Moral Trial and Error

by Mario Rizzo

The recent discussion-thread at the blog Coordination Problem regarding a Hayekian case for same-sex marriage got me thinking more generally about moral evolution.

In a market there is a process of trial and error. New products or methods of production come into existence. Some fail; others succeed. Some speculators make successful predictions of the future course of prices; others make mistakes. In general, the filter for these decisions is the profit and loss mechanism.

F.A. Hayek famously argued that the evolution of institutions, including moral and legal rules, follows a similar course, that is, trial and error. And yet the analogy with market processes is far from perfect. How do we view the trial and error process of moral rules? What is the filtering mechanism?

Right off, let me say that I do not have definite answers to these questions. I simply have some relevant thoughts.

We are always living during a process of institutional evolution. The process is literally unending. Despite the claims of some, there is no One Eternal Moral Truth. Even if there were eternal moral principles, for example, the rules that would come out of them would necessarily change as knowledge changed. This is because some principles are result-oriented. If we are enjoined to help the poor our understanding of what is likely to help the poor will change as, say, economic knowledge grows.

Importantly, moral principles can change, as they have. For example, there has been an evolution in the conception of justice. In earlier times justice was scarcely distinguishable from revenge. The tit for tat infliction of death and other destruction produced a downward spiral. In addition, guilt and punishment was at the level of the tribe and later of the family rather than the individual. Justice is a limitation of pure revenge. It is also more predictable, less arbitrary and less costly from a social perspective to administer. It has grown in the context of greater awareness of the importance of social cooperation and the need to use justice to further social cooperation rather than to destroy it through some form of “mutually assured destruction.”

I cannot here recount the many examples of moral evolution. The main point is that moral evolution has happened and will continue.

From the perspective of our own lives of relatively short duration we must realize that it will not be easy to distinguish between changing moral rules and simple violation of moral rules. In an important sense, the distinction will depend on what happens later. Moral rules must be violated for them to change. As time goes on, it will become clearer that people no longer hold certain behavior to be moral or immoral. Subjectively, they feel no guilt in the violation of the dying moral norm.

This is an important reason that we do not want to rigidify all moral norms by giving them the force of law. We want to let the waxing and waning of social sanctions to act now as a strong disincentive and then as a weak disincentive to putatively immoral behavior.

Fundamentally, moral rules and institutions develop from the bottom up. This is a sociological or cultural “fact” which I realize is inconsistent with the idea that they come ready-made and eternal from religion or God. But look at the evidence.

Just as many theists have made their peace with biological evolution, they ought to make their peace with moral evolution. The Hebrews of old favored revenge. The Jews and Christians of today have rejected it.

Is moral evolution good? This is a tricky question. In one sense, the standard by which it is to be judged is itself. In another sense, we might posit a meta-standard like social cooperation. Does the new norm promote or inhibit social cooperation? If so, then it is good; if not, then it is bad.

Why is social cooperation the standard (or perhaps the filter of the trial-and-error process)? Well that is a big story. Suffice it to say now that social cooperation is the sine qua non of just about everything we think beneficial to humankind.

40 thoughts on “Moral Trial and Error

  1. I ran the discussion on The Constitution of Liberty for Pete’s constitutional econ class last year and I was completely enamored by Hayek’s conception of social trial and error.

    I don’t think it is really possible to have a metastandard, as I think such standards are themselves nothing but the product of the same trial and error process. I would be interested in your “big story” for why social cooperation is your chosen standard, but I suspect that the explanation likely lies more in the social tradition of which you are a participant–of the ideas, arguments, and frameworks floating around the Austrian tradition in its modern form and as you have experienced it.

  2. I’m not sure social cooperation is the one and only criterion for good institutions. Highly cooperative people can nevertheless destroy themselves– Jared Diamond gives several examples.

    Re Adam Gurri “such standards are themselves nothing but the product of the same trial and error process.” Yes, institutions, including moral standards, determine what we think is good or bad. Still, we can study past societies that had different institutions and look at what happened to them.

  3. This is something about which I have thought a lot. I have published on it in NOMOI (pg 3), “From the Sensory Order to the Moral Order: Bridging Hayek to Hayek (Part I)”:

    Click to access NOMOI_5_2010.pdf

    And wrote extensively about it on my blog, most notably:


    I think Mario is right that social cooperation should be understood as, if not a metastandard, the very foundation of ethical behavior for a social mammal. It is another way of saying that the universe tends toward complexity, and that moral behavior helps people to live their lives according to that tendency. Ethical action makes for a more complex (not complicated) world, while unethical action reduces complexity. To make humans act in a less social or more animalistic manner is unethical.

    For me, there is little doubt that there is an evolving moral order. However, it is tethered to our evolved nature.

  4. I’ve shown how norms are selected over in the filter process of scientific advance as described by Thomas Kuhn in my paper on Kuhn and The Selection of Desiderata for theory choice linked at my Taking Hayek Seriously web site.

    A somewhat similar filter process could be imaged for competing moral communities — especially in the earlier prior period of smaller competing groups.

  5. In Morality, decision, and social organization: toward a logic of ethics, Karl Menger endorses ethical pluralism and a competition among moralities. The son or “our” Menger argues against the categorical imperative. He points out that is perfectly possible to have compatible groups of people who are not all following the same norms. Certain mixes work and others do not. Menger says:

    If various groups work under a common system at cross-purposes, then even some of the most important features of the underlying ideas may fail to materialize so that their effects may never be clearly seen or the basic plan fully judged. A system of several self-imposed regulations of various groups makes the effects of their diverse ideas manifest. Only on this basis have moral plans, removed from primary passions and from the happenstances of many struggles, the opportunity to prove themselves. Only in this way do ideas and their effects become comparable with one another. Only the organization proposed in the thought-experiments [of co-existing voluntary communities] opens the possibility of a contest of plans. (Menger 1974 [1932], p. 91)

  6. Roger Koppl–
    Re “…perfectly possible to have compatible groups of people who are not all following the same norms. Certain mixes work and others do not. ” That’s a great way of looking at it.

    It is useful to separate the two points, namely whether people in a society subscribe to the same norm and whether that norm makes for social cooperation. A society of thugs may all believe in thuggery and apply this norm in mutual destruction.

  7. Roger and chdemkurdas:

    Hayek’s notion of social trial and error involved small minority groups trying out something new, and a handful of those things eventually end up gaining wider adoption.

    I always thought that corollary to this that he never explored was that there are some things that end up getting adopted beyond the initial group but never get so big as to be “mainstream”. In short, there are varying levels of adoption; and social trial and error in a free society leads to more, rather than less, heterogeneity.

  8. A crucial part of “trial and error” is understanding when an action is indeed erroneous.

    There’s a strong tendency, when discussing morality, to declare that the error in a particular action is one of insufficient degree, or time, or stringency, or whatever. In other words, “Stay the Course, give it time and it’ll work” or “Tragedy of the Commons, this is failing because not enough people are doing it”.

  9. This would seem to add a dimension to “moral relativism” – not just place, but time.

    That this is valid – even required – seems obvious after a moment (or two)’s reflection.

  10. DD:

    From the perspective of what’s being discussed, “error” really just means failure to gain widespread adoption.

  11. Does private property aid social cooperation? That debate has raged for 300 years. Socialists argue that it impedes, and they have won the hearts of the majority. That’s why we have been experimenting with destroying property rights for the past century.

  12. “Ethical action makes for a more complex (not complicated) world, while unethical action reduces complexity.”

    Why in the world would that give moral injunctions any force? Why isn’t the crushing response of Hitler or Stalin to this understanding of morality simply, “But *I* like simplicity”?

  13. Then they are oing against the natural tendecy on the univrse, the natural end result of which is nihilism and destruction.

    What gives any moral injunctions other than “because God said so” any force? I’m interested in scientific explanations.

  14. Elliott Sober’s book _The Nature of Selection_ is very helpful in thinking about the logic of how selection processes work.

    Note that there are many different variations on how to implement a selection filter process.

    The immune system has a selection process.

    Sexual selection is different than natural selection.

    Likely, there is a selection process at work over competitive neural networks — e.g. as suggested by Gerald Edelman.

  15. Troy: “What gives any moral injunctions other than “because God said so” any force? I’m interested in scientific explanations.”

    That is the question with which philosophers have struggled for centuries. Greg is the philosophy expert, so correct me where I’m out of line because I haven’t revisited the stuff in years, but the great atheist philosophers from Nietzsche on wrote that without God morals don’t and can’t exist.

    Of course, they meant something very different from what most people mean when the use the word “morals” today. By morals they meant objective morality determined by reason or revelation and universally applicable across cultures and time. What we call morals today people used to call mores (with an accent over the e), or consensus opinion but without moral authority. They don’t have authority because no man or group of men has moral authority over others.

    So if we define morality as consensus on what improves social cooperation, we have conceded the points of the great atheist philosophers.

  16. Greg, what do you think of Adam’s comment on what Hayek meant by experimentation?

    I’m not sure. In Fatal Conceit he applauds religion for persuading people to follow principles for which no short term benefits could be discovered.

  17. Roger: I guess what I should have said was that “error” means failure to gain widespread adoption that persists in the long run.

    On what gives moral notions any force, with or without God–I have a few thoughts on that. To summarize, there’s no more or less “force” to moral notions with a God than without one. The fact that a God exists and commands us to stick to certain rules does not mean that anyone has to obey Him.

  18. Adam: “nothing about these arguments actually compels anyone to obey moral rules.”

    You’re right. Only immediate force will compel people to obey rules they don’t like. I think that’s one reason that the American version of karma is so popular. People think they can see immediate consequences for good and bad actions.

    But that’s not what philosophers meant by morality. The issue of enforcement is a practical matter of government. They were concerned first with determining what is moral and isn’t using reason so that is was objective and having authority so that it would be universal.

    Objectivity and universality are important because if you’re going to punish someone for being immoral you have to have some authority to do so. Otherwise you have no right to punish someone other than to exclude them from the club.

    Without authority, morality is nothing but a home owner’s covenant.

    Social cooperation was actually one of the fundamental assumptions of natural law theory. But it was hemmed in by other assumptions about God and creation so that it was useful. Without those assumptions, social cooperation becomes such a vague term that it is little more than an empty vessel into which anyone can put any pet idea.

  19. In other words, before you can worry about application and enforcement, you have to know what you’re going to apply and enforce and why.

  20. Well, more that any argument for why you should listen to what God says can just as easily be applied to why you should obey X principles, without any God being involved. There’s no logical connection between the existence (or nonexistence) of a God and the need to obey moral rules.

  21. Adam, that’s true, but it has a lot of bearing on what is moral. In the traditional definition of morality, morality had to have universal authority, which meant that it could only come from god.

    Under the new definition of morality anyone can assert that something is moral and if they can get enough like minded people to agree then for them it is moral. It may not be moral for other people who disagree but who cares?

    For me, the acid test or morality is the real life example of people in the East who murder infant baby girls because they want male children. What would your “morality” say to them?

  22. Roger,

    I think the problem is that you’re coming at this from a philosopher point of view. When you talk about “the traditional definition of morality”, you are talking about morality as traditionally defined by philosophers.

    Non-philosophers–that is, the vast majority of the human race in the present and through all of history–have never had to rely on explicit definitions of morality or justifications based on universality. Something just was or was not considered right or wrong.

    You say “Under the new definition of morality anyone can assert that something is moral and if they can get enough like minded people to agree then for them it is moral.” But this is how it has always been, from the perspective of someone observing humanity from above. The only difference is that most people believe their particular moral code has some endorsement from God or some basis in Reason and science, and so therefore people who believe differently are just ignorant, evil, or blind to the will of God.

    But if you’re trying to explain what morality is, there’s little more you can say than that it is an emergent phenomena among different cultures, countries, and communities. You can talk about the underlying biological basis–and talk of the moral sentiments that everyone shares, as Smith and Hume did. You can say that variation between cultures won’t go beyond certain parameters set by biology.

    But unless you yourself hold some faith in God or rationalism, there’s no point in talking about universality as envisioned by those ideologies.

    For me, the acid test or morality is the real life example of people in the East who murder infant baby girls because they want male children. What would your “morality” say to them?

    I think it’s abhorrent, but I think a lot of things that are going on in the world are abhorrent. That doesn’t really stop them from happening.

    I also wonder if those people really believe that what they are doing is moral; if they would try to justify it on moral grounds or if they’re just doing it for self-serving reasons and are quite aware of it. But that’s another matter.

  23. Adam, you’re right. I am looking at it from a philosophical point of view. But the whole point of philosophy used to be to bring some kind of order out of chaos. Lots of people claim that their morality is from God. Philosophy used to try to come up with objective means of sorting out the claims.

    Modern moral philosophy did not solve the problem of authority; it merely redefined morality as needing no authority. It’s not particularly clever to end an argument by simply redefining the terms.

    As for the anthropological issue of how humans actually arrive at what they refer to as morality, I think institutional economists would argue that historically it has come from religious beliefs. Someone convinces other that they have heard from god.

    The really curious thing is how little the basic morality (murder, theft, fraud, etc.) has changed over the millennia. We simply don’t know anything about prehistorical morality at all. That’s pure imagination. Within written history we have a pretty good record and morality tends to vary within a narrow channel.

    As for those who kill infant girls, I can only take their word for it that they think it is moral, or if they think it is immoral it is a very minor violation. But I was hoping for something more than personal disgust. What objective or rational grounds would you use to explain to them that the practice is more than personally disgusting; it is immoral.

  24. Philosophy used to try to come up with objective means of sorting out the claims.

    And I think philosophers were always mistaken in their assumption that an objective means exists.

    I disagree with you that morality traditionally has come from religion. Rationalists are fond of pointing out all of the contradictions inherent in the way religions are practiced vs. the way they are preached; I believe that a lot of the things we associate with religion often predate the specific religion.

    I would even argue that many moral developments that occurred while one religion was the dominant one in an area could nevertheless be said to be orthogonal; to have developed in response to local conditions and changing times rather than in response to the prevailing doctrines preached by clergy.

    Hayek’s argument in the Fatal Conceit, for instance, is not that religion in itself encourages people to buy into tradition, but that the kind of people who stick with traditional religions are the kind of people who defend traditions in general, beyond the specifically religious sort.

    But I was hoping for something more than personal disgust. What objective or rational grounds would you use to explain to them that the practice is more than personally disgusting; it is immoral.

    Sorry to disappoint you, but I don’t believe that any such objective or rational grounds exist.

    I would appeal to their sense of shame at murdering a helpless child, their own flesh and blood, just because the gender happened to be less convenient for them. There isn’t much more you can do beyond that.

    If I was religious I might say that God would judge them for their actions, but if they weren’t religious too I hardly see how that would be any more likely to sway them than my argument based on sentiments.

  25. Adam, “I think philosophers were always mistaken in their assumption that an objective means exists.”

    So you agree with the great atheist philosophers. That was my main point.

    Modern thinking on morality is nothing more than getting together with people who have similar tastes and ridiculing those who don’t.

    If people thinking killing children is OK, why would they experience any shame? The Spartans practiced that, too.

    “Hayek’s argument in the Fatal Conceit, for instance, is not that religion in itself encourages people to buy into tradition, but that the kind of people who stick with traditional religions are the kind of people who defend traditions in general, beyond the specifically religious sort.”

    I don’t think that’s what he meant at all. He was explaining how it was possible to maintain certain principles for which no one could defend in the short run.

    Considering that relatively wide spread atheism atheism (less than 10% of the population) is a recent development in history, I think you underestimate the role of religion.

    The orientalist Bernard Lewis says the same thing in one of his books about the spread of Islam.

  26. The issue brings up another one – epistemology. We know that in economics one can find support in history for any economic theory because the data are so vast and contradictory. So as Mises wrote we have to approach history with sound theory in mind in order to not get lost.

    I suggest that the same thing is true of morality. Whatever our theory of how people form morals, we will find support for it in history. In order to not get lost, we need a sound theory of morality, a praxeology of morality, before we can investigate history.

    Actually, that may have been what the philosophers were trying to do. I think that was the goal of natural law philosophy.

  27. Yes but that only takes the problem one step back. Whatever “sound theory of morality” we want, we can find support for in history–or, if you like, other people whose prior beliefs are already sympathetic to it.

  28. We have the same problem in economics, though. Austrians have sound theory with which to interpret history, but we haven’t succeeded in convincing many people to follow.

  29. 1. How did we arrive at that sound theory, other than by reading things that we personally found convincing?

    2. I think that academic theories/schools of thought, like traditions of morality, are emergent process-driven phenomena.

  30. The same way we have arrived at sound economic theory – observation of how humans act and the use of logic. That’s what Aristotle and Thomas Aquinas did.

  31. Right, and that took them to–the equivalent value theory of trade, and the just price doctrine. Not very compelling examples.

    Truth is not manifest–it’s not as though you can simply observe people and the explanations for their behavior will just come to you, logic or not.

    Just like moral traditions, there is a social trial and error that occurs with theories. For subjects like economics, there isn’t a very strong feedback against inaccuracy–it’s not like I can go out and apply Austrian theory in a bunch of lab trials in easily observable ways. So the success of theories in gaining adoption tends to depend, at least in part, on factors that are orthogonal to their accuracy.

  32. Actually, equivalent value came from Aristotle. Aquinas and the scholastics rejected it. The search for a just price led the late scholastics of the School of Salamanca to conclude that just prices exist only in a free market and that gave birth to modern capitalism.

    The popularity of ideas is often in inverse proportion to their truthfulness. If I cared about being popular, or sitting with the majority, I certainly wouldn’t be interested in Austrian economics.

  33. Forget popularity; I’m talking about adoption among groups of any size. There are many academic schools the size (that is, as popular/unpopular) of the Austrian school that subscribe to theories which contradict the theories of the Austrian school.

    What’s now called the Austrian school was in fact more prevalent among the mainstream of economics early in the 20th century–the ideas have grown no more or less accurate with its upswing and decline in popularity.

    The vital question, as Hayek and Sowell would say, is what the feedback mechanism is, and how focused is it? Is there really anything that punishes inaccurate theories in a field like economics? If there is, how big a range of inaccuracy does it allow before it kicks in?

  34. I guess I don’t understand you. Are you arguing for empirical support?

  35. I’m saying that ideas aren’t validated through evidence, because there is never enough evidence to logically prove anything. Rather, ideas are validated through processes.

    How good we are at validating particular ideas depends on the subject matter of the idea and the quality of the process it goes through.

    For instance: if an idea involves a prediction which can be tested in a hard-and-fast, easily observable, easily replicated way, it’s more likely to be punished for inaccuracy than an idea which cannot be tested in ways that are easy to observe and replicate. A simple example: claiming that a pencil will fly to the ceiling when I release my grip on it is easily invalidated.

    The process also matters, though, since it would be prohibitively costly to ask every individual scientist (or anyone who wanted to believe something accurate) to replicate every test of every theory that their particular corner of science relies on. So in any system, the majority of participants aren’t going to test most theories most of the time, no matter how easy it might be in any individual case.

    So the fact that people are replicating the tests and communicating their results is essential for the whole system to work. Scientific knowledge advances by trial and error only if the process of science is dependable at rooting out the errors.

    Now things like persuasion, preconceived ideas, and so on all have a part in the spread of any ideas, regardless. But the tighter the feedback mechanism is–the easier it is to root out errors–the less these more subjective factors matter in the long run.

    Economics is a subject where the feedback mechanism for accuracy is not very tight; tests are difficult, conditions hard to replicate, and errors, therefore, hard to root out.

    That’s all I’m saying.

  36. I think I understand. It seems to me that Hayek’s method was to use the fictional construct of equilibrium theory as a substitute in economics for the controlled experiment in the natural sciences. He derived basic ideas through logic from the state of equilibrium. Then he gradually removed assumptions to show the effects. Hayek wrote in his Nobel Prize speech that the data to validate those theories would probably never exist. His validation process came from the reality of the assumptions and consistency of the logic.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s