Discourse is a social behavior with cooperative (dove) and selfish (hawk) strategies. If we want discourse to be cooperative, we need to recognize, avoid, and punish discourse hawks.

For years I have been turning the phrase “the nature of discourse” over in my head, like a tumbler polishing a stone. Discourse is a verbal interaction among people. As such, it is a social behavior that might benefit from evolutionary analysis. Never mind that discourse has already been studied by great minds for centuries and even millennia. So have other major topics such as religion, morality, economics and sexual behavior. Evolution provides a new conceptual framework for understanding these other topics, and perhaps it can do the same for the nature of discourse.

Let’s begin with the elementary observation that a verbal interaction can reflect either cooperation or selfishness. We know from general evolutionary models of social behavior that when individuals behave selfishly, the outcome is seldom good for the group as a whole. As a simple example, consider the famous hawk-dove model from evolutionary game theory. Doves cooperative with each other, hawks exploit doves for their own selfish gain, and hawks fight with each other to their mutual detriment. Whenever the hawk strategy evolves in the hawk-dove game, the outcome is a waste of time and energy for everyone. Nothing good can come from it.

Dovish strategies can evolve in a game theory model in a number of ways. If doves can choose their social partners and if it sufficiently easy for them to find each other, then they can robustly evolve because pairs of doves fare better than pairs of hawks and mixed pairings seldom take place. Alternatively, we can introduce conditional strategies into the model (e.g., behave like a dove toward doves and like a hawk toward hawks) or strategies that include punishment (e.g., behave like a dove and punish hawks).

Readers familiar with evolutionary game theory will recognize these models as standard fare, but what happens when we interpret hawkish and dovish strategies as different types of discourse? Nothing good can come from hawkish discourse strategies. We should try to stamp them out, just as we try to stamp out other forms of antisocial selfishness. We should do this by excluding discourse hawks from our conversations, by reflecting their hawkishness back at them while remaining dovish toward doves, or by flat out punishing the hawks. If we don’t protect ourselves from discourse hawks, they’ll take over and verbal communication, the lifeblood of human social life, will be poisoned.

This is not how most people think about discourse. True, we all lament how bad it has become and use terms such as gridlock to say that nothing good can come from it, but we seldom make an effort to distinguish between discourse hawks and discourse doves, or to punish discourse hawks in the same way that we would punish other acts of antisocial selfishness. If we did, then maybe discourse would become more cooperative and we’d have less gridlock.

Avoiding and/or punishing discourse hawks requires recognizing them for what they are. This is not as simple as it might seem. Punishment behavior often looks like hawkish behavior and we must be careful not to shoot the sheriff. In addition, productive discourse doesn’t necessarily resemble an episode of Mr. Roger’s Neighborhood. It can be a competitive process where opponents do everything they can to defeat each other and winning is based upon who has the best argument. That’s what debates, the legal system, and scientific controversies are all about. Productive competitive discourse is designed to be cooperative as a system, even if the opponents don’t look like they are cooperating.

How can we distinguish productive competitive discourse strategies that are acceptable from hawkish discourse strategies that need to be stamped out? Judgment calls are made all the time in formalized competitive discourse. Debating societies, legal systems, and scientists all have rules that keep the competition productive and are enforced when they are broken. The same goes for competitive sports. Every second of a sports match is overseen by referees to keep the competition clean.

The same cannot be said for political discourse or the popular media. There are no formal referees, many people are not exhibiting self-control, and there are no effective mechanisms of mutual policing. If you fed those conditions into an evolutionary game theory model, the hawks would win. And the discourse hawks have won in those domains of discourse.

Another elementary insight from evolution is that you needn’t regard yourself as a discourse hawk to act like one. The behavioral strategies that we employ and how we think about them need not resemble each other in any way at all; it’s only necessary for the latter to cause the former. That’s the fundamental distinction that evolutionists make between ultimate and proximate causation. Discourse hawks need to be recognized by their actions, not by how they think about it.

The problem is aggravated by the currently dominant economic narrative, which teaches us that laissez faire leads to the common good. If we can make society work by all independently trying to get rich, then why not make society work by all independently yelling at each other? In this fashion, well-meaning people who think they are part of the solution can become part of the problem. We need to replace the currently dominant economic narrative with one that recognizes the basic evolutionary dynamics of cooperation, before we can see hawkish discourse strategies for what they are.

When hawkish strategies win in a game theory model, they are often locally stable, which means that work is required to move the system beyond the selfish domain of attraction into a more cooperative domain of attraction. Fortunately, the cooperative domain of attraction can also be stable, even more stable than the selfish domain, so that only an initial effort is required to accomplish the domain shift. Not only is this theoretically possible, but I think it is also feasible. The first steps are to get people to think about discourse as a kind of social behavior with dovish and hawkish strategies, to recognize hawkish discourse strategies for what they are, and to combat them with avoidance and punishment. That’s what we do for hawkish strategies of all sorts, and discourse hawks should be no exception.

David Sloan Wilson

David Sloan Wilson

David Sloan Wilson is SUNY Distinguished Professor of Biology and Anthropology at Binghamton University. He applies evolutionary theory to all aspects of humanity in addition to the rest of life, both in his own research and as director of EvoS, a unique campus-wide evolutionary studies program that recently received NSF funding to expand into a nationwide consortium. His books include Darwin’s Cathedral: Evolution, Religion, and the Nature of Society, Evolution for Everyone: How Darwin’s Theory Can Change the Way We Think About Our Lives, and The Neighborhood Project: Using Evolution to Improve My City, One Block at a Time and Does Altruism Exist? Culture, Genes, and the Welfare of Others. .


  • Doc B says:

    A great article, but just to sweeten the cerebral pot a little, let me complicate David’s definition of “discourse”?  In an article I’m working up for EVOS describing my time in Afghanistan, I put it like this:

    “Discourse:  Just a Fancy-Ass Way of Saying ‘Discussion’?

    Actually, it’s not.  A discourse, in the world of literary and cultural studies, is more than just the sum of the words that we use to describe or talk about something:  it is that, too, but it becomes coupled with the social contexts in which they occur, the formal and informal rules that guide our use of language in those social contexts, and all of the “in between the words” nuances that govern how and what we feel we are authorized to say, can say, might say, and might not even say but merely imply (through body language, for instance, or facial expression). 

    To illustrate, imagine a baseball discourse.  It is a way of talking about baseball that includes much more than just two people using terms like “home run” and “bat” and “major league.”  Those are all denotative terms.  Connotatively, major league implies minor leagues, and so it inherently also implies “better” and “worse” degrees of baseball.  This, in turn, implies better and worse qualities of players playing it.  It also implies social status, the economics of baseball, audiences who can afford tickets (and those who cannot), and more.  It implies the whole history of baseball, and of what we think about where today’s baseball fits in the long historical picture of all things baseball.  So here, the discourse would be everything that we say, think, and feel about baseball when we talk about it.  It also involves our perceptions of what those around us might think, feel, and say, and about how we may or may not feel as though we can speak frankly.  Is it OK to say that the poorer classes get the worst seats at a ball game?  Or would that be bad form?  Why is it bad form?  Because those who can afford good seats feel guilty about their privileged position relative to others?  Because we were taught not to gloat? 

    There is a discourse for cars, for cooking, for gender, for holidays, for almost everything; discourses are the result of our conscious and subconscious, social and cultural ways of communicating.  Discourses act like normative guideposts for what we feel we’re allowed to say about something.  At times, they dictate what we are even capable of imagining. 

    So it is that the term discourse is particularly useful for describing influence activities.  Western discourses are foreign to those in a country who have been living, by and large, and for millennia, an agrarian, subsistence-based, tribal existence:  and who do so in their own languages.  But even then, we’re talking much more than just language here, much more than just the gap between, say, English and Dari or Pashto, the two official languages of Afghanistan.  We’re talking about fundamental differences in perception, about an immense gap between what is even consciously and subconsciously available to be talked about in the first place.  About what exists in the very realm of cognitive possibility.  We were so far “outside” of Afghan discourses that at times it seemed like we were completely and utterly stabbing in the dark in trying to across a message like “everyone has a right to a fair shake.”  The very premise is not part of a common discourse, a discourse dominated by another message:  “inshallah.”   

    For people whose average life expectancy is about 45, and who have been under the throes of conflict for over 30 years, the discourses of “hope” or “peace” are virtually non-existent: or at the least badly fading.  Lacking a collective cultural memory of either concept, many Afghans view messages that positive change is possible not necessarily with skeptical disbelief but rather with some form of silent incredulity.  Those discourses, those available ways of talking about such ideas, have all but been lost amid decades of sheer survival.”

    Thanks for hanging on this long, but I raise this point in detail to complicate the notion that discourse may “merely” be what people say.  David is right to say that it is “kind of a social behavior” – I would go even further and say that discourse actually IS social behavior.

    What are the implications for game theory as discussed?  Well, could one of them be that the words that we use to describe things aren’t what need to be combated but rather the perception that language is invisible?  Can I ever achieve cooperation as a strategy if I focus on only the illusory surface of the meaning embodied in language?  Or do I need to get at all of the informal “stuff”—overtly ideological, social, political, etc, yes, but also covertly subconscious (innate, primal)—underlying any hawkish (or dovish) articulation?

    All just some discursive food for thought.


  • Dave Gerstle says:

    Doc B – thank you for what you’ve written here. It complicates things in the ways they need to be complicated.

    David – This particular ‘stone’ needs to be ‘polished’ some more (that’s a pretty unflattering metaphor you’ve chosen, by the way). I’m not exactly sure what you’re attempting here, but please consider:

    1) You’re trying to contribute to (what you acknowledge as) an incredibly ancient and nuanced problem. But you sidestep everything that has ever been said about it. Your definition of discourse (“a verbal interaction among people”) is suited to fit the needs of your model. But, as Doc B’s post hints, that is not how 99% of the literature on discourse (from philosophy, linguistics, history, sociology, literary theory, etc.) would define it. You’ve often suggested that others need to do their homework… take your own advice.

    2) You are unable to draw a single example from real life that fits your model. This is because the motives, actions, and language of real people are not simply ‘cooperative’ or ‘selfish’. Let’s take the one example you touch upon – legislative gridlock (I’m guessing you mean). The people involved in congressional or parliamentary debates represent huge swaths of people, beliefs, needs, and interests. Of course their debating styles are different and sometimes vitriolic. But thinking of any of their actions as unambiguously ‘selfish’ (save for gross corruption) only foregrounds the weakness of the game theory model: The dynamics of interpersonal debate (in government, or science, sports, religion, etc) are not occurring within an ahistorical and (ultimately) unearthly vacuum.

    3) A more bizarre argument comes within your judgment of the hawk-dove dynamic. Who, exactly, are the hawks? And since when is a person’s resistance to compromise inherently a form of “antisocial selfishness” that needs to be “stamped out”? You might think you’re describing some congressional bullyboy, but you’re also describing the leaders of any number of civil rights movements. If an individual (or group) believes the opinions of the majority to be unfair or cruel, they would be fools to conform to those opinions.

    4) The ‘refereed’ institutions that you romanticize here also tend maintain inequality through biased forms of ‘policing’ their participants behaviors. The history of the US legal system or competitive sports will show that these are among the institutions that are most prone to racism, sexism, class discrimination, homophobia, and a dozen other human failings. Science is not far behind them, either. It might be convenient to think about the beliefs and actions of Lysenko, Mengle, or the Tuskeegee Public Health researchers as ‘self-serving’, but they were all operating according the rules of their institutions – inhumane and misguided though they were.

    It’s not that I don’t think that evolutionary science has nothing to add to knowledge about interpersonal communication. I know that it does. But your essay is a diatribe that ignores a vast body of literature and follows the calculus of game theory to some (I’m sure) unintended conclusions – promoting censorship, suppression of minority positions, and unrealistic methods of policing communication… which most in our society would agree is a bad idea anyway.

    With all respect to your right to express YOUR opinions, please reconsider what you’ve put forth here.



Leave a Reply