Is morality an emergent behavior?

I have been thinking about the question of what exactly is it that develops in moral development, and as a result I want to put forth some inconclusive thoughts. Cognitive structuralism’s approach to this question suggests that the answer is reason, that as people’s reasoning abilities develop, so do their morals. Piaget, for instance, mapped his stages of mental growth to heteronomous and autonomous stages in the development of moral reasoning. Kohlberg, following on Piaget’s footsteps, outlined six stages of moral reasoning from early childhood to adult life (heteronomous morality; individualistic/instrumental morality; impersonally normative morality; social system morality; human rights/social welfare morality; and morality of universalizable, reversible, and prescriptive general principles). The idea in both cases in that as people’s mental abilities develop, they are able to implement more complex and less self-centered models of morality.

This might make instinctive sense. After all, one could argue, aren’t adults better equipped to distinguish moral nuances than children? But careful consideration reveals some problems with this perspective. For example, does cognitive structuralism’s approach to moral development imply that organisms with higher reasoning skills are more capable of moral behavior than organisms with lower reasoning skills? Or to put it in more crass terms: Are smarter people more moral than their counterparts? Do humans behave more morally than jellyfish?

Gilligan (1977), among others, has presented a critique of this approach by contrasting Kohlberg’s idea of the moral subject as an individual who can think formally and act autonomously with a model of the moral subject as someone who thinks contextually and acts socially. Similarly, Hoffman (2002) tries to elucidate the difference between these two perspectives by contrasting morality as justice versus morality as caring. And Dreyfus (1990) argues that intellectualism is of little use to an ethical expert who responds “instinctively and appropriately to each ethical situation” (p. 11; more on this to follow).

In this paper, however, I want to present a different type of critique to the cognitive structuralist view of moral development by making two claims: 1) that Reason (as defined from a Western, Humanist perspective) actually impedes moral development, and 2) that this is so because morality is actually an emergent behavior-in other words, a behavior exhibited by organisms acting according to very simple rules requiring little reasoning, but behavior that results in a complex system, a system which is, in fact, the basis for the order of the Universe, and which is replicated even by organisms without brains… How’s that for outlandish claims?!

To begin, I would like to make it clear that my argument does not rely on a renunciation of reasoning or logic as the basis for morality. To the contrary. While it is a particular conceptualization of reasoning that is the focus of my critique (the Humanist, Individualistic definition of reasoning, or Reasoning), my entire argument rests on the foundation that morality is a form of logic. In this, I take a page directly out of Piaget, who produced one of the most elegant memes about the relationship between morality and logic that I have found, quoted below by Gibbs:

The intertwining of morality with logic is expressed in Piaget’s famous assertion: “Morality is the logic of action just as logic is the morality of thought.” In other words, the two intimately interrelate: Moral reciprocity is rational just as rationality is prescriptive. (Gibbs, p. 36)

Usually, Piaget’s statement is taken as a strong argument for the case that, as his next line suggests, “pure reason [is] the arbitrer both of theoretical reflection and daily practice” (quoted in Dreyfus, 1990). But I would like to take the liberty of using Piaget’s words against his own position by interpreting his statement to mean, simply, that moral reciprocity (“Do unto others as you would have them do unto you”) makes logical sense-it’s just how the Universe works. According to this interpretation of Piaget, logic is moral in that there are right and wrong answers (2+2=4, not 3 or 5). Likewise, moral action is logical in that moral reciprocity makes as much sense as 2+2=4, and moral irreciprocity makes as much sense as 2+2=5. But this has more to do with the way the Universe works than with the particular characteristics of ‘pure reason.’ The fact that moral reciprocity does not require pure reason has been exemplified, among other instances, by the Prisoners’ Dilemma competitions. In this tournament, simple software routines that learn to cooperate with each other do better than those that focus on competing with each other (for a recount of these tournaments, see for example Grossman, 2004). This kind of behavior is referred to as emergence.

But before I discuss how the logic of moral reciprocity is evident in the emergent behavior of organisms in the Universe, I would like to discuss what happens when this order is disrupted by Individualistic Reasoning. My thesis is that the Universe would work much better without this brand of ‘logic’ and that Individualistic Reasoning is in fact a deviation from the type of logic that actually promotes moral behavior.

If a scapegoat must be named, his name is RenΓ© Descartes. The problem is that Descartes convinced himself that all we have access to in the world is our own private experience. Descartes, following on the footsteps of the Skeptics but armed with the new language of modern science, questioned the reality of perception. He did this on the grounds that our sensory organs, such as eyes, ears, skin, etc., are very imperfect transmitters of information to the brain, which is the only organ capable of interpreting and acting on that information. For example, it is the brain that activates signals of pain received by a particular part of the body, or creates an itch where an amputated limb used to be. It is also the brain that makes things seem real during dreaming, when in reality they don’t exist. So our access to reality is indirect, mediated by the imperfect senses and actualized only by the brain. This line of thinking lead Descartes to believe that the only thing we could be certain of was therefore the content of our brains, and everything in the outside world was consequently less real, or not real at all.

This Skeptical view was eventually contested (after three centuries!) by various schools, including the Pragmatics and the Existential Phenomenologists, who argued that there was no point in even asking how we perceive the ‘external’ world because we are embedded right into it, inseparable from it. As Heidegger argued, there is no such thing as a subject who is not being-in-the-world. “Taking the skeptic seriously and attempting to prove that there is an external world presupposes a separation of the mind from the world of things and other people that defies a phenomenological description of how human beings make sense of everyday things and of themselves” (Dreyfus, 2000, p. 53).

What interests me here is the particular anti-social way in which Reasoning was defined by Descartes and adopted by Western Humanism. Under this rubric, logic (including moral logic) has been defined in the West as something the individual does in isolation, not as part of a system. Norbert Elias describes the antisocial consequences of Descartes’ philosophy as follows:

Descartes’ Cogito [“I think therefore I am”], with its accent on the I, was also a sign of this change in the position of the individual person in his society… The isolated thinker perceived himself-or more precisely, his own thought, his ‘reason’-as the only real, indubitable thing. All else might possibly be an illusion conjured up by the Devil, but not this, not his own existence as thinker. This form of I-identity, the perception of one’s own person as a we-less I, has spread wide and deep since then. (Elias, 1998, p. 231-232)

Individual Reasoning subverts morality by disassociating the acts of the individual from the emergent acts of the ecosystem, of the we. Humanism, in its rush to liberate humankind from “illogical” (read: religiously imposed) morals, made it practically impossible to act in accordance with the logic of the Universe, a logic that Humanist Science itself claims to try to understand! In order to substantiate this claim before I am labeled an obscurantist, I need to finally turn to my statement that morality is an emergent phenomenon. What is emergence?

Emergence is what happens when the whole is smarter than the sum of its parts. It’s what happens when you have a system of relatively simple-minded component parts-often there are thousands or millions of them-and they interact in relatively simple ways. And yet somehow out of all this interaction some higher level structure or intelligence appears, usually without any master planner calling the shots. These kinds of systems tend to evolve from the ground up. (Steven Johnson, in an interview with Sims & Dornfest, 2002)

Johnson argues that emergent systems

solve problems by drawing on masses of relatively stupid elements, rather than a single, intelligent “executive branch.” They are bottom up systems, not top-down. They get their smarts from below… In these systems, agents residing on one scale start producing behavior that lies one scale above them: ants create colonies; urbanites create neighborhoods; simple pattern-recognition software learns how to recommend new books. The movement from low-level rules to higher-level sophistication is what we call emergence. (2002, p. 18)

Johnson indicates that “[e]mergent behaviors… are all about living within the boundaries defined by rules, but also using that space to create something greater than the sum of its parts” (2002, p. 181).

How does morality fit into this model? Well, the simple rule is moral reciprocity. The “stupid” agents are all living things (regardless of their level of reasoning). The complex emergent behavior, the sum greater than the parts, is Universal Order. One of the things that makes emergent systems durable and easy to propagate is that they are adaptive. Moral reciprocity is universal because there is no “executive branch” that needs to tell everything in the Universe how to behave; rather, the ‘DNA’ of the behavior is widely spread, and organisms-from simple jellyfish to complex humans-can adapt the rules and work out contextually what the logical/morally-right thing to do is.

This does not mean, obviously, that humans have as easy of a time as jellyfish in applying moral reciprocity. Humans are complex organisms living in complex social settings. However, as Dreyfus (1990) argues, the idea that therefore an intellectual approach to moral reasoning is bound to be superior than an intuitive approach might have more to do with our Cartesian biases than with the way things actually work. Dreyfus puts forth a model of moral development that resembles more the process of gaining mastery in driving a vehicle or playing chess than the process of philosophizing: expertise does not constitute deep pondering and analyzing of each move, but comes intuitively:

The intellectualist account of self-sufficient cognition fails to distinguish the involved deliberation of an intuitive expert facing a familiar but problematic situation from the detached deliberation of an expert facing a novel situation in which he has no intuition and can best resort to abstract principles… [I]n familiar but problematic situations, rather than standing back and applying abstract principles, the expert deliberates about the appropriateness of his intuitions. (p. 13)

This constitutes, in effect, a reversal of the “Western and male belief in the maturity and superiority of critical detachment” (p. 23). Instead of the ideal of a detached, uninvolved brain making sense of suspect sensory signals, “[t]he highest form of ethical comportment is seen to consist in being able to stay involved [in the world] and to refine one’s intuitions” (p. 23). I interpret this to mean that, unlike jellyfish, we constantly create and encounter new moral dilemmas, and thus have to “reason” our way back to emergent moral reciprocity-to the logic of the Universe-not by applying abstract principles, but by contextualizing our intuitions. Only in situations that are completely alien to us, argues Dreyfus, do we fall back on abstract moral rules, but “it should be no surprise if falling back on them produces inferior responses. The resulting decisions are necessarily crude since they have not been refined by the experience of the results of a variety of intuitive responses to emotion-laden situations and the learning that comes from subsequent satisfaction and regret” (p. 13).

In short, when individuals apply Individualistic Reasoning to define morals (in an attempt to become the “executive branch” of morality), they stop being part of the emergent system, of the universal order. Individualistic Reasoning presupposes that morality is a function of the rational elite, those organisms with advanced reasoning skills (who for some strange reason are mostly white adult males). Emergent moral reasoning, on the other hand, presupposes that moral reciprocity is a function of the Universe. Everything and everyone acts morally in the sense that their interactions are part of the logic of the Universe, the logic of moral reciprocity. Moral reciprocity just makes logical sense, like 2+2=4; it just happens. It is encoded into everything in the Universe.

Individualistic Reasoning, which assumes that higher reasoning results in higher morality, disrupts this balance by trying to make the system top-down, not bottom-up. In other words, Humanism has placed rational humans as the source of morality. Morality is what highly rational humans say it is, not what the rest of the Universe is telling us it is. Thus, Individual Reasoning-even if it spouses the highest ideals-ends up disturbing the logic of moral action by limiting the domain and practice of morality to the actions of “mature” rational beings.

Some might wonder: if moral reciprocity was really the order of the Universe, how come there is evil in the world? How come killer whales kill senselessly, cats torture mice, and humans commit the most atrocious acts against each other-all of which make it hard to believe that moral reciprocity rules the Universe? The answer to this question is that immorality, in the form of moral irreciprocity, is also part of the emergent system. In fact, it actually serves a very important pedagogical function. It ensures that moral reciprocity spreads virally, in the sense that by suffering or observing moral irreciprocity, everything in the Universe learns-using the most basic reasoning skills, if not mere instincts-that moral reciprocity is the only strategy that guarantees survival. Even the Prisoners’ Dilemma software can figure that out quickly. If we didn’t have deviations (in the form of moral irreciprocity), we would not be aware that moral reciprocity makes logical sense. The exception proves the rule.

Of course, chaos is part of an emergent system, which means that moral reciprocity and moral irreciprocity are not in perfect balance at all times and in all places. In fact, because moral irreciprocity channeled through Individualistic Reasoning-although illogical-satisfies the needs of the individual, and because we live at a time in which Humanism has made the Individual the center of the Universe, we are currently experiencing a larger proportion of moral irreciprocity (ironically, under the guise of Humanism). However, as this places an inordinate stress in the emergent system, we can expect the laws of chaos to eventually enact an adjustment. It’s just our job to help it along the way by surrendering ourselves to emergent moral reasoning, by letting go of our egocentric
belief in ourselves as superior moral beings πŸ˜‰


Dreyfus, H. (1990). What is moral maturity? A phenomenological account of the development of ethical expertise. Retrieved on December 17, 2004 from

Dreyfus, H. (2000). Telepistemology: Descartes’ last stand. In K. Goldberg (Ed.), The Robot in the Garden (pp. 48-63). Massachusetts: MIT Press

Elias, N. (1998). The Norbert Elias reader: a biographical selection. (J. Goudsblom, Ed.), Oxford, UK: Blackwell Publishers

Gibbs, J.C. (2003). Moral development and reality: beyond the theories of Kohlberg and Hoffman. Thousand Oaks, Calif.: SAGE

Gilligan, C. (1977). In a different voice: Women’s conceptions of self and of morality. Harvard Educational Review, 47, 481-517.

Grossman, W. (2004). New tack wins prisoner’s dilemma. Retrieved on December 17, 2004 from,1284,65317,00.html

Hoffman, M.L. (2002). Empathy and moral development: implications for caring and justice. Cambridge: Cambridge University Press

Johnson, S. (2002). Emergence: the connected lives of ants, brains, cities, and software. 1st Touchstone ed. New York: Simon & Schuster

Sims, D., & Dornfest, R. (2002). Steven Johnson on “Emergence.” Retrieved on December 16, 2004 from

[Note: Originally submitted as a term paper for a Moral Development class]

Join the Conversation


  1. Blogs are not the only fruit

    Blogs may have been a word of the year for 2004, but a wider variety of social software tools and group structures will start to gain widespread adoption in 2005, which will present both challenges and opportunities for those of us who are implemen…

  2. Wow. This is an impressive article. It took me a quite a while to digest it, and I’m still not entirely sure I got the point.

    I would love to catch you in person to talk about this, because I think you are using subtly different definitions than I have of many of the words that are critical to your argument – like, for example, “emergent” :).

    Disclosure: I consider myself a Humanist. I assumed that you were of this school of thought as I read the subject, because the idea of morality as emergent seemed completely congruent with a humanistic worldview. For much of the article, you seem to be saying almost the opposite, however – that far from being emergent, moral behavior does not arise from simpler behavior, but there is a fundamental “pull” towards moral behavior because the universe is ordered in such a way as to promote it directly, not indirectly through a series of simpler rules. Except for the penultimate paragraph, which I think I agree with completely…

    At any rate, one of the fundamental properties of emergent systems is that they are complex overall tendencies which emerge from locally defineable behaviors. In other words, if morality is an emergent behavior, then it must be true that a moral actor has no global knowledge or understanding of the system (the universe) as a whole, or even of morality as itself a concept. If one of the premises you cite is true – that moral reciprocity is in fact the only large-scale viable survival strategy (which I heartily agree with) – then actors locally *should* be concerned only with themselves, in particular, with self-preservation, because by following some simple rules to preserve themselves they will become moral if they select the best available strategy at random (and if they do not, they will die and be replaced with those who do).

    I believe this is true, but with an ironic twist. A postmodern critic would observe that by stating things in this manner, e.g. that “actors *should* do X” to promote moral reciprocity, you are identifying yourself as an observer that recognizes moral reciprocity to have some intrinsic value. Does it, then, in fact have intrinsic value to other actors in the system? Would they sacrifice self-preservation as a primary value for it? If so… it must not be (locally, at least) the best possible self-preservation system.

    Of course, maybe it is not an emergent behavior among humans, but among memes: moral reciprocity has memetic properties which make it extremely good for the survival of *populations*, if not necessarily of individuals in all situations. The continued existence of such populations ensures the viral spread of the reciprocity meme, which conditions individuals within it to accept moral reciprocity as the *only* value which trumps self-preservation (as the self-preservation meme is intuitive, “built in”, and makes for a good initial vector to propagate the meme).

    In any event, none of these scenarios where moral reciprocity is emergent require moral actors to use intuition, as long as they are evolutionarily selecting for good survival strategies (and memes are evolutionarily selecting for survival strategies of their own).

    The other thing is that the moral actors in the system have to have enough memory to be able to identify what actions to take to lead towards moral reciprocity, and enough cognitive power to project scenarios based on the current situation. This is also a good argument for more complex organisms being more morally evolved. Far from being disruptive to this process, the ability to apply abstract moral reasoning would be extremely important, because it would allow the actors to place decisions into new and alien contexts more quickly than simple trial-and error. Abstract reasoning is also a positive survival trait.

    The inability of abstract moral reasoning to select a good decision in many cases is due to the fact that the genetic algorithm that selects the optimal moral code is still running, and we probably still have a lot of evolution to go (both humans and memes) before it’s done. The fact that intuition succeeds where abstract reasoning appears to fail is that it is far easier to formulate a solution to a situation you’ve seen before by copying, regardless of whether you reason about it abstractly or emotionally. Abstract reasoning is most required in situations where the actor lacks information and experience, and is therefore predetermined to be more difficult than one where they had a concrete example to reason about and compare to.

  3. Thank you for your comments, Glyph. Below, I have tried to address various parts of your critique.

    Yes, I can see why you would assume that I am a Humanist, and then be confused by the rest of my argument. This is intentional, although perhaps unwise. In essence, I am using arguments from one camp to make a point that is aligned with the view of another camp. This is not motivated by a desire to deceive, but by a desire to make both camps find grounds for dialogue. Most probably I have not succeeded entirely (I consider this a work in progress), but you can’t blame me for not trying πŸ˜‰

    Although I do not claim to be an expert on emergence, I know enough to realize that I am definitely pushing the concept in directions which it was not meant to be taken. I guess from your perspective I am trying to have my cake and eat it too by arguing that moral behavior arises from simpler behavior, AND that this in fact constitutes “a fundamental “pull” towards moral behavior because the universe is ordered in such a way as to promote it directly,” as you paraphrase. Yes, I am arguing that morality is a set of simple behaviors performed by agents at a micro level (with no necessary knowledge of the macro), but I am also saying that this behavior amounts to a universal (macro) order.

    However, despite some inconsistencies in my rhetoric, I am not arguing that “actors should do X.” In fact, I am trying to make the case that agents can’t help themselves but to act morally, that morality is a natural trend in the universe, and that it is only when certain ideas about abstract rationality are introduced that the balance is upset.

    Yes, I’m identifying myself as an observer that recognizes moral reciprocity to have intrinsic value, and expect that it has intrinsic value for other (human and non-human) beings as well. However, I’m afraid I don’t understand your point about intrinsic value and the failure to be the best possible self-preservation system. You lost me there.

    Finally, while I agree that complex organisms that feature more advanced memory and cognitive systems are capable of more sophisticated moral behavior, I would still like to question what you mean by saying that these organism are more ‘morally evolved.’ You seem to be arguing that memory and cognition are continually and linearly ‘evolving’ in humans (and memes, although I don’t know enough about meme theory to follow this line in your argument), and that this linear evolution produces better abstract reasoning, and hence more morally advanced beings. This is indeed the opposite of what I am suggesting. I am arguing that while some beings are capable of more morally sophisticated reasoning (i.e., they are able to cast their emphatic nets more widely, to encompass beings outside their own group), moral reciprocity is a ‘low common denominator’ that is replicated by all living organisms, without regard to their capacity for abstract reasoning. Perhaps I need to develop this part of the argument further in the future.

    As far as the role of experience vs. abstract reasoning in determining moral behavior, I would suggest that you read the Dreyfus article I cite. He seems to be arguing that abstract reasoning is not necessarily the best way to go.

    You’ve given me a lot to think about, and I hope we can continue this dialogue (and perhaps get others involved in it!).

  4. A recent article from the NYT titled _Unintelligent Design_ claims:

    “The gravest imperfections in nature, though, are moral ones. Consider how humans and other animals are intermittently tortured by pain throughout their lives, especially near the end. Our pain mechanism may have been designed to serve as a warning signal to protect our bodies from damage, but in the majority of diseases — cancer, for instance, or coronary thrombosis — the signal comes too late to do much good, and the horrible suffering that ensues is completely useless.”

    The author’s point is that the design of the universe is not very intelligent (intelligence equated in his mind with efficiency, and that from an anthrocentric perspective as well). I wonder what the theory of emergence has to say about outcomes or designs being the most efficient as opposed to the simplest or easiest to disseminate, which is not always the same thing. If anybody knows, let me know.

    Link to article:

  5. Ever since I was first inroduced to the idea of emergent properties I have been wondering if there is a way to make a sound argument for morality as an emergent property of society. I would even say a necessary emergent property of society. I have always had difficulty with the notion of a single human in isolation developing morality. It is only through our interaction with other’s that morality is born. The morality of a given society is an emergent property. I think that there are some necessary morals that emerge in this way. For example, “thou shall not kill” In all socities, a form of this rule has emerged. It can take on different forms in the exceptions and boundaries of the rule but it is a necessary rule for a society to function.

    Did you do this paper for a class at Cornell? What teacher was it? Are you aware of any other papers or publications that cover the idea of morality as an emergent property? I studied philosophy at Cornell.

    I am also interested in the notion of emergent properties as a solution to the mind-body problem.

    I have to say I find it amazing that you just posted the article as this idea has been pushing its way out of my mind lately.

  6. Hi Christina,

    I am not aware of any other books or papers on morality and emergence. The idea of morality as a social phenomena, however, is quite old. So it was only a matter of time before people applied the concept of emergence to moral development as a way to expand on the social perspective. I am a big believer in the collective unconscious (enhanced by the internet), so I don’t find it that amazing that you also had the same idea πŸ˜‰ I think it’s great!

    This was not a class at Cornell, but at Columbia. Specifically at Teachers College. We did not really talk about emergence during class, but for this paper I decided to go out on a limb. If you are interested in the field, Gibbs and Hoffman are a good place to start (see citations in my paper).

    I’d be interested to hear what other research you uncover, and if you ever write something on the topic yourself.


  7. Ulises,

    Excellent article and discussion. I was eating a burrito today and suddenly the idea of emergent morality occured to me as a way to explain the moral fabric of a large group, say the population of the United States. I was delighted to see people thinking about the idea.

    Emergence theory does an excellent job of explaining morality on a system wide level. Consider that emergence demands individuals to share information. The changes in the individual belief or behavior that results from the sharing then impacts the system wide belief or fabric. The overall moral system of a society is comprised of the individual moral systems of individual people.

    I don’t think you need to worry that emergence occurs from the interaction of simple rules to form complex behavior. It could very easily result from complex rules generating even more complex behavior. Essentially, the whole process of human innovation is emergent.

    Emergence could also help explain society wide behavior with regard to morals. Take for example, the seeming geographical bifurcation of “moral” systems within the United States. If we all act individually, forming our own system of morals, these local morals would be most effected by other local morals we come in contact with (i.e our neighbors). This would create a feedback loop that would reinforce local morals, while making it hard to understand foreign morals and even system wide morals. Still, the overall system of morals results from the collection of individual systems.

    This accounts for outliers, who might impact morals, pushing them one way or the other by screaming louder. Still the overall system would even out and could eventually respond to campaigns by individuals or groups.

    Eventually, the system will even out the population shifts around and new information is exchanged.

    I know that I am stretching this argument a bit and turning a philisophical arguement into a sociological and political one.


  8. Thanks for your comment, Greg. Some of my best ideas have come while eating Mexican food πŸ˜‰

    I think emergence would make it harder to talk about ‘local’ and ‘foreign’ morals, don’t you think? If anything, it would be an argument for universal moral values. But you raise a good question: how would we explain minor differences in morality across societies? Are they cultural? How do moral values develop and adapt through emergent processes? Looks like that is the question on your mind. I’ve written a couple of papers (see ‘best of ideant’ section) on how the internet, for example, could be said to facilitate the equilibrium or standardization of morals by putting people from different backgrounds together. On the other hand, it could be said that the internet allows for the creation of small moral communities, which would then be open to the processes you describe.

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.