+ Reply to Thread
Results 1 to 14 of 14

  1. #1
    Join Date
    Mar 2006
    Location
    Sydney
    Posts
    1,167

    Game Theory, Reciprocal Altruism, Tit-For-Tat

    Coincidentally I am reading The Origins of Virtue and just reached the section on Game Theory and today I was browsing videos in Google and came across this:
    http://video.google.com.au/videoplay...ichard+dawkins

    Game theory is such an interesting concept, and I think it really does capture the concept of altruism.

    I was really captured by this summary of 'the winning strategy' in a competition of game theory programs. They had people write a basic strategy for how to approach the prisoners dilema, and then they played all of those startegies against each other and kept score. the strategy that always won was "Tit for tat".

    Anyway, I like this description Dawkins provides of it:
    Firstly, Tit for tat is nice. It's never the first to defect. Now this came as quite a surprise to some of the experts who sent in programs, because they had imagined that the way to succeed in this game was to make subtle opening moves of .
    subtle moves of defection

    Secondly, tit for tat is not envious. It doesn't measure its success by how well its doing against its immediate opponent but by its total winnings at the end of the competition. tit-for-tat infact cannot win in an individual game against an opponent, it can only do at least as well. and tit-for-tat only does well only when its opponent is doing well too. it is fundamentally co-operative strategy.

    Thirdly tit-for-tat is a forgiving strategy, it forgives as swiftly as it retaliates.

    Fourthly, its simple. easy to read. Uncomplicated.
    Now I'm not really new to the prisoners dilema or to tit-for-tat, but this summary really made me understand how simply this describes sooo much moral theory. I mean look at what he is saying this incredibly simple strategy does.

    First of all it invokes the golden rule. It starts by offering co-operation. It offers what it wants. An identical version of the strategy which starts with defection won't be as successful in the longrun as the version which start with co-operation.

    Secondly, it punishes wrongs done to it. Not harshly, but it withdraws the priveledges generally offered. Nonetheless it forgives instantly, at the first signs of co-operation. it doesn't carry a grudge. it is a forgiving strategy.

    After all of the discussions I have had in the religious forums lately, it is funny to see Jesus reduced to such a simple mathematical strategy....

  2. #2
    Join Date
    Nov 2006
    Location
    East of Eden
    Posts
    115

    Re: Game Theory, Reciprocal Altruism, Tit-For-Tat

    But Christ did not teach tit for tat. He said not to do to others what they have done to you, e.g., his statement: "Whoever slaps you on your right cheek, turn the other to him also."

  3. #3
    Join Date
    Mar 2006
    Location
    Sydney
    Posts
    1,167

    Re: Game Theory, Reciprocal Altruism, Tit-For-Tat

    I'm not saying tit for tat is exactly what christ taught. My point was that the first principle of tit for tat is (I think) the primary moral taught by Christ: Do un to others as you want them to do to you. Tit for tat employs that strategy by offering cooperation first (which is what it wants). It does what it wants the other to do....

    Secondly, it forgives mercilessly. There is no grudge keeping, which was also very much like what I understand Jesus to represent... a forgiving person.

  4. #4
    Join Date
    Nov 2006
    Location
    East of Eden
    Posts
    115

    Re: Game Theory, Reciprocal Altruism, Tit-For-Tat

    "My point was that the first principle of tit for tat is (I think) the primary moral taught by Christ: Do unto others as you want them to do to you."

    But see, I don't think that's true either. I believe Christ taught that one should do what is just, treat all people as your brothers and sisters REGARDLESS of how they treat you. So it's a step beyond tit for tat because you don't base your decisions and actions on how you want others to treat you, you base them on an altruistic principle that is, frankly, rather impractical, which is why you don't find many people practicing it. Tit for tat is more pragmatic.

  5. #5
    Join Date
    Dec 2004
    Posts
    464

    Re: Game Theory, Reciprocal Altruism, Tit-For-Tat

    Tit for tat is more Jewish, I think.

    E.g., in Judaism the court system is very biased towards the defendant--but there is a court system.

    As I understand Christianity, there wouldn't be as much as that--so if someone was a rapist, murderer, or other violent criminal, you'd let them keep doing it. Turn the other cheek?

    But then there's the part in Mark where Jesus tells his disciples to slaughter all the enemies of Jesus.

    So... :confused:

  6. #6
    Join Date
    May 2006
    Location
    USA
    Posts
    3,330

    Re: Game Theory, Reciprocal Altruism, Tit-For-Tat

    Quote Originally Posted by ComplexKid
    Tit for tat is more Jewish, I think.

    E.g., in Judaism the court system is very biased towards the defendant--but there is a court system.

    As I understand Christianity, there wouldn't be as much as that--so if someone was a rapist, murderer, or other violent criminal, you'd let them keep doing it. Turn the other cheek?

    But then there's the part in Mark where Jesus tells his disciples to slaughter all the enemies of Jesus.

    So... :confused:
    CK, this post I made indirectly address your question:

    http://scam.com/showthread.php?p=296748#post296748.

    Hope this helps.

    Hot

  7. #7
    Join Date
    Dec 2004
    Posts
    464

    Re: Game Theory, Reciprocal Altruism, Tit-For-Tat

    It doesn't work that way in Judaism, HP. Yes, capital punishment is allowed, but more than once in 70 years is seriously overdoing it by Jewish legal standards.

    It's almost impossible to get a conviction, in other words. And don't forget the Cities of Refuge that G-d ordered to be set up for anyone who accidentally killed someone so that a family member of the deceased wouldn't go kill that person in a rage.

    It was the same with eye-for-an-eye, etc. What that meant was that you had to pay for lost wages, medical expenses, and pain and suffering if you hurt somebody--not that if they put out your eye, you put out theirs.

    And on thievery, it was assumed that the thief needed the money. Now, it still wasn't right, but this is how it was solved. There isn't a different word in biblical Hebrew for slavery and employment in this situation, so here's how it goes: If you were found guilty of stealing, you got sold to the person you stole from. Now, if you had a trade, they had to put you to work doing that trade, and you paid back 120% of what you'd stolen. But you lived in their house, ate at their table, and had all the rights a regular employed person did, save that your wages went to restitution until that was paid off.

    If you didn't have a trade, they'd be respoonsible for teaching you one, or having someone else teach you one, and those wages would go to paying back the damages plus 20% extra.

    At the end of the term of 'slavery', your master had to make sure you were set up in business and had enough money to get yourself settled before you could be released. That's why there are also provisions for what to do if a 'slave' wanted to stay a 'slave'.

    It was actually a pretty progressive system for the time. If property crimes were involved, then the rich got to know how bad the poor had it, and had to treat them like human beings, and the poor got a chance to learn to support themselves.

    In the cases of capital crimes, it did everything it could to make sure that nobody got hot-headed and started a vengeance war.

  8. #8
    Join Date
    May 2006
    Location
    USA
    Posts
    3,330

    Re: Game Theory, Reciprocal Altruism, Tit-For-Tat

    Quote Originally Posted by ComplexKid
    It doesn't work that way in Judaism, HP. Yes, capital punishment is allowed, but more than once in 70 years is seriously overdoing it by Jewish legal standards.

    It's almost impossible to get a conviction, in other words. And don't forget the Cities of Refuge that G-d ordered to be set up for anyone who accidentally killed someone so that a family member of the deceased wouldn't go kill that person in a rage.

    It was the same with eye-for-an-eye, etc. What that meant was that you had to pay for lost wages, medical expenses, and pain and suffering if you hurt somebody--not that if they put out your eye, you put out theirs.

    And on thievery, it was assumed that the thief needed the money. Now, it still wasn't right, but this is how it was solved. There isn't a different word in biblical Hebrew for slavery and employment in this situation, so here's how it goes: If you were found guilty of stealing, you got sold to the person you stole from. Now, if you had a trade, they had to put you to work doing that trade, and you paid back 120% of what you'd stolen. But you lived in their house, ate at their table, and had all the rights a regular employed person did, save that your wages went to restitution until that was paid off.

    If you didn't have a trade, they'd be respoonsible for teaching you one, or having someone else teach you one, and those wages would go to paying back the damages plus 20% extra.

    At the end of the term of 'slavery', your master had to make sure you were set up in business and had enough money to get yourself settled before you could be released. That's why there are also provisions for what to do if a 'slave' wanted to stay a 'slave'.

    It was actually a pretty progressive system for the time. If property crimes were involved, then the rich got to know how bad the poor had it, and had to treat them like human beings, and the poor got a chance to learn to support themselves.

    In the cases of capital crimes, it did everything it could to make sure that nobody got hot-headed and started a vengeance war.
    Oh, right, and I agree, it was a progressive system. :)

    I should have mentioned, though, that I was actually addressing just this question that you posed regarding Christianity:
    Quote Originally Posted by ComplexKid
    As I understand Christianity, there wouldn't be as much as that--so if someone was a rapist, murderer, or other violent criminal, you'd let them keep doing it. Turn the other cheek?But then there's the part in Mark where Jesus tells his disciples to slaughter all the enemies of Jesus.

    So... :confused:
    I thought the link might help you with this.

  9. #9
    Join Date
    Dec 2004
    Posts
    464

    Re: Game Theory, Reciprocal Altruism, Tit-For-Tat

    Gotcha. Now let's stop talking religion in the science forum before somebody else boots us.

  10. #10
    Join Date
    May 2006
    Location
    USA
    Posts
    3,330

    Re: Game Theory, Reciprocal Altruism, Tit-For-Tat

    Quote Originally Posted by ComplexKid
    Gotcha. Now let's stop talking religion in the science forum before somebody else boots us.
    Haha!
    It took me 5 minutes to find this darn thread; I forgot what the name of it actually was and I've been looking for it, like the fool that I am, in the Religious forum. What the heck am I doing in Science, anyway? :eek:

  11. #11
    Join Date
    Mar 2006
    Location
    Sydney
    Posts
    1,167

    Re: Game Theory, Reciprocal Altruism, Tit-For-Tat

    *shrugs*

    So..the science.

    I have progressed a bit further along in the book and obviously things become much more complicated. They start to factor in the consequence of "accidents". Tit for tat is bad when one of the tit-for-tat'ers accidently defects...because then you end up with both constantly defecting. (resulting in losses)

    Secondly, they start to do non-synchronous runnings of the problem.

    Under some of these variations they found another strategy which beats tit-for-tat which they called "Pavlov".

    Basically it's strategy is to start cooperating. What it then evaluates is whether it wins or loses (win = both co-operate, or it defects and the other cooperates, while Lose = Cooperate vs defect and defect vs defect). It's strategy consists of "If it worked last round and I won, do it again. If last round = Loss, then change. So when it comes up against a tit-for-tat, it actually does equally as well as tit-for-tat (all cooperate), and it compensates in scenarios where an opponent accidently defects.... anyway. This is probably more difficult to understand that I can convey in a couple of sentences.

    Its really interesting though.

  12. #12
    Join Date
    Nov 2006
    Location
    East of Eden
    Posts
    115

    Re: Game Theory, Reciprocal Altruism, Tit-For-Tat

    Watch yourself, you may end up founding a new religion!

  13. #13
    Join Date
    May 2006
    Location
    toronto, ontario
    Posts
    34

    Re: Game Theory, Reciprocal Altruism, Tit-For-Tat

    The dilemma game is much easier to solve than tit for tat, which really isn't a moral principle at all.

    Simply change the labels of the strategies in the 2 person dilemma game, ie snitch v squeal, or something like that, to get the cooperative behaviour.
    Michael Webster, Ph.D LL.B.

    Suite 1250, 180 Dundas St West
    Toronto, ON

    Misleading Advertising Law

  14. #14
    Join Date
    Sep 2004
    Posts
    26,310

    Re: Game Theory, Reciprocal Altruism, Tit-For-Tat

    Quote Originally Posted by aegist
    *shrugs*

    So..the science.

    I have progressed a bit further along in the book and obviously things become much more complicated. They start to factor in the consequence of "accidents". Tit for tat is bad when one of the tit-for-tat'ers accidently defects...because then you end up with both constantly defecting. (resulting in losses)

    Secondly, they start to do non-synchronous runnings of the problem.

    Under some of these variations they found another strategy which beats tit-for-tat which they called "Pavlov".

    Basically it's strategy is to start cooperating. What it then evaluates is whether it wins or loses (win = both co-operate, or it defects and the other cooperates, while Lose = Cooperate vs defect and defect vs defect). It's strategy consists of "If it worked last round and I won, do it again. If last round = Loss, then change. So when it comes up against a tit-for-tat, it actually does equally as well as tit-for-tat (all cooperate), and it compensates in scenarios where an opponent accidently defects.... anyway. This is probably more difficult to understand that I can convey in a couple of sentences.

    Its really interesting though.
    anything outside of ones own sphere of experience/knowledge is/can be interesting!?like watching the games of rome from a tv set!?or seeing the big/complicated reduced to a small/simple eqation or theory of operation.

    what does prisoner mean!?is anyone free from prison if prison exists!?and if so,how!?what does the idea/reality of prison say about man!?how do you see your own life in relation to the game of tit for tat!?how do you see those in the real realm of a prison concerning that game!?why is the game necessary!?can it be favorably enhanced to add to the principle of altruism!?what would be the basis of this favoring!?why would/should it be employed!?the marines stormed guadacanal!?and the first in got wiped out!?you ready for the beachead!?or does interesting mean,i'm glad I dont have to play that game for my well being!?or at least it'
    s so sublimated, i can ignore it's real terror!?does the real world offer a clean game playing field!?if not,why!?hehe!!......just askin

Similar Threads

  1. Submit your website for free no reciprocal needed
    By Administrator in forum Mail Order Scams
    Replies: 0
    Last Post: 11-17-2013, 02:24 PM
  2. In theory...
    By Yeah Well Fine Then in forum General Chat
    Replies: 38
    Last Post: 01-05-2009, 09:36 AM
  3. The One Up theory
    By Mike_Edwards in forum Mail Order Scams
    Replies: 11
    Last Post: 06-27-2008, 07:07 PM
  4. Stoned Age Man Theory
    By enlightenment in forum Science Scams
    Replies: 30
    Last Post: 03-03-2007, 06:32 PM
  5. Ultra-Game.org or Ultra Game Ltd. Reshipping Scam
    By landofoz in forum Work at Home Scams
    Replies: 0
    Last Post: 12-06-2006, 10:45 AM

Tags for this Thread

Bookmarks

Bookmarks

Posting Permissions

  • You may post new threads
  • You may post replies
  • You may not post attachments
  • You may edit your posts
  •