A Crime in the Cancer Lab | Hacker News Reader

A Crime in the Cancer Lab

nytimes.com

73 pointsazuajef30 comments

I have a buddy that went on an archaeology dig in Israel. He saw first hand that artifacts that contradicted what the leader of the dig was trying to prove were destroyed.

Scientists are people. But we base so verymuch on their work that integrity needs to be the highest priority. Maybe we need a science UBI for grants??

I would hope that intentional destruction of artifacts would result in jail time and also the offender never being allowed near artifacts again in their lifetime. Hopefully as recording tools become more prevalent, such crimes become easy to expose. It's one thing to dump alcohol in an experiment that costs $10k and can be replicated. It's another order of magnitude to destroy something that can't be recovered.

People some times talk about alternatives to grants that are like UBI, but the mechanics seem odd and hard to get right since you need some sort of filter, unlike UBI. I think a better way may be to find more career staff scientists and fewer "trainees." The idea being that PIs can no longer just use up the formational years of dozens of smart people and then foist them onto a job market with no prospects, and that PIs would have to hire career scientists that get paid enough to, say, raise a child or put a small percentage away for retirement. The disadvantage being that a 2-3 year fundinf cycle can still be fairly destructive if somebody has to find a new lab potentially that quickly. So longer term commitments of smaller amounts of money may also may make sense...

There's no easy solution but fortunately the NIH and NSF seem more interested in changing these thing than they did 10 years ago.

The most obvious solution is to in increase the grad school stipend. If grad students made more then you avoid those 'wasted years'. Yes, this means we would have fewer post grads, but that seems like a reasonable trade off.

Archeology has been weaponised there and it's now part of the settlers side of the conflict.

https://www.google.co.nz/amp/www.independent.co.uk/news/worl...

http://www.aljazeera.com/indepth/features/2013/11/israeli-la...

The Al-Jazeera article is reasonably balanced for Arab media though I would advise people reading it to understand that it does present the Palestinian viewpoint. Listen to both sides and make up your own mind.

Some background on Silwan/Shiloach: https://en.wikipedia.org/wiki/Silwan

By the way, putting aside the conflict, if you get a chance to go visit the tunnel ( https://en.wikipedia.org/wiki/Siloam_tunnel ) I highly recommend it. It's an amazing feat of engineering going back thousands of years.

I find it extremely hard to believe archeological artifacts would be destroyed in a dig in Israel. Israel has very strict laws regulating archeological digs. You can go to jail for 3 years for destroying archeological artifacts. Get your buddy to file a complaint?

I find it interesting that the majority of the "outside assistance" in this case came back to Heather as the culprit. On one hand, it makes a lot of sense that the person most immediately involved would likely be the prime suspect. On the other hand, it also leaves you wondering just how willing to step outside their own pre-existing biases these people are?

> On the other hand, it also leaves you wondering just how willing to step outside their own pre-existing biases these people are?

If (as is true in cases like this one) a large database of solved cases reveal that the crimes are often self-inflicted, then it's not bias, it's common sense and the dispassionate result of statistical analysis.

Consider the murder of marriage partners -- in very many cases, the perpetrator is the other marriage partner, and that's the first person investigated. Is that bias, or common sense?

"Over 71% of Wives Kill Their Husbands" : https://rmaxgenactivepua.wordpress.com/2013/07/19/over-71-of...

(The title is somewhat misleading -- it means that, when a husband is murdered, in 71% of cases, the perpetrator is the partner.)

Or perhaps:

  71% or married men, who
  are murdered, are found 
  to have been killed by 
  their spouses.

True, the point where it becomes bias in my mind is when they decide that the assumption is true and don't look further than the first guess.

Yes, this is definitely the true issue.

Suspect whomever you want; carry out your investigation with an open mind regardless. To disregard a pattern is stupidity, but to enslave yourself to it is laziness.

> To disregard a pattern is stupidity, but to enslave yourself to it is laziness.

I may have to quote you on that one ;)

That multiple involved people were totally willing to suspect/blame Heather is rather more troubling to me than the actual act of sabotage (if only because a tendency to assume that such sabotage is self-inflicted can let culprits get away without even being suspected).

Its the most likely pattern from prior cases, just as if a man's wife gets murdered then the man is suspect one.

Experience, and patterns in general, often trouble our souls, but there is a reason they are patterns. Also, as this case shows, the previous pattern can be wrong. Its an easy time saver to pursue the things that have happened in the past before spiraling out to new suspects.

The real problem is if you do not do the actual investigation and assume this follows previous experience. The assumptions without proof almost always lead to tragedy.

Developers use this technique every day ("Well, I saw X so its probably Y in this code") so we follow the same investigative behavior and I hope developers put the time in to make sure the problem is really what happened just as I hope all law enforcement do the same.

The author did not punish Heather based on his entirely reasonable suspicions. There's nothing wrong with having a reasonable a priori guilt distribution for a crime, so long as you don't punish people based on weak evidence and you allow for new evidence to update your guilt distribution.

I was floored at the repeated circling back to suspect Heather as the suspect, and the lack of a "mea culpa" by the author for ever having seriously suspected her in the first place.

We are so, so quick to blame the victim - it lets us believe that the one thing that is "out of alignment" with our belief in a good and just world is the only thing out of alignment, even if we have to blame someone who is clearly a victim, instead of having to search for even more evidence that dispels our too-perfect mirage of the world we live in.

It's because the author was being honest about their thought process. Heather was the most likely suspect, a priori. There is nothing wrong with recognizing that, so the author has nothing to apologize for. If the author had punished Heather based on his suspicions, that would have been wrong.

I fail to understand why a person in a lab who approaches a superior to lodge a complaint that their work has been sabotaged is "a priori" the most likely suspect. Perhaps you have some kind of lab-sabotage data you're referring to? Do you have a link.

I think if someone comes to you and tells you their work has been sabotaged, the most straightforward thing to do is to believe them.

Someone with a failing business is going to be the prime suspect when it burns down, a researcher whose project appears to be going nowhere will be the prime suspect when their samples are sabotaged. It's fairly straight-forward reasoning.

You seem to draw a parallel to other types of victim blaming. Normally, people mean by that blaming the victim not because they are the perpetrator, but because they should have protected themselves better (or have hidden their vulnerability better, etc).

This incident has nothing to do with that kind of vicim blaming - so I'm not sure what you're trying to point out here.

Furthermore, applying a label like "victim blaming" pejoratively is harmful to a rational discussion because you imply that it is intrinsically harmful, even though that's impossible to state categorically - the notion of blame and victim are so broad you can apply thing to all kinds of beneficial behavior. It's no better than the cliche "think of the kids".

The "victim" came to their superior to complain that their work had been sabotaged. After talking about it, her superior came to the conclusion that the most likely person to "blame" was the "victim". Hence, the straightforward use of the term "victim-blaming".

Since my wife has done a lot of work with abused children, I was thinking most directly of that - when someone alleges abuse in a school or church community, it's quite common for people to bend over backwards to find ways to disbelieve or blame the victim. In the case of child abuse, almost nobody blames them "because they should have protected themselves better (or have hidden their vulnerability better, etc.)". They generally find ways to disbelieve them, or (quite commonly), say things like "you know how children lie about these things" (spoiler: they generally don't).

The parallel I drew between two different (and vastly different in terms of personal damage done) types of blaming the victim was positing that it's the impulse to preserve the ideal, imagined just world that would drive people to go to such lengths to discount the straightforward path.

Your last paragraph is a muddle. I think you're claiming that using the term "victim blaming" is an appeal to emotionally-loaded terms in a heretofore "rational" discussion, and claiming I'm jumping the discussion from the (presumably usually "rational" hackernews norm - hah!) to some kind of mindless emotional appeal. I think that's my cue to exit…

I think the issue here is that the term victim blaming is not a simple sum of the meanings of victim and blaming. It's become something of an idiom. As the parent said, when someone cites "victim blaming" they're nearly always talking about situations where there's a clear victim and clear perpetrator (even if the perpetrator hasn't been identified), yet people claim the victim is largely at fault because they should have better protected themselves.

I don't often hear the term used in cases of child abuse; or, when I do, it's a response to people claiming that the child shouldn't have been so annoying or provocative. I think most people would agree that discrediting or ignoring child abuse complaints is a separate issue that's not what people mean when they say victim blaming.

In the case of the cancer lab, I would say this is a third type of scenario where people suspect the victim. In fact no one actually blamed the victim--merely suspected them--and everyone went forward with the investigation and eventually arrived at the truth. (Saying it's blaming because they're thought to be to blame is a conflation of idioms, in my opinion.) Once again, this definitely isn't what most people mean when they say victim blaming, not just because the victim was never blamed but because of the specific meaning we already talked about.

It's totally okay to use the words for their definitions instead of going with what most people think the terms mean--that's your right. Just know that if you use the term like that, you're probably going to confuse people. Because victim blaming is a hot-button issue right now, it might start some pointless arguments that could be avoided by using different semantics.

For what it's worth, I think it's okay to doubt, so long as you still follow through with the investigation. I'd say even in the cases of child abuse the issue is that no one bothers to find out the truth--not that they initially doubt the children. Clearly the victim could be lying or they could be telling the truth, and your personal estimation of those probabilities are based on all kinds of things. What matters is that you come to the correct conclusion in the end. Unfortunately, people often don't bother to investigate at all because it's messy and inconvenient.

It's a pretty bizarre misuse of the term. The obvious person to suspect in this situation is the person who has the most to benefit; for example, it's not victim-blaming to suspect a homeowner in an arson case when they got a large insurance payout, it's choosing the obvious suspect. Graduate students'/post-docs whose research is going nowhere have done all sorts of stupid shit in the past (like sabotage another student's work for no apparent reason).

I was surprised to see the actual culprit's name show up frequently on Google, as apparently this is a well-known case:

http://www.nature.com/news/2010/100929/full/467516a.html

And apparently, such sabotage is so rare that the Ames case is frequently cited whenever sabotage is suspected:

https://www.uam.es/personal_pdi/ciencias/jmsierra/documents/...

I don't think that sabotages are rare (as the lab head mentions in the NYT article). Though reporting to police certainly is.

It's also going to be damn hard to even notice sabotage.

Sure, alcohol in media is easy to smell. But there are a zillion ways to hose someone without leaving any trace.

These kinds of experiments tend to be subtle and fiddly. Getting a result even without interference is difficult enough.

And if the story sounds hard to believe, the Nature blog entry from 2011:

http://blogs.nature.com/news/2011/04/lab_sabotage_deemed_res...

"Nature releases exclusive surveillance tapes, obtained through the Freedom of Information Act in Michigan."

The finger pointing at the graduate student Heather Ames reminds me of the concept of gas lighting, derived from the 1944 movie Gaslight:

gas·light

ˈɡaslīt/

verb

gerund or present participle: gaslighting manipulate (someone) by psychological means into questioning their own sanity.

https://en.wikipedia.org/wiki/Gaslighting

This is a fairly common bullying tactic where the bully both bullies, usually secretively, and tries to make the victim look like a nut case.

The story, with others pointing the finger at Heather Ames, raises the question whether more people were involved than the single researcher who was caught on video. While self-sabotage was certainly a possibility, sabotage by a colleague or co-worker was clearly a substantial possibility and reportedly proved to be the case.

Found a picture of the victim and perpetrator - http://mabuhaycity.com/photoplog/images/205/large/1_1073_000...

And this one has a shot from the hidden camera... https://image.slidesharecdn.com/researchmisconduct-110506045...

Its always good to put a face to the name. It makes it more real.

The end of the article

> This axiom is especially important in a “post-truth world” where we must work even harder to question our assumptions, whether in science, medicine, justice or any other field of problem solving.

tied in perfectly with the ad that followed below

> The New York Times. Truth. It's hard to find. But easier with 1000+ journalists looking. Subscribe to The Times.

With new administration that is hostile to truth, we need real journalists now more than ever. I've tried to do my part to support them with subscriptions to NYT and Washington Post.

Reply on news.ycombinator.com