Morality and Evolution 1: The Milvian Bridge

[Morality and Evolution 1 2 3 4 5 6 7]

A while back I gave a talk to a group of theologians on the question of Darwinian accidents. It had no ethics content. The first question I was asked was “If you are an atheist, how can you have moral rules?” Like many others who talk about Darwin and evolution, I have been asked this a lot, and my answer is always the same:

“I am an ape. That is what apes do.”

Social apes (arguably all the apes) are evolved to function in social groups and norm-following is a crucial aspect of this. It would be remarkable if humans, who evolved from the other apes, did not follow group norms, not that they do.

Ethical philosophers (I mean, philosophers who do ethics, not good people who do philosophy) call the view that all moral content comes from a God or Divine source as the Command Theory. It seems to be the default view in western nations and probably many others. It boils down to the following claims:

  1. Moral values are absolute and real
  2. If you aren’t told to live by these values by an authority, you will act savagely and horribly.

Whether or not moral values are real (a view known as, obviously enough, “moral realism”) or not (a view called “error theory” on the basis that it is just an error to think moral values are real), it is the second claim that is seriously in question.

It appears to imply that we are all sociopaths at best and psychopaths at worst, and that without Divine Command and Threat of Punishment, we would all be rapists, murderers and thieves. My response is that I really do not want to be around people who, if they lost their faith for any reason, would default to such behaviours. Ordinary folk, however, will not. And atheists are ordinary folk, nearly all of the time.

So, since atheists are apes, and apes follow group norms, it appears the moral monsters here are the theists who think they’d automatically become rapists and murderers, etc., if they ceased being Christian. The real question is, “Why would a Christian need to think only theists are moral?” And the answer to that is “Because it makes being Christian (or theist) more important in their eyes.”

I am not an atheist in the philosophical sense, I’m an agnostic, and the question came out of the blue, but the question highlights the real concern people have about moral questions and evolution. If we evolved, what does this mean for our moral values?

Many people believe that we can have morals only if God commands them (and enforces them). Others believe that morals are what any rational being, human or not, would choose to enact. Still others think that morals are facts about the world. It’s confusing and complicated.

Still, I think the main issue is easy: are moral values real or are they constructed? If we evolved, many may think that moral values are constructed by organisms, and yet a good many thinkers believe that any rational evolved creature will trick onto the same moral values. If those aliens are coming, we might well hope they share our values, although that inference didn’t work so well in the film Mars Attacks.

It boils down to this: the world is thus and so. Does it include moral facts or not? If you say it does, whether these facts are facts about the natural world or about what rational agents will converge upon, then you hold moral realism. If you say it does not, and that moral values are constructed (that is, you think they are at most facts about us), then you are a moral antirealist. The term “antirealism” is basically just a word we apply to those who deny realism about some issue or other, and so it can have many different ways of being accepted.

Let’s start with moral realism. There is an argument, due to the ethicist Guy Kahane at Oxford, called an evolutionary debunking argument. It runs roughly like this:

• Something evolved through natural selection

• Natural selection tracks fitness

• Truth is not the same as fitness

• So if that thing evolved, it can be said to be fitter, but not true.

He calls this the problem of “truthtracking”. Consider this argument:

• We evolved our ideas about the world and God

• If the idea of God has evolved then it is a fit idea, but not necessarily true

• Hence we do not have a reason to think God is real because people tend to believe in God

That is, the idea of God is to some extent debunked by explaining belief in God as the outcome of evolution. Now apply this to moral values:

• We evolved moral values and our ideas about them

• If the idea of moral values evolved, then they are fit ideas, but not necessarily true

• Hence we do not have a reason to think that moral values are real.

Note that while an evolutionary debunking argument does not disprove the ideas that have evolved, it does tend to undercut our reasons for believing it to be true, because its success can be explained by increasing fitness not truth. Evolution tracks fitness, not truth as such (I’ll discuss whether that means we must think our ideas gained through evolution must be questioned later in the series on what evolution means).

Milvian

Paul Griffiths and I have called this a Milvian Bridge.[1] On October 28, 312, the contenders for the post of emperor of the Roman Empire, Constantinus and Maxentius, fought a crucial battle on the Milvian bridge over the Tiber in Rome, which Constantinus won, eventually becoming the emperor Constantine. The church chronicler Eusebius recounts the story:

Being convinced, however, that he needed some more powerful aid than his military forces could afford him, on account of the wicked and magical enchantments which were so diligently practiced by the tyrant … Accordingly he called on him with earnest prayer and supplications that he would reveal to him who he was, and stretch forth his right hand to help him in his present difficulties. And while he was thus praying with fervent entreaty, a most marvelous sign appeared to him from heaven, the account of which it might have been hard to believe had it been related by any other person. … He said that about noon, when the day was already beginning to decline, he saw with his own eyes the trophy of a cross of light in the heavens, above the sun, and bearing the inscription, Conquer by this. At this sight he himself was struck with amazement, and his whole army also, which followed him on this expedition, and witnessed the miracle.

Consequently he won, and later formally converted to Christianity, making it the official a legitimate religion of the empire. Now one might say that because he won, God supported him and thus caused the victory, but equally a more skeptical historian might note that most of his forces were Christian, and they won because they thought they had a divine mandate and fought harder than their opponents. One cannot argue from the success of the belief to the truth of the belief. The Milvian bridge will not cross over from success to truth.

The success of common moral values in human societies means only that those who hold them will tend to flourish in societies that reward those values. Does it mean that societies that hold those values are more closely approaching moral truth? Darwin tried an argument like this in his Descent of Man(1871).

No tribe could hold together if murder, robbery, treachery, &c., were common; consequently such crimes within the limits of the same tribe “are branded with everlasting infamy…” [I.93]

It must not be forgotten that although a high standard of morality gives but a slight or no advantage to each individual man and his children over the other men of the same tribe, yet that an advancement in the standard of morality and an increase in the number of well-endowed men will certainly give an immense advantage to one tribe over another. There can be no doubt that a tribe including many members who, from possessing in a high degree the spirit of patriotism, fidelity, obedience, courage, and sympathy, were always ready to give aid to each other and to sacrifice themselves for the common good, would be victorious over most other tribes; and this would be natural selection. At all times throughout the world tribes have supplanted other tribes; and as morality is one element in their success, the standard of morality and the number of well-endowed men will thus everywhere tend to rise and increase. [I.166]Darwin explains the origin of moral values as the outcome of the success of the groups that have “a high standard of morality”. This is a kind of natural selection, only of groups rather than of individuals. It is very similar to a view that dates at least from Aristotle, that moral values increase the “flourishing” of human societies.

However, this gives us no reason to think that moral values are real, only that they have a kind of instrumental value. Darwin’s target is a view, quite popular at the time he wrote and since, that what drives ethical decisions is individual selfishness. In the ethical philosophy known as “utilitarianism”, ethical choices should be aimed at maximising some sought good, like the avoidance of pain or the achievement of pleasure. To seek these out is a matter of personal value, not value in the world at large. Modern versions of utilitarianism, such as Peter Singer’s, however, treat the minimising of suffering and the maximising of pleasure as good in themselves for all beings, and so I don’t want to suggest that all utilitarians are selfish. Singer, for example, extends utilitarian values to all sensitive creatures (i.e., those that can feel pain), not just humans. Darwin’s nineteenth century targets are more like modern libertarians or neo-conservatives, sometimes called (wrongly) “social Darwinians” today.

A rival view in ethics is the Kantian view, derived of course from the late eighteenth century philosopher Immanuel Kant. He held that what is true for moral value is that any rational (or reasonable) person or moral agent would choose to do what they would want others to do to and for them. Thus, as we do not want to be killed, so we should not kill others. Kant’s view is a philosophical version of the Sermon on the Mount: do unto others as you would have them do unto you, also called the Golden Rule.[3] These are facts about the world: rational agents will always converge on the same solutions.

In evolutionary thinking, this is discussed under the heading of “game theory”. In the middle of the twentieth century, mathematicians like John von Neumann and others worked out a mathematics of social interactions. Starting with a problem called the Prisoner’s Dilemma, they assumed that rational agents are self-interested, and so developed a procedure for working out, as in cases where the Soviets and Americans faced off against each other with nuclear weapons, a way to predict what the competing sides would choose to do.

The Prisoner’s Dilemma works like this. Two criminals are being interrogated separately, and they cannot communicate with each other. They are both offered the following options: rat on the other and go free. If neither rat out, then they will both be convicted of a lesser offence and get less jail time, but if both rat, they both get the same large jail sentence. Their individual choices form what game theorists call a “payoff matrix” [3]:

  Prisoner A doesn’t rat (cooperates) Prisoner A rats (defects)
Prison B doesn’t rat (cooperates) Each serves 2 years Prisoner A goes free Prisoner B serves 3 years
Prisoner B rats (defects) Prisoner B goes free Prisoner A serves 3 years Each serves 10 years

Now rationally, neither should rat, since by doing so they would minimise the time anyone spends in prison. But A will know that B knows this, and prefers that he gets no jail time, not caring about the other. However, A knows that B knows this too, and will therefore rat. But if they both rat they get a big sentence, and so on. They have to defect to minimise their own personal outcome. There is no solution that they can come up with that doesn’t end up in both getting a long spell in jail. This simple game can be used to describe many kinds of “transactions”, from the interactions of DNA in evolution or in social interactions between economic or political actors. That’s in a single case. In many cases, though, the dynamics are more interesting. When “iterated” or repeated cases occur, it turns out that if most people are inclined to cooperate, then someone who simply does to another agent what that agent did to them previously, but who starts off by cooperating on the first “move”, can on average tend to do better than many other strategies. This strategy is called “tit-for-tat” after a computer simulation in which economist Robert Axelrod submitted just this strategy.

Later work, though, showed that even tit-for-tat can do very badly in a population of noncooperators. In short, if you live among Chicago gangsters, it pays to not cooperate as a default. The end result is that cooperators get eaten alive, and you end up with nothing left but selfish economists. Greed is only good if that’s the society you operate in.

So it seems that game theory won’t solve our problem. What succeeds depends on whether your group is largely Hawks or Doves. A militant tribe might do very well if the surrounding tribes are too nice, to return to Darwin’s metaphor. One might cynically see this playing out in the modern world.

Many evolutionary writers, such as Richard Dawkins and Michael Ghiselin, made a lot of the early results of game theory in the 1970s. Ghiselin wrote,

Scratch an ‘altruist’, and watch a ‘hypocrite’ bleed.

And Dawkins argued that we are driven by “selfish genes”, which he meant that all our moral altruistic choices are in the end based upon genetic interests. We help those who are more closely related to us, genetically. This doesn’t mean that individuals are not psychologically altruistic, but that the reason they are is because cooperation helps the genes that make them psychologically inclined to help others. We are genetically selfish, but psychologically altruistic. Many evolutionary psychologists hold to the contrary that we are “eusocial”, meaning inclined to be more cooperative than a game theory account might suggest we would be, because we evolved in small groups of related people, and that now, in a larger society of less related people, we have a moral module in our heads that misfires, so to speak. This is the “faculty of empathy” that Darwin discussed at length in the Descent.

So we haven’t been able to cross the Milvian bridge for morality yet. In the next post, I will discuss whether selection can track a kind of moral fact.

Note

Commenter Enon noted that Constantine did not make Christianity the official religion of Rome, but that Theodosius in 380 did. I had not read my Gibbon…

References

  1. Paul E Griffiths and John S. Wilkins, “When Do Evolutionary Explanations of Belief Debunk Belief?,” in Darwin in the 21st Century: Nature, Humanity, and God, ed. Phillip R. Sloan (Notre Dame, IN: Notre Dame University Press, In Press); John S. Wilkins and Paul E. Griffiths, “Evolutionary Debunking Arguments in Three Domains: Fact, Value, and Religion,” in A New Science of Religion, ed. J. Maclaurin and G. Dawes (Chicago: University of Chicago Press, 2013).
  2. Similar ideas are proposed in other philosophies, such as Jewish and Buddhist thought. Gautama Buddha (5th century BCE) said “Hurt not others in ways that you yourself would find hurtful.” [Udanavarga 5:18] The Jewish philosopher Hillel (1st century CE) wrote “That which is hateful to you, do not do to your fellow. ” [Talmud, Shabbat 31a]
  3. The payoff can be any amount, so long as the individual choice is that ratting is preferable to not ratting for each possibility.

33 thoughts on “Morality and Evolution 1: The Milvian Bridge

  1. What makes the command theory hard to get away from is the way that the psychological question of how we become moral individuals is intimately intertwined with the logically distinct question of how we decide what we ought to do. We are blackmailed into controlling our own behavior and observing social norms by the threat of the loss of parental love. That’s how we develop a conscience and learn to feel guilt over bad behavior. We become adults who want to be good, though wanting to be good, strictly speaking, is not a moral motive at all. I ought to want to promote the greatest happiness for the greatest number or perhaps act so that I can at the same time will the maxim of my act to be a universal law of nature. Being a good little boy or good is rather beside the point; and, for that matter, feeling guilt over bad behavior is rather illogical since the rational response to making an error is acting better in the future, not inflicting additional pain on yourself while boring your friends and relations. Anyhow, remorse and repentance are actually rather selfish since they are ways of making moral issues all about me, which is rather problematic, especially for those for whom ethical behavior has something to do with love.

    Of course the trouble with the Mr. Spock version of morality is the fact that people who don’t feel guilt are generally pretty scary. It has often been observed that the conscience or superego or whatever else you call the internalized parent would be a sadistic monster if it appeared by itself, but we run from those who don’t inwardly harbor such a creature. It may be that some of the hostility aimed at philosophers who think of ethics outside of the command framework arises from the perception that they are advising us to act like sociopaths, albeit virtuous sociopaths.

       0 likes

  2. “The Milvian bridge”

    Constantine was rather prone to visions (or his court does seem rather prone to retrospectively attaching them to make events memorable).

    By sheer coincidence he had an identical dream relating to Apollo before his sight is tempered by christian images and interpretation.

       0 likes

  3. The “evolutionary debunking argument” seems to be very much like an argument by CS Lewis and the “Evolutionary Argument Against Naturalism”. I confess that I am not favorably impressed by the argument. I don’t see how the argument does not apply with (at least as) equal force in the inverse direction: How does a supernatural or divine origin ensure truth, any better than an evolutionary or natural origin? Supernatural beings can be as unreliable as natural ones. Perhaps more so: Natural objects may be difficult to understand, but they cannot be devious.

       0 likes

  4. This post is a fascinating approach to the argument and questions you state. Conflict exists between the views of atheists and theists about the concept of mortality but aren’t moral values based on perception anyway? What we human refer to as “good” doesn’t necessarily mean that it is moral in terms of evolution, natural selection or the natural cycle of life.

    “The command theory” may be true or not but it is a convenient view to gain a lot of support in religion as religion is the perfect scapegoat people use to justify their supposedly “immoral actions”. They claim that as their religion views this action as moral and true therefore in any context this action is moral and true.

    Evolution and moral rules are connected but not a natural sense. It seems that moral rules have only been molded to fit the preferences and perceptions of people within society thereby intermixing with religious views as well as they hold great weight to the majority of the world’s population.

       0 likes

  5. The Prisoner’s Dilemma table does not seem right.

    Shouldn’t it be:
    Pay off for B if B doesn’t rat:

    - if A doesn’t rat : each 2 years
    - if A rats: A free, B 3 years.

       0 likes

  6. “because we evolved in small groups of related people, and that now, in a larger society of less related people, we have a moral module in our heads that misfires, so to speak.”

    I find this interesting as I would link the early development of ethnicity to the manner in which customary law (evolved to regulate and police small groups of related people) can be simply adapted to eradicate the status and economic and reproductive viability of outside groups.

    “A Briton is still a Briton even if he has a gold hilted sword.”

    Not my subject but I cant help shaking of the notion of a potentially strong link between ethics, morality and the development of ethnicity.

       0 likes

  7. It’s a common misconception that Constantine made Christianity the state religion, but that didn’t happen until 380 CE under Emperor Theodosius.

       0 likes

      1. “He also issued decrees that effectively made orthodox Nicene Christianity the official state church of the Roman Empire.”

        https://en.wikipedia.org/wiki/Emperor_Theodosius

        I picked this up from Professor Bart Ehrman, an expert in the history of early Christianity. (Like you and me, an agnostic on gods, an atheist on God. Unlike you and me, he really bought into the ‘born again’ lifestyle for a number of years and studied at the Moody Bible Institute.)

        He has written on misconceptions about Constantine and the Nicean Council on his blog. He also writes both scholarly and popular books about Jesus, early Christianity and the New Testament.

        A partial blog post on this issue is here (full blogposts are behind a paywall, all funds go to charity):

        http://ehrmanblog.org/constantine-christianity-2/

           0 likes

        1. I did buy into evangelicalism for a number of years. I even studied theology at the local Church of England theology college for two years before I ceased finding it credible. It was somewhat more mainstream than Moody, though. We even studied Hooker’s Ecclesiastical Polity.

             0 likes

  8. As always, John, clear (as can be in a confused debate), entertaining and presented with je ne sais quoi. I can’t wait to baffle the next overbearing command theorist with “I am an ape. That is what apes do at the Milvian Bridge.”

       0 likes

  9. The idea that you need a divine commander in order to have morals is not as common in my experience as it is in yours. But there are several other ways in which religious folk connect religion to morality.

    One is the idea that while pagans/atheists are indeed moral up to a point (after all, the Bible explicitely states this to be the case: Luke 6:32, Romans 1:14-15), members of favoured religion have supplementary morality. Often the idea here is not that you need a commander, but that you need a divine spirit inside you in order to be capable of morality on that level.

    Another is the preception that an atheistic universe would be discouraging, that without a divine being on your side you’d end up feeling that nothing good you do makes any difference to the world anyway.

    On the subject of absolute vs relative morality, I have for years divided the spectrum into: (a) naiive absolutism, which holds that the absolute moral laws not only exist but are finite and can in principle be written down in a book; (b) complex absolutism, which holds that morality is absolute but that the laws are not wholly knowable unless you are God, context being arbitrarily subtle and all; (c) pragmatic relativism, which holds that morality may well be absolute, but since it is unknowable we might as well act *as if* it is relative; (d) (e) various forms of moral relativism.

       0 likes

  10. John, thank you, great article series….

    “Darwin explains the origin of moral values as the outcome of the success of the groups that have “a high standard of morality”. This is a kind of natural selection, only of groups rather than of individuals. It is very similar to a view that dates at least from Aristotle, that moral values increase the “flourishing” of human societies.”

    “However, this gives us no reason to think that moral values are real, only that they have a kind of instrumental value.”

    John, this looks like objective reality about ethics. Saying this is not really moral realism looks odd, at least to me.

       0 likes

    1. As Michael said, moral values are real as values pertaining to that environment (and that species) which have instrumental value. But from instrumental value to the reality of objective moral values is a long leap. Suppose our social dispositions are selected such that we become something akin to Rand’s moral bastards because that really is fitness-enhancing, and cooperation is for the weak. In that case, it really would be true that selfish individualism would be “right” for that version of humans. So the moral values are just a reflection of the more successful rules or dispositions, and are not real as moral values.

         0 likes

      1. This model of advantageoues selfish individualism (moral bastards) could never exist in a continuous population of reproductive agents capable of contemplating philosophy. For example, any population of contemplative agents requires extensive parental nurturing of the young agents, which excludes likeliness of a population of contemplative agents with primary disposition of selfish individualism.

           0 likes

          1. How did I change the subject? If John meant no reference to a population of contemplative agents, then I missed his point and he can clarify that. But if he tried to propose a hypothetical population of contemplative agents who have a primary disposition of selfish individualism, then I did not change the subject but focused on a weakness of his proposal.

               0 likes

              1. I read some summaries and reviews of A CASE OF CONSCIENCE. I suppose that you refer to the sci-fi alien race of intelligent bipedal reptile-like creatures, the Lithians. They are a creation of Satan. These satanic spawn live in an atheistic utopia without crime, conflict, ignorance, and poverty. Each Lithina inherits moral knowledge from their father’s DNA. Satan spawned this race for no other reason than to piss off the established Roman Catholic church that officially teaches that such a godless race could never exist in peace and logic.

                This could turn into the basis for a great sci-fi movie, but I do not see this supporting your proposal of a perpetual race of contemplative agents with a primary disposition for selfish individualism who are generated by evolutionary mechanics. However, I will keep my eye out for the book at used book stores….

                   0 likes

              2. The Lithians are not presented as spawn of Satan, but that is how the protagonist interprets them, because they are moral but without emotions. Instead they cooperate because of rational reflection. It’s a nice conundrum. Great book by the way.

                   0 likes

      2. What would this say about beings which are not in a population which rose by natural selection. Intelligently designed and/or manufactured. Eternally existing. Imaginary (shmoos).

           0 likes

        1. What about beings which are in a population which was favored by national selection for cooperation in groups of a certain kind, and the opposite between groups of another kind?
          For example, groups of a certain size. Or groups of similar appearance. Or groups of similar culture.
          Would it then be “right” for of that population to have difference of cooperation, according to the group character?

             0 likes

  11. I don’t think you’re really fair to Divine Command Theory. The more reasonable versions of it would hold not that we behave morally because we believe in God and fear his punishment, but because God gave us a conscience. This would explain how even atheists can be moral paragons.

    I’m a moral anti-realist myself.

       0 likes

    1. If a moral conscience is divine, then Christians have no reason to claim that one can only be an oath keeper if one believes in a higher power (as Locke thought). As such they have no foundation for arguing that a moral society relies upon religious faith, as many do argue. It’s a conundrum.

         0 likes

  12. I propose that all moral facts come from God but not God’s command. For example, God has always known all moral facts. Moral facts apart from any particular circumstance are not created by God but always existed. In this view, there is no ultimate conflict between deontology and consequentialism.

       0 likes

      1. Well, to agnostics, I say, if God exists, I propose that all moral facts come from God but not God’s command…. And, yes, I have nothing but consequentialism for the strong atheist. At least, I have something to offer to most everybody :-)

           0 likes

  13. Thanks for the discussion and some of the roots of the Milvian Bridge. Do you have a post that traces the history of emphasizing that evolution tracks fitness and not truth further back? It is an obvious statement, but it is often surprising how long if takes to remind people of it. For example, in cognitive science — more specifically, study of perception — it seems to only have been introduced in the last 15 years with Hoffman’s interface theory of perception. I would be interested in seeing how the same idea has re-emerged in different fields.

    I also wanted to make some more technical comments on your game-theory section, since this is what I specialize in. I think this statement is misleading:

    So it seems that game theory won’t solve our problem. What succeeds depends on whether your group is largely Hawks or Doves.

    First of all, your preceding discussion was of PD, and Hawk-Dove is a completely different (and much more friendly) game that can sustain cooperation at a mixed-strategy equilibrium. But more importantly, although iterated PD is the oldest approach to forcing evolution of cooperation,it is far from the only and (in my opinion) not the most interesting.

    My favorite approach is to focus on spatial/social structure, since it is often enough to get cooperation to emerge. This can also be brought back to religion, and tension between ‘true’ and ‘fit’ ideas. For example, my colleagues and I recently showed how (under certain settings) spatial structure is enough for agents to evolve misrepresentations of objective reality that serve as a social interface for tracking inclusive fitness and promoting social welfare. You can <a href="“>link this back to religion, provide an explanation in an (evolutionary) game theoretic setting, and not need to rely on moral gods or threats of punishment (i.e. you can use this reasoning to trace back further in the history of religion than punishing Gods).

    So I don’t think that you should be as dismissive of game theory, especially the recent developments.

       0 likes

  14. Thanks John, I enjoyed your thoughts here. However, I wanted to point out that you are not quite correct about the meaning of “error theory” in moral philosophy. “Error Theory” is most commonly associated with John Mackie, who did reject moral realism. However, what really distinguishes error theory is its approach to moral language. Unlike earlier philosophers who rejeced moral realism (notably the logical positivists and the various noncognitivist theories that followed them, e.g. Hare, Stephenson, etc.), Mackie agreed with the moral realists that our moral language attempted to refer to real properties (he was a moral cognitivist).

    The earlier philosophers who rejected moral realism had attempted to understand moral language as non-propositional, as expressing an emotion or attitude, or as commanding or prescribing some kind of action rather than as describing reality. Mackie thought that moral language did try to describe reality, but did so by referring to non-existent properties and entities. So this is the reason why Mackie’s view is called error theory. He thinks there is a fundamental error in our moral language, which is that we make claims about entities and properties that don’t exist. We say that x is good, but “good” fails to refer, and so we fail to say anything true (on this account moral statements are by default false).

    Anyway, the proper contrast to moral realism would be moral anti-realism or some cognate, as it should properly include both cognitive and noncognitive opponents to moral realism.

       0 likes

Leave a Reply