top | item 30298732

Feynman on group decision-making at Los Alamos (1985)

412 points| areoform | 4 years ago |cs.au.dk | reply

150 comments

order
[+] dkarl|4 years ago|reply
What strikes me is not a bunch of people keeping their ego in check -- that helps because it ensures that all points will be heard, but it doesn't ensure coherent decision-making. What strikes me is that everybody was able to keep track of what were the major points and minor points, so they could talk through the whole range of possible objections and decide that they were all less valid or less important than the points in favor. This is how engineers talk, searching for flaws and weighing the ones they find.

I think it's important for engineering decision-makers to be comfortable with this culture. It's disheartening when you have a great conversation like the one Feynman describes, the engineers are in agreement, and then the person in charge stands up and says, "Look, if everybody hates this idea, we can't be lazy and go forward with it just because we don't have a better one," or, "I guess we can't make a decision on this, because there's too much negativity and people are just going to nit-pick every idea to death!" No, no, talking through every possible objection to an idea is how you can tell that people are starting to zero in on it as the most likely best solution. The closer an idea gets to acceptance, the more thoroughly you vet it. The more it becomes clear that an idea is the best way forward, the more you start to prepare for dealing with the reality of it.

This makes practical sense, but not political sense, so it seems weird and unnatural to a lot of outside observers, who expect respect to be communicated by agreement, and emerging consensus around an idea to be expressed by a shared avoidance of mentioning the downsides.

[+] jerf|4 years ago|reply
I think one of the most important life hacks of all is to search out the reasons why you are wrong, not the reasons why you are right. Your brain is primed to tell you why you are right already. It takes active cognitive effort to find the reasons why you are wrong.

I've characterized this before as the single most important aspect of science [1]. It is obviously very important in engineering. It's important in navigating the noise of propaganda and lies out there in the world. It's important in day to day life, e.g., it's hard to scam someone who is actively trying to figure out why your scam offer is not going to work rather than helpfully talking themselves into going along with it.

I suppose there is an excessive negativity trap. In the end you must decide something. But for many people that would still be an upgrade, vs. letting their natural inclination to self-justify every decision with the full power of the human brain run rampant.

[1]: https://news.ycombinator.com/item?id=5680397

[+] N1H1L|4 years ago|reply
Tangential to this is I have observed one thing. I work in science, and a lot of our hires are fresh PhDs/postdocs. The best question to ask is the limitations of their approach. The more thoroughly they have worked on it, the better they know the limitations.

On the other hand, people who are super confident and wave away any limitations of their approach have invariably been problematic hires.

There is a cultural aspect to this too, because America for some reasons worships confidence and hates ambiguity.

[+] Datenstrom|4 years ago|reply
> talking through every possible objection to an idea is how you can tell that people are starting to zero in on it as the most likely best solution. The closer an idea gets to acceptance, the more thoroughly you vet it. The more it becomes clear that an idea is the best way forward, the more you start to prepare for dealing with the reality of it.

Sounds a lot like the Braintrust described in the book "Creativity, Inc." which was accredited with a large portion of the credit for Pixar's success.

> This makes practical sense, but not political sense

The book says that to create the kind of environment that a Braintrust can function in takes a deal of care and one thing was all politics needed to be removed. Great book though highly recommend it.

[+] hammock|4 years ago|reply
There are at two paths when you vet an idea to death (life?) like that. One is that you come to the optimal solution. Another is that enough people get exhausted defending their objections and eventually give in to what is a suboptimal solution.

So there is something still missing from the recipe for coherent decision-making. It's the reason Feynman says these are "great men" - it's NOT the process which makes it work, but rather a quality inherent in the character of these men.

I don't have an answer for what that may have been, but in the spirit of your second paragraph I just wanted to bring up this thought.

[+] pjungwir|4 years ago|reply
Even in literary criticism, this balance and specificity is such a pleasure. It's one of the things I love about Samuel Johnson's essays. (Contrast with what people are saying in the article about failing literary magazines.)

I just started reading Boswell's biography, and the introduction to the Everyman's Library edition makes the same point:

> Part of this comes from a literal-mindedness that won't mistake one fault for another or all others. This worked even when his dominant feeling was not of fondness (as for Smart) but of dislike (as for Sterne). Boswell reports Goldsmith toadying up to Johnson, in a conversation on Sterne, by saying 'And a very dull fellow': Johnson replied 'Why, no, sir.' Fact.

[+] W0lf|4 years ago|reply
In my experience (and coming just out of a meeting alike) these types of respectful and forward-moving meetings are typically in a case where each attendant has roughly the same level expertise in a given domain and knows the basics of scientific principles. Hence, everybody in the meeting brings in his or her unique view of the topic at hand and knows intrinsically when it makes sense to contribute something to the discussion. Usually things get out of hands, if somebody is talking just for the noise or, and this is worse, a clearly bad idea wins the discussion because politics or the messenger has shouted the loudest.
[+] cousin_it|4 years ago|reply
Maybe it's not necessary for everyone to be at the same level, you just need to have everyone knowing everyone else's level and be humble about it. "Bob says the tractor can't pull this load, and he knows more about tractors than I do."
[+] systemvoltage|4 years ago|reply
Even if participants have same level of domain understanding, it takes more to be willing to work towards a solution without their egos, their achievements, their alpha-ness getting in the way. I’ve seen the opposite more often than not. Where I do see this level of camaraderie is amongst close friends who know better about each other.
[+] foobarian|4 years ago|reply
I suspect it may partly also be cultural for that segment of people. Repeating yourself at seminars or other academic meetings was highly frowned upon in several instances I've seen, it was not just greatness/respect/intrinsic etc.

Perhaps sufficient greatness of attention span is a prerequisite to remember what everyone has said since the start of the meeting :-)

[+] m463|4 years ago|reply
I suspect the men in the feynman meeting had been rigorously or even ruthlessly chosen to form an effective group (for such an important mission)
[+] dijit|4 years ago|reply
I have been in such environments and I while I hadn't considered that the reason they worked so well was that they were great people: they were, in fact, great people.

An interesting thing happens when there's a high level of respect.

People stop being so anxious about not "winning", or "being heard", people just remove themselves from their ideas and feel comfortable and confident.

I left that company because the company around that team was a bit shit, but I wish I could foster an environment of high trust and mutual respect like that again.

[+] watwut|4 years ago|reply
I dont think it is just the question of quality of individual. It is result of the way leadership shapes the place, what it rewards or not. People react greatly to the environment they are in.

Whether people worry about being heard, whether being winning is important or not influences a lot how they will discuss. Whether you can afford to remove yourself from the idea is artefact of the environment too.

[+] jart|4 years ago|reply
Well I imagine when you're an Arthur Compton or a Richard Feynman or an Oppenheimer, you don't have a whole lot to prove.
[+] csours|4 years ago|reply
Several comments here about what is the missing X factor - why did these guys get along. The quoted story takes place during World War II, and they were developing nuclear technology, specifically The Bomb. It's very easy to look back and criticize the outcome, but AT THE TIME, these people thought that they had to do this to preserve their way of life and defeat evil.

I think this is a critical component of any moonshot - you have to have competition or some motivation that feels like an external compulsion.

[+] PaulHoule|4 years ago|reply
What I hear is that these people are really listening and paying attention.

In so many normal conversations people are repeating over and over again because they assume it goes in one ear and out the other and they are right.

[+] searealist|4 years ago|reply
An alternative theory is that they all got along because they were mostly Ashkenazi Jews from central Europe, especially Hungary.
[+] jdrc|4 years ago|reply
Also, these were the only guys who were trying seriously, they were on a mission,and in a military environment
[+] avazhi|4 years ago|reply
Not sure who, besides the Japanese at the time, have really criticised the outcome. Even now the Japanese seem to at least understand why it had to happen. It was the difference between Nazi fascism winning global hegemony or not. I don’t think there’s a serious argument that it shouldn’t have happened.
[+] narrator|4 years ago|reply
If I want to make a serious effort to know the truth of a controversial matter, I make it a point to argue without using fallacies or relying on cognitive biases. This kind of arguing tends to work better in traditional forums instead of newsy forums like HN and Reddit. One can spend a long time researching one's argument and come back several days later with a well researched and formed argument and bump the thread back to the top. In HN and Reddit, by the time one has done this research, everyone has moved on.
[+] suction|4 years ago|reply
This also describes companies during the "we're creating a great product / service" phase, when meetings aren't "won" by those who can sell themselves or their ideas best. Companies on the decline will always have very charismatic / well-spoken / authoritarian characters at the top who'll put their own ego before sustainable success. Think Ballmer at Microsoft.
[+] spiderfarmer|4 years ago|reply
I was at a conference in Amsterdam once with a lot of talks from CEO's of fast growing companies. The day was very inspirational. The speakers were obviously very competent and the questions from the CEO's in the audience were interesting and they all received great answers.

To top it off the chairman of the day summarised the whole event in a speech that I still remember to this day. It was so charming, intelligent and erudite. I was floored.

[+] Max_Ehrlich|4 years ago|reply
I recommend everyone read "Surely You're Joking Mr Feynman."

The parts on Brazil were particularly interesting.

[+] KKKKkkkk1|4 years ago|reply
Narrator: What these guys didn't know is that Compton talked it over with the chair before the meeting.

If you've worked in a bureaucracy, you'll know that the important decisions are not made in meetings.

[+] dav_Oz|4 years ago|reply
It's a great angle on the communication side of science.

I think the notion of "great man" like the genius of "Einstein" or "Newton" is seriously skewed. If you take a closer look at their correspondence and upbringing, they both were very well "connected" in their own ways beforehand.

This picture of a loner sitting bored at a Patent Office and all by himself coming up with 4 groundbreaking papers (photoelectric effect, brownian motion, special relativity) in 1905 - [or Newtons "annus mirabilis" after fleeing Cambridge University because of the plague in 1665 and "quarantining" himself for 2 years in his native village (by his own account (!) there he singlehandendly developed integral calculus, dicovered the composite nature of light and finally refined his gravitational theory)] - readily highlights the individual instances of "genius" which I do not want to dismiss here - but entirely blanks out how those free floating ideas were able to circulate and bounce off each other giving rise to those great insights in the "quiet" moments of those "geniuses".

Only after their recognition of "genius" they seemed to develop some really unhealthy stubbornness.

Newton - beginning in 1686 with his publication of the Principia - "disputing" any influence Hooke might have had and at his height of power letting his "Royal Society" create a comittee to decide against Leibniz "priority" regarding "calculus" in 1712).

Einstein beginning with the Bohr debate (1927 onwards) and productively contributing to the EPR paradox in 1934 (further developed by Bell to the infamous "Bell's theorem") felt more and more "misunderstood" and isolated himself to work on his own "unified field theory" and from then on did not participate in any of the "quantum mechanics" of his time.

[+] elb2020|4 years ago|reply
Judging from his book, Surely You're Joking, Mr. Feynman, he was not a very modest person himself, with regard to his own accomplishments and ideas, maybe that's why he was shocked by the modesty and restraint of other men with (presumably) similarly great talents.
[+] PoignardAzur|4 years ago|reply
An idea I'm coming around to is that the "great men" dynamic Feynman is talking about isn't actually a property of individual people, it's situational.

Like, being humble, competent in your field, mature enough to actively listen to other people, etc, all help. But there are also more subtle factors like compatibility, company culture, project parameters, etc, that are also necessary to have that dynamic where everybody cooperates in discussions.

Getting those subtle factors to align is hard, sometimes it's down to conditions you can't control, and it's definitely not as simple as "just" getting great people in a room.

[+] thenerdhead|4 years ago|reply
There is something magical when everyone leaves their ego at the door and is willing to build upon each other’s ideas like improv.

In the end, you’re left committing to only one idea and everyone is supporting it because that’s the only way it’s going to be successful. This is known intrinsically by everyone participating.

What this quote is missing is all the context. They were fighting for time and had to make a decision to get started on experiments. Many of these contributors were doing experimental work and Feynman was doing theoretical work.

Also he was 24 years old at the time.

[+] smitty1e|4 years ago|reply
My takeaway from the anecdote is that greatness != omniscience. These smart fellows brought experience and biases to difficult questions and had to have enough trust and confidence in the group to let the group function as a seminar.

The best teachers never really quit being students.

[+] Synaesthesia|4 years ago|reply
The more you learn the more ignorant you feel. It was refreshing to hear that even eg Noam Chomsky feels this way.
[+] lbriner|4 years ago|reply
I'm unsure of what is surprising to Feynman. Is he saying that he assumed clever men would be obnoxious and self-righteous when they weren't?

Although an anecdote, I would imagine there are plenty of examples where the opposite would happen to people who could still fairly be called great men/people.

In fact, some times, great people are necessarily obnoxious - or at least cold. I can't imagine that kind/warm people could have got Tesla or Apple to where they are, it takes a certain hard nose to make hard choices in business.

[+] vasco|4 years ago|reply
I interpreted it as it merely being a side point, this aspect of obnoxiousness. What Feynman was impressed it was the fact that they didn't need to repeat the ideas more than once. I think you're trying to see a deeper argument when this is the argument itself. Even in groups of smart people I've found that the group needs to write ideas down as it discusses them and repeat some of them to clarify to one person in the group that didn't get it the first time. If you don't do the writing down / repeating most people will lose track of most ideas, the same way most people can't play blind chess and remember all the pieces on the board. It's a comment on mental capacity moreso than manners.
[+] rawoke083600|4 years ago|reply
I guess there really are a spectrum of the graph if you plot: intelligence-VS-nice-guy-rating.

I was lucky enough to work with some 'really really clever people (math+physics+cs)' in my career. Bar one or two generally the nicer they were the higher their intelligence seems to be at least from my point of view.

The reverse were also true. I've also had the misfortune to work with some really 'difficult individuals' and you can spot their lack of basic knowledge + intelligence from a mile away.

Maybe it's an evolution thing ? The "dumber" you are... the more you need to be aggressive (I.e Caveman: No charm, but I got a club to seduce females + competitors) to have your genes continue.

Versus if you clever, you don't need all that aggression, you can finder clever ways to "woo" the women to carry on your genes :) ?

[+] ricardo81|4 years ago|reply
For me, it was the 'correctness being unsaid', he was thinking to himself the ideas being shared were running away from what seemed right.

Another interpretation might be that he considers this a rare trait in people.

[+] nefitty|4 years ago|reply
It's interesting to me that when I imagine the trope of an idealized executive, I imagine a cold, stonefaced person.

When I imagine a manager, my imagined ideal is a warm and conscientious person.

For individual contributors, I imagine lukewarm and single-minded. For team members, warm and cooperative.

So if most of humanity values warmth, how did executive types end up at the top of the pile? I thought that I would eventually move into executive levels when I was focused on a management career, but the managers I admired who could move their people always seemed distinct from the hardasses who could move mountains. I'm just now realizing this.

Anyways, your comment piqued my interest because of the big impact the mental model of "warm/cold" has had on my social life. I'm more compassionate of cold people because I assume that they just haven't learned that mental model. I also literally sometimes imagine myself as cozy warm to help me open up and welcome in.

[+] dguest|4 years ago|reply
There are plenty of kind / warn / "great" people who got Tesla and Apple to where they are. They aren't all figureheads, but they are there.
[+] somishere|4 years ago|reply
Honestly, my reading of this wasn't as clear cut as most of the other comments here. That said, I'm a devotee of the likes of Evelyn Waugh and Joseph Heller .. satirists both
[+] neuro_image2|4 years ago|reply
I wish there was anything close to this in healthcare.

Most medical group meetings are:

a)A small number of people that understand the issue (generally specialists) being drowned out by:

b)A large number of people with a vague understanding of the issue (related specialists and generalists)

c)A number of people with no understanding of the issue with the most political clout (administrators)

And then a final decision made by the least knowledgeable, most political administrator with the most vested interests.

[+] max_|4 years ago|reply
>It was such a shock to me to see that a committee of men could present a whole lot of ideas, each one thinking of a new facet, while remembering what the other fella said, so that, at the end, the decision is made as to which idea was the best---summing it all up---without having to say it three times.

Can we attribute this to the fact that in physics, everything can be validated mathematically? Hence less room for debate.

[+] zb|4 years ago|reply
No, because he’s clearly describing an engineering problem, not a physics problem.
[+] morelandjs|4 years ago|reply
This was the value I saw in my PhD. It has opened doors to jobs where the people I work with are brilliant and generally objective in their evaluation of competing ideas. It is an absolute treat to work in an environment like that. On many occasions, I worked with people who just knew more than me on a certain topic, and I was enthusiastic to take a back seat and learn from them.
[+] HWR_14|4 years ago|reply
I'm curious, does one really need a PhD outside academia? Does sufficient experience make up for it?