Selfishness Isn’t A Winning Strategy

I hoped I would have some time to read (and write!) during vacation, but it didn’t happen. However I was able to catch up on some of the articles I’d saved with Instapaper. One I particularly enjoyed was from the slow web, by which I mean it was an article with long-term relevance that required more time to read and digest than a bite-sized blog post,

Edge master class 2011: The Science of Human Nature. The Evolution Of Cooperation

This article is from 2011, and I don’t remember how it got into my Instapaper. It’s the kind of article that makes me wish I were back in school. Martin Nowak is a mathematical biologist, and the article is a transcript of a talk describing his research into how cooperation can emerge through the process of evolution and natural selection.

The topic is well-known to any student of evolution, a subject that interested me greatly when I took a series of classes on the history of science as electives while completing my Bachelor Degree. The importance of this subject derives from the apparent contradiction between cooperation and evolution: in an evolutionary process in which natural selection favors survival of the fittest, cooperation among individuals should not emerge. To understand why this might be so requires first understanding what is meant by cooperation: one individual’s fitness is diminished by cooperating with a second individual whose fitness is improved. If natural selection favors survival of the fittest, then individuals tending to help others at their own expense will be less fit and the expressed trait (collaboration) will disappear.

Obviously, it makes a big difference whether natural selection operates on individuals, on groups or on populations. Nowak discusses this and I won’t elaborate on that point here. Of course, cooperation does occur so there must be some way to explain how it could evolve. He goes on to discuss several possible mechanisms that would favor collaboration and cites a lovely passage from Darwin, which I will reproduce here,

Here is a beautiful quote from Charles Darwin from his 1871 book, “There can be no doubt that the tribe including many members who are always ready to give aid to each other, and to sacrifice themselves for the common good, would be victorious over other tribes. And this would be natural selection.”

The original quote is from chapter five of The Descent Of Man.

The most fascinating part about the transcript was Nowak’s description of a series of experiments he undertook revolving around a game called The Prisoner’s Dilemma devised to test survival strategies. The game has two players and each can choose between cooperation and defection. It costs a player to cooperate, but if the other player cooperates with him, he receives a payoff. If the other player defects, then he receives no payoff and thus incurs a net loss. It is easy to see that the optimal strategy for both players in a single round of the game is to defect. At worse they neither gain or lose. At best they receive a payoff if the other player cooperates. In any case, cooperation seems to be a risky strategy because of the chance it results in a loss if the other player defects.

Or does it? Since a series of repeated encounters is a better description of interactions between living beings than a single isolated one, Nowak studied strategies for repeated rounds of the Prisoner’s Dilemma. When he looked at the best strategies for playing repeated rounds of the game, he found something quite different: the best strategies for winning a series of encounters involved cooperation and integration of feedback concerning the other player’s actions. In fact, the best strategy was one called “Tit-for-Tat.” To play the game following this strategy, on the first round you start with cooperation, and on subsequent rounds you do the same thing the other player did on the previous round. For example, if you start cooperating in round one and the second player defects, then you defect on round two. If in round one the second player cooperates instead, then on round two, you cooperate again.

Nowak described several different variations of the game and strategies. I’m not going to go into a detailed description of them here. If you’re interested, I highly recommend reading the transcript; it’s time well-spent.

The importance of this work is that it shows so clearly that selfishness is not the best strategy. The person who defects (cheats you) in your first encounter is not being smart. Although he may make off with a temporary gain, in the long-term that strategy will cost him. This statement is true not because such behavior seems morally wrong or because we’d like to think the world is based on fairness, but because Martin Nowak’s study objectively demonstrates that this is so,

What I find very interesting in these games of conditional reciprocity, direct and indirect reciprocity, we can make the point that winning strategies have the following three properties: they must be generous, hopeful and forgiving.

Hopeful is that if there is a new person coming, I start with cooperation. My first move has to be cooperation. If a strategy starts with defection, it’s not a winning strategy.

And forgiving, in the sense that if the other person makes a mistake, there must be a mechanism to get over this and to reestablish cooperation.

This is analysis of just what is a winning strategy. This is not an analysis of what is a nice strategy, this is just an analysis of what is a winning strategy, and it has these properties.

That is simply beautiful. Anyone who’s ever needed to answer the question “Everyone else is acting selfishly so why shouldn’t I?” should find inspiration in these words.

In fact, if everyone had the chance to read and understand this, just maybe the world would be a better place.

Further reading:

Martin Nowak’s homepage at Harvard
The complete works of Charles Darwin available online

This entry was posted in Awesome, Science, Society. Bookmark the permalink.

16 Responses to Selfishness Isn’t A Winning Strategy

  1. Corinne says:

    Thanks for quoting this book from Darwin – I also read about his 1871 work earlier this month, by reading the book from Thierry Janssen I quoted on my blog a few months ago, where there’s a whole chapter summarizing a number of recent psychology & sociology publications on optimism, altruism and empathy… => I noted a few references of interest on that topic (I still need to read them myself so can’t summarize):
    – Frans de Waal on empathy in animals, and in particular groups of mammals:
    – Carolyn Zahn-Wayxler on empathy in toddlers,, and the interesting complementary findings by Felix Warneken -
    – possible evidence from neurosciences (in relation with mirror neurons) and the findings of genes associated with more frequent altruism behavior (
    Maybe even more interesting, Janssen mentions Ayn Rand’s 1964 book “The virtue of selfishness” as a key influencer to the doctrine of “live for yourself” that still drives individualism and capitalism in our societies. But this book has to be put back into context: Rand was an immigrant from Russia in 1926, and she actually promoted selfishness against… communism! this ideology war reads so obsolete today…

    • laura says:

      Thanks for all these interesting links for further reading. The article about Dr. Martin Reuter’s research is fascinating. I want to read more!

      Interesting too that you mention Ayn Rand. It seems that many of her works have come back into fashion, but having never read any of them, it’s difficult for me to catch all the nuances. I agree with you that it’s important to consider the context in which she was writing, on the other hand, from what I can tell, in her case I’m not sure that made a wit of difference.

      I hope you will do a post on these when you have finished reading them.

    • mikecane says:

      >>>Maybe even more interesting, Janssen mentions Ayn Rand’s 1964 book “The virtue of selfishness” as a key influencer to the doctrine of “live for yourself” that still drives individualism and capitalism in our societies. But this book has to be put back into context: Rand was an immigrant from Russia in 1926, and she actually promoted selfishness against… communism! this ideology war reads so obsolete today…

      No. That’s a dodge I can’t let stand. Rand never “qualified” her advocacy of selfishness as a political stance against Communism. Although she brought up the “evils” of collectivism again and again, her stance on selfishness was against that of what she perceived as “altruism.” To Rand, everything had to “compute logically.” Human beings are not computing machines and very rarely act in accordance with any rules of logic. The biggest mistake she ever made in her personal life was based on her “logic” that she and Branden should have an affair. In her artistic life, her biggest mistake was upholding as heroes Suits. Suits are the natural enemy of what she claimed to be: an Artist.

      She was a self-alienated person whose thinking could not be trusted because it was constantly undercut by her daily intake of amphetamines.

  2. Ric Day says:


    Interesting, as always! I have a personal interest in sharks and a few years ago came across the work of Alta de Vos at the University of Cape Town; in particular her experiments to see if fur seals exhibit selfish herd characteristics when under threat from Great White sharks. Here’s the publication:

    I have dived most of that coast, from Cape Town up to Durban, as well as lots of other locations on other continents. One common element was always how marine animals “clustered” when under threat from sharks or other predators. Which would seem to support the selfish herd theory. Though I would be inclined to call the behaviour of the fur seals “cooperation” rather than selfishness: they gather as a group in safe (shallow) water, then make a mad dash across (deep) dangerous water, swimming very close to each other and often touching. de Vos’s work suggests they have a fairly equal chance of being attacked.

    • laura says:

      Thanks for pointing me to the article on fur seals. I read the abstract, and saved it to read later. Diving along the coast of South Africa sounds wonderful. Unfortunately, I must content myself to watching underwater videos and letting my imagination do the rest.

      Coïncidentally, just after I posted this article, Mario Livio, whom I know from the time I spent as a postdoc at the Space Telescope Science Institute, tweeted a link to new work suggesting that animals like fish and birds may have evolved to flock together as protection against predators.

      It feels like several different seemingly unrelated parts of my life are coming together all at one time.

  3. Ric Day says:

    I was lucky to have done most of that diving in the 60s, when water pollution was not really an issue. As a kid in the 50s I read the Hans Hass book, Menschen unter Haien, which got me interested in sharks and diving and I did some snorkeling in the Mediterranean and Red Sea while travelling with my parents. A family friend was JLB Smith, a famous ichthyologist (coeleocanth) and he influenced my interest in the African east coast, which I was fortunate enough to dive in many places from the Cape to Mogadishu during the 60s. One would often see fish forming “balls” when under threat from predators and t seemed obvious they were clustering to lessen the individual statistical risk. Now I mostly watch documentaries, feeling a bit old for scuba and the local waters are cold!

    It is ironic that we know less about what goes on in the oceans than we do about deep space. By the time we figure it out, the seas will be a desert, I fear.

  4. Fluff says:

    Hi Ric
    Thanks for the interesting blog and the link to the Nowak video which I have not seen before. (Cheers)
    I did want to just highlight that much of what you describe above (and which is perhaps not as clear as it could be in the video) was not Nowak’s work. It was actually the work of Robert Axelrod (and Bill Hamilton). The work was published in 1981 in the Science magazine under the title “The Evolution of Cooperation” and is now a very famous piece of work which has inspired countless other works. Subsequently Axelrod wrote a book on this work which had a huge impact – not the least on Richard Dawkins who was being asked to do a rewrite of the Selfish Gene at the time. Dawkins deliberately changed very little in his update ….other than to provide a whole chapter to discuss Axelrod’s work (and an additional chapter to introduce concepts of extended phenotype). Dawkins also promoted the Axelrod work in the foreword of the new edition of his book, mithered Penguin to release the book in the UK, wrote the foreword for the UK publication, and went on to present a Horizon documentary about the work (it’s on YouTube called “Nice Guys Finish First”.
    I think this alone gives a measure of just how significant the Axelrod work is…. that Dawkins, a man whose work is significantly based on the self-regarding focus of the gene, should be that absorbed by this apparent sudden addition to the field of evolution: namely that… In a world of self-regarding individuals, cooperative strategies make sense from an evolutionary perspective.
    Tit for Tat as mentioned in the video, was the winning strategy sent in to Axelrod’s tournaments by the “mathematical psychologist” Anatol Rapaport.
    Nowak has made immense contributions to this field, subsequent to Axelrod’s original work (with inspiration and support from giants in their respective fields such as Robert May and Karl Sigmund) and has also recently written a popular science book on the subject (aswell as a very highly respected textbook) however, the original credit for the opening up of the field and the original work is Axelrod’s
    It would be fair to say that Nowak and Axelrod have not been without their critics. And particularly in terms of Axelrod’s original work, the situation is nowhere near as clear cut or simple as he presented it…”the rabbit hole goes much deeper” :-) Having said that, both fully deserve the credit they have received regarding their contributions to the field.
    It has become a fascinating cross-disciplinary topic which connects to game theory, political science, economics, language, cultural development, computer science, ecology, ethology and behavioural (and other) evolutionary studies.
    The Axelrod book, should you be interested, is a great read, although as I hinted at above, it perhaps over extrapolates (and anthropomorphizes) simulation results which aren’t as straightforward as Axelrod portrayed them.
    For a more grounded view on where this work has led over the past 30 years in just one particular area of science (and one it seems you might be interested in) the Stanford videos on behavioural evolution by Robert Sapolsky are fascinating
    Here’s a link to part 1

    PS I have just seen your comments on Apple’s design philosophy. I think your opening statement is beautifully observed:
    “When you have to explain your design philosophy to people, when you have to explain to them that your products are special because of how they make people feel, when you have to explain to people what they’re supposed to feel, instead of just letting people experience that feeling, then chances are something’s not going right”

    • laura says:

      Thanks for correcting the record about the original research on this subject and for sharing these interesting references and links. I’m adding them to my reading list.

      • Fluff says:

        Laura – My apologies for addressing my above post to ” Ric” :-( It should obviously be addressed to yourself.
        Thanks again for posting the video link which has proved very helpful re something I have been working on today.
        I spotted that one of your repliers mentioned Franz de Waals. Not really about cooperation specifically ….but a very interesting, thought-provoking (and entertaining) short presentation by him about morality (and the evolution of) in animals:

      • laura says:

        No worries. Thanks for the links.

    • RicDay says:

      Just a quick note: Laura writes this blog; I am just a regular reader and occasional commenter ;)

  5. RicDay says:

    An interesting aspect of cooperation among animals when they are prey is their tendency to cluster (per the de Vos research on fur seals). Good examples would be sardines, which cluster into huge “balls” in the ocean when under attack by predators, reducing the statistical odds that any one individual will be consumed by their prey. Large herds of animals in the African savannah cluster together when threatened by lions, seeking to keep their young safe in the middle; similarly, herds will cross crocodile-infested rivers in large groups, again letting statistical probability reduce the risk to each individual.

    On the other side of the coin, cooperation in hunting is commonly seen among predators such as lions and hyenas. My favourite is the way Orcas work together to hunt seals. There is a good, short National Geographic rendering of how they work on YouTube:

  6. richfinck says:

    The recently deceased John Nash of Princeton has also published on this. It turns out game theory is applicable to any things in this world. Interesting theory of his is this link.
    This world does desperately need cooperation for the common good.

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 )

Google+ photo

You are commenting using your Google+ 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