Human Beings

Two Different Means to an End

Human beings have wishes and goals. To achieve a goal, we have to find a way to get there. We have to decide which means can be used to reach the goal. This is not always easy. The more complex the environment, the more difficult it becomes. We learned in Section 2.5 that an action in a complex system always has more than one effect.

The sociologist and economist Franz Oppenheimer divided these means into two opposing groups [Opp27]:

• the economic means
• the political means

The economic means is one’s own labor or voluntary exchange, because the economy is founded on these two activities.

The political means is the use of violence or coercion to obtain a good. Oppenheimer called it the political means because goods were mainly appropriated through wars and other military operations (he wrote this in 1927). Taxes and levies also fall under the political means.¹

This distinction is minimal and somewhat provocative, but it provides a way to structure the topics of this book. At first, this book considers only the economic means. Only later, in the final Chapter 11, which deals among other things with state regulation, will the political means be discussed. The political means can be applied both “top-down” and “bottom-up.”

Bounded Rationality

Until the 1970s, many scientists assumed that people generally behave rationally and that their thinking is usually consistent and logical. Deviations from this were explained by strong emotions such as fear, love, or hate [Kah12]. This changed in the 1970s, when so-called behavioral economists began to examine economic decisions from a psychological perspective. Researchers found that human thinking works quite differently from what had previously been assumed. One part of the brain is responsible for “fast thinking,” for spontaneity and ideas. But when nothing comes to mind, “slow thinking” switches on. We ponder, reflect, and try to make an effort. For this reason, Daniel Kahneman titled his book “Thinking, Fast and Slow” [Kah12]. Kahneman calls the fast part System 1 and the slow part System 2. System 1 handles everyday tasks, such as:

• Understanding simple sentences in everyday language
• Spatial vision
• Driving a car in calm traffic situations
• Trained behavior and memorized knowledge
• Innate behavior inherited from animals, such as running away from fire

System 1 works automatically and quickly. Using it does not strain us. It is constantly in operation, and we cannot control it directly. For example, we cannot switch off language comprehension. If you do not want to hear what someone is saying, you have to cover your ears. System 1 can be trained. You can learn new languages and, after some practice, understand them immediately without much thought. Energy consumption then falls, and you can listen longer without getting tired. Because System 1 always runs automatically, it is difficult to notice wrong decisions. Sometimes it decides so quickly that we are not even aware that a question has been answered at all.

System 2 performs slow and systematic thinking. It requires concentration and can be disturbed from the outside. This kind of thinking consumes energy; over time, it is tiring. System 2 is responsible for human self-control. If you can pull yourself together when you are angry, that is System 2 at work. With the help of System 2, you can focus or “program” System 1 for specific tasks, such as looking for a particular person in photos or particular objects in hidden-picture puzzles. When someone says “Pay attention” or “Listen carefully,” they want the other person to concentrate fully with System 2.

Both systems are basically active. System 1 is always working; System 2 is normally asleep and is activated only when needed. We may then wake up with a start: “What was that noise?” Unfortunately, we cannot switch off System 1’s errors. We can only alert System 2 to check everything carefully, for example when solving an arithmetic problem with many numbers. We can, however, gradually retrain System 1. In summary: System 1 performs heuristic work, while System 2 performs systematic work [Pea15].

Important: Human beings have only “bounded rationality.”

People behave rationally from their own perspective, but they are limited by their brains. The degree of rationality also depends on how much time a person has to solve a task. If things have to happen very quickly, the person uses System 1, and there is a risk that they will do something “intuitive” but wrong. We will return to this later in Chapter 8 in order to answer the question of whether artificial intelligence and robots will be able to replace human beings.

Game Theory and Human Behavior

In the children’s game “Rock, Paper, Scissors,” two children play against each other. Each child can represent scissors, a rock, or paper with one hand. Scissors are shown with two spread fingers, rock with a fist, and paper with a flat hand. The game is usually played over several rounds. Both children show their hands at the same time after saying a rhyme such as “Rock, Paper, Scissors.” The winner of the round is determined by the following rules: scissors cut paper, paper covers rock, rock crushes scissors. Games like this can be represented as a table or, in mathematical terms, as a matrix:

The rows show the options available to the first player, and the columns show the options available to the second player. Each cell contains two numbers: the first is the payoff for the first player, the second the payoff for the second player. If Player A chooses “Rock” and Player B chooses “Scissors,” the result is the gray field at the top of the right column, with the content “1 -1.” This means that Player A wins 1 and Player B loses 1. Here, a win is formalized as 1 and a loss as -1.

This game is a simple example of the kinds of games studied by game theory. What these games have in common is that two players have the same actions available and cannot coordinate with each other. In game theory, one might ask, for example, whether there is a strategy in the game above that allows one player to defeat the other. Game theory studies games using logical and mathematical methods. This mathematical game theory is a normative theory that identifies optimal strategies. How “real” people behave in such games, however, is the subject of “behavioral game theory” [Gin14]. Game theory is an analytical method for studying social interactions. It applies to all living organisms, which is why it is also used in biology, economics, computer science, artificial intelligence, psychology, and sociology.

The children’s game “Rock, Paper, Scissors” is a so-called zero-sum game. One player’s gain is the other player’s loss. Such zero-sum games often occur in reality, for example in competition. In sports, one side always wins at the expense of the other. If two companies offer a very similar product, they are in a zero-sum game. Every additional sale comes at the competitor’s expense. In some countries, “haggling” over the price is allowed. In a shoe store, for example, you may be able to demand a lower price than the one on the price tag. This “haggling” is also a zero-sum game. One side gains at the expense of the other.

The Prisoner’s Dilemma is an example of a non-zero-sum game [Fre14]. Two people, A and B, are caught together by the police during a burglary and are put into separate cells. They cannot communicate with each other. Both are to be interrogated by the police, and the sentence depends on whether they confess or not. The following table expresses the game:

The evidence is quite thin, and if both players remain silent, they will both be imprisoned for only two years. If both players confess, they will both be imprisoned for five years. If only one player confesses while the other remains silent, the one who confesses goes to prison for only one year, while the one who stays silent spends ten years behind bars.

What should the players do? Unlike in the previous game, “Rock, Paper, Scissors,” the players are not simply playing against each other; they would benefit if they cooperated. It is a non-zero-sum game. The smartest move is to remain silent. But then there is the danger that the other person will betray you, because one year in prison is better than two. Can one player trust the other? It is a dilemma. The Prisoner’s Dilemma is an example in which egoism does not pay. If both act egoistically, both are definitely worse off than if they act socially and cooperate.

The Prisoner’s Dilemma can be generalized in the following table:

The two players can cooperate or boycott [Sig09]. The cells contain four different values:

• Reward R for cooperation
• Punishment P for defection
• Temptation T
• Sucker’s payoff S

Many games can be formulated with these values. If T > R > P > S, then the game is a variant of the Prisoner’s Dilemma.

Such dilemmas often appear in the real world. A well-known example is the “Cold War” in the 1980s between the West and the Eastern Bloc. At the time, the question was whether to rearm, that is, to produce and deploy more nuclear weapons, or whether to disarm. In this example, cooperation means disarmament, while boycott means rearmament.

Another game is the Stag Hunt [Sky03]. Imagine two prehistoric cave dwellers thinking about what they should hunt that day for their families. If each goes out alone, each can hunt two hares, enough to feed two people. If they join forces, they can hunt a stag, which feeds 10 people. If one wants to hunt stag but the other wants to hunt hares, there are only two hares to eat. The following table describes the game:

Here too, both are better off through cooperation. Here R > P = T > S.

In the Prisoner’s Dilemma or the Stag Hunt, it makes a big difference whether the game is played once, twice, or indefinitely often [Gin14, Sig09, Sky03]. Human beings have a memory and do not forget when another player treats them unfairly. What would happen if one burglar in the Prisoner’s Dilemma came out of prison after ten years? Would he want revenge? Would he make the same mistake again? People also remember who has treated them well and form friendships. In the Stag Hunt, friendships almost become visible, because hunting together has great advantages.

Which strategy is optimal for a game repeated indefinitely?

From a mathematical perspective, it is best to cooperate on the first move and then always do what the opponent did. You “copy” the opponent’s behavior. This strategy is called “Tit for Tat.” But human beings are only boundedly rational. Behavioral game theory has investigated how people actually behave [Gin14]. If someone always maximizes only their own advantage, they are egoistic (“self-regarding”). Another term for people who have no empathy for others and no regard for others’ situations is sociopath. The opposite, someone who always maximizes the other person’s advantage, is self-sacrificing (“other-regarding”) or altruistically cooperative. If the opponent turns out to be an egoist, you can punish them on the next move by refusing to cooperate as well. If a player is willing to accept personal disadvantages in order to punish, this is called “altruistic punishment.”

Scientists have found that human beings are usually conditional cooperators and altruistic punishers [Gin14]. They normally tend toward cooperation, but when someone misbehaves, violates social norms, and acts egoistically, they tend toward altruistic punishment.

Important: Most people are conditional cooperators and altruistic punishers.

However, a small share of people also have an aversion to inequality (“inequality aversion”). These people act in ways that produce as little inequality as possible. If they see themselves disadvantaged by inequality, however, they tend toward stronger measures than when they themselves disadvantage others.

Human beings are therefore neither the selfless altruists of utopian political theories nor the egoists of economics. They seek win-win situations and punish those who behave differently.

Two Visions of Humanity

There are many different “theories” about the ideal human society and about how justice, equality, freedom, power, and rights should be handled. Examples include socialism, communism, and anarchism. Strictly speaking, these political worldviews should not really be called “theories,” because a theory must be scientifically proven. These political “theories” are more like conjectures or hypotheses. Interestingly, many of them are incompatible with one another. Not only are very different theories, such as socialism and the market economy, incompatible; even specific subgroups within these schools differ greatly.

The economist Thomas Sowell traced these different “theories” back to two different “visions” [Sow07]:

• the unconstrained vision
• the constrained vision

In the unconstrained vision, human beings are basically good. Circumstances and society, however, do not guide them to show this “goodness” in real life. Human beings can be improved; they are “perfectible.” In principle, they would be capable of becoming perfectly moral people who think only of the common good and act altruistically. This ideal can also be established on earth; heaven on earth is attainable. No sacrifices or costs may be spared in order to reach this goal. The costs are often justified with sayings such as “That is simply the price of equality.” Since the costs do not matter, representatives of this vision are rarely willing to compromise. As a typical philosophical work, Sowell cites “Enquiry Concerning Political Justice” by William Godwin (1756 - 1836), published in 1793. William Godwin divided human actions according to two criteria: whether they were intentional or unintentional, and whether the result was good or bad [Sow07]. The following table provides an overview:

If one intentionally does good, it is virtue. Intentionally doing something bad is vice. For Godwin, one could only do bad unintentionally. There was no unintentional good. Yet unintentional good is an emergent property of the system. It arises “bottom-up” from people seeking win-win situations. It is more than the sum of its parts. Even today, people who follow the unconstrained vision often fail to see the existence of this “bottom-up” dynamic.

In the constrained vision, by contrast, human beings cannot be fundamentally improved. They are mostly egoistic and sometimes altruistic. Human beings are not morally perfect; some people are even immoral. A society has to expect these constraints and provide appropriate institutions for dealing with them, such as a legal system and the police. Society must also create incentives for good behavior so that society as a whole can still develop positively. As an example, Sowell cites “The Theory of Moral Sentiments” by Adam Smith (1723 - 1790), published in 1759.

The two great revolutions of the 18th century can be seen as realizations of these two visions: the unconstrained vision behind the French Revolution (1789–1799) and the constrained vision behind the American Revolution (1763–1789).

The constrained vision will be examined more closely over the course of this book.