Recapping the Challenger Disaster

Is Feynman above fooling himself? Was Feynman fooling himself when he looked at the flexibility of O-rings at low temperatures?

Probably not. But I try.

Who was Behe fooling when he published an article using a model that showed evolution can’t happen, but only if he assumed every single mutation occurring in a duplicated gene except for the ones he decided to exclude is deleterious? A number of people, it seems, including possibly himself. But no one who really counts in terms of his goal of overthrowing evolution.

4 Likes

The fundamental challenge at the center of much of the history and practice of engineering is containment of potential energy. Disasters most often take the shape of catastrophic release of gravitational, inertial, pressure, or chemical energy across the gradient. The corollary of this is that containment of energy inherently involves risk.

The first question addressed at the top engineering level is, “can the potential be eliminated”? Sometimes the answer is yes, a simple relocation or alternate approach takes the risk to zero. Most often, the answer is no, containment is essential and inherent; you need pressure, say, to drive the turbine. The next question would be “is any degree of risk tolerable”? If not, the only solution is to cancel the entire project. Too bad, so sad. This is the general political reality in regards to nuclear power.

The usual answer is yes, society wants the reward and is willing to tolerate (or turn a blind eye to) considered risk. So now attention turns to risk mitigation. The first step is to evaluate likelihood vs consequence; there are varying approaches, but they all reduce to a matrix of likelihood on one axis and consequence in terms of cost and mortality on the other. The intersection of high likelihood and high consequence is unacceptable, but generally never makes it past the initial proposal. The most contentious square is low likelihood and high consequence. How realistic is the estimated low?

Where there be dragons is processes which tend to snowball, such as exothermic reactions; as opposed to those which tend to dissipate, such as endothermic reactions. A high temperature - high pressure unit is much harder to shut down if it is dominated by positive feedback. Examples would include Chernobyl with its positive void coefficient, or the Deepwater blowout with decreasing pressure on a rising column of gas.

To tame the beast, the usual approach is redundancy, such as enclosing reactor core vessels in secondary containment, and duplicating instrumentation and valves so that no single point of failure results in loss of control. As well, the control system is designed so that anything that fails does so in a way that alerts to the failure and is safe. This is very challenging. Disasters such as the Buncefield fire and Texas City Refinery explosion were directly due inadequate implementation of those measures.

Looking at these factors in the Challenger disaster, the inherent risk is as high as it gets. You have a violently exothermic process propelling the mission. The potential consequence is close to binary; you make it back safely or there is complete loss of mission and crew. What could be controlled was the vulnerability to a single point of failure. When the sealing problems came to light, the same redundancy and control fixes that were enacted after the disaster could have been done beforehand. That would have incurred considerable time and expense however, and the rest is history.

In light of this, it really is astonishing that NASA placed a school teacher on board. That itself was clearly an administrative PR stunt and had nothing to do with engineering. Ironically, this was postured as enhancing the science education of american students.

3 Likes

LOL! Abiogenesis is a disaster event!

Way off-topic even for this discussion. I’ll shut up now. :slight_smile:

No. Do you? Does Behe? Shall we look at his claims and compare them to the evidence, or is this endless round and round about how people can fool themselves all you’re willing to talk about?

Okay, anyone can fool themselves. Let’s proceed to find out who those people are. You in?

1 Like

Sorry Dan, but it doesn’t seem likely that this thread is going to sustain any discussion on the Challenger disaster.

1 Like

I “liked” this post in error and there seems no way I can correct it. Moderator help please.

It was your derail of another thread.

1 Like

Sincere question to you - Have I specifically blamed scientists/engineers for the bureaucratic problems?
I think I have been consistent in trying to applaud and commend the view expressed by Feynman including his warning to delivered particularly to scientists that can as well be extended to everyone and any organization. Particularly, I say anyone or any organization that can have self-interest. Who doesn’t want to be right? Who doesn’t chafe at being shown to be wrong?
And I’ve tried to rebut any claim similar to, “As scientists, we rigorously test our hypotheses to ensure that we are not fooling ourselves”
Note that in that quote from an actual poster that there is no mention that most scientists ‘attempt’ to do this. Rather it reveals nothing but hubris with the claim that they ‘ensure’ that they are not fooling themselves. If NASA is susceptible to such when people’s lives are on the line (not to mention billions of dollars in hardware), how are we not all?
I find the attempt to make distinctions between management and engineers and scientists disingenuous. There was a line from Report of the PRESIDENTIAL COMMISSION on February 25, 1986, Mr. Lund testified as follows regarding why he changed his position on launching Challenger during the management caucus when he was asked by Mr. Mason “To take off his engineering hat and put on his management hat”:
According to the poster of “As scientists, we rigorously test our hypotheses to ensure that we are not fooling ourselves” when I encounter a person who errs and is both an engineer and a manager, I am supposed to blame the manager part of the person and not the engineer part of the person? I blame the person, period. We don’t live in a world where there are no pressures to perform and worse to conform.
Bottom line is I give credence to Feynman’s warning about the propensity of all of us to fool ourselves and no credence to any blowhard suggesting that scientists are above such foibles.

Point to the post where the thread was derailed.
And while you’re at it point to what you’ve posted on the Challenger Disaster.

“Science is a way of trying not to fool yourself. The principle is that you must not fool yourself, and you are the easiest person to fool.”–Richard Feynman

We have the scientific method because it is easy to fool ourselves. As Feynman says, science is the way we try not to fool ourselves. We all agree with Feynman that it is easy to fool ourselves which is why we objectively test our hypotheses which is the best way we know of to not fool ourselves.

Feynman is not saying that we are always fooled. Feynman is saying that if we are careful we won’t be fooled.

1 Like

You have said nothing to rebut Mercer’s claim that scientists do their best to ensure they do not fool themselves. Can scientists make mistakes or reach false conclusions? A big yes. Do they ever realize they may have made mistakes? Another big yes and that’s what you need to understand.

Through rigorous hypothesis testing with the highest quality of data available, we rid ourselves of false ideas.

1 Like

@sam, its clear you are sympathetic to Behe’s position, so I would like to know how you know Behe is not fooling himself?

It would be best if everyone allowed him to answer this question

Odd complaint, coming as it does just under my post detailing how risk assessment and mitigation relates to the Challenger explosion. What are you looking for?

Your insistence that it is possible for a scientist to fool himself is becoming repetitive in light that posters here have agreed with you repeatedly on that point. Again, what is it you are looking for? You have not engaged with the response that the methodologies of science are designed to correct for exactly such.

1 Like

In fairness, I wanted this as a sub-thread because Feynman and wanted to share my Statistical Methods story. Call it vanity?

1 Like

My wife says it’s my ego. :innocent:

1 Like

Nothing. Why do you ask?

Wiith Dan moving comments it’s hard to reconstruct events. Forget I mentioned it.

1 Like

Sincere answer: In my opinion you had not expressed the role bureaucracy played as a contributor to the problem, blaming the arrogance of scientists instead. I could be asking too much (and I could have missed it?); no one can expect you to recount the entire event before these questions arise. The point is, we are discussing it now, and you acknowledge the problem. :slight_smile:

AND sure people can be arrogant and prideful, no question there. But the NASA response was not to hire less arrogant scientists. NASA made major organizational changes to enable cross-team information sharing, so that everyone would be better able to answer questions using science (and opinions to be heard).

2 Likes

Interesting that Sam omitted that first sentence. Why did you do that, Sam?

1 Like

To be fair, there are many versions of the quote on the interwebs, many of which do not have the reference to science.

Added in edit:

@Sam and others should give this Feynman speech a read:

A few quotes:

3 Likes