March 17, 2011

Diversity vs. Ruin in Small Groups and Elsewhere




There's another advantage that accrues to groups without hierarchy and persistent leaders: They don't all do things the same way. Perhaps this seems a strange thing to describe as an advantage? Stand by!

In 1970 there was a severe corn blight in the U.S.. Because of it's high-yield, a genetically engineered hybrid strain of corn had come to dominate 85% of U.S. corn production. Even though only the high-yield corn strain was susceptible, the disastrous potential existed to lose 85% of U.S. corn production. The problem was that as far as corn went, the U.S. agricultural industry had put most of it's eggs in one basket. Almost noone recognized the danger - - - until after the blight began to spread.

But what's that have to do with sex?

If you stop to think about it, sexual reproduction is not only not the only way to reproduce (parthenogenesis by wasps, "fission" by amobas, etc.), but very complex and chancy. Suppose you can't find a partner, for example? One wasp or amoeba can reproduce and thus keep its species going. It takes two more complex sexually reproducing ogranisms --- which includes us humans. And they have to be of opposite sexes, an appropriate age, and in the right mood! And even then, it doesn't always work.

The question of why such a relatively complex and chancy process as sexual reproduction became not only important to "nature" -- but essential for the continuation of most "higher" animal species -- stumped evolutionists until rather recently.

The suggested explanation is that sexual reproduction scrambles genes in unpredictable ways. The essence of sexual reproduction (meiosis and fertilization) then, produces variety and change in organisms -- and thus diversity -- by scrambling genes.

The Values of Diversity

But why go to all this trouble for variety and change -- while giving up the "security" that your species can survive with only one individual and/or without the necessity of mating?

One answer is, "Because other organisms, generally regarded as 'parasites' from the viewpoint of someone using that word, are perpetually zeroing in on your species, trying to exploit your specifications for their own ends." Viruses, bacteria, fungus, worms, etc. -- not to mention lions, tigers, and bears -- are all trying to feed on you and your brethren. And we on them. Like the 1970 corn blight fed on corn. Change -- and thus variation and diversity -- brought about through sexual reproduction's meiosis, fertilization, and the occasional viable mutation, makes a species a moving target for such parasites.

One explanation evolution thinkers give for the extreme increase in the complication sexual reproduction introduced into the world, then, is that it is a novelty producer -- that is, sexual reproduction, since it mixes the DNA of each individual during meiosis -- and two different individuals during fertilization -- creates diversity in the gene pool. This produces phenotypic diversity -- some people are blond, some brunette -- and there are other more substantive differences, even down to the biochemical level. Mosquitos and fleas prefer certain people over others as their feeding grounds.

And there's a second advantage to variability and diversity: Even physical world circumstances not only constantly vary from place to place -- they also change in the same place over time, remember. As a result, organisms must constantly change and adapt to these variations as well in order to survive. For the last 300,000 years or so, for example, the Sahara changes from fertile well-watered plain into a desert and back every 20,000 years or so, apparently in tune with the precession of the earth's axis. Among the diverse plants living there before the cycle to desert starts, the ones that have differences which conserve water, etc. -- such as the thicker leaves characteristic of what we call "cactus," -- gradually take over.

And there's a third value to diversity: Dfferences more radical than thicker leaves happen via viable mutations -- we could call them "genetic innovations" -- (eyes, claws, wings, etc.) which may bestow additional though, at the time of development, unknown survival advantages. An archaic giraffe, for example, could get leaves that grow higher up il the trees during a drought and thus lives to pass his longer neck on. [1]

There's also a fourth overlapping value to diversity - - -

Gambler's Ruin

Perhaps the easiest way to explain one of the main advantages of diversity is to take a quick look at what's known as "gambler's ruin." Let's suppose you have $100 dollars and you're going to take a chance with it, gamble it - - - at the roulette tables in Las Vegas.

Suppose this is a promotion at Bob Stupack's old "Vegas World" casino and the zero and double zero (they are the only green numbers usually) don't cause you to lose but only mean you tie. In other words, when the spin counts, half the time the ball will land on red and you'll win and half the time on black and you'll lose. Nothing at all will happen when the ball lands on green.

How often will you lose the whole $100 (get "ruined") if you bet it all on red on just one spin of the wheel? We'll logically call this bet-it-all strategy "double or nothing." Suppose instead you diversify and bet $50 twice? We'll call this strategy "slightly cautious." Suppose you diversify further and bet $25 four times? (Call this "normal.") $10 ten times? (Call this "conservative.")

If you bet it all on one spin, half the time you'll lose the whole $100 --- that is, half the time the ball will land on black and you'll be "ruined" (and the other half it'll land on red and you'll double your money -- "double or nothing," remember.) If you take the same $100 and bet it $50 at a time on two separate spins (the "slightly cautious" strategy), one quarter of the time you'll win both bets (red comes in twice in a row), one quarter of the time you'll win the first time (red comes up) and lose the second (black lands), one quarter of the time you'll lose the first (black) and win the second (red), and one quarter of the time black will come in twice in a row and you'll lose the whole $100. That is, one in four times on average, you'll have experienced, once again, "gambler's ruin."

In both cases ("double or nothing" and "slightly cautious") notice, you'll bet the same total amount of money over the betting cycle: $100. In the first case however (betting it all at once), you have a 50% chance of losing it all, while in the second case, you only have a 25% chance of losing it all (and being "ruined").

By "betting cycle," in this case we mean betting a total of $100 whether you do it in one bet or 100 bets.

Rather than experimenting, you can calculate the chance of this "ruin" happening in any given betting cycle in advance by multiplying the chance that a given event will occur (here, the ball landing on black -- which in this case is one out of two times or 1/2) raised to the-number-of events (in this case the number-of-spins-you-will-bet-on) power. In the first case, betting it on one spin, it's 1/2 raised to the first power -- that is, 1/2 x 1 which just equals 1/2.That is, betting "double or nothing," one out of two times the ball will land on black. Half (1/2) the time you'll lose all your money and be "ruined."

In the second case -- "slightly cautious" -- it's 1/2 raised to the number-of-spins (two in this case) or 1/2 squared and it looks like this: 1/2 x 1/2 = 1/4. That's the one-in-four times the ball lands on black on two consecutive spins and you lose both bets and your whole $100. That is, the ball lands on black twice in a row in only one out of four betting cycles. You'll only be "ruined" one in four betting cycles.

You probably already have the idea, but let's carry it through the four $25 bets and the 10 $10 bets and see if we can learn anything else. In the case of four bets the math looks like this: 1/2 x 1/2 x 1/2 x 1/2. In this case we get 1/16 or, one out of sixteen. That is, placing four $25 bets, you'd lose the whole $100 dollars only one time out of sixteen.

Let's look at the ten $10 bets, which should clarify things even further. The math is: 1/2 x 1/2 x 1/2 x 1/2 x 1/2 x 1/2 x 1/2 x 1/2 x 1/2 x 1/2 = 1/1024. Making 10 separate bets of $10 each, you would expect to lose everything only once in one-thousand- and-twenty-four or so times. That's because black would have to come up ten times in a row. Think of flipping a coin ten times and having tails come up all ten times. This is very rare indeed, and explains why spreading your bets over many events is a very sound strategy if you want to avoid "gambler's ruin."

So, with double-or-nothing, you could be ruined in one spin. With slightly cautious, it would take at least two "bad" spins in a row to ruin you. With normal, it would take at least four bad spins in a row and with conservative, it would take at least ten "bad" spins in a row to ruin you. And, of course, the odds against having that many bad spins in a row increases geometrically as the number of spins increases. Thus you have a 1-in-2 chance of ruin with "double or nothing" and only a 1-in-1024 chance of ruin with "conservative."

This effect is why "Don't put all your eggs in one basket" is usually good advice. [2] If we generalize then, the more we spread out or diversify our resources, the less likely we'll be "ruined." And increasing the number of times we can play not only decreases the risk of ruin, it does so geometrically, not just linearly. In our example, the risk of ruin went from one in two when we put all the eggs in one basket to less than one in a thousand when we spread the risk out over ten trials or events.

Now, with the above discussion and the addition of the "conservative" betting strategy, we have a specific numerical reason it's good to diversify -- and a fourth, if overlapping, reason to do it: Diversity avoids the "all the eggs in one nest" biological version of gambler's ruin. . And the possibilities in the flip of a coin or a red-or- black bet on a roulette wheel are exceedingly simple compared to the infinite variety of possibilities inherent in the nature of a universally dynamic and incredibally complex reality, remember.

So, while there may be exceptions, as a general rule and in the long run, diverstiy is how life survives the unexpected.

Organisms change, then, 1. to be a moving target to parasites, 2. to adapt to a constantly changing physical world, 3. to develop "innovations," and 4. to avoid "gambler's ruin." So from "nature's" viewpoint, change and diversity in organisms are not only good, but essential to long-term species survival. In fact, change and diversity are the driving forces behind that process known as "biological evolution" in all cases.

If the process of evolution produced such a convoluted system as sexual reproduction to produce diversity, diversity must be a very valuable addition to life's adaptation methods indeed.

But, why?

Predictability. NOT!

To start with, no one knows ahead of time what circumstance a given difference may be good for -- or when, if ever, that circumstance may occur. No one knew in advance that silly-looking long neck would pay off.

So, the over-all answer to that latest "why" is that anticipating "the future" isn't a slam dunk. In fact, just the opposite. The unexpected is always among us. As seminal Austrian economist Ludwig von Mises put it, "... to acting man the future is hidden." [3] Or in the words of baseball great (and philosopher?) Yogi Berra -- perhaps echoing seminal quantum physicist Niels Bohr -- "Prediction is very difficult, especially of the future."

And that's the bottom line on why, given the dynamic and complex nature of reality, diversity and variety are not just good, they're essential.

So, in the face of that inescapable unpredictability, variety is not only the spice of life, it preserves life. That is, in essence life copes with the unpredictibility inherent in real-world change and variety with its own potentially counteracting change and variety through genetic variation. A variation in physical make-up caused by sexual reproduction (mitosis and meiosis) -- or more rarely, chance mutation -- may enable the organism having that variation (our giraffe, for example) to survive unexpected change.

How This Applies to Us

In addition to long necks, Giraffe's also have the habit of standing on their hind legs, which gives them access to even higher leaves than just the long neck alone. Clearly then, behavioral factors are also important to survival and thus subject to the process of genetic evolution -- all those reflexes instincts and drives, remember. And differences in the genes that control these behaviors are another source of diversity. [4]

Parallel to genes, memes too have behavioral implications, remember. In the case of us humans, our capacity for variety and diversity is greatly extended because of the memetic nature of our species, remember. We don't have to wait for genetic change to change our behaviors. Often just thinking some up and/or describing them creates new behaviors. We can make it up as we go -- often using language in our thinking -- and with language and symbols, pass it around to our buddies and on to our kids.

So, in memetic organisms such as ourselves -- and mirroring the value of the change and diversity sexual reproduction causes -- it stands to reason that memetic variation -- and the resultant behavioral variation -- also contribute to life's potentially adaptive diversity. On steroids.

There's another note-worthy evolutionary survival advantage to the apparent "anarchy" by which our ancestral groups operated. As noted in the intro paragraph to this chapter, it's that without hierarchy, members of small groups don't all do things the same way. There are at least two advantages. You'll recognize both.

First, since the individuals in a human group possessed a wide variety of diverse knowledge and no one could boss anyone else, all sorts of diverse approaches to a given situation -- particularly a new situation -- were nearly guaranteed, limited only by different ideas, enthusiasm and available man-hours. Because of the "free coming and going" approach, folks discover all sorts of new resources and new hunting techniques, etc. This is clearly the ancestral version of research and development, and another example of number 3. above, that is, it serves "to develop innovations."

Clearly, this contributes to human behavioral diversity. Second, remember, as noted in Chapter 12, "all sorts of interesting, useful, but untested and thus sometimes dangerous and counterproductive behaviors sometimes become established in a human population." Because of the nature of "hierarchy," -- one dominant individual "dictating" or "commanding" one action -- this can be particularly dangerous should such a behavior be initiated or prematurely commanded by an insistent "one size fits all" hierarchical leader. Luckily, our "instincts for freedom" once again come to our defense and strongly counteract this source of "all eggs in one nest" behavior. {This sometimes keeps the whole group from being wiped out (ruined) by everyone {forced to take | taking} the same risky, dictated, chance. }

How does this apply to us and our ancestors?

If our ancestors were all coming and going just as they pleased, and had no persistent hierarchical leader demanding they all do this or that same thing, they would less often have all their eggs in one basket. They would not often bet the entire band, tribe, or group on one spin of the wheel or flip of the coin.

Just like genetically powered diversity, then, protected by our "instincts for freedom," memetic diversity helps us 1. to be a moving target to parasites, 2. to adapt to a constantly changing physical world, 3. to develop "innovations," and 4. to avoid "gambler's ruin."

Thus in the real world, more diversity is even more important. This nearly always being the case, we should nearly always avoid putting all our eggs in one nest. AND we should value and appreciate diversity -- as our ancestors did, remember.

All the Eggs, Only One Basket

Now it's true there are some circumstances when concentrating your forces may be desirable. The question is, "Will a concerted defense by all band members save more lives -- thus preserving more of the information base -- than diverse efforts by different individuals and sub-groups -- or perhaps an 'every man for himself' dispersal?" Often the world isn't kind enough to make this a clear-cut decision, but there are provisions in our genome for the "all eggs in one basket" approach. We already know a few of them from previous chapters: ancestral democracy as practiced during "emergencies," putting one person in charge in other situations such as during hunting parties and droughts, and temporarily following a confident group mate (again often during emergencies), based on the strength of his alpha confidence signals.

And, as we know, often when our ancestors did take such a concentrated risk, it involved co-Operation and the input of all group members, thus still potentially using all available (distributed) information, rather than having behavior dictated by an alpha-dictator with limited information.

Functional Hierarchy

So, despite the fact that our ancestors kept hierarchical behavior amongst themselves to a minimum, we can see that certain aspects of hierarchy were, none-the-less, useful under certain conditions. Under those conditions, those "certain aspects" of hierarchy were part of "How They Got Things Done."

So while generalized hierarchy was not acceptable in our ancestral groups, it still had perhaps five limited functions in those groups: 1. Maximum intimidating displays may repell enemies and attacking animals, 2. hierarchy may be useful in training and controlling the young, 3. hierarchy and domination may work in sexual bonding, 4. to temporarily coordinate people during "emergencies" and, 5. moderated alpha-complex dominance signals may be used to signal who is most confident of their knowledge and information -- which allows temporary control to shift easily and seamlessly from person to person as is appropriate based on each person's subconscious, honest, self- evaluation of his knowledge and information -- thus greatly facilitating co-Operation.

If hierarchy and domination were only tolerated for five limited purposes by our ancestors, echos of earlier questions arise: "Have The Great Transitions from our ancestors' egalitarian societies to today's hierarchical ones led to necessary -- or even useful -- hierarchical cultural artifacts in the modern world?" And, "If so, what are they?"

Our Instinct(s) for Freedom: A Summation

Given that the notion of our "instincts for freedom" wasn't widely accepted as proven when I originally wrote The HI-JACKING of Civilization," I feel it's a good idea to sum-up the arguments supporting their existence. This is the place I've chosen to do that. If you're already completely convinced they exist, you can safely skip this section.

In Chapter 11, An Instinct(s) for Freedom, we noted that we humans entertain not only tendencies toward hierarchy but also opposing anti-hierarchy and anti-authority egalitarian "instincts for freedom" within our genome. This apparent genetic contradiction prompted a few questions, particularly:

"What was the survival value to our ancestors of our instincts for freedom?"
"What could possibly be more important to survival than the 'physical strength, dominance, and warfare' associated with hierarchy?" And, in contrast,
"Why would having your group dominated by an alpha hurt your -- and his -- chances of survival compared to groups which didn't permit alpha domination?"

By way of answers to these questions, in Chapter 13, we tallied-up the value of our instincts for freedom -- including our "problems with authority" -- this way:

While altruism protects our unique distributed memetic information from being lost, our instincts for freedom -- and other anti-hierarchy anti-authority genetic predispositions -- protect its use from being blocked or degraded by hierarchy and authority.This safe-guards our behavioral flexibility. ...this alone is probably enough to explain our instinctive aversion to both hierarchy and "authority."

[THERE MAY BE AN ADVANTAGE TO CENTRALIZED HIERARCHICAL CONTROL WHEN YOU'RE STEALING STUFF THAT'S ALREADY BEEN PRODUCED BY OTHERS, BUT FOR PRODUCING STUFF IN THE FIRST PLACE -- AND PARTICULARLY FOR INNOVATION -- IT'S DEMONSTRABLY COUNTERPRODUCTIVE.]

Further, in Chapter 15, we recognized that as long as our instincts for freedom, etc. prevented persistent hierarchical "leaders" from dominating everyone with constant "alpha-confidence" signalling, our ancestors could use that signalling to make efficient moment to moment use of their distributed information to make distributed decisions, resulting in true Co-Operation with all the information-use advantages that implies.

Then in Chapter 16, we discovered that our instincts for freedom were strong enough to keep our ancestors egalitarian, and thus un-enslaved by "winner-take-all" democracy, and, at least until the intrusion of U.S. Government "Indian Agents," etc., free of insistent hierarchical "chiefs" as well. This enabled these ancestors to avoid "centralized political authority" so they could use their distributed information to make better longer-range decisions in meetings such as tribal councils etc. Thus, rather than permitting the hierarchically controlled and dictated behavior typical of "centralized political authority," our instincts for freedom allowed our ancestors to engage in co-Operation and consensus, giving them the 5X advantages strongly implied by Axelrod's simulations and Bob Waldrop's experience.

In this chapter we pointed out additional advantages of avoiding domination by persistent hierarchical "leaders:" Without a persistent leader regularly "dictating" or "commanding" one action, no one would habitually be causing everyone to chance "gambler's ruin" by constantly putting all their eggs in one behavioral basket, often based on limited information and knowledge -- or, simply, for the pure love of domination and hierarchy. And there's the related value: While obviating "all the eggs in one basket" by promoting diversity of behavior, this approach also facilitates ancestral research and development.

So, with this chapter, we can add the advantages of diversity of behavior to our list explaining the increased survival value of egalitarianism and our "instincts for freedom" over hierarchy -- and two more reasons (increased R&D and avoiding "gambler's ruin") alpha-dominance would have been detrimental to our information-intensive ancestors. There may, of course, be other reasons as well, but, if so, that's in your hands now.

I believe these likely explanations are quite adaquate to explain the practical reasons why the process of evolution produced ancestors who were anti-hierarchical, opposed to persistent leaders, "all but obsessively--concerned with being free from the authority of others," why they so dearly cherished "independence of thought and action" -- and thus where our "instincts for freedom" came from and why we have them. Thus it was that early 16th century "urban literate observers" of the New World often discovered that "every individual seemed to come and go just as he or she pleased."


Summary

In this chapter, we suggested another advantage our ancestors derived from being anti-hierarchical: They didn't all do things the same way. This means diversity, and we pointed out several basic ways that "nature" builds such diversity into plants and animals, particularly by the rather radical process of sexual reproduction (thru meiosis, fertilization and an occasional viable mutation). All inherently scramble genes and thus the characteristics they encode, producing diversity.

We pointed out that one advantage to such diversity is that it makes us moving targets for parasites.

A second, that it facilitates adaptation to physical world circumstances which not only constantly vary from place to place -- but also change in the same place over time, remember. The Sahara's 20,000 year cycle from lush forests to parching deserts, for example.

A third advantage of diversity is that it results in organisms innovating -- eyes, wings, or claws, etc. Or the long neck of the giraffe - - -

AND, there is a fourth overlapping value to diversity. We demonstrated the up-side of diversity (and the down-side of the lack of it) using the model of placing roulette bets of the same total value but placed in different ways. We discovered that the fewer the number of bets we placed (and thus the greater part of our bank roll risked on each bet) -- meaning less diversity -- the greater the probability of losing everything. We called that unfortunate eventuality "gambler's ruin," and discovered, conversely, that the advantages of spreading risk over a greater number of events -- more diversity -- increase geometrically with the number of events over which the risk is spread: If you bet it all on one spin, you have a 50 - 50 chance of gambler's ruin, while if you spread it over 10 spins, you only have a one in a thousand or so chance of ruin.

This is why "Don't put all your eggs in one basket" is nearly always good advice.

As a result, we observed from "nature's" viewpoint that change and diversity in organisms are not only good, but essential to long-term species survival. This helps explain why our ancestors not all doing things the same way was a survival advantage.

As we noted, the unexpected is always among us, and as Ludwig von Mises put it, "... to acting man the future is hidden." Or in the words of Yogi Berra, perhaps echoing quantum physicist Neils Bohr, "Prediction is very difficult, especially of the future." This observation is the ultimate reason variety is not only the spice of life, over time it is necessary to preserve life. We used the evolution of the long neck of the giraffe as an example of the advantages of diversity.

We pointed out that we have the added advantage of the diversity inherent in our "memetic" nature, which enables us, using language and symbols, to not only rapidly create new and innovative behavior, but to pass innovated behavior and diverse knowledge around to our buddies and on to our kids.

Next we attempted to answer the questions raised in Chapter 11, explaining why, despite the survival advantages of hierarchies in other species (it tends to make them more agressive, etc.), we, with our instincts for freedom, are anti-hierarchical. The reasons developed before this chapter all had to do with preventing hierarchical behavior from interfering with the distributed decision making made optimum by our distributed information. In the case of moment-to-moment co-Operation, this is accomplished by keeping persistent leaders from monopolizing control while in the case of "consensus seeking," it involves harnessing distributed information, though in a more leisurely fashion, using methods such as Quaker Process.

In this chapter we added the advantages of increased diversity of behavior which resulted when one hierarchical leader wasn't regularly calling the shots and thus "every individual seemed to come and go just as he or she pleased." This situation lessens the probability of everyone doing things the same way and thus greatly reduces the probability of a behavioral "gambler's ruin." It also facilitates "ancestral research and development."

We pointed out that while there are circumstances where putting all the ancestral eggs in one nest may have been desirable -- in particular we mentioned ancestral democracy as practiced during "emergencies," putting one person in charge during hunting parties or droughts, and momentarily following a confident group mate -- the world isn't always kind enough to make this a clear-cut decision.

Next we noted that despite the downside, there were five practical uses for hierarchy in ancestral groups: To repell enemies, control and train children, for mating, to temporarily coordinate people during "emergencies," and finally and most importantly thru alpha confidence signalling, as the basis of co-operation.

Finally -- echoing earlier similar questions -- we asked, "Have The Great Transitions from our ancestral egalitarian societies to today's hierarchical ones led to necessary -- or even useful -- cultural artifacts in the modern world?" And, "If so, what are they?"


NOTES:

[1] The differences between what we call "mutations" -- and normal sexually related diversity -- become blurred where they meet in the middle. Were the thicker leaves inspired by the Sahara mutations -- or just the result of "normal" sexually related genetic variation? Ditto the giraffe's longer neck, especially in the early stages of its evolution. return

[2] Just recently (January, 2002) the Enron bankruptcy, the largest in history thus far, spot-lighted the dangers of putting all the "bets" in your 401k in just one company, usually your own. Sixty percent of Enron employees' 401K retirement money was invested in Enron stock. Studies have shown that such concentration of capital is a very common and ill-conceived practice by today's "investors." return

[3] -Ludwig von Mises, Human Action (Chicago, Illinois: Contemporary Books, Inc. 1966), pg. 105 return

[4] Behavior itself happened because it was a clear survival advantage. A tree has to stay and take it; a fox on the other hand, can flee the fire. return


Total Contents
In Clips
Summaries
Appendices
L's Articles