Universal Ethics
Universal Ethics > Stories and Games > About Bumper Cars

Bumper Cars - how experience leads to rules

Tiny game image Before reading this, try out the simulation game of Bumper Cars. There are a few questions at the bottom of the game page. Take a moment to think about those questions, and then read on (below)...

Learning Ethics from Bumper Cars?

A "bumper car" ride is a common attraction at summer fairs. The ride has slow-moving electric cars on a flat platform. The cars are oval with big rubber bumpers all around, so that children or other guests can ride in them without being hurt if the cars crash into each other.

Lady on bumpercar

In the traditional ride, there isn't any place in particular to go, so the kids just drive the cars around randomly, and it doesn't matter much if they get stuck between other cars or blocked from going to any particular part of the platform.

So at the fair, there isn't much the kids will learn at bumper cars, except perhaps to get practice at steering. But this bumper car game is a bit different, and there is something about ethics we can learn from it...

Suppose we were going to use bumper cars for our everyday life, in order to get to destinations that we want to go to? That is the scenario shown in the bumper car land game. These cars move horizontally or vertically in order to reach the coloured destinations shown on the four sides of the grid, which represent places that people want to go: to home, to work, to a shopping centre, or to a park.

Each driver picks where he wants to go next, and tries to get there as fast as possible, but without any regard to traffic rules. In the world of bumper car land rules haven't been invented yet! So the computer simulates that by moving cars randomly, and from time to time the cars bump. If the cars just pull back and try to resume the same direction after bumping, they'll just bump again. So, not knowing what to do, the drivers just choose a new direction randomly after bumping and head off that way. Sometimes this causes them to bump again, and other times they are free to proceed.

I think you can see that this is not a very efficient way to travel! Not only are bumps inconvenient, but because of the constant risk of bumping, the cars must be limited to going very slowly. Otherwise the drivers would be injured. So we can see that although the drivers are all different, with different choices about where they want to go, nevertheless they would all be happier if they could travel without bumping!

Ethics typically involves making people happier via the use of rules, so if the drivers can solve their problem by adopting rules, that is an exercise in ethical decision making!

Potential rules

Finding potential rules is a matter of invention rather than discovery. There is no place one can look with a telescope or a microscope to discover the appropriate rule within nature. So let's use our inventive genius to consider how we might prevent collisions.

There are two kinds of collisions that might occur: head-on collisions, and collisions where one car hits the other from the side. Let's consider these separately.

Avoiding head-on collisions

In the real world we already have a rule to prevent head-on collisions on our roads. The road is divided down the middle, and all car drivers stay on their right, so that they can pass each other without hitting. In some countries they do the opposite and stay on the left. It really doesn't matter as long as the drivers all follow the same rule and do it consistently.

In bumper car land there are no streets, but we could accomplish the same thing by numbering the rows and columns and designating directions of travel. For example, we could then designate that:

Or we could do the opposite, swapping "even" for "odd". Either way would work if all drivers pick the same rule and follow it consistently.

Avoiding side collisions

Sometimes a car going horizontally will meet another car that is going vertically. To avoid hitting, one of them needs to stop, but which one? Again we can take a clue from rules that have been invented already: we could assign the right-of-way to the car on the right. Or we could follow the example of countries like England and Australia, and designate that the car on the left should proceed. It doesn't matter so long as the drivers choose a rule and follow it consistently.

Adopting rules by experiment

The problem is that in bumper car land nobody learned any rules. Individual drivers might think up some of the above rules, but they won't necessarily each come up with the same rule. So how will this ever be decided? Won't chaos prevail forever, until somebody tells everybody else what they must do?

Not necessarily! There is a way that the rule may be adopted without anybody deciding it. It can happen using random choices and an evolutionary learning process.

The Random Walk

Let's consider the left-hand vs right-hand rule for avoiding collisions at intersections. Suppose the car drivers randomly pick one rule or the other, and follow it. If a driver has more success choosing one rule than the other, he will tend to pick that rule in the future. Even though the drivers choices are initially random, after a while one or other of the two rules will become prevalent.

The reason this happens is because of a statistical phenonmenon known as the "random walk." In a random walk, a series of random events eventually leads in a particular direction.

Checker board

To demonstrate this principle, you can do a simple experiment using a chess or checker board, a single chess piece or checker, and a die. The die will be used to make some random choices about moving the checker. You might suppose that the checker would just move around in the centre of the board, but that is not what happens. It will gradually move to one of the four sides. To try this out, click here for detailed instructions.

The same is true for vehicle drivers choosing the "left hand" vs "right hand" rules independently. Sooner or later, they will all be using one rule or the other. We can't predict which rule it will be, but we can predict that one of them will be adopted. And it will happen without the need for any single "authority figure" to dictate to the others what they ought to do.

To see a demonstration of this, click here to see the "Rules of the Road" computer simulation.

Throughout the ages, philosophers have argued about what rules are appropriate for morality, often emphasizing differences in their opinions rather than similarities. Trying to figure out what is moral by reading such debates can be very frustrating, leading some people to conclude that there is no such thing as morality at all. However, what they didn't understand is that rules of morality will inevitably emerge regardless. It emerges slowly, as an evolutionary process.

What about teaching?

Although rules will emerge in an evolutionary process, that is a slow process that has not yet resolved every kind of conflict that can occur. Moreover, there can be a lot of suffering involved in "trial and error." Rather than continually repeating the mistakes of the past, wouldn't it be useful to make decisions based on what has been discovered already?

Moreover, applying a bit of intelligence to a problem, one can often predict results without conducting a dangerous experiment, or by subsituting a less-risky experiment in place of it.

In the bumper car land game, the simulated drivers never learn. In the rules of the road simulation they do, but still there are collisions at first. But if we have real people as the drivers, they could do better yet by meeting beforehand, discussing their ideas for the potential rules, and then picking one or more rules. That way they could all go to their chosen destinations without the need for a learning process that involves some collisions!

That's one reason why people believe that it's important to teach rules to their children: so that the children can be spared from some undesirable experiences.

Keep in mind that the real world isn't limited to safe interactions such as bumper-cars colliding. Imagine putting children in full-size, powerful automobiles without them knowing any rules. By leaving the children to "learn by experience" that way, some might learn but most would end up dead.

Children are learning machines. Watch a little toddler and you will see how they explore to discover the world. Everything the see, they touch it or put it in their mouth. So that the child won't hurt themselves, parents give them a controlled learning environment. Small objects that they could choke on are put away. Electrical outlets in the play area are covered. Entries to stairways are closed off. So the child can learn by experience safely.

Then as the children get older, they learn to understand a language. Then they can read and listen to discover the rules that society has created. Some of these rules involve arbitrary standards, like the left-hand vs right-hand rule described above. Nevertheless, it makes sense to follow such rules because of the benefits they produce--at least until such time as better rules are invented and people agree upon them.

If you start to enumerate all the rules in a society, you will discover that there are many thousands of rules. The law books are filled with rules, and typically law only covers a subset of the ethical standards of a society. As children grow up, they learn the rules in the same way as they learn a language: one little piece at a time.

This cannot all be taught to children by means of a lecture. It is also achieved by creating a safe environment where children can experiment and learn on their own. And it is achieved by "vicarious learning" through stories and movies, and by observation of people whom the child admires. Therefore a parent also teaches by setting a good example for their child.

Improving on the Rules

As children grow up, they try to integrate and make sense of all the rules. Social scientists do that too, but at a more advanced level. Nevertheless, the rules themselves did not arise from a single brilliant philosopher long ago. The rules evolved, and typically the ethical theories serve more as rationalizations of the rules than as the true foundation of them. However, even as rationalizations, theories can help people to improve the rules.

For example, one theory is that each person seeks to improve their happiness (sometimes called "utility") and that the best result for a society is maximize the sum of the happiness of all individuals.

Do the bumper car rules described above do that? Not by themselves they don't. The drivers get satisfaction by going to the recreation park, to the shopping center, and other places in bumper car world. Maybe some of the drivers would have a happier life if they spent more time at the recreation park. Others might be happier if they spent more time relaxing at home. The rules of bumper car land are not sufficient to ensure that everyone is happy.

The reasons why bumper car land doesn't have rules on how long people should spend working, shopping, resting at home, or enjoying recreation are that: a) different people have different preferences, so no one prescription will work perfectly for everyone, and b) we have no means of measuring their satisfaction anyway, in order to optimize it. So instead, in bumper car land the drivers simply go where they please, and they develop rules so they don't get in each other's way.

But even though we can't measure the individual happiness of the drivers, the concept of improving their happiness is still useful. The drivers each seek their own happiness, and by being free to go where they want, quickly and without collisions, this would likely improve their happiness.

Moreover, if we understand that each driver has a goal to improve their happiness, we can spot inconsistencies between rules and situations where new rules are needed. For example, suppose that in bumper car land the drivers adopted an effective rule for preventing collisions at intersections, but for cars approaching head-on they still crash occasionally. Obviously this would be inconsistent with the goal of happiness for the car drivers. If one of them were to propose an improvement to cure that situation, the others would be pleased to accept it, thus resulting in adoption of the improved rule.

As you can see, ethical behavior involves cooperation. Individuals rely on each other and interact with each other. The most effective way for an individual to enhance his (or her) own personal satisfaction is to seek solutions that also enhance the happiness of others with whom he interacts. Moreover, in order for this to work effectively, it needs to have the broadest agreement possible. Leaving some people out for arbitrary reasons (eg: "I don't like that car's color.") is counter-productive to the desired results.


What do you think of the content on this web page?

Your Vote (click one)

Search icon Site Search    Home icon Return to Universal Ethics home page

About Bumper Cars, copyright Arthur de Leyssac, 2014, All rights Reserved.