Tuesday, February 28. 2006
I just watched a documentary entitled Enron: The Smartest Guys in the Room. This documentary was fascinating, especially from an evolutionary point of view. It provides a great cautionary tale illustrating at least one of the reasons why compulsory eugenics doesn't work.
One of my favorite papers in evolutionary biology, which I have mentioned here before, is this:
Muir, W.M., and D.L. Liggett, 1995a. Group selection for adaptation to multiple-hen cages: selection program and responses. Poultry Sci. 74: s1:101
It outlines the group selection effects observed when trying to breed chickens for increased egg production in multiple-hen cage environments. In short, selecting individual chickens for increased productivity in a group environment didn't select for increased productivity. Instead, it selected for mean chickens. The result was an overall reduction in productivity. Only by selecting at the group level was productivity increased.
This is a great experiment because it illustrates why evolutionary theory cannot be reduced to the phrase "survival of the fittest." That phrase isn't technically wrong, but it neglects so much that it might as well be. "Survival of the fittest" is either meaningless or misleading. It's like saying that mountain climbing is just "walking upward" while neglecting to discuss proper supplies, fitness training, establishment of base camps, selecting the proper climbing group, atmospheric oxygen considerations, and... the fact that you don't always walk upward. Sometimes you have to walk sideways, or downward, to get to the top.
So how does this chicken paper relate to Enron? Well, it turns out that Enron sorta reproduced this experiment through their corporate human resources policies. (Are you shuddering yet?)
Apparently one of the Enron CEOs was a big fan of Richard Dawkins' book The Selfish Gene. He took Dawkins' (in my opinion) overly reductionistic view of evolution and proceeded to even further reduce it in his own mind to social Darwinism of the knuckle-dragging "survival of the fittest" (grunt, grunt) variety. Enron's HR policy included an iterative performance evaluation and firing step reminiscient of a reality TV show like Survivor or The Weakest Link. Basically, they would evaluate the traders and most other employees based on performance metrics and then fire the lowest 10-15% of the company population.
Think chickens and trading floors folks. Enron was a trading, brokering, and investment company. (Go ahead, shudder some more.)
Everyone knows that there are many things you can do in any corporate environment to give the appearance and impression of being productive. Enron's corporate environment was particularly conductive to this: it's principal business was energy trading, and it had large densely populated trading floors peopled by high-powered traders that would sit and play the markets all day. There were, I'm sure, many things that a trader could do to up his performance numbers, either by cheating or by gaming the system. This gaming of the system probably included gaming his fellow traders, many of whom were close enough to rub elbows with.
So Enron was applying selection at the individual level according to metrics like individual trading performance to a group system whose performance was, like the henhouses, an emergent property of group dynamics as well as a result of individual fitness. The result was more or less the same. Instead of increasing overall productivity, they got mean chickens and actual productivity declined. They were selecting for traits like aggressiveness, sociopathic tendencies, and dishonesty.
After a couple rounds of this selection experiment, these mean chickens could be heard on recorded intra-office phone communications laughing about "those poor grandmothers" they were ripping off via market scams. They changed the company motto internally from "Enron: Ask Why?" to "Enron: Ask Why, Asshole."
Of course, everyone knows the rest of the story. While these mean chickens weren't terribly productive (the company was losing money hand over fist), they managed to peck their trading consoles so as to give the impression of increasing productivity. This worked, for a while. Then this whole monument to Darwinian fundamentalism collapsed rather spectacularly.
I propose that the most significant root cause of Enron's collapse was this HR policy. Maybe CEOs everywhere should read more about poultry science?
This is one of several reasons that I am aware of why compulsory eugenics programs do not and cannot work. I'm sure the same thing would have eventually happened to Nazi Germany: a master race of mean, lazy, sociopathic Aryan supermen.
If you are curious about other reasons why compulsory eugenics won't work, here's another one. Evolution is a process that adapts to the environment through direct empirical sampling of fitness. Now consider what would happen if empirically determined fitness, with all it's multidimensional complexity and subtilety, were replaced by a fitness evaluation performed by agents within the evolving system itself. Think the fallacy of self-referential criteria as applied to fitness determination. "It is fit because the fit say it's fit" is a recipe for gradual population "devolution" through random drift and mutation load. The end result of self-referential fitness evaluation would be to spiral away from reality. Something like this could be seen in Enron as well. Think about Enron's accounting practices and self-referential criteria.
So now you know why eugenics doesn't work, asshole.
Tangled Bank #48
Welcome to Tangled Bank #48! Tons of good stuff to share with you today. I considered a number of themes, including the invisible theme (aka, none at all), but decided on "songs by groups that may have been in my...
Tracked: Mar 01, 11:00
Enron, chickens, and the problem with eugenics: or why self-referential fitness evaluation doesn't work
Enron, chickens, and the problem with eugenics: or why self-referential fitness evaluation doesn't work
Weblog: discarded lies - hyperlinkopotamus
Tracked: May 15, 22:45
Midday Line, May 17
Go for the win in 2006: Townhall's Lorie Byrd and The Anchoress are of one mind on whether or not conservatives should blow up the Republicans in 2006. Sociologist Frank FurediÂ pulls back the curtain on the politics of happiness at...
Weblog: Pajamas Media
Tracked: May 17, 14:14
Display comments as (Linear | Threaded)
Great stuff, asshole!
I fell off my chair laughing!
But, 'tis no laughing matter...really!
It's really interesting that eugenic-type thinking played a role in the Enron disaster. Eugenics is basically evolutionary/population genetic quackery.
I think nazi Germany also ended for darwinian reasons. They made an environment which was so bad that people who were creative (scientist) just left, and the bomb was built somwhere else. If Germany was not invaded, it would still have lost. End for purely "survival of those that attracts the fittest" reasons.
Jews are the most racist community in the world. They don't allow other people to join their faith or group easily. They don't mix with any community among which they live, if this isn't racism then what is?
Ever wonder why the Federal Reserve Chairman is always a Jew, when jews are only 3% of American population. What happened in spain, italy, uk, germany (jews were kicked out for appropriating wealth disproportional to their numbers) is going to repeat soon in usa also isn't it? Wealth is a zero sum game, when somone take more others has less!
How does small group of people appropriate wealth disproportionate to their numbers?
1. Control currency
2. Control banking
3. Control media
4. Control education
5. Subvert democracy
I know your just trolling... but you should do some research... wealth is not a zero sum game. Wealth can be created and destroyed. Also if you've figured out how the Jews control all the wealth why don't you do it yourself and stop being a pest.
While in a strictly classical economic sense, wealth is not a zero-sum game, in an important sense it is, because trying to acheive relative wealth is a zero-sum game and in an advanced society, relative wealth is what matters. It has been shown that above a certain low level, the wealth of a society is not correlated to average happiness, but relative wealth is strongly correlated to happiness. Sorry, I don't have the reference off the top of my head.
Bingo. "Fitness" is always dependant on the specific environment, not on some godlike vision of "perfection". And human beings have the remarable ability to actually change and create our own environment.
Create an environment by assholes for assholes, and you breed assholes.
Yup. You've hit the nail right on the head.
Manual Trackback: http://www.conceptualguerilla.com/showthread.php?id=218
But, couldn't all this simply boil down to the fact that they selected for the wrong traits?
Even in the chicken experiment, you state that they did not select for productivity. Instead they (inadvertently) selected for mean chickens.
The same could be said of the Enron scenario. They might have thought they were selecting for productivity but actually weren't.
The real lesson to take away from these two stories (chickens and Enron) is that it is difficult to define a selection function. (A well accepted fact in evolutionary computing.) Granted, there are complexities involved in whether an agent (within the environment) can accurately define a selection function given its myopic perspective; but this anecdotal evidence does not prove that an agent (or set of agents) cannot accurately define and implement a selection function.
A proof would have to be along the lines of your "fallacy of self-referential criteria", showing that implicit selection functions somehow must be inadequate. But, I'll take the opportunity to remind you that impredicative definitions are not really a problem in most cases. Similarly, a self-referential (or co-evolutionary) selection function is probably not a problem... at least not a big enough problem to prove that eugenics doesn't work. At best, an argument like that would show that there are paradoxes in efforts like eugenics. But, it doesn't mean that those paradoxes are unresolvable, especially through a tranform to some other representation or formalism.
Unfortunately, even with a theoretical proof showing that eugenics was problematic, we'd still have ignorant narcissists (a.k.a. "hicks") who'll believe in eugenics because it suits them to believe it. And the only way to combat that is through an accumulation of eugenics case studies like that of the chickens and Enron.
Hmm... haven't read this post in a while, and Glen made some interesting points I'd like to respond to.
First of all, I think they did select for performance in the egg laying experiment. They were selecting at the level of individual chickens on the basis of how many eggs they produced. They inadvertently selected for mean chickens because mean chickens, it turned out, were better at obtaining food and therefore produced more eggs. However, an entire cage of mean chickens produced less eggs as they caused a high level of stress and in some cases harm to one another.
Yes, selection functions are challenging. I've been getting a taste of this myself recently. There's definitely a fine art to it.
I agree that there has been no rigorous proof that eugenics could never work. However, I am intensely skeptical of it for the reasons outlined above, and I would predict that if someone dedicated a lot of intelligence to examining this problem they'd probably be able to produce such a proof. I'll say that the fundamental infeasibility of eugenics is a hypothesis that I would strongly put forward.
I think you will however agree that even if there were some theoretical way to make compulsory eugenics work that a real human political system would be very unlikely to be this wise or subtile. Thus, eugenics would almost certainly always fail in practice.
It's an interesting comment that selecting the bottom 10% can have unintended consequences. That may have indeed hurt Enron. But the reason Enron failed did not have to do with lack of profits in energy trading. That was the place they were making all their money.
Enron went under because of greed that led them to attempt to duplicate their huge energy trading success in areas they were not expert in, like trading electricity, building power plants, and acquiring and running new businesses.
Perhaps the eugenics factor made it worse. But any company that went as far off the reservation as Enron did is asking for disaster.
Unfortunately, eugenics is still one of the most ridiculous and extended way of thinking of many people. Simply ask whether they would prefer to get rid of anything that doesn't suit their concept of "normal", and since for most human beings a normal human being is the summit of evolution, the non-normal (unproductive, coloured, uncoloured, religious, areligious, ...) is purely debris. The only of way of showing this people their stupidity is with sound science, so thanks for the post.
Brilliant. Thanks for posting this. Great movie, great interpretation.
One way to think of would-be eugenicists is that they're a hazard to the general population which it will deal with using ordinary mechanisms (such as defeating a bunch of them in WWII).
Any population will include power structures which select (e.g. reward or punish) their subjects based on various arbitrary criteria. E.g. in a football team, more gifted athletes get more important positions, play more, and presumably get paid more and have sex with more partners. Standing outside and looking in, Eugenicists are just a power structure with a specific -- essentially religious -- agenda, and no more or less likely to succeed and/or be dangerous than many others.
You know what's funny? That when you're describing self-referential fitness selection, you're precisely describing sexual selection. Nice one.
Do you really think the peacock is fit? I rather think not. What about testosterone poisoned macho cavemen? Well, I think not.
And on the flip side, do you really think those mean son of a bitch sociopathic bull walruses are unfit? Nature disagrees.
The problem with eugenics isn't that humans reduce the wonderful complexity of nature. No, the problem with eugenics is that evolution sucks and nature sucks hard.
Nowhere, in Origin of the Species, is the concept of 'Survival of the Fittest' put forward. It is, and always has been, 'Survival of the Fit', which is a much simpler, less exclusionary idea. Thus, while I agree with some of what you say, the entire thing -- especially the "proof" against "Darwinian Fundamentalism" -- is silly.
Hmm... gets posted on Reddit, quality of replies goes down...
It's pretty important to separate Social Darwinism from the actual Theory of Evolution. One is hard science, and the other is pseudoscientific social policy that allows the rich to claim superior fitness over others, often hiding racism, classism, cronyism, and a bunch of other -isms.
If by eugenics you refer to selecting based on identifiable traits (phenotype) in the hopes of obtaining pure genetic stock (genotype) for reproduction, then I think it is safe to say that, short of mapping every trait to their respective genes, eugenics has long been dead. First, most traits having an underlying genetic component have multiple genes (multiple loci). Second, most genes come in different "flavours" (alleles) that are not just GOOD gene / BAD gene, but contribute in a graded fashion to the phenotype. Not even eye colour, a simple, obvious PHYSICAL trait is black and white (or brown and blue). The complexity involved in complex psychological "traits" with underlying genetic components is truly astounding. As a final nail in the coffin, to have truly pure stock, you need to reduce the population to single alleles. Since there are two copies of each gene, the multiple alleles for each gene are distributed throughout the population such that almost all people have two different alleles of particular genes. If you don't kill them all, the "unwanted alleles" survive through sexual reproduction and spread back through the population. And all of this is on top of the fact that the traits you are selecting for may be linked, as in the egg laying, to non-desirable traits. And these will differ for each individual as well. Oh, and the belief that all traits are solely genetic is a lost cause. There is even a specific term for how "genetic" a trait appears to be: genetic penetrance.
This really isn't anything new. Companies have been doing this for years with their sales forces. There once was a time the concept of a "Gentleman Saleman" wasn't thought of nostalgicly. However, decreasing profit margins and tougher competition has lead to the decline of the salesperson who actually cares about filling a customers needs and brought us "Enron-esque" assholes who would stab their children in the eye if it meant they'd meet quota this quarter. Enron took it to a whole new level, granted, by clearly defining this as their goal. But most companies practice this to some level, perhaps not consciously.
Even worse is what I've noticed and described as "layoff evolution". When a company starts to do poorly, the rumors of a round of layoffs are not too far behind. If a company has a history of jettisoning their staff in order to save money, the rumors become thought of as fact. What happens next? The smarter people find other jobs before they're forced out, leaving just the people who were too stupid or not skilled enough to leave. Worse, the company will then still go through with the layoffs, leaving these people to do the work of two or more people. The downward spiral begins. All the while, the people truely responsible for the failures of the company (the managers) are typically left in place to continue to make the same bad decisions over and over again. And people wonder why I'm so bitter. :)
Doesn't the very existence of so many different dog breeds prove eugenics/selective breeding can work?
But all of these dog breeds have so many weaknesses as well that mongrels ODN'T have. Bad backs, high cataract rate, high blood pressure, etc. etc.
Pure bred dogs are a good example of why selective breeding is a failure. Any good trait of induvidual breeds would be considered a fault in other breeds.
(English Bull Dog comes to mind) Some of the pure bred dogs are not capable of breeding without human assistance.
The collie breed has genetic faults that produce eye problems...some terrier breeds will eat their new born pups....displastic hips are endemic in large breeds...deafness recessives are common in Boxers...bad backs, bad eyes, bad temperments, bad feet...bad bites...defective jaws and breathing problems, and so on and on....
These are not traits of a good proof in favor of eugenics or selective breeding.
What I find so refreshing about this essay is the respect given to HR policy in forming the character of a company. I think that's a point many people studying organizational behavior tend to dismiss. Kudos on bringing it out so brilliantly.
Seems like the best way to select for the right traits is not to select an individual based on that individual's traits. Behavior is group-oriented and emergent, so you must "break the barrier" between individuals in order to gain the correct metric.
Do not select a chicken based on how well that chicken lays.
Select for the chickens around whom all the other chickens lay more.
Select Chicken A based on the behaviors of all chickens not-A.
If you're selecting for group traits and emergent behavior, that's all you can do. That one thing alone would select for all the traits you want, and select against the meanness and sociopathy that surface when you select Chicken A based on Chicken A's performance.
To bind individuals into a group, select across the boundary between individuals. Poof! Done.
I think you capture the important point of observing the benefits of the surrounding chickens, but you cannot so easily discount the individual either. Chicken A itself is integral to the system.
For example: If you were indeed to "Select Chicken A based on the behaviors of all chickens not-A" then you might have the opposite effect of what happened in the paper cited. All the chickens would be passive and could theoretically not care if they get food or not. Thus producing less eggs.
I think the whole point behind Adam's post is that the dynamics of the system can't be broken down into a simple answer that can accordingly be maximized/taken advantage of.
As a side note there is some interesting work being done in some corporations along the lines of this type of thinking. I read about just such a lab at HP in Scientific American May 2006. Forget the researcher's name.
absolutely brilliant. very simple and correct links made from Dawkins to Enron.
Sounds like it was more of a Coup than a Coop!
At Enron many middle-managers and executives ( VP Sherron Watkins ) had serious doubts about Enron's direction. Furthemore, at least one top executive - Rebecca Mark - had a very public clash with Skilling that led to her departure. She is on record about being very uncomfortable with Skilling.
The fact that many employees raised concerns internally and later went to the regulators and also sought counsel is a contradiction of the chicken thesis.
Both Watkins and Marks had decades at Enron as compared to Skilling who worked at McKinsey & Co and did not go to Enron until 1990.
The chicken experiment itself is flawed on several levels. Its held captive to its initial conditions and design biases.
Seems to me that there is a relatively simple dynamical systems theorem hidden in your observations. Simply put: any forcing function that favors a single variable X, to the extent that X feeds positively back upon itself and becomes asymptotically exponential, will lead to a collapse of the entire system (or at least several other "local" variables). In other words, if you stack the cards to favor a single variable in a particularly bad way, you will endanger the health of the entire system.
I have no idea if such a theorem has been proven, or how to prove it (after defining things properly) if it has not.
This fundamental flaw was inevitably adopted with the acceptance of evolution as a scientific theorem. This is not to say that Darwins work is not worthy of more, even than that, rather that science itself's inextricability from society may cumulatively render the entire persuation irrelevent.
I think that one important fact that your overlooking is that survival is not as much due to fitness as it a concrete probablity given all the variables in a situation. For example, since humans are highly social we are able to enhance our survival by our aggregate effect, or reduce our survival by our aggregate effect via nuclear war or pollution type scenarios. If we only took one human based on his intelligence to survive in the wilderness naked, an earthworm stands a better chance of survival. The point is very valid about using individual benchmarks not being as reliable as using group benchmarks. You can see the same effect when sports teams in a championship series, where the team of superstars are beaten by the team without superstarts that plays very well. The human social construct is far too complicated to make accurate predictions of group performance, ie an entire company, based soley on individual performance criteria. There are a variety of group work situations that require vastly different management structure purely based on the personality of the group, and even though management styles differ the output is nearly equal. I/O psychology is much better at predicting success than fad selection criteria.
The acceptance of the theory of evolution over creationism has had some nasty side effects.
Sadly, the first widely accepted 'understanding' of evolution is that if we were not created directly by the hand of GOD then somehow we are nothing more than animals with no purpose or moral restrictions.
However if you read this:
it will argue that while genetic evolution may be a 'survival of the fittest', of sorts, SOCIAL evolution is NOT.
And you've proved it once again by showing how a 'survival of the fittest' HR policy did nothing more than create a society of 'mean chickens' that were not more productive.
The greatest social evolution happens when great men and women open our eyes and show us that truth, freedom, and fairness are needed by all. When slavery ended, when the poor could vote, when women could vote, and when blacks could vote...we evolved. Those moments are remembered with fondness and pride by us all.
And how do we remember the Vietnam war, and how will we remember this Iraq war?
Did we evolve from it?
No. We divided the country by hating the men and women that were willing to fight and die for us simply because their cause, which was not theirs to design, was one of consolidating power and not improving society. Now we have a constitution that has been overruled, and a strong division in the entire world. Our ideals of a democracy are being replaced with that of an elitist society where only the rich and ruthless have the right to move the world.
We have seen this before with the royal family and nobility that existed in Europe for centuries. We are falling back to that because the average citizen has become inwardly focused, not attentive of how each individual is responsible for, not a victim of, this day and age. We are replacing love with fear, knowledge with fanaticism, and truth with deception.
And if we really decide that the truth is not as important as the fantasy, then God help us, and our children. Help us, because forgiveness is not what we need...this is the future of all mankind that we are destroying, and there is no forgiving when there is nothing left at all.
Blogs We Like
Syndicate This Blog