedia.org/wiki/Nuremberg_Code

[2] https://en.wikipedia.org/wiki/Nazi_human_experimentation#Mod...

[luckylion]

> Simply? Yes. Most of it was scientifically poor research, of no value, anyway.

That's a different reasoning though, one where nobody would disagree, I'm pretty sure. If they have value, what then?

[NeedMoreTea]

The scientifically invalid point was in addition to the straight no. :)

To expand: I still think it should not be used or even acknowledged. You're on untenably thin ice, ethically speaking. "people were forcibly experimented on or killed, but we got some great data" doesn't pass the most basic ethical sniff test.

The results should be case for the prosecution, not for citation. Citation should have you hauled before an ethics panel. Find a way to conduct the research that's achieved safely with willing consent, find an alternative methodology, or find an alternative career.

It gets greyer if you go back far enough into history that we hadn't established those ethical and human rights standards. You can't really hold standards and laws against someone in a time they didn't yet exist. Given world history, that probably means pre-Nazi and eugenics eras.

[luckylion]

I didn't mean to get into the Nazis too much, just pointing to them as an extreme example. Certainly, they did research that wasn't unethical (think climate research, material science, architecture etc), but it was still done by Nazis.

The OP talked about Minsky, and I assume nobody thinks that whatever he did in connection to Epstein has anything to do with his scientific work.

I think there's a somewhat good argument to sever ties with a scientist that behaves unethically outside of science (but then again: do you -hypothetically- disregard a cure for cancer because the researcher is racist?), but to reject the results afterwards? Should we throw out everything that Tim Hunt ever worked on (and everything that cites/is based on his work) because some people perceived remarks he made to be sexist?

[NeedMoreTea]

To the wider example, it gets much, much more difficult.

There is certainly a need for some separation between work and personal life, but it can never be totally distinct. I'd find it very difficult to codify where the line should be. Context, extent and motivation becomes hugely relevant. That makes it far more suited to an ethics panel or court of law, and completely unsuited for the Twitter court of outrage.

We could paint different shades of grey all day, but for a flavour:

A racist researcher discovering a major cancer breakthrough has reprehensible beliefs, as does Minsky - if allegations are proven, but it need not impact on their previous work's ability or validity. I'd accept the previous work - if that were conducted ethically, just as I take the lunchtime sandwich from the guy I never spoke to about politics, race or what have you. But...

It starts to matter if it goes to law - was it a $50 fine or a 25 years conviction? A drunk weekend party with a one off incident of incredibly poor judgement is very different to a sustained predatory history over a decade. Employing them in the next job, even continuing to employ them, or placing in a position of visibility or policy, could indeed be a different matter. If they crossed the line to illegality, how often and how far, might they bring the organisation into disrepute? If it gets so heinous they are convicted and/or struck off, e.g. Andrew Wakefield with the MMR autism fraud, perhaps everything they researched is taitned too and should be struck.

Hence it should be for an independent and hopefully considered judgement, with proportionate penalty, not a headline and outrage fest on twitter.

Tim Hunt seems easier - he made a bad call on what I think was intended to be simply a joke. Poor choice for a specialist conference, but the outrage afterwards seemed way over the top for the offence. More a case of needing a quiet word with HR or management to engage brain before mouth, rather than discounting all his work or firing.