"Freddie, what's wrong?"

"The world's turning upside down on me.  I don't know what to think."

"Is it your Sociobiology 101 again?"

"Yeah.  My term paper.  I keep getting all the wrong answers.  I turned it in this morning, and I'm not happy with the way it turned out.  I thought it was a smart idea to go for a computer simulation of population fluctuations simply because I'm a computer major."

"You mean you didn't you write the perfect program?"

"It's not that.  I kept getting opposite answers."

"Opposite from what?"

"Opposite from the right answers!"

I felt sorry for Freddie.  He almost never gets things wrong.  He's not used to failure.  I offered to treat him to lunch so he could explain what happened.  It turned out to be a long lunch.  I became engrossed, and forgot about the time.  We must have talked for several hours.

It seems he started out with a simple model for something called "mutational load," which a Soviet scientist described in a 1988 Nature article.  He chose a Monte Carlo simulation, which he found to be very effective in many previous situations.  IQ was used as a "dummy parameter" to stand for anything that was heritable and affected survival.

One of his first results is that offspring average IQ must be about 2 points lower than their parent's.  But those who survive a "culling" effect, in which it was typical for 75% to not survive to reproductive age, exhibit IQ averages the same as their parents.
That seemed straightforward, though it seemed odd that nobody had previously called attention to such an interesting phenomenon.

The Monte Carlo simulations connected parental generation IQ to something called "genomic IQ."  Each generation of surviving offspring had an average IQ that was intermediate between that of the parental generation and that of previous generations.  Low IQ parents tended to have offspring brighter than themselves, while high IQ parents tended to have offspring less bright than themselves.  The offspring IQ tended to "spring back" toward a preferred "genome IQ," which changed very slowly with time.  This effect was included in the Monte Carlo model.

Freddie's sociobiology professor suggested that the model include a relationship between the survival rate of offspring versus the population's per capita accumulated wealth, such that wealthy societies afforded better survival rates.  Birth rate was related to per capita wealth in a similar manner.  This was modeled using the standard relation having decreasing birth rate with increasing wealth.

One key hurdle was to express a population's change in accumulated wealth during a given generation.  A power law was finally chosen, which related the wealth change to the population's average IQ and their per capita wealth.

Freddie's model simulations showed very little to generate concern, at this point.  The genomic IQ exhibited slow excursions of typically 10 points with timescales of about 2000 years.  He made several runs to deduce the feedback forces, and concluded the obvious:  "low IQ populations" could not sustain "high per capita wealth societies," while "high IQ populations" could not sustain high population levels (because their low birth rates were not sufficiently compensated by their higher survival rates).  The reassuring thing about these runs is that IQ movements did not go forever in any one direction, but oscillated with periods of roughly 2000 years.

"That's great, Freddie!" I exclaimed, trying to cheer him up.  "You've shown another way to account for the spacing of the golden eras of the Greek/Roman civilization and the Renaissance!  And also the Minoan in relation to the Greek, perhaps."  He didn't seem impressed by this attempt at congratulation.  Instead, he was glum.

"Wait til you've heard my next simulation; it's not so wonderful!"  And he proceeded to explain what the professor had suggested for the next model improvement.

The unsatisfied professor stated that populations that are Gaussian for one trait may be non-Gaussian for another.  What this meant, for the model, is that Freddie had to account for the fact that the higher IQ parents within a generation might be more successful in assuring the survival of their offspring than were the lower IQ ones.  In other words, the model had to treat offspring survival rate as differentiated within a generation such that individual parental memberships adhered to the specially specified survival rate relations.

Moreover, the same differentiation within a generation had to be applied to the birth rates.  So, during any given generation time-step, Freddie had to take into account how many offspring each parent combination produced, and how many of them survived to adulthood.  Freddie's a computer program major, so this was easy.  No problem!

The new model simulations had different dynamics.  Population oscillations had shorter timescales.  Probably due to the Monte Carlo paradigm there didn't seem to be any predominant period of oscillation.  Dramatic changes in population IQ, typically 10 points, occurred in 200 years!

This, by itself, could have been disturbing.  But what seemed to really bother Freddie was the sign of the correlation of IQ changes with population, and with per capita wealth.  In every instance IQ rose when populations fell, which also coincided with per capita wealth values that worsened.  The converse was true:  when populations rose, and per capita wealth increased, IQ declined!

"So why does this bother you, Freddie?  Couldn't you have predicted that from the model assumptions?"

"Yeah, the model did what it had to.  But think of it!  When times are good, and getting better, the people were getting dumber!  That's wrong!  I don't care if the computer thinks that's right!"

"Well maybe you left something out of the model that might 'correct' that problem," I suggested.  "Everyone knows that world models are notorious for leaving out dynamically important factors."  Maybe I said the wrong wrong thing.

"It gets worse!  Save your bright ideas for when I really need them!"

Freddie proceeded to describe what the professor suggested next.  It had to do with the tendency for humans to be polygamous, as borne out by a preponderance of polygamous primitive societies.  Somehow, polygamy had to be incorporated into the model in a way that accounted for its influence on fitness, as represented by the dummy parameter IQ.  That was a harder challenge for Freddie, for it took him further afield from computer science than any of the previous model representations.  But since this was a class in sociobiology, it was an appropriate exercise, and since the professor thought Freddie might actually learn something from it, poor Freddie was on his own to figure out a solution.

The first thing he did was to create another independent variable, because he wanted to retain IQ as a collection of things related to only intelligence.  He invented MQ, or Monogamy Quotient.  High MQ meant there was a strong tendency to be monogamous, etc.  At the professor's suggestion he began keeping track of the fate of genes for MQ and IQ separately.  Thus, individuals became carriers for gene alleles; and the model focused on the fate of the gene alleles rather than individuals.  However, the "user" of the model still monitored the aggregate properties of the population, such as total population, average IQ, average MQ, per capita wealth, etc.  In this sense the model user viewed model results in the same way, but the model was achieving them with a more detailed representation of properties below the aggregate level.

The first experimental runs were used to impose different initial values of MQ, in order to study population behavior.  Nothing interesting happened.  Populations fluctuated as before, in similar accordance with different combinations of initial conditions for societal wealth, IQ, etc.

The professor came to Freddie's help at this point, and suggested a bold new improvement to the model.  Since it was a Monte Carlo model, it would be easy to add the realistic property of uncorrelated inheritance of alleles at different gene loci.  For example, a specific individual could be specified to inherit the gene for high MQ as well as any hypothetical mutation at the site of any other gene.  To make this model upgrade work it was necessary to explicitly specify the presence of a large number of genes.

Genes were hypothesized to exist that coded for such arbitrary qualities as skin "color," body type, and others, which collectively came to be referred to as "anatomy genes."  Other genes were hypothesized for such qualities as blood type, metabolism type, immune system function, which came to be referred to as "physiology genes."

The professor insisted that IQ be sub-categorized, since IQ is basically a behavioral descriptor, and other behavioral properties must be explicitly incorporated in the model.  MQ was an example of a behavioral descriptor.  It was fun devising new behavioral properties to ascribe to genes, and Freddie and the professor spent a lot of time sorting through the candidates.  A final set had to be selected which was manageably small in number, yet large enough to represent important hypothetical contributors to population fluctuations.  This set of genes came to be referred to as "behavior genes."  IQ was just one of the behavior genes, as was MQ.

Having selected a half-dozen anatomy genes, another half-dozen physiology genes, and a dozen behavior genes, it was necessary to devise realistic interrelationships.  For example, skin color could be related to sunlight exposure, which in turn was related to geographical location (i.e., latitude and cloudiness of climate), amount of clothing worn (i.e., coldness of climate), amount of time spent outdoors (i.e., lifestyle), and many other things.  It was not possible to incorporate all of these factors explicitly, so a "lumping" strategy was adopted.  When the model was running the user could arbitrarily impose a change in the aggregate of factors affecting "skin color," for example.  The user would simply alter the payoffs and penalties for having various skin colors at some arbitrary time during the model run.  This might correspond to a migration, or a change of lifestyle, or the adoption of clothing in response to a climate change.

The same process was used to reward or penalize blood types.  It was arbitrarily assumed that one blood type had properties which were desirable for a long interval of time, and after some "change in the environment" (such as the appearance of a virus) another blood type was more desirable.

Behavior genes were a greater challenge.  For example, a philandering gene was hypothesized.  Freddie and the professor had long arguments about how to specify the payoffs and penalties of philandering.  Freddie would state that he thought philandering would never pay off, and the professor argued that it would always pay off provided it did not jeopardize a monogamous relationship.  Then they'd argue the merits of monogamy versus polygamy.  Then the "rape gene" was brought up, and they'd argue about the merits and demerits of rape.  The arguments got complicated, and the professor kept having to remind Freddie that the merits and demerits of any particular gene had to be measured by the way it fared in an environment of other genes, as opposed to how the individual carrier of the gene fared.

Freddie seemed to have the additional problem of confusing different measures of "good."  Whereas he could accept that a gene was "good" in the sense of surviving many generations, he could not accept it as "good" from a "personal morality" standpoint.  The professor had to keep reminding Freddie to not be influenced by his moral preferences when the task at hand was to develop a population model that was as realistic as possible.

They eventually began to develop a hierarchy of behavior gene groups.  In their initial attempt, the first level of the hierarchy consisted of 1) IQ, to represent "posterior" brain functions, and 2) EQ, or "executive quotient," used to measure the capacity for those "anterior" brain functions that pertain to purposeful decision-making.   They had great arguments about how to incorporate another scheme for gene grouping, wherein LB, or "Left Brain" index, would be used as a measure of the capabilities that are most strongly represented in the Left Brain, and RB, or Right Brain index, would be used as a measure of the capabilities that are most strongly represented in the Right Brain.  There was no easy way to impose such a left/right grouping upon a pre-existing anterior/posterior grouping paradigm, due to the belief that a particular talent drew upon the  participation of many specific locations from both sides.  After many futile attempts to conceive of systems for cleanly-separated gene effects, they arrived at an unusual solution.

They would start thinking about genes in terms of subsets.  For example, verbal skill depends on competence in Broca's Area (left/frontal) as well as Wernicke's Area (left/posterior).  Reading skill is even more complicated, for it depends on  competence in Wernicke's Area (left/posterior), character recognition areas (right/posterior), serial combining areas (left/posterior), and context-dependent meaning search areas (right/posterior).  Since it is unlikely that any single gene would produce improvements in all these locations there probably is no single "reading gene."  Instead, any gene that produces an improvement in one of these areas could be called a "reading gene component."  This is a multi-gene view of reading competence.  They decided that all complex competencies are to be thought of as multi-gene.

Another feature of their subset gene theory is that every gene usually has more than one effect.  Thus, a gene that improves the "serial combining area" used in reading could lead to improvements in other tasks that required serial combining.  Just as important, genes that improve competence in one thing might decrease competence in other things.

Thus "armed" with their "subset gene theory," Freddie and the professor set out to create overlapping gene groups, without worrying about the details of the multi-gene origin of capabilities or multi-effects of each gene.  This was a breakthrough, for it allowed them to postulate many gene groups that would not otherwise appear acceptable.

For example, they postulated a group of genes for PR, a "producer" index, used to measure the capacity for personal accomplishment, or productivity.  They became bold, and conceived of PR's opposite, called PA, representing a "parasite" index, used to measure the abilities an individual might have for engaging in socially parasitic life styles.

At this point they had four behavioral gene groups:  IQ, EQ, PR, and PA.  Each one was acknowledged to consist of many individual genes.  It was further acknowledged that in some instances a gene could also belong to other gene groups.  This was a detail that wasn't important to worry about for the objective of developing a sophisticated computer model of population fluctuations, but it was an accommodation to the way genetics works.

It was clear that IQ consisted of many components.  IQ tests suggest that at least a dozen components are recognizable.  Some are even identified as residing in one or the other cerebral hemisphere.  Since a person's IQ is generally high, or low, in all areas (of a hemisphere) together, there is a temptation to think that the genes that influence IQ are probably close together on the same chromosome.  However, genes interact in ways that can make them appear to have been inherited together, and it is not necessary for the population model to explicitly contain these details.  The model treated IQ as if it were created by a small group of genes, mostly inherited together.

EQ was treated the same way.  MQ was classified as a subset of EQ.  PR and PA, strictly speaking, are also subsets of EQ.  The attractiveness of their subset gene theory was that it didn't matter which gene group was a subset of whichever other gene group.  Thus, Freddie and the professor stopped worrying about the hierarchical relationship of gene groups.

However, they did start worrying about the functional relationship of gene groups.  For example, specific PA strategies cannot be effectively practiced without the requisite anatomical inheritance.  The socially parasitic strategy of philandery is more successful when the individual male is viewed by the females as sexually attractive.  But what is it that causes females to view a particular male as sexually attractive?

"We had long arguments about this,"  Freddie told me.  "Sue would argue that men did have the equivalent of peacock tails by which women judged them, and I'd argue that that was ridiculous!"

"Tell me Freddie, was it uncomfortable talking about these things with a woman?"

"Well, not really.  Since I don't find her attractive, I don't care much what I say, or what she thinks of what I think.  She is my professor, so I really don't have any choice."

"It must be useful being able to bring both perspectives to bear on problems that involve sexual strategies, isn't it?"

"We did have clashes, but I don't know if her sex had anything to do with it.  Especially when she raised the question of why men have to exist!"

"You mean she actually asked such a question?"

"Oh yeah, and now I understand that it's a legitimate question."

And Freddie went on to explain why it's appropriate to ask why do men have to exist?  The argument goes something like this.  Men usually don't invest in rearing children.  "Paternal parental investment" is always smaller than "maternal parental investment."  The major contribution men make in being a father is one successful sperm!  If the female could reproduce asexually, as happens in a few species, the male gender would not be needed.  Without men, the argument goes, all offspring would be daughters, and all daughters would invest in raising offspring.  More children could be raised each generation, which would double genetic impact.  Furthermore, the woman's genes would make up 100% of the daughter's genome, instead of 50%, and all granddaughters would be made from 100% of her genes, instead of 25%, and so forth.  From the standpoint of the genes, having daughters asexually would be a more efficient way for them to reproduce themselves than the slower and more diluting process of reproducing sexually.  So why are there males in so many species?  And why do men have to exist?

"This question arose when we were trying to specify the rewards and penalties of monogamy and polygamy.  In polygamy the father's parental investment is less than for monogamy, so it is even more difficult to account for why polygamy was so common for our distant ancestors.  And if we couldn't account for this, we probably wouldn't be able to properly model the influences of polygamy and monogamy on population changes."

"Did you point out that modern men, who are monogamous, actually do contribute significantly to child rearing?  Both directly, through being with them and teaching, and indirectly, by providing home, food, protection, and probably many other incalculably valuable things?"

"Yes, but she said such facts could only account for modern monogamy, not prehistoric polygamy.  And whatever accounted for prehistoric polygamy has to be an important factor to not overlook in the model."

Freddie and the professor proceeded with their best estimates for the functional relationships between the behavioral gene groups and all other parameter's of an individual's life.  They created a few additional behavioral gene groups, and a few additional anatomy and physiology gene groups, but these were mere refinements which did not affect the gross behavior of population fluctuation results.  The greatest weakness in their model was identified to be the unrealistic functional relationships between the identified behavior gene groups.

They were surprised by an intriguing and unexpected result for the relationship between PA and PR.  There was a category of population fluctuation which was produced by this relationship.  When societal wealth accumulated, and the population rose, the incidence of individuals choosing the socially parasitic PA strategies increased!  A time came when too many people were engaged in PA, and the remaining population of PR individuals could not sustain the level of per capita wealth.  The non-productive individuals were "dragging" the living standard of everyone else down, and eventually the accumulated societal wealth began to decrease.  By this time the lower IQ portion of the population had outbred the upper IQ portion, and these new members were further "sapping strength" from society at large.  The dynamic was unstable, and always led to a population crash.

After the population crash, life was hard.  Any individual who chose a PA strategy was not tolerated, and most individuals chose PR.  Survival rates were low, so IQ began to rise.  Another population cycle began.  The seeds of every great age were sewn during the dark age preceding it.

Sue Barker was excited by this finding.  She admitted that part of her excitement could be traced to the discrediting of Human Nature, which apparently had become a childhood need that lived semi-subconsciously into adulthood.

Freddie was less excited by this finding.  He was troubled each time his moral values were "insulted," as he put it.  "The model led to the wrong answer every time!" he would exclaim.

Sue was still troubled by the issue of why men existed, though.  At times this unanswered question seemed more important than the unexpected new findings.  She would badger Freddie about it, and imply that the model was incomplete because it didn't provide any clues.

"Sometimes I think she was obsessed by the need to explain the enigma of sexual reproduction, in order to be considered for a Nobel prize, or something.  Then, at other times, I'd think she wanted to discredit men.  I really didn't know why it was so important to her, but I found myself taking it on as my goal also."

"Tell me, Freddie, why was she spending so much time with you?  After all, this was just Psychology 101, not a graduate level course.  She was treating it as if it was your PhD thesis!"

"I think it was because she saw in me a chance to get answers that other people couldn't provide.  I'm a computer major, and she could see how fast I could put the model together.  She had never seen a Monte Carlo approach applied to sociobiological population problems, and it probably struck her as a unique opportunity."

"So how do you know she wasn't really interested in you, Freddie?"

"She took every opportunity to discredit men, and I'm sure she could see that it didn't make me happy."

"For example?"

"The time she pointed out an obscure detail in my computer output."

And Freddie proceeded to give an accounting of a remarkable finding.  They had been studying various PA strategies.  One dynamic that had been programmed into the model, at Sue's request, was the strategy of rape.  She noted that several dozen animal species exhibit male rape of females.  The females raise their offspring, which are in fact half hers regardless of whether the biological father was the female's mate or another male who raped her.  It is in her interest to raise the offspring, provided she succeeds in concealing the true paternity.  If her mate were to find out that an offspring was not his, he would proceed to harass it to death, thus making more parental resources available to his offspring.  All this was well understood and explainable by standard biological theory.

The entire scenario of rape, concealment of paternity, possible discovery by the male mate, and subsequent harassment and possible death of the illegitimate offspring, all of this had to be somehow incorporated into the model.  It was accomplished by adjusting the survival rate of rape offspring.

This was one piece of the puzzle that led to Sue's eventual finding.  The other one has to do with "philandery," another dynamic that Sue insisted on incorporating.  Married men commonly philander, so this was modeled by including a specified probability of philander offspring for each man.  If the male adopted a PR strategy, his probability of having philander offspring was reduced compared with men who adopted PA strategies.  This was meant to account for the fact that successful PR men jeopardize their offspring's welfare by taking the risks of philandering, whereas PA men have little at risk by philandering.

Indeed, parasitic men often do not marry.  They become "pirates," figuratively, and sew their seeds wherever opportunities occur.  They often never see their offspring, so cannot provide parental investment, as Sue carefully pointed out to Freddie.  "What good were such men?" she asked, rhetorically.

She described "Attila the Hun" as the ultimate parasitic male, a marauding pillager of other productive men's labors, killer of other men's children, and raper of other men's wives.  She could not decide if such men adopted this strategy in response to perceived environmental conditions or because they inherited genes that predispose them to be like that.  Sociobiologically, she stated, it is understandable that such male behavior should exist.  She speculated that their numbers might increase when there was more to pillage, when conditions were good for the majority of people.  The argument she used is that when wealth existed in abundance there was more to lose in defending the excess wealth than in allowing it to be stolen.  Hence, parasitic behaviors should occur during the "best of times" in the affairs of men.

This was her finding, and it pleased her.  She was pleased that the model showed an intuitively plausible dynamic between degrees of male parasitic behavior and societal per capita wealth.  The two factors had correlated fluctuations.

Whereas this finding pleased Sue, it really bothered Freddie.  He kept seeing the model give the message that during the best of times the worst of human nature came out.

Freddie saw another pattern emerging which he suspected Sue was also aware of.  Namely, that the worst aspects of human nature seemed to be exhibited by men.  It was the men who were parasitic, who murdered, stole, raped, abandoned, and who instigated pillage and mayhem.  The women bore children, were caring, endeavored to maintain stability, and protected offspring from murderous stepfathers.  It was looking like all the ills of Man were caused by men.

"But I got even with her," Freddie said triumphantly.  "I beat her at her own game.  I pointed out that it was the women who maintained the worst aspects of men's nature by seeking out exactly those men who lived parasitically for their illegitimate liaisons.  The fact that women find such men sexually attractive attests to their co-conspiratorial role in maintaining the undesired traits."

"But what would be in it for the women?" I protested.  "What would be the payoff for the woman who seeks out parasitic men for surreptitios sexual affairs?"

"Easy," said Freddie.  "Their son's are likely to be parasitic, just like their parasitic father, and they will carry the mother's genes into the future using proven parasitic strategies.  The daughters, meantime, will give birth to sons who will be parasitic, etc.  So the women who think pirates are sexy are playing a double strategy game.  They're hedging their genetic bets by going with both competing futures."

"What did Sue say to that?"

"I think she was upset.  She didn't dispute my interpretation, and even volunteered that if women didn't play the male pirate game they could put such men out of business in only a few generations.  But since it would only take one uncooperating woman to undo things, it would not be possible for any of the women to liberate themselves from the encumbrance of being attracted to pirates once the dynamic had established itself."

Freddie's glee was moderated.  He had succeeded in forcing Sue to see that men were not all bad and women all good.  But he also thought that the bad that could be seen in some male strategies was ugly.

Freddie's major coup occurred, however, shortly after this minor one.  Freddie solved the problem of why men should exist!

He had tried playing-off some groups against others.  It was part of his attempt to understand why some populations exhibited more robustness against environmental assaults than other groups.  He had controlled all factors except certain ones in order to try to isolate the key factor affording robustness.  Freddie noticed that his polygamous populations were more robust than the monogamous ones.  Of course he didn't like that finding, so he kept checking it for other initial conditions.  All runs showed the same thing.  The more monogamous, the more robust.

Instead of telling Sue, and seeking her advice about it, he proceeded with a detailed inspection to find out why polygamy conferred robustness.  He noted that in polygamous groups, the ones with higher group survival rates, the individual male survival rates were lower when hypothetical environmental assaults occurred.  This seemed paradoxical, but then Freddie was getting used to paradoxical model results.

The paradox was explained when he looked in greater detail.  The individuals who survived became the polygamous heads of families, and the offspring of the lucky man bore his traits, which were presumably the same traits that enabled him to survive.  Thus, in a polygamous society, a larger fraction of the offspring acquire rare desirable traits faster than in the monogamous society.

Freddie extended this finding to explain what men are for.  Men's job is to explore mutation space, and provide a few good "winner take all" survivors.  Female offspring are the beneficiaries of traits received from their fathers, who were the "winner take all" victors of their generation.  The children of each generation benefit for the same reason.  Everyone benefits (except the loser males).

Freddie worried about the morality of this theory, as usual.  His explanation made it look like most men were meant to die.  As if most males were just cruel experiments, never meant to have families, and often not meant to survive childhood.  Freddie thought it would be a better world if every man and every woman could grow up, get married, have children, and live a peaceful life.  Freddie couldn't accept the fact that in the normal state human males had about a 20% chance of reaching adulthood, as did about 30% of women.  And that only a small percentage of those 20% of male survivors succeeded in gaining access to women.

Freddie explained his findings to Sue.  At first she was stunned!  He had succeeded where she had tried, and failed!  It was her question that he had answered.  The irony is that he didn't like the answer as much as she did.  It would have been more fitting for her to have made the winning speculation.  But Freddie succeeded, not because he got the answer he was looking for, but because he had the tools with which to probe, and the courage to face the facts that his model results presented.

Sue congratulated Freddie.  She said his term paper was now complete, and an A+ grade was too small a reward for what he had done.  She offered to co-author a paper presenting their model, and presenting their results.  And Freddie was happy about this.

Still, he was not happy about the messages he found in those results.  He was upset!  And Sue understood his emotional reaction.  She had become more insulated from the emotional meanings of her work, as many practicing sociobiologists do.

She tried to explain to Freddie that every living species should give thanks to an innumerable number of unfortunate ancestors.  Or rather, to contemporaries of one's actual ancestors, to the contemporaries who were the cruel and unsuccessful experiments of an uncaring Nature.  But only through such "wasteful" and painful mutational experiments can Nature produce the wondrousness that each species indisputably exhibits.  For every beautiful peacock there are many, many ugly peacocks.  For every human brain that thinks, there are many, many brains that did not think so well.  All these forgotten beings exist somewhere in the past.  We, the living, are indebted to them.

"That was Sue's consoling message to me," Freddie Sole explained.  "I don't know what to think.  My world seems upside down.  And that, Bruce, is what's wrong with me!  Thanks for listening."
.____________________________________________________________________

This site opened:  October 30, 1998.   Last Update: October 30, 1998