Customer Reviews

The Evolution of Cooperation: Revised Edition

5 star:		(42)
4 star:		(4)
3 star:		(4)
2 star:		(1)
1 star:		(1)

 

 

If you read this book as long ago as I did, you probably first heard about it from Douglas Hofstadter's "Metamagical Themas" column in _Scientific American_, or the book in which his columns were collected. (If you're just now being introduced to this book, check out Hofstadter's too; his discussion of it is very helpful and insightful.)

What Robert Axelrod describes in this book is a novel round-robin tournament (actually two such tournaments) in which various game-theoretic strategies were pitted against one another in the game known as the Iterated Prisoner's Dilemma. Each strategy was scored, not according to how many times it "beat" its "opponent," but according to how many points it accumulated for itself. The surprising result: a strategy dubbed TIT FOR TAT, submitted by Anatol Rapaport, cleaned everybody's clocks in both tournaments.

Why was this surprising? First, because TIT FOR TAT was such a simple strategy. It didn't try to figure out what its "opponent" was going to do, or even keep much track of what its "opponent" had _already_ done. All it did was cooperate on the first move, and thereafter do whatever its "opponent" had done on the previous move. And second, because this strategy can _never_ do better than its "opponent" in any single game; the best result it could achieve, in terms of comparison with the other player, is a tie.

So it was odd that such a simple strategy, one that went up against all sorts of sophisticated strategies that spent a lot of time trying to dope out what their "opponents" were up to, should do so much better than all the "clever" strategies. And it was also odd that a strategy that could never, ever "beat" its "opponent" should nevertheless do so much better _overall_ than any other strategy.

As Axelrod is careful to point out, this isn't always true; how well TIT FOR TAT does depends on the population with which it's surrounded, and in fact it wouldn't have won even _these_ tournaments if certain other strategies had participated. But TIT FOR TAT is surprisingly robust, and its success does offer some tentative political lessons.

Axelrod spells them out, in the form of principles like "Be nice and forgiving" -- which means: never be the first to defect; be quick to forget what your "opponent" has done in the past. And in a follow-up computer simulation, he shows that it's possible -- under some conditions -- for a little cadre of "cooperators" to increase their numbers and "take over" a population that practices other strategies.

Axelrod's research was and is important for several reasons, one of which has to do with evolutionary theory: it shows that, under the right conditions, natural selection can tend to generate cooperation rather than competition, even among actors who act solely out of self-interest. Another has to do with the spontaneous growth of cooperative behavior in predominantly competitive or hostile environments (Axelrod's examples include holiday cease-fires in the trenches during the First World War). Yet another has to do with the need (or otherwise) for external authorities to _enforce_ cooperative behavior -- a point not lost on Axelrod's libertarian and/or Hayekian readers, including myself.

Nevertheless, as groundbreaking as this work is, the results are modest and Axelrod states them very cautiously. TIT FOR TAT doesn't _always_ "win," and in any case not all of our social interactions can be modelled as Iterated Prisoner's Dilemmas. It's a _very_ hopeful book, but readers will want to be careful not to claim more for Axelrod's results than he claims for them himself.

In short, this volume is a solid piece of political-sociological-mathematical research that has stood the test of time and stimulated all sorts of follow-up work. I expect it will be read for a long time to come -- this conclusion being a simple extrapolation from the fact that I've been reading it for almost two decades now myself. It's fascinating.

This book has information for military theorists, biologists exploring gene regulation, antitrust policy-makers, and Miss Manners. It is a wonderfully clear explanation of how almost any two entities, interacting over time, develop a mutualism more profitable than greed.

The experimental support for these claims comes from a series of contests. Dozens of authors provided computer programs to play in the Iterated Prisoner's Dilemma - a simple model, but one that describes a surprising number of real-world phenomena. Most importantly, it's a testable model. It almost puts a common aspect of social interaction into a test tube. What came out of that test tube was startling in its clarity and simplicity.

The book is very readable. Axelrod segregates the mathematical and non-mathematical discussions with some care. Math-free readers see the whole set of experiments and conclusions, clearly explained, and need to skip only a few paragraphs during the main discussion. The last few chapters reward math-positive readers with additional precision and rigor. Even then, the math is accessible to someone with good high-school algebra skills.

Axelrod's discussion truly timeless, except for references to the Cold War as current events. I can accept that. Even though that un-war is mostly over, it's a critical part of modern history and it still informs current policy. Any insight into that madness helps, and Axelrod is very helpful.

This book stands above any one category. It's one of very few that I recommend to the bookshelves of every educated person.

Sometimes, the individual benefit seems to conflict with the benefit of the community as whole, even though the community includes this very individuum. One such example has been formulated as the Prisonner's Dilemma: two suspects, A and B, are arrested, and kept separated so that they cannot communicate. If they continue to cooperate, they will be both sentenced to one year. However, if suspect A cooperates, but suspect B defects, A is going to be sentenced to five years, and suspect B will be released. Vice versa, if B cooperates and A defects, A will be released and B sentenced to five years. Finally, if both defect, they will both be sentenced to three years each.

It is clear that the best solution for both of them is cooperation. On the other hand, each individual is also tempted to maximize his own individual benefit. And each of them benefits most if he decides to defect, which in turn brings the worst possible outcome for both (six years total). So one-shot Prisonner's Dilemma rarely leads to cooperation. Now, what if the very two chaps are later arrested again? Will they cooperate when given another chance? Or if they know they will face the same situation every five years? Professor Axelrod tested the iterated Prisonner's Dilemma with computer programs, and investigated under which circumstances cooperation can emerge.

The book is nicely scattered with fragments of game theory and examples from world politics. All in all, as Richard Dawkins has commented in the foreword to its British edition, in breathes with optimism, and is a delight to read. Still, it has one problem, and actually shares it with Dawkins: the book reaches its climax right at the beginning. The book starts with a strong and very convincing idea, but later fails to keep the same pace of dynamic. The idea is splendid, but the structure of the book could be enhanced.

This book is a must-read not only for students (broadly defined) of the social sciences, but also for politicians and bureaucrats, especially those in charge of military and foreign affairs. Axelrod's book is a tour-de-force in multi-method approaches. Although the author is a trifle repetitive and occasionally laborious, I think the profound content of the book far outweighs the minor inadequacies of its form. At the risk of sounding like a logical positivist, I would venture to say that Axelrod's approach offers hope for a bottom-up construction of cooperation in an uncertain world without a central authority.

In 20 to 50 years people will look back on the experiment described in this book as one of the very important events at the end of the 20th century. There is a section of the book that nearly forces you to abandon it, the mathematical description of the famous prisoners dilema game. But if you get past that point , it just gets more fascinating the more you read it and think about the consequences. Short descriptions of this experiment will appear in a lot of good books, but it is nice to get the complete description of the experiment, along with the author's projections of its implications in such a short wonderful book.

If you're an intellectual and want to read a book that will change your perception of many facets of the world forever, this is the book for you. It's not a long read, but you will spend a lot of time thinking about all its implications as you read it. I found it applicable to everything from inviting people to parties, to business and personal relationships, to species competition, to wondering whether a theoretical race of super-powerful extraterrestrials would enslave us, to... Well, you just have to check it out!
I'm reading the sequel ("The Complexity of Cooperation") right now, which is also amazing. In it he quotes a letter written to him about EoC by a woman who claims that the principle developed in it helped her with her divorce proceedings! How can you miss a book with such broad applications.

This book has a wonderful first chapter that Axelrod repeats throughout the book. Axelrod uses computer simulations to develop game theory and devise strategies that result in collaborative behaiviors. While Axelrod's findings can be applied to other fields, such as politics or interpersonal relationships, it should be noted that Axelrod is dealing with very precisely (and arbitrarily) defined initial conditions and constraints that are not at all gauranteed to exist outside the world of computer simulations. Generally speaking, the first chapter is excellent and offers a ray of hope that collaborative behaiviors can and will evolve on their own; on the other hand, the rest of the book is repetative and highly arbitrary.

Amidst the glowing 5-star reviews I feel the need to interject some concerns and outright criticisms of this book, although I do recommend its reading. Certainly the book provides a relatively good starting point in a very complex area, but it should not be construed as the final word on a much more complex subject. I do think that R. Axelrod provides an excellent, if at times overbearing, presentation of how game theory, specifically in the realm of an Iterated Prisoner's Dilemma (IPD) scenrario, can explain a number of historical as well as daily situations. He goes on to expound on some good generalizations on how people might act to inspire more cooperation and these generalizations, if implemented by people (groups, governments, etc.), might result in a more cooperative world. For this, there are good things to say. But I would caution not to create from this any utopian potential for the real world.

The `winning strategy' of TIT FOR TAT (TFT) works because it starts out `nice' but it retaliates immediately if someone else does not `play nice', too. So this is not vision of world cooperation. It is a realistic vision of maximizing cooperation under specific conditions which he covers relatively well towards the later chapters. Ultimately, the natural take away is a hopeful view of a potentially more cooperative environment, with perhaps a bit of forgetting that unkind retaliation is an integral part of his winning `cooperative' strategy.

Here are some aspects which Mr. Axelrod alludes to but somewhat minimizes, in my opinion, in their impact on TFT's potential for success in real world interactions:

a) The IPD strategies, which were submitted by experts from around the world, were submitted to computerized testing to determine which strategies `win' the most. This is based on an established point system that awards different points for different actions by two players. (This is summarized in one review already, so I will not repeat it here.) While this makes the playing of the game easy and consistent, it does not reflect the real world conditions which often exist. For example, if, in one turn, one player `defects' and the other `cooperates', the point system says the defector `wins' 5 points and the cooperator gets nothing. Are all defections `equal' in the real world? Or are some interactions far more important than others, so therefore cooperating on small things but defecting on bigger things might result in different outcomes than would ever be accounted for in the point systems used in these studies? Intuitively, a major defection may have far more significance (i.e. point value) than a minor one, but to accomplish his analysis, Axelrod's point structure is always the same.
b) Related to the above, are all joint cooperative efforts or joint defections `equal' as implied by the point values? (When two strategies cooperate, each is awarded 3 points. When each defect, they each get 1.) At the time this book was being published, an interesting `real world' example was playing out which would question this very assumption. At that time, Reagan was building up the military and the Soviet Union was `matching' the build up. You can either view this as mutual defection or mutual cooperation, depending upon whether you view cooperation as always a `positive' thing. In any event, the results of these mutual actions were NOT equal. The US had far more resources to invest in the game so each time `points' were awarded, the US actually gained more than the Soviets who eventually had to stop playing. Consistent point values simply do not account for this, unless one wants to interject additional elements not presented in Axelrod's work.
c) Related to point b), the IPD study presents things from a vantage point that each actor starts out from an equal footing and therefore the only functional question is whether one strategy consistently wins in such a way as to keep near the top of the point standings. In reality, rarely does any person or group begin interaction on an equal footing. What I am trying to raise is not the same as his discussions in Chapters 8 and 9 on the strength and growth of various strategies in a world starting out with many different strategies. In those scenarios, some strategies die out because they stop having sources of points to take from others. But what if each strategy starts out from an unequal basis, some having significantly more `capital' to expend and coupling this to a more realistic scenario where each interaction is not free, but each has a cost to the participant (somewhat like putting your bet down before you play poker - you may win or lose, but if you lose your position is not the same as when you started, it is less than when you started). In such a scenario, which is more realistic in terms of how companies, governments and even people interact, the results would be materially different. So the starting premise of Axelrod's IPD scenario and point schemes may be so well constructed as to make for interesting descriptions of some specific and even common interactions, but it may be too well constructed to be able to be extrapolated to many more complex situations.
d) Another weakness is the assumption that all defections are of the same magnitude. If, in an otherwise nice political campaign (ever seen one of those?) a small, third party candidate launches an attack ad and the attacked major party retaliates, is the retaliation equal? Or, again, does their starting point allow them to annihilate the attacker? (Since I have never seen a positive political campaign, you can interject any other similar scenario which might actually come up in real life!) The `equal points for equal actions' premise is inherently flawed. Yet this is the basis of much of the book's conclusions.
e) What is the end result of a universe where everyone uses a TFT strategy? This is only marginally considered, at best. While the issue had been in my mind throughout much of the book, it was not until fairly late that Axelrod makes clear, albeit briefly, that in any set of two player interactions, TFT will at best result in the same total number of points as the other player and, for a number of reasons, probably slightly less. In other words, the person, company, group or government using a pure TFT strategy must be happy being close to the best in whatever the interaction. Personally, I am fine with that, thank you! But to assume that everyone is and that nobody will come up with a disruptive strategy that, perhaps only for a time, garners more points but winds up putting them on top of the heap in whatever competition may be in play. If the end game is the Superbowl, being second is not gratifying. Ask Philadelphia.
f) Finally, I found some of the descriptors attached to the strategies interesting, especially in light of Axelrod's ending recommendations on cooperation. Some strategies which attempted to use planned defection as part of their strategy were labeled as `meanies' while other strategies that started their initial moves with cooperation were labeled as `nice'. I find it interesting that a `scientist' would use such descriptors to classify things given that they bring with them connotations. Even TFT could be construed as a `meanie' since one of its logical outcomes is that it cooperates once, and once only, only to defect from then on based on the other parties action. Are all non-cooperative interactions except one inherently more `nice'? This is not important in the scope of my real concerns, but I always sense a red flag rising when an otherwise objective presentation resorts to affect-linked labels to make part of its case.

All of the above is NOT to imply that I think Axelrod's work is wrong and should be thrown out. I simply felt that, unlike the rest of the glowing reviews, Axelrod's work does not go far enough and leaves some gaping holes yet to be explored. Read this book, but keep your eyes and mind open for the flaws which seem to be thinly covered with papier-māché.

Rarely do you get to read a heads up account of a scientific project, understand the science behind it, fathom the possible impact on society, and enjoy the story to boot. This is one of those times. For a bonus, the use of the word evolution in the title will not throw many into paroxysms of indignity upon reading the text, but instead find it appropriate.

"When should a person cooperate and when should a person be selfish?" Not the biggest existential question but perhaps the one most ask and answered. This book addresses this most human conundrum. The writing is direct and the subject interesting. I recommend this book to anyone who has ask themselves the above question. Which means I recommend this book to everyone.

This book is a stunningly lucid combination of game theory, computer science, strategy and evolutionary psychology. I always eschew any kind of 'business' book, and encountered this in the course of a computer research project, but it has had great influence on my thinking in a variety of personal and professional fields.

From first principles, and using ingenious empirical techniques, the author extrapolates from a simple so-called 'prisoner's dilemma' (would you betray a friend to save your neck?) right out to some extremely persuasive and general lessons on the conduct of potentially adversarial relationships.

The conclusions he draws are both powerful in their application, and refreshingly humane in what they imply for optimal behaviour in stable societies.

I'd rate this above even Kuhn's 'structure of scientific revolutions' as a piece of nominally scientific writing that has widespread relevance beyond the field for which it was intended.