3. Sociobiology

The main business of this post is to introduce sociobiology as a second source, alongside utility theory, of fundamental insights about human nature, which the economist can use as an input to theories that elucidate how human beings behave and interact in the aggregate. The main innovation of this post will be to adapt indifference curve maps, turning them into isofitness curves, and use isofitness curve maps as a way to make graphical arguments in the sociobiology space. I am not aware that anyone has ever done that before, but it is only the style of graphical argument that is (possibly) original. Most of the principles of sociobiology that I will be teaching in this post are, as far as I understand, quite standard and uncontroversial, though I take a few liberties with my examples.

For an economist to argue in the biology space at all may seem odd, but it really shouldn’t. Human beings are life-forms, after all, and it stands to reason that some principles of biology which apply to other life-forms, would apply to human beings as well. So it’s not surprising that there has been a long commerce of ideas between economics and biology. Long before economics broke off to become a separate discipline, Aristotle, the great polymath Greek philosopher who wrote about almost everything but above all was a biologist, wrote a treatise on “economics,” or household management. Since modern economics was founded in 1776 with Adam Smith’s Wealth of Nations, a few economists have made major contributions to biology. Thomas Malthus, one of the three great classical economists along with Adam Smith and David Ricardo, inspired Charles Darwin by showing how population growth in the context of competition for scarce resources tends to drive living standards down to subsistence and below, until it is checked by starvation. Herbert Spencer, another polymath but this time one who could be classified as an economist as much as anything else, coined the phrase that has long been one of the pithiest summaries of Darwinism: “survival of the fittest.” More recently, game theory first made its mark in economics, e.g., in the work of John von Neumann, before it spread into biology as a tool for thinking about mechanisms of evolution. If it has become unfashionable for economists to learn from biologists in return, that is partly because utility theory has struck so many economists as such a satisfactory paradigm through which to look at human nature that biology could be ignored, but I showed in the last post why this is a mistake. The weaknesses of utility theory are a good reason to let biology, and in particular the branch of it sometimes called sociobiology, back into the room.

First, let me explain what I want sociobiology for. I want to invoke sociobiology as a way to answer a skeptic who doubts whether it’s possible to justify strong, general claims about what people want. Utility theory based on revealed preference says that people must want something because they choose it, and while this approach has some merit, we saw in the last post what its limitations are, given that most enjoyment is social in nature. Sociobiology, sometimes called evolutionary psychology, says that people must want something, or at least that they must have a powerful instinctive tendency to want it and to feel satisfaction from having it, because evolutionary fitness requires it. A third method of finding out what people want is survey research, i.e., simply asking them, and that has its merits and weaknesses as well, which I might address in another post. But suffice it to say that it is not such a panacea that we should disdain to learn what we can about human preferences from revealed preference and sociobiology.

Evolutionary psychology makes predictions about human nature based on the fact that once upon a time, in a period sometimes given the technical name of the environment of evolutionary adaptation, or colloquially, the Stone Age, human genes competed for space in the gene pool, and some people had traits more conducive to survival and reproduction than others did. The word fit is used as shorthand for traits conducive to survival and reproduction. Fitness can be represented graphically using a chart such as Figure 1:

Figure 1. An isofitness curve map of preferences over wealth vs. safety

In its shape, Figure 1 will be familiar to anyone who has read the last post at this blog, or studied utility theory in college. It has the shape of an indifference curve map. But the labels have changed. Instead of being labeled U, for utility, these curves are labeled SAEO, for “Stone Age expected offspring.” And instead of “indifference curves,” I am calling them “isofitness curves,” since what the points in each curve have in common is not that they are equally preferred by an individual, but that they represent states of equal evolutionary fitness, i.e., they would result in equal numbers of offspring for a Stone Age man. Man, not woman. That’s an important point, to which we’ll return.

(Technical digression) Like indifference curves, isofitness curves adumbrate, in two-dimensional space, a phenomenon that really needs to be understood in a far more complex space of n dimensions. Many, many factors, not just wealth and safety, affect fitness. Isofitness curves differ from indifference curves in that the numbers attached to them have cardinal as well as ordinal significance. To say that indifference curve I yields 100 utils and indifference curve J yields 50 utils only means that consumption bundles on I are preferred to consumption bundles on J, and arbitrary numbers have been assigned for convenience in order to express this. But to say that isofitness curve K corresponds to SAEO=10 and isofitness curve L to SAEO=2 is to say that a man on K would have had, on average, ten offspring in Stone Age conditions, and the man on L, two. (End technical digression)

We now know what the isofitness curves in Figure 1 mean, but are they true? Since no units are specified on the axes, it’s not a question of the accuracy of specific numbers, but rather, of the qualitative features of the chart. In particular, why do the isofitness curves slope down?

Like indifference curves, isofitness curves show trade-offs, though with a slightly different meaning. It’s not that a man is willing to sacrifice some safety for more wealth, but that his ability to beget and rear offspring remains steady if as his environment or lifestyle becomes somewhat more dangerous, his wealth also increases by a certain amount. Safety and wealth are both good for fitness. He can beget and rear more children if he doesn’t get killed. He can also beget and rear more children if he can get enough food to feed his wife and a few children, or better yet, if he can maintain many wives and dozens of children in comfort. And so if a risky opportunity to get more wealth and status comes along, for example, if he has an opportunity to become chief of a tribe through combat with a rival or by courage in battle against the tribe’s enemy, it may or may not be worth it, depending on how great the danger is, and how much wealth and status he stands to gain.

Now, just because certain conduct will maximize a man’s expected offspring, it does not follow that he will choose that conduct. His preferences might, logically, be somewhat or even completely different than what serves the interests of his selfish genes. But if his preferences induce him to act in a way that does not maximize his offspring, his genes will lose market share in the gene pool. Consider Figure 2:

Figure 2. Maladapted Preferences

In Figure 2, the solid curves represent, as before, wealth/safety combinations corresponding to equal numbers of Stone Age expected offspring, while the dotted lines represent the preferences of some hypothetical man living in a Stone Age environment, whom we might call Peasant. Since there is no reason to think that life will offer Peasant a continuous distribution of wealth/safety combinations, I have represented his choices, not as a budget constraint, but as seven points, A, B, C, D, E, F, and G, each of which represents some life option available to Peasant, among which he has to choose. Peasant, with his indifference curves as they are, will pick B, and can expect two surviving offspring. But he could have picked F, and expected ten.

Now, suppose there is another man in the same population, named Warrior, whose preferences correspond exactly to the SAEO curves. What he likes to do coincides exactly with what maximizes his selfish genes’ market share in the next generation. Warrior will choose F, and can expect to have ten offspring to Peasant’s two. What happens next? Warrior’s offspring in the next generation outnumber Peasant’s five to one, and inasmuch as Peasant’s offspring inherit his risk-averse propensities, they will continue to do less than their share of procreation. Eventually, the genes that underlay Peasant’s attitude to risk will disappear from the gene pool, and Warrior’s descendants will inherit the earth. Thus evolution should ensure that indifference curve maps adapt to match isofitness curve maps. This is what Herbert Spencer aptly called survival of the fittest.

But what if the environment changes?

Pre-Christian societies were either polygamous or tolerated concubinage, so wealth and status could yield handsome genetic payoffs for men, and risky, violent behavior was more likely to be worth it. But first polygamy, with the Greeks and Romans, and then concubinage, starting with Christianity, was rendered increasingly unacceptable by the moral progress of civilization, so that today, using money to acquire more wives and concubines isn’t really an option. Men who choose safer lifestyles won’t miss out on so many opportunities, and are likely to have as many or more offspring as men who live more dangerously. Shouldn’t that erase and reverse the influence of Stone Age conditions on men’s instincts?

It might, given enough time. But instincts must have some foothold in the gene pool if they are to expand their market share under favorable conditions. An extinct gene won’t be revived just because, had it still existed, it would now have proven useful. Anyway, modern society doesn’t select for fitness as ruthlessly as Stone Age society did. Religious customs and welfare states protect the weak from starving, and give most men, not just the fittest and strongest, opportunities to reproduce if they want to. Indeed, through most of the history of civilization, Darwinian pressures have been muted. And the history of civilization has been quite short relative to the glacial time scales over which evolution operates. So while our rational minds have been colonized by civilized and modern ideas, our bodies and instincts are probably still largely or wholly adapted to the Stone Age.

Fast forward to modern times, and each man is, after a fashion, two men, the man of reason, and the man of impulses and instincts, and these two men are often at war. A version of this problem is depicted in Figure 3, where a new “rational utility function” is juxtaposed on the old isofitness curves. In principle, the isofitness curves shouldn’t be relevant to a modern man, since he doesn’t live in the Stone Age and can’t expect the reproductive payoffs they describe. Yet he has been honed by evolutionary forces to have those instincts and impulses, and he can’t help but feel the pressure of them.

Figure 3. The modern man’s self-control problem

I have relabeled the horizontal axis “Camaraderie, Command, Audacity,” because if I labeled it “Wealth & Status,” as before, a crucial point would be missed. The selfish genes may, in feeding us instincts, aim at wealth and status, and ultimately at the offspring which they help a man to beget and rear, but they can’t do our thinking for us. They don’t have the full mental apparatus that we do. They can only feed us instincts, desires, passions, likes, dislikes, etc. And they’re absurdly out-of-date. So what the selfish genes give us are not preferences for wealth so much as instincts to engage in certain kinds of conduct which would have tended to make us wealthy under Stone Age conditions.

I have guessed, for the sake of argument, that the traits that would tend to bring a man wealth and status under Stone Age conditions are camaraderie with his fellow warriors, audacity in battle and counsel, and a commanding presence and a tendency to assert dominance, or in short, “camaraderie, command, audacity.” Furthermore, I assume that these traits are also conducive to flourishing in modern conditions, up to a point. That may be seen in the downward slope of the left side of the indifference curve maps. But later the indifference curve maps bend up, signifying that past a certain point, camaraderie, command, and audacity become detrimental under modern conditions. Perhaps there was such a thing as too much camaraderie, command, and audacity even in Stone Age conditions, in which case the isofitness curves would eventually bend upward. The point is that the isofitness curves are steeper than the indifference curves, which signifies that under modern conditions, the relative payoff to steady, routine work, compared to camaraderie, command, and audacity is higher than it was in the Stone Age.

Points A, B, C, D, E, F and G again represent different lifestyle choices available to an individual. Just for fun, I have labeled two of them. B, which I previously labeled Peasant, is now labeled Nerd. F, which I previously labeled Warrior, is now labeled Cool Dude. How much modern nerds resemble Stone Age peasants, or modern cool dudes resemble Stone Age warriors, I won’t attempt to assess. At any rate, I certainly do not mean to suggest that modern nerds are descended from Stone Age peasants. On the contrary, the point of the Peasant vs. Warrior story was that Peasant leaves no descendants in the long run, and we are all descendants of Warrior. The difference between Cool Dude and Nerd in Figure 3-- and in this respect, I think the model may capture a key difference between cool dudes and nerds in real life-- is that Cool Dude is governed by his impulses and instincts, and achieves, for a time, a form of flourishing that human beings, including Nerd, instinctively desire and envy, while Nerd overrides his instincts in order to maximize his long-term utility, and ultimately leads a life that rational people, including perhaps a regretful Cool Dude in his later years, recognize as more genuinely prosperous.

But what does that rational utility function consist in? What is this more flourishing life that reason can perceive and pursue, at the cost of having to struggle against instincts and impulses? Here there can be an endless variety of answers, but an interesting distinction can be made between answers that would tie the ultimate content of human happiness to our genetically programmed desires and preferences, and answers that would see much or most of true human happiness in forms of enjoyment proper to the soul rather than the body, and not reducible to any sociobiological basis. A man might resist his impulses to eat junk food, hang out with the cool kids, and above all, chase the girls, because by hitting the books, he’ll likely earn for himself, later in life, an abundance of fine dining, a chance to hobnob with the high-society elite, and a good marriage. Or, he might resist base, hedonistic impulses because he wants to master philosophy or poetry or music, or preach the gospel. But I only stop to note this interesting question, which would take us too far afield.

For now, I want to show the value of sociobiology by two straightforward successful predictions. The first, illustrated in Figure 4, concerns gender differences in attitudes to risk.

Figure 4: Gender differences in preferences over risk

Figure 4 shows two isofitness curve maps in wealth/safety space, one for men (red), and one for women (green). The isofitness curves for men are much steeper, signifying that men are, by nature, more willing than women to sacrifice safety for the sake of wealth and status. The reasons for this difference lie in the reproductive opportunities facing the two sexes in the Stone Age. For men, the reproductive payoff to wealth and status could be very high, since a high-status male, a victorious warrior for example, might get himself a harem, seduce female fans, and/or rape female captives. For women, the potential reproductive payoff was far lower, because she is anatomically limited by the time required for pregnancy and nursing. Moreover, a man who dies for the sake of having a child (e.g., to save his pregnant wife, or to rape a girl) might still get a viable offspring out of it, while a pregnant woman’s death is also that of the baby. Risky opportunism, then, is far more conducive to male than to female genetic fitness, and this is the evolutionary and instinctive basis for the stereotype that men are more daring and women more cautious, and the observed facts that women tend to invest their money more conservatively, and men are vastly overrepresented in risky professions.

By the way, I’ve been mentioning rape in a rather casual way, as one of the strategic options which it might be conducive to evolutionary fitness for a man to pursue. Obviously, I absolutely do not regard rape in a casual light. It is a horrific crime which I utterly deplore. Yet men may still have an instinct to do it. Some may feel it’s a little wicked of me even to suggest that, because it might plant the seeds of a deadly fascination in some male readers, or even supply an excuse for the act. "Instinct made me do it." But I think we have to face the dark side of human nature, and fight against it. Rape is evil. It is also natural, or to be more precise at the cost of introducing theological premises in which I believe but on which I will usually not rely for purposes of this blog, it is not natural to men as originally fashioned by God, but is part of what human nature has become as a consequence of the Fall. Men and women, be warned!

Second, Figure 5 shows a stylized representation of people’s evolved preferences for salt, sugar, and fat. In Stone Age conditions, sugar, salt, and fat were very scarce. Since they’re all useful to the body, evolution equipped us with a strong liking for these food elements, since the best policy, under Stone Age conditions, probably was to eat them whenever we could. Yet the optimal consumption of salt, sugar and fat for purposes of health is not infinity. It’s quite limited. Stone Age genetic competition never systematically ran up against this issue, because the scarcity of sugar, salt, and fat ensured that people could hardly ever overindulge. But in modern times, sugar, salt, and fat have become abundant and cheap, so we can afford much more sugar, salt, and fat than is good for us. Therefore, unless we restrain ourselves, we overeat them, and end up obese, with high blood pressure. In Figure 5, the budget constraint shifts out, and consumption increases past the line representing the healthy optimum.

Figure 5. The evolutionary roots of unhealthy eating

Gender differences in attitude towards risk, and people’s maladaptive propensities to eat junk food, are somewhat subtle predictions of sociobiology. Many other predictions of sociobiology are so obvious that it seems stupid to mention them. Sociobiology predicts that people often find sex pleasurable, that parents feel a special attachment to their children, that people feel natural desires for many things useful to the sustenance of the body, such as food and drink and comfortable temperatures and shelter, that people have an instinct for self-preservation, that they fear fierce wild beasts and falling from great heights, and so forth. Yet it is worth mentioning, after all, that common sense in all these cases is backed up by sociobiology, since common sense itself is not necessarily infallible. Sociobiology gives us more of a warrant to generalize such traits of human nature than would common sense alone.

As an instance of what sociobiology might contribute to the understanding of human welfare, consider the question of whether life has improved for Americans in the past 50-60 years in light of (a) ongoing increases in GDP per capita and proliferation of technology, and (b) a breakdown of the family, with increases in divorce and illegitimacy and a drop in the married share of the adult population. Economists tend to think of it as a "fact" that (a) has increased Americans' well-being, and a "matter of opinion" that (b) has reduced it. That, in part, is because the doctrine of revealed preference implies that people are always made better off by having more choices, but doesn't easily support the widespread popular assumption that divorce is bad. A very naive analysis of divorce from a revealed preference standpoint is that it must be good or people wouldn't do it. This obviously fails because divorce is jointly decided in complicated ways, but that only leaves the utility theorist agnostic. Sociobiology can get around the interpersonal complexities by insisting that yes, human beings must desire sex and family, evolution ensures that, and can go from there to... well, I'll revisit the topic in later posts. But it promises a path out of the unsatisfactory situation where economists persist in overconfident praises of GDP growth while responding to a phenomenon as vast and transformative as the Sexual Revolution with a kind of agnostic shrug.

I have made a lot of rather hasty graphical arguments in this post, so I should clarify that I don’t think they prove much if anything. In general, graphical arguments can be very effective tools of exposition, but they rarely if ever rise to the level of proof. Rather, they are mnemonics and placeholders, and are usually open to cogent critiques to which the most rigorous and compelling answers would consist not in more graphs, but in careful verbal arguments, or formal mathematical models, or-- a theoretical methodology hardly used in economics but of which I’m an advocate-- computer simulations.

Sociobiology and revealed preference are key sources of insight about individual human motivation. More could we said, but by now we know enough about why humans act to start examining what happens when they interact. And so we turn to GAME THEORY.