Enable JavaScript.

On the Evolution of Ethics


Ethics is the study of right and wrong conduct. There is great variety in the ethical systems dictated by the world's religions and cultures. Two observations can be made. First, faith and reason affect all of these systems, and second, ethical systems evolve; they change over time. In this paper, ideas from the science of Complex Adaptive Systems are applied to ethical evolution. It begins with a brief description of the key ideas of this science, introduces the concept of fitness landscapes, discusses the roles of faith and reason in the evolution of ethics, and then applies these concepts to the evolution of personal and community ethics.


  • Introduction
  • Complex Adaptive Systems
    • Properties and Mechanisms
    • Fitness Landscapes
    • Internal Models
  • Ethical Systems
    • Properties and Mechanisms
    • Internal Models
      • The Faith Variable
      • The Reason Variable
      • Faith and Reason
  • Ethical Evolution
    • Personal Ethical Evolution
    • Community Ethical Evolution
  • Summary
  • Bibliography


Ethics is the study of right and wrong conduct; in other words, morality. To act ethically is to act according to standards. But how are standards determined, for individuals and for communities? And do those standards vary by place and time?

For many people, the answer is simple. Ethics is dictated by religion. The response to condition A is action X, to condition B is action Y, etc. The response is automatic, the result of intensive training in that religion's ethical standards. Any attempt at the justification of an ethical standard is of the form "Because that is what God wants," or "Because that is what the Bible (or Koran, or other scripture) says," or "Because that is what the Church, God's intermediary on Earth, wants." Coupled with these answers are the carrot and stick of heaven and hell, and reinforcement through massive doses of fear and guilt.

The problem is that there are so many different religions, they promote a wide variety of ethical standards, those standards change over time, and there is no way to determine what the ultimate set of standards should be. Less than 160 years ago, slavery was seen as ethical behavior by large segments of the American population. It was condoned by a variety of Christian religions, and justified on biblical grounds. Today, slavery is universally considered evil, condemned by the same religions, and denounced on biblical grounds. It would be difficult to argue that all ethical systems are equivalent; that an ethical system that values slavery is as good as one that values freedom.

A conclusion that can be drawn is that ethical behavior is not dependent on any particular religion. Instead, it is dependent on something that underlays all of the world's religions; something that changes over time. One task of this paper is to show how morality is affected by faith and reason. A second is to show how ethical systems evolve over time.

As a starting point, we can ask which ethical standards are qualitatively the best, at the highest possible level of correctness. A person at any given ethical level cooperates and competes with other people, some of whom are at a higher level, and some of whom are at a lower level. Each person argues that his own ethics are obviously the best, at least for a particular situation. This can lead to so-called "situational ethics," where it is up to every person to determine his own standards of behavior in every situation. However, most people base their personal ethical standards on those of a community. That moves the question from individual ethics to community ethics. An alternative is to ignore the question as irrelevant, and to assume that we are all, as individuals and communities, engaged in a process of ethical coevolution.

It may seem strange to apply evolutionary ideas to ethics, but why should ethics be different from other complex systems? Consider the general problem of the correction and punishment of criminals. Today, a few of the American states have abolished the death penalty (Minnesota); most states use it sparingly; and a few use it frequently (Texas). These positions all reflect local ethical standards; so which one is ethically the best? We will just have to wait and see.

Given today's advances in communications and transportation, the Earth appears to be shrinking. This is now a key factor in ethical evolution. Ethics previously developed in isolation must now compete on the world stage. High ethical standards appear to be positively correlated with cultural competitiveness. While the United States has many problems, political corruption is the exception rather than the rule, as it is in many countries.

Complex Adaptive Systems

The history of science has been largely one of reductionism, of describing real world entities in terms of their parts and the interactions of those parts. For example, atoms are made up of particles, which are made up of quarks; and animals are made up of organs, which are made up of cells, which are made up of organelles, etc. This method has been very successful in telling us what the world is, but it does not address the question of how it got that way.

Over the last few decades, an interdisciplinary group of scientists have been working on this problem. Under the auspices of the Santa Fe Institute [Waldrop, 1992], this group has developed theories about how the world generates complex adaptive systems (CAS). They have also developed new ways of thinking about complexity, and they have created highly persuasive computer simulations to test out their ideas.

Properties and Mechanisms

John H. Holland summarizes the following four properties of complex adaptive systems [Holland, 1995, p. 10-37]:

  • Aggregation. "...the emergence of complex large-scale behaviors from the aggregate interactions of less complex agents."
  • Nonlinearity. Interactions among entities are nonlinear if the number of possible interactions among them is the product of the ways in which they can interact. "...nonlinear interactions almost always make the behavior of the aggregate more complicated than would be predicted by summing or averaging."
  • Flows. "In cas, the flows through these networks [of interacting entities] vary over time... neither the flows nor the networks are fixed in time. They are patterns that reflect changing adaptations as time elapses and experience accumulates."
  • Diversity. In cas, a wide variety of entities compete in each environmental niche. "If we remove one kind of agent from the system, creating a 'hole,' the system typically responds with a cascade of adaptations resulting in a new agent that 'fills the hole'."
and the following three mechanisms:
  • Tagging. "...a pervasive mechanism for aggregation and boundary formation in cas." All of the entities that form a subsystem share a common tag, or flag, to identify each other as elements of the subsystem, and to control interactions both within the subsystem and with entities outside of the subsystem.
  • Internal Models. "... the organisms chances of survival are enhanced by the predictions, implicit or explicit, that the model entails."
  • Building Blocks. "In realistic situations an internal model must be based on limited samples of a perpetually novel environment. Yet the model can only be useful if there is some kind of repetition of the situations modeled." Building Blocks are repetitions of entities in a variety of contexts, "even though they may never twice appear in exactly the same combination."

These properties and mechanisms have been observed in a variety of domains known for their complexity: chemistry, biology, immune systems, nervous systems, language, technology, and economics. The conclusion being drawn is that complex adaptive systems exist because they have these properties and mechanisms - because they obey a law of nature that enables and causes them to evolve to ever greater complexity, to ever greater order. As Stuart Kauffman puts it: "Order vast, order ordained, order for free. We may be at home in the universe in ways we have hardly begun to comprehend." [Kauffman, 1995, p. 92] Why shouldn't ethics evolve to greater complexity and order too?

Fitness Landscapes

One of the ways that scientists have developed for thinking about complex adaptive systems is in terms of fitness landscapes. Abstractly, a fitness landscape is a multidimensional space in which a fitness value exists for each point in a base coordinate system.

In mathematical terms, the base coordinates are independent variables, and fitness is the dependent variable. But this is not a mathematical function; there is no equation that can be evaluated to determine fitness at each base point. Instead, the fitness value of each base point is determined by other factors, often of great complexity.

In the simplest case, we can plot a curve on a 2-dimensional plane defined by a base variable x and a fitness value z. For every x there is a value of z on the curve, even though we cannot say that z = f(x). Similarly, a system with two base coordinates, x and y and fitness value z, defines a 3-dimensional landscape. Connecting the fitness points to their neighbors, we see a fitness landscape, complete with valleys, plains, plateaus, saddles, peaks and pits. While spaces with more base coordinates can be conceptualized, they cannot be easily visualized or grafted; so we will continue to talk in terms of 3-dimensional landscapes.

The concept of fitness is abstract, and must be defined for each domain, but it generally means ability to act correctly in a complex, changing environment. Biological fitness is the ability to act correctly in a physical ecology. In this paper, ethical fitness is the ability to act correctly in a complex, changing social environment.

One of the features of fitness landscapes is that an entity at a given base point has no way of knowing or determining if it is at minimum fitness, maximum fitness, or somewhere in between. It can even be at a local maximum, a peak lower than the maximum peak that it might be possible to attain. The entity has no way to determine where a peak may be, except by going there. There is no mathematical function that can be evaluated, and no way to examine the fitness of neighboring points in the base plane.

Evolution is the search for increasing fitness. Evolution occurs because entities in complex systems exist on the edge between order and chaos. Some entities are frozen at local fitness peaks, resistant to all but the most drastic of changes. Other entities move radically on the fitness landscape in response to the smallest of changes, never able to attain any degree of stability. Between order and chaos is a realm where entities are stable enough to persist for some period of time, but unstable enough that small changes can move them on the fitness landscape. Natural selection then determines whether such a move is toward increased fitness.

As if this were not enough, complex environments contain many kinds of entities, all attempting to maximize their own fitness while their landscape is affected by all other entities. Thus, the landscape changes as the entities seek to maximize their fitness, making each entity more or less fit without any changes to the entity, itself. This is called coevolution, the cooperative evolution of all of the species in an ecology. Evolution, the search for fitness, is an attempt to blindly hit a moving target.

This brief introduction to fitness landscapes will have to suffice for this paper. See [Kauffman, 1995] for a much deeper treatment.

Internal Models

One of the key mechanisms of complex adaptive systems is the internal model of each entity. These models enable an entity to analyze its environment, predict environmental changes, and simulate the effects of the various actions it can take. If it survives, the entity's internal model learns, and enhances its ability to handle future events.

Internal models vary greatly in size, scope and complexity. A single-cell amoeba moves to areas of increased concentration of its food supply. It does this because its internal model has learned that increased food leads to survival. A low-ranked male gorilla learns to defer to the tribe's alpha male. He does this because his internal model of gorilla society has learned that this reduces the alpha male's violence. In human societies, people are generally polite to each other because each person's internal model has learned that politeness reduces friction and enables better relations.

Human internal models, however, go far beyond the needs of everyday living. With our big brains, we build (and learn) models of great sophistication. Some models relate to professional concerns, such as those of a medical doctor or a mathematician; some relate to art, aesthetics, logic, or science; some relate to religion; and some relate to ethics. While conceptually discrete, there are many interconnections among these models. New models are constantly being created and destroyed, and connections made or broken. Minsky calls these models agents and their aggregation a "society of the mind" [Minsky, 1985].

Not all models are of the same value in the search for fitness. It is all too possible to create models that have little or no relation to reality. An author creating a fictional universe can define the physical and cultural laws of a story. The characters of the story then follow the logic of those laws. This can be great fun, as seen in Tolkein's Lord of the Rings trilogy, or in Star Trek. Most people do not treat these model universes as anything but entertainment, but there is always a fringe group that gets caught up in them, and even holds conventions.

But what if a model is presented not as fiction, but as TRUTH that demands acceptance of a rigidly defined set of dogmas and laws? Without some criteria to winnow the wheat from the chaff, people seem willing to accept almost anything if it is preached loudly, persistently, and emotionally enough. Thus, we have all the models promoted by the great religions of the world, along with a great variety of pseudo-sciences and pseudo-religions.

At least in the domain of the physical universe, there is a criteria for evaluating the truth of propositions. Science requires the testing of theories (models) against reality, and it requires that everyone be able to do the same tests and arrive at the same conclusions. Otherwise, either the theory, the data, or the scientist is open to criticism. The result is self-correcting models in which we can have a great deal of confidence.

But science has its limitations. Many aspects of human experience and emotion have not been open to scientific examination. In these areas, the models of the world's many religions continue to hold sway. People adhere to a religion because it gives them answers to life's great questions, however unsubstantiated those answers may be.

Ethical Systems

Properties and Mechanisms

Is it reasonable to apply the ideas of Complex Adaptive Systems to a discussion of ethics? It is if ethical systems have the general properties and mechanisms of a cas.

To begin, we must decide the level of detail of the system that we want to consider. In preface to an example that applies cas ideas to economics Paul Cilulers wrote:

In order to frame our description, we have to decide what our 'distance' from the system will be: in other words, what level of detail are we going to consider? If we stand far enough away, we could only consider the activity of large financial institutions - banks, large corporations, even countries. Obviously a lot of smaller detail will get lost in the process. If we are going to examine the system in microscopic detail, we may have to keep track of the status of every individual penny. In that case we will run the risk of all meaningful patterns being obscured by the buzzing activity at the lower level. Let us, for argument's sake, frame the system in a way that will allow us to consider individual human beings - in their capacity as economic agents - as the elements of our complex system. [Cilulers, 1998, p. 5]

In this paper, we will also "frame the system in a way that will allow us to consider individual human beings" - in their capacity as ethical agents - "as the elements of our complex system."

Looking at cas properties:

  • Aggregation. With the world population currently exceeding 6 billion, we are clearly considering a system with a large number of elements that aggregate in more-or-less well defined groups; i.e., in religions and cultures. All of them live according to ethical principles, however varying these principles may be.
  • Nonlinearity. All of these people interact with each other in a wide variety of ways, locally and as members of groups. These interactions include competition, cooperation, and feedback loops. Paraphrasing Cilulers, without feedback there would be no ethical system. Who would act in anything but immediate self gratification if there were no consequences.
  • Flows. An ethical system is necessarily open, with new situations constantly requiring new ways of determining correct solutions. For example, new human fertility technologies raise ethical issues that have never before been faced.
  • Diversity. What is considered ethical in each of the world's religions and cultures varies widely. There is agreement in regard to some principles, but there is also a lot of disagreement. Is the death penalty ethical? It is in some cultures, but not in all.

And looking at cas mechanisms:

  • Tagging. Ethical tagging is indirect, in terms of religion and culture. A member of a particular religion or culture proclaims his membership by his actions. A Catholic makes the sign of the cross; a Jew refrains from eating pork, a Moslem prays facing Mecca. These actions also signify acceptance of the ethical principles of the group.
  • Internal models. Each person has his own internal model of reality. Minsky characterizes this model as a society of agents that are continually created, destroyed, reinforced or degraded in response to events [Minsky, 1985]. One such agent could be an ethical filter of the actions that can be taken in response to an event, such as a prohibition against killing for revenge. Another agent could be one that demands action in response to events, such as voluntary aid to the victims of an earthquake.
  • Building Blocks. It is not necessary for each person to individually create all of the ethical agents required for life in a culture. Most of these agents are learned from parents, friends and teachers; others are acquired through personal experience. These building blocks are combined, generalized, and specialized by religion and culture into easily assimilated packages.

Internal Models

As compelling as Minsky's agents are, it is not yet possible to tease out the precise set of building blocks required to implement a person's ethical system. Baring this, we must take another approach. Assuming that an ethical landscape overlays a person's internal model, then that model can be examined in terms of some base coordinate system - if we choose the base variables correctly.

In this paper, we will look at two variables, faith and reason. Other variables may also be at work, but they are outside the scope of this paper. Initially, we will assume these variables are independent and orthogonal in their effect on ethical evolution. In a subsequent section, we will discuss their relationship.

The Faith Variable

Let us put aside reason and focus on the everyday kind of faith that enables amoebas, gorillas and people to survive in the real world. This faith is confidence in an entity's internal models, and in the ability of those models to make correct decisions.

This definition of faith will undoubtedly seem strange to someone who is accustomed to thinking in terms of "Faith in God," but there is no real conflict. Such a person's faith in God is confidence in their understanding of the objective reality of God; that is, that God will act in the ways the person believes God acts. Whether God actually exists and acts in those ways is a different issue.

The faith axis measures the amount of confidence a person has in his internal models. Above the faith axis, there is a curve of ethical fitness, but there is no mathematical correspondence of faith with ethical fitness. The problem is to determine the amount of faith required to find the highest stable peak of ethical fitness.

At low amounts of faith, every individual is on his own in every situation. Without a minimum amount of faith in other people, everyone acts only for his own immediate benefit: social life is impossible.

Moving to a higher level of faith, people share a common set of assumptions about the world and how other people will react to a set of circumstances. Local communities, tribes, are now possible. The members of the tribe are organized in a pecking order (usually through violent means), dictating the rights and responsibilities of each member. Members act according to those standards. A tribe of nomads is a good example.

At a still higher level of faith, individual faith expands beyond the tribe to confidence that larger groups of people will make decisions in ways deemed to be correct, at least in the context of a specific community. For example, driving on the highways and streets of America requires a huge amount of faith that other drivers will act in predictable, lawful ways - especially given constant evidence to the contrary. Another example is the faith that scientists have in the scientific process; that theories are validated by experimentation and corrected as new information becomes available.

Beyond some point, faith becomes absolutist faith, the blind acceptance of dogma and standards of behavior. In this way, religions provide a fixed set of rules (for example, the Ten Commandments) that help people live at a high level of ethical fitness. This makes it easy for people to determine what their ethical standards should be.

But are these ethical standards truly at optimum peaks, or are they just local peaks? Given the number of major religions, it would be hard to argue that they are all optimum peaks. The problem is that organized religions make it difficult to evolve to higher ethical fitness; they are too far into the domain of order and stability. One need only consider the human sacrifices performed on a massive scale for centuries by the Aztecs. It took a brutal outside force, the Spanish Conquistadores and their missionaries, to stop that practice.

Extreme levels of faith also exist with low levels of ethical fitness. Over confidence in the teachings of a charismatic guru can lead to tragic results. The Jamestown and Branch-Dividian suicides are good examples. Similarly, excessive faith in political demigods can drive whole countries into barbarism. Hitler and The Holocaust are a prime example.

The Reason Variable

We defined faith as confidence in one's internal model of the world. Reason is the ability to form judgment and draw conclusions and predictions from that model. For complex entities in a complex environment, this becomes increasingly difficult, requiring higher orders of abstraction, structure, and connectivity. The reason axis corresponds to an entity's ability to use, and change its internal model. However, there are well known limits to reason. Kurt Godel proved that in any logical system there are statements that can be made that cannot be proved within that system.

The reasoning performed by an amoeba in its search for food is implicit in the blind reactions of its internal model to external stimuli. If we could tease the model apart, we would likely see logical rules in operation, of the form "IF food concentration increases, THEN move in that direction." This is static, physical reasoning learned over many generations; clearly not dynamic, symbolic reasoning.

A low level male gorilla must take into account many physical and social factors, and somehow factor in his own goals for food, status and access to females. We now know that gorillas are capable of a limited form of symbolic communication, and the manipulation of symbols. We would not expect a gorilla to be able to state any rules of logic, but gorillas clearly use reasoning in their everyday lives to determine correct behavior.

The internal models of primitive humans are full of explanations about why the world is as it is. Many are based on generations of accumulated experience and learning. Others are based on religious traditions that include multitudes of gods, superstitions, and rituals. Empirical and religious knowledge are intermixed, forming a single model from which ethical conclusions are drawn. There is no criteria for determining which results from their internal models lead to ethical peaks, and which lead to ethical pits. Altruism and cannibalism are both found.

The internal model of a modern, well-educated person goes far beyond those of a primitive tribesman. Thousands of years of thinking has trimmed the roster of gods, and eliminated many superstitions. Formal logic, mathematics, the sciences, and a variety of philosophies have been developed that make thinking easier and more reliable. And, we now have computers, powerful tools for creating formal models and simulations.

These tools allow us to see the world in much broader terms than has ever been possible. Who can forget the view of the Earth taken from the Moon by Apollo astronauts? We are one small planet and one humanity, however much we fight among ourselves. There is no telling how far new ways of thinking will take us in our drive for ethical fitness.

However, reason can also be misused to draw conclusions from internal models that are fundamentally flawed. There are numerous parasitic models that continue to plague us. The pseudo-sciences disguise themselves as rational models, and claim to be able to tell us how to live. Some examples are astrology, demonology, mysticism, and the paranormal in its many forms. We cannot ignore these parasitic models; all of them use reason to drive ethical fitness into deep pits. We must follow Carl Sagan's lead and seek to understand their position in and effect on the ethical landscape [Sagan, 1995]. High levels of reason can produce both peaks and pits of ethical fitness.

What if we push reason to some extreme of absolute reason, where people are able to always make correct decisions for every situation? This was the premiss of the novel The World of Null-A [van Vogt, 1945]. It is an entertaining story, but the complexity of the real world precludes any such scenario. If a game like chess can have an almost infinite number of solutions, how much more complex is the whole world. The map is truly not the place.

A person can ascend to a higher level of ethical fitness by increasing his repertoire of thinking skills, and applying those skills to the evaluation of his internal models. But if the models are bad to begin with, the results of all that thinking will not be any better. The computer programming saying for this phenomenon is "Garbage in, garbage out." Skepticism is a necessary precondition to ethical evolution.

Faith and Reason

So far, faith and reason have been examined as if they are independent, unrelated variables. Faith and reason have been discussed in terms of the models people use to determine correct action. It would be nice to be able to say that faith and reason are directly related; that faith and reason are interdependent; that increases in faith necessarily lead to increases in reason, and increases in reason necessarily lead to increases in faith. But reality does not appear to be that simple. Extremes of faith result in diminishment, and often abandonment, of reason, and vice versa.

The ancient Greek philosophers brought reason to a high peak, while maintaining faith in their many capricious gods. They based their ethics on telos, the goal of fully realized human nature. The Hebrews and the early Christians focused on faith in a single god, and reoriented ethics to what that single god wanted of imperfect human nature. The problem, of course, is reliable knowledge of what God (one or many) wants, as opposed to the politically motivated wants of a dominant religious hierarchy.

There is a long tradition of scholarship in all of the world's major religions, and there have been many attempts to use reason to justify faith. The results have been elaborate but ultimately founded on tautologies, circular reasoning that requires faith to reason about faith.

During the Renaissance, new ways of thinking about the world led to the development of science, to the view that we should test our models against the real world, and correct those models as needed. This philosophy was seen as a direct attack on religion, but it only questioned secondary aspects of the predominant religious model.

Does the Sun circle the Earth, as was then taught as dogma, or does the Earth circle the Sun, as deduced by Copernicus? Is the number of species of plants and animals fixed for all time, as claimed by biblical fundamentalists, or does a process of evolution create and destroy species, as theorized by Darwin? Does the human species have a central role in the universe, or is it just an insignificant species living on an minor planet on the edge of an insignificant galaxy? The traditional, religious answers to these questions have mostly given way to scientific explanations, although there remain stubborn groups of fundamentalists mentally frozen in static faiths.

For these questions, faith has shifted from religion to science, from confidence in dogmas preached by religions to confidence in the processes and values of science. But there are other aspects of a religious model that are not subject to reason and experimentation; that must be accepted or rejected by each person, either individually or as part of a community.

Is there a god? Is God a personal god? Is there an afterlife? Is prayer in any sense effective or necessary? The answers to these questions are important to many people, regardless of the fact that there is no way to answer them using any of the tools of science and reason. They provide patterns for daily life, build communities, inspire great art, and comfort people in times of personal distress, none of which science and reason are very good at doing.

No longer is there only one source for answers to the problems of living ethically. Educated people must decide to what extent faith and reason should be applied to each situation. Unfortunately, the answers to the problems of living correctly provided by religion and reason are often in conflict. For example, some religions say that abortion is always wrong, while reason (in the face of the global population explosion) says it is acceptable, and while practical living says it is repugnant but sometimes necessary. The search for appropriate standards is never easy.

Ethical Evolution

Corresponding to each point in the faith/reason base plane is a point in the landscape of ethical fitness. The problem of ethical evolution is to search the base plain in a way that finds ethical peaks and avoids ethical pits. Because ethical systems are complex, this search occurs automatically. Situations are encountered and actions taken day by day, century by century. Slowly but surely, standards are found and adopted.

At any given time, there are people and groups at different points in the ethical landscape; from saints to serial killers, from humanitarians to hate-mongers. Good and evil, saints and sinners, exist not as absolutes, but relative to the ethical landscape of a culture. Individuals and groups compete, cooperate and provide feedback to each other. Over time, the ethical landscape slopes upward. Paraphrasing Kauffman, we have "Ethics vast, ethics ordained, ethics for free."

Personal Ethical Evolution

A person searching for a peak of ethical fitness must be willing to question the assumptions of his internal models. Faith and reason can be manipulated in this search. This is done by changing the underlaying internal model, and adjusting the amount of faith and reason that is required. Some changes require the use of advanced reasoning skills, and others require additional faith. A healthy amount of skepticism is required.

An important idea from the sciences of complexity is to be at the edge of order and chaos. This is at a point in the base plane where relatively small changes in a person's internal model can cause jumps to other points that may have a higher level of ethical fitness. A person rigidly locked into a dogmatic model cannot conceive of any possible reasons for change, and therefore, cannot grow or even conceive of wanting to grow.

One possibility is to jump from the ethical system of one religion to that of another religion by adopting the model of the new religion. This is a difficult thing to do, but it can be done given education and examples to follow. For example, a Christian might study the Koran and admire certain Moslems enough to adopt their religion, or vice versa. (Conversions by the sword are also effective, but of questionable ethical fitness.) A problem is that religions tend to be all or nothing affairs. The more elaborate and dogmatic the religion, the harder it is to blindly make a leap of faith or to escape from the religion.

A willingness to distinguish between the core dogmas and the peripheral dogmas of the religion is required. The core dogmas are those that affect ethics, and the peripheral dogmas are those that do not. In general, belief in life after death, whether Christian heaven and hell or Buddhist reincarnation, is often used to justify ethical standards. On the other hand, peripheral dogmas, such as virgin births, angels, prophecy, and miracles, add a rich texture to a religion, but do not affect ethical standards. As long as the core dogmas are acceptable, conversion to a religion is possible.

Another possibility is to replace a religion with a personally created internal model, one that evolves as a person's view of the world matures. This takes a lot of work, both to sift through the person's current religion (and other religions), and assemble a new model. Without community support, this can be difficult to accomplish. Religions have always been really good at educating, encouraging, and policing their members. The loss of religion does not necessarily mean the loss of ethical fitness. It is the person's internal model that has changed, not necessarily his ethical fitness.

A special case is a jump from a religion to a scientific view of the world. Science is not a religion; it is not a fixed, dogmatic, absolute set of beliefs. Science is a dynamic, self-validating, self-correcting way of understanding the world as it actually is rather than as we might want it to be. Physicist Richard Feynman remarked that it is often impossible for a person trained in science to retain the religion of his youth [Feynman, 1988]. It is not so much a rejection of that religion as the adoption of the philosophy, processes, and community of science; there is no supernatural to muddy up the works!

A final possibility is to adopt a religion when there previously was none. There are many people who do not belong, and have never belonged, to any religion. Their internal models are initially learned from their parents and teachers, who have a variety of internal models of their own. If these models are of low ethical quality, then the person's model may be as well. The propagation of criminal behaviors, such as the sexual abuse of children, from generation to generation is well known. But a person does not have to adopt that model; it is possible to learn from the wider community, and to evolve ethically.

Community Ethical Evolution

No man is an island. This cliche applies to ethics as well as to other areas of human life. A person searching for an ethical peak does so within the context of a community, actually numerous communities: family, religion, locality, nation, local culture, and global culture.

Some ethical standards are formalized as laws enacted and enforced by various levels of government, but ethical behavior goes beyond the requirements of the law. There can never be enough laws to cover every possible set of conditions. Instead, individuals must act according to general rules at the roots of their internal models. Most often, a person's conscience is formed through family modeling and in participation in a religion.

In some societies, a single religion is overwhelmingly dominant, and generally intolerant of other religions. Catholic Spain in the Fifteenth century, with its Grand Inquisition and torture chambers, is a good historical example. Today, Iran and Afghanistan are good examples of religiously dominated societies in which people are executed for heresy or blasphemy. For any individual in such a society, ethical evolution is extremely difficult (and risky). Frozen deeply in the realm of order, it is impossible for the society to evolve ethically, without some avalanche of revolution.

The opposite kind of society is one in which there is little religion. After more than seventy years of Communist rule, in which participation in any religion was actively discouraged, Russia is ethically in the pits. Its government is corrupt, black markets dominate the economy, drug and alcohol use have skyrocketed, and kleptocrats rule the land. Something is needed to ethically stabilize Russian society, to move it out of the realm of chaos. The Russian Orthodox Church tries to fill this void, but it's models are too frozen to appeal to many people. While there are certainly local ethical peaks, Russian society as a whole appears to be at a low, chaotic level.

Between these extremes, other societies have ethical landscapes that have successfully evolved to higher levels. A good example is the United States of America. Initially populated by deeply religious Christian sects, wave after wave of immigration from other countries has exposed the population to a wide variety of religions. These religions have competed for members, and they have interacted in numerous ways in support for, or in opposition to, various national trends. For example, in the 1960s, racial discrimination was finally seen as a great injustice. People from many religions gathered to protest and change the attitudes and laws that enabled discrimination. Today, racism still exists in many forms, but it is generally viewed by most americans as wrong.

The ethical landscape of the overall American society has clearly risen over the years since the American Revolution. The dominant Christian view of man as master of the world, has given way to man as caretaker of the world, and man as part of the world. And intolerance of others for reasons of religion, race, or sexual orientation has given way to the celebration of human diversity. Of course, not all people buy into these new standards, but enough people do to allow us to claim progress. Examples of regression to lower levels can also be cited, such as the amount of violence, drug use, gambling, and pornography, but there is a growing awareness that they are harmful to human life; problems to be dealt with.

It appears that it is this diversity in American society that enables ethical evolution. With many stable religions interacting and competing, there is a general ratcheting-up of the American ethical landscape. But there is no guarantee that everyone will join in, or that progress will continue.

Each change to a society's shared internal model takes it to a new point in the ethical landscape, which can be at either a higher or lower level of fitness. Nazi Germany should serve as an example of what can happen when the dominant model of a society falls into an ethical abyss. A Christian society known for its high culture quickly descended into militaristic barbarism at the cost of millions of lives. The only effective guard against this is pluralism.


It is often asserted that religion is the necessary foundation for ethics. But religion is only one type of model that affects ethics. The new science of complex adaptive systems provides tools for a more complete analysis. Fitness landscapes and internal models were used in this paper as a way to conceptualize the ethical landscape of faith and reason. Religion has an important role to play, but not an exclusive one. A wide variety of ethical fitnesses can be seen in the relationship of faith and reason. Ethical evolution happens because ethical systems are complex and adaptive.


[Cilulers, 1998]
Cilulers, Paul, Complexity and Postmodernism: Understanding Complex Systems, Routledge, New york, 1998.
[Feynman, 1998]
Feynman, Richard, The Meaning of it All: Thoughts of a Citizen-Scientist, Perseus Books, Reading, MA, 1998.
[Holland, 1995]
Holland, John J., Hidden Order: How Adaptation Builds Complexity, Addison-Wesley, 1995.
[John Paul II, 1998]
Pope John Paul II, Fides et Ratio, a papal encyclical, 1998.
[Kaufmann, 1995]
Kauffman, Stuart, At Home in the Universe: The Search for Laws of Self-Organization and Complexity,Oxford University Press,New York, 1995.
[Minsky, 1985]
Minsky, Marvin, The Society of Mind, Simon & Schuster, New York, 1985.
[Sagan, 1995]
Sagan, Carl, The Demon-Haunted World: Science as a Candle in the Dark, Random House, New York, 1995.
[van Vogt, 1945]
van Vogt, A. E., The World of Null-A, Berkley, 1945.
[Waldrop, 1992]
Waldrop, M. Mitchel, Complexity: The Emerging Science at the Edge of Order and Chaos, Simon & Schuster, New York, 1992.

Sur l'évolution des éthiques


Désolé, pas encore traduit.