Objectivist response to the "Prisoner's Dilemna"

[RationalBiker]

I found this in the 2003 Encyclopedia Britannica CD edition.

Unfortunately for ethical egoism, the claim that we will all be better off if every one of us does what is in his or her own interest is incorrect. This is shown by what are known as “prisoner's dilemma” situations, which are playing anincreasingly important role in discussions of ethical theory. The basic prisoner's dilemma is an imaginary situation in which two prisoners are accused of a crime. If one confesses and the other does not, the prisoner who confesses will be released immediately and the other who does not will spend the next 20 years in prison. If neither confesses, each will be held for a few months and then both will be released. And if both confess, they will each be jailed for 15 years. The prisoners cannot communicate with one another. If each of them does a purely self-interested calculation, the result will be that it is better to confess than not to confess no matter what the other prisoner does. Paradoxical as it might seem, two prisoners, each pursuing his own interest, will end up worse than they would if they were not egoists.

The example might seem bizarre, but analogous situations occur quite frequently on a larger scale. Consider the dilemma of the commuter. Suppose that each commuter finds his or her private car a little more convenient than thebus; but when each of them drives a car, the traffic becomes so congested that everyone would be better off if they all took the bus and the buses moved quickly without traffic holdups. Because private cars are somewhat more convenient than buses, however, and the overall volume of traffic is not appreciably affected by one more car on the road, it is in the interest of each to continue using a private car. At least on the collective level, therefore, egoism is self-defeating—a conclusion well brought out by Parfit in his aforementioned Reasons and Persons.

By "ethical egoism", the author was referring to Objectivism. What is the (or an) Objectivist response to this criticism? I have thought about it some, and with my limited knowledge so far, these are the issues I have with this position:

1. Is there in fact a claim that we would ALL be better off with an Objectivist society?

2. On what do they base that the confessor will be freed instantly? Confessions USUALLY result in convictions, then time in prison, unless plea bargained or the confession is thrown out. In fact the whole premise of the figures they come up with based on the iterations of confess/no confession doesn't make sense to me.

3. The author assumes "purely" self-interested in motive, not "rationally" self-interested in motive. He then therefore assumes the self-interested response would be not to confess. If the objectivist values integrity and honesty, then the moral choice would be to confess if in fact one committed the crime. This is being responsible for one's actions.

4. In the second problem, he again ignores rationality and the open ended nature of objectivity. As the objectivist learns or observes that the traffic is becoming congested, it would be in his rational self-interest to reconsider mass transportation, if that would indeed reduce his time getting to work.

Are these valid points that I raise? Are there other issues with these scenarios?

VES

[MinorityOfOne]

The problem with the first point is that it disregards principles in ethics. It presents the situation as though the context matters: as though the ethical egoist necessarily thinks that one can say whether or not they ought to confess without knowing whether they actually have anything to confess to.

You hit the nail on the head about the second one. Each individual car won't matter much, but there will come a point when people will reconsider. Actually, there's probably a parallel to economics here: there's some equilibrium state that people will tend to center around. If traffic gets worse than that state, more people will tend to take public transportation, thus bringing it back toward equilibrium; if traffic lessens significantly, more people will drive.

[y_feldblum]

The prisoners' dilemma bears no relation to reality. It is a highly constrained, highly artificial imposition, and it can tell you nothing of the validity of either principles or applications. It's not even a specific situation; it's an impossible situation and a mind game.

And, since morality ends at the point of a gun, the case cannot properly be considered an ethical dilemma.

The exact same is true of the commuters' dilemma. Granted, it's not highly artificial, but it is imposed and constrained, and morality still ends at the point of a gun.

[rainoffire]

I find this is a typical tactic of people critical of Objectivism. Arbitrary situations plucked out of thin air, without context or any relationship with reality.

By trying to present an 'Objectivist' position to their 'dilemma' you have already succumbed to the trap; you have granted it the validity it does not deserve.

[Prometheus]

You answered your own question with the second point you made. Why would someone confessing a crime be left scot free? What sort of society would that be, where an innocent individual, who the officials know has not committed the crime, be left to rot in jail?

So the point they make is purely irrational. And in the case of the commuter, what would happen if all of them decided to take the bus and the "state owned" bus breaks down, or someone calls a "strike"? I would rather take my chances with the car than go on the bus, it's as simple as that.

dinesh.

[Capitalism Forever]

Unfortunately for ethical egoism, the claim that we will all be better off if every one of us does what is in his or her own interest is incorrect.

Unfortunately for the author, I don't give a damn. I want myself to be better off, not "all of us." I'm rational because I'm selfish, not the other way round, and therefore no amount of reasoning will talk me out of my selfishness.

Besides, the claim is not that we will all be better off in every possible situation. The claim is that we will all be better off in most of the situations relevant to our lives, and therefore, all in all, we will all be better off in the whole of our lives.

I was going to say that the prisoner's [prisoners'?] dilemma is not one of the situations relevant to our lives--but, come to think of it, the opposite is true: It is THE kind of ethical choice we all have to make in a social context. Although the scenario is designed to make the point that, if we all act in our self-interest, we will all be worse off, it actually shows how, if we all act in our self-interest, we will all be better off than if we all acted on the opposite principle.

Assuming that the prisoners are innocent, how would a selfish man in an all-selfish society approach the dilemma?

"If I confess, I will get 15 years or I'll walk; if I don't, I will get 20 years or a couple of months:

  | C    N

--+-------

C | 15  20

N |  0  .2[/code]

"Both figures in the 'confess' column are lower than the corresponding numbers for pleading innocent, so if I don't know what the other guy will do, I am better off confessing. Yeah, but the other guy will just as easily arrive at the same conclusion, so what would actually happen is that we would both confess and both get 15 years. The premise that confession is in my self-interest leads to 15 years in jail for both of us, which is definitely not in my self-interest. This is a contradiction, therefore the premise is false; confession is [i]not[/i] in my self-interest.


"So I will follow my self-interest and not confess, and--since in this society we [b]all[/b] act in our self-interest--so will my mate. Which is great, because it will get us released in a couple of months."


The key assumption is that the individuals [i]all[/i] act on the same principle. If they do, it means that they will take the same action, and thus the outcome will be along the main diagonal in the matrix: 15 or .2. The 0 and the 20 that result from the prisoners acting differently will be outside the range of possible outcomes. The choice is not between 15/0 or 20/.2; it is between 15 and .2--and selfish people will choose the latter. Thus, all individuals acting in their self-interest does indeed lead to the "group optimal" outcome.


It is only when some people break the principle that the optimal outcome does not result. In the absence of selfless people--when the selfish prisoner can [i]count[/i] on the other guy making the same choice as he does--there is no need to worry about the possibility of becoming a victim and being jailed for 20 years. And, while in the example the victimizer would happen to end up very well off, in real life the values gained by victimizing innocents are always offset by the other consequences of the act.


So the prisoner's (prisoners' whatever) dilemma is actually a rather good analogy of how selfishness does NOT mean theft or fraud (= confessing to get the perceived "0" in the lower right corner) but voluntary trade and cooperation (= going for the ".2" by not confessing).

[RationalBiker]

Thanks all for the responses. It's pretty much a wash as I figured, but for more reasons than I suspected.

This philosophy requires that I view things in a different light than I had in the past so exercise is necessary to change my thinking muscle.

VES

[Bearster]

The commuter question is based on the same Marxist premise that plagues most discussions of politics or economics today. Wealth is zero sum, and it can never be increases or decreased, only distributed and redistributed.

If roads were selflessly, altruistically controlled and subsidized by the government for the public good (wait, that's how it is now!) then they would be in desperate need of repair, and in very short supply.

If roads were in selfish private hands, then there would be a profit in building more roads long before 20-mile commutes turn into 2-hour traffic jams!

[y_feldblum]

Bearster's answer explains what I meant by "artificial, imposed, constrained" and by ethical-egoist morality not being applicable (to the situation as a whole).

The dilemma is based on the premise that morality is possible under force. Bearster shows exactly where the force lies. (In the second dilemma, anyway; the first dilemma isn't hard to figure out.)

[Censure the Opposition]

The problem I see here is that the prisoner's dilemma offered in the original post has a few errors to it (At least compared to the one I have seen utilized time and time again by economists). It's supposed to be fairly straightforward.

First, the prisoners are supposed to be guilty, there's no question of whether or not they committed the crime (At least in their minds). With that, you eliminate the possibility that they are confessing to a crime they didn't commit and the ethical dilemma inherent in doing that.

Most importantly, the confession is tied to ratting out the other prisoner, so hopefully that eliminates some of your doubts as to how realistic the scenario is. With that plea bargain, it's plausible to assume the prisoner would get no jail time.

You also have to assume that the prisoners each have no idea what the other is going to decide.

So, the best outcome for the individual prisoner is no jail time. The only way to achieve that is to confess and rat out the other prisoner. Therefore, if both act in their OWN self-interest trying to get no jail time, it would backfire and they would both end up with 15 years (Or whatever figure, it really doesn't matter).

The best outcome for both is a handful of months jail time (Since they're both guilty anyway) and the scenario is designed to show that in SOME cases, cooperating for the "greater good" is better for everyone. There's no inherent premise about all of society acting a certain way because that gives insight into what the other prisoner would do and negates the whole exercise (as capitalism forever has shown).

Also keep in mind that it's an economic construction and economic models rarely truly reflect every detail of reality.

Hopefully that helps with some of the confusion.

[source]

I found this in the 2003 Encyclopedia Britannica CD edition.

By "ethical egoism", the author was referring to Objectivism. What is the (or an) Objectivist response to this criticism?

First of all, we can only guess how things would work in a laissez-faire capitalism. To say a few things about the example above, of commuters travelling by car or by bus. Clearly, in laissez-faire capitalism, which is by the way the only ethically sound system according to rational egoism, roads would be private property. I am no road specialist, but for a rational egoist who also happens to be a road specialist, a congested traffic system would be like heaven (if I can borrow the meaning of this term). There is no knowing what solutions would a free mind come up with, unless it was set free. It is also hard to tell whether it would be a roads specialist who would solve the problem. Perhaps someone would invent a new kind of transport which wouldn't require roads. Then again, who says that there would be any commuters at all? Perhaps architects who weren't so limited by today's prevailingly socialist standards would be able to design such buildings which would at the same time serve as industrial buildings and as home for the workers in that industry.

But there are two errors in the above argument. The less important one is that all of that is science fiction. The more important one is that by saying that, I would actually be accepting an invitation of the socialist/altruist to fight the battle on the ground of his choice, instead of stating the relevant facts which concern the issue.

What is the alternative to a free society? Slavery. Socialism, communism, feudalism, etc. What is the alternative to rational egoism? Cynicism? Altruism? Self-sacrifice. The world you have today. Neither of the alternatives is moral. Even though a free society is not a functional/practical society in socialist/altruist view, it is still the only morally acceptable society.

I have thought about it some, and with my limited knowledge so far, these are the issues I have with this position:

1. Is there in fact a claim that we would ALL be better off with an Objectivist society?

There is no such thing as an Objectivist society. I assume what you mean is a capitalist society, namely laissez-faire capitalism. History has already given the answer. Look at America in 18-19th century.

2. On what do they base that the confessor will be freed instantly? Confessions USUALLY result in convictions, then time in prison, unless plea bargained or the confession is thrown out. In fact the whole premise of the figures they come up with based on the iterations of confess/no confession doesn't make sense to me.

It's not a bizzare example as they say, but a stupid example. They try to prove their point by resorting to examples which aren't and can't be real, pretty much as the priests base ethical principles on unreal ideas which they call supernatural forces.

3. The author assumes "purely" self-interested in motive, not "rationally" self-interested in motive. He then therefore assumes the self-interested response would be not to confess. If the objectivist values integrity and honesty, then the moral choice would be to confess if in fact one committed the crime. This is being responsible for one's actions.

Howard Roark's one example.

4. In the second problem, he again ignores rationality and the open ended nature of objectivity. As the objectivist learns or observes that the traffic is becoming congested, it would be in his rational self-interest to reconsider mass transportation, if that would indeed reduce his time getting to work.

See my above argument.

[Richard_Halley]

It seems to me that the Prisoner's Dilema only provides an example in which it would be hard to tell what is in your own self interest, it does not provide an example of time when it is better to act against that interest.

[Tom]

It seems to me that the Prisoner's Dilema only provides an example in which it would be hard to tell what is in your own self interest, it does not provide an example of time when it is better to act against that interest.

Yes, but the altruist regards "self-interest" as a context-less absolute. That's one of the reasons WHY they're altruist.

[Living Student]

another thing:

Paradoxical as it might seem, two prisoners, each pursuing their his own interest, will end up worse than they would if they were not egoists.

But niether of them are responsible for this. Only by the other's sacrifice, did either of them benefit. Their actions were largely unconsquential when it came to their own future. And to the extent that they were consquential, to confess would have still been the better choice.

The author incorrectly concluded that to act unselfishly is in your own self-interest.

Gee, how did he do that?

He should have concluded (based on his "dilemma") that to have your fellows act unselfishly is in your own self-interest. And that you should act selfishly.

Most importantly of all, as most everyone else has already said, is that this "dilemma" isn't a useful thought-experiement in dealing with the real world.

In fact, its a very bad one. For a number of reasons. But I don't need to preach to the choir

[Bearster]

there's no question of whether or not they committed the crime

The only question, therefore, why in the name of Mary Magdalene would anyone believe the prisoner's dilemma says something out egoism?

Two criminals, caught, in separate jail cells pragmatically plotting to get off with the least punishment and trying to outwit eachother and their jailor. Yup, sounds like producers and property owners and investors trying to live selfishly!

[DavidV]

I am a big fan of game theory, perhaps because I was often exposed while getting my economics and political science degrees. To the best of my knowledge, there can be no contradiction between game theory and ethics for the same reason that there can be no conflict between the values of rational men. Game theory is a method of logically evaluating the potential value of the actions of rational actors by assigning numerical values to the potential benefits of those actions. As such, a game theory model accurately represents reality as long as the model correctly represents the values as stake. Since they accept altruism, game-theoreticians themselves are often unable to appreciate this.

The challenge is then to correctly represent the values at stake. The main limitation of game theory is that human interactions have consequences beyond the immediate gain/loss. In other words, human beings act according to their values and judge others according to theirs. The difficulty of assigning numerical values to value judgments of oneself and others limits the usefulness of game theory models.

For example, in the prisoner scenario mentioned above, the error would be to consider it a one-time interaction. In reality, the actions of rational actors have consequences far beyond the immediate loss/gain. In other words, pragmatism is not in one’s self-interest. If we consider the prisoner scenario as a long-run equilibrium, game theory would indicate that betraying a partner in crime is not in one’s self-interest. Within the context of that model, that is the correct outcome and a useful lesson, as it shows that a short-term benefit from cheating someone will result in a long-term loss.

It would be incorrect to say that the above example demonstrates that dishonesty is in one’s self interest, because that would require a great deal of additional factors to consider, such as the long-run results of being a liar. It is currently impossible to quantify the psychological harm that comes from being dishonest – but that does not disqualify game theory. Rather, it limits the scope under which that model is valid, just as Newtonian physics is limited in its application.

And if the above didn’t make much sense, the error in the Encyclopedia Britannica entry is that it defines selfishness as whim-worshipping pragmatism.

Suppose that each commuter finds his or her private car a little more convenient than the bus; but when each of them drives a car, the traffic becomes so congested that everyone would be better off if they all took the bus and the buses moved quickly without traffic holdups.

The error here is that it is an example of the “tragedy of the commons,” or the lack of well-defined property rights. In a private road system, the efficient solution would be to make each driver bear the full cost of driving. The more basic answer is that there is no conflict of interest between rational men.

[RationalBiker]

To clarify a couple things that some may have missed. The version of the Prisoner's Dilemna was copy/pasted right off of the 2003 Encyclopedia Britannica CD. It's reproduction has not been altered by me.

Secondly, the original author lumped Objectivism in with the crowd he/she referred to as "ethical egoists". Therefore, when I referred to an objectivist society, I was only doing so in response to the premise put forth by the original author in referring to "ethical egoists" (not capitalists). Since I'm new to this, I may make mistakes from time to time in my terminology.

That said, I appreciate the wide and varied responses pointing out all the problems with this "dilemna" in a much better fashion than I could.

VES

[Bearster]

Thus, stated another way, the problem in using criminals in an example showing how to determine the selfish choice among several alternatives is that criminals are criminals because of their profoundly selfLESS actions.

[The Sailor]

This is basically John Nash's equilibrium theory put in another way., is it not.

The Nash equilibrium concept solves in a plausible and fairly tractable manner the

basic problem of human interaction, which is to determine the behaviours and actions

of two or more individuals or organisations (such as firms, political parties and

governments) when the behaviours and actions of all these parties potentially affects

each party's welfare (or payoff). Much human interaction (be it economic, political or

social) is of this type; that is, a party's welfare (or payoff) depends not only on his or

her actions but also on the actions of other parties. This type of situation is known

as a `game' situation. In such a strategic situation, the best (or optimal) action for a

player (such as an individual, a firm or a government) depends on the actions chosen

by the other players, and vice-versa. Which action, then, should a player choose to

achieve her objectives? The Nash equilibrium concept provides the answer to this

essential question.

[DavidV]

This is basically John Nash's equilibrium theory put in another way. is it not.

Actually, I don't think so. The Nash equilibrium is the dominant scenario, which in this case would be for the suspects to rat each other out.

Nash’s theory itself however, is valid and interesting, at least when properly applied. For more, see http://www.gametheory.net/

[BreathofLife]

The problem with the Prisoner's Dilemma is that they are trying to make you practice rational concepts in an irrational world. That's always the way people attack rationality: by trying to disprove it with irrationality.

[source]

BreathofLife,

You are so correct!

But notice also that they actually say then that acting irrationally is rational. I've seen in many motion pictures now (especially star trek movies and series), how certain characters are told that it is irrational of individuals to fight for ideals against the whole society, that it is much simpler and rational to "look at the world objectively and give in," meaning by this that they should stop pursuing their happiness, because some tradition forbids it or there is a law against it.

[BreathofLife]

Have you noticed that evil usually disguises itself in the robes of Good? It makes sense then that those same people would try to make irrationality seem rational. It's because they are evil yet they know that man is a rational being and any man who actualizes his potential to be rational, seeks the Good.

[Elle]

The Prisoner's Dilemna

You and your partner rob a bank and you get caught. You are seperated and given the following options (you can not collaborate and it is assumed that you did not determine how you would respond to this situation beforehand):

-If you both confess you each get 5 years in jail.

-If neither of you confess you each get 2 years in jail.

-If one confesses and the other does not then the one who confesses goes free and the one who does not gets 10 years in jail.

What would you do?

This just irked the hell out of me. In what situation of justice would this EVER be a realistic scenerio? Why is the punishment determined by the confession, as opposed to the evidence? Why do we compare a bank robbery to a business strategy when one is a destruction of value through taking of the unearned, and the other is the creation of value? This damn prisoner's dilemna just keeps coming up. Economics classes have so many problems and I'd be interested to hear from anyone with more of a background in econ - since I am just beginning to study it. Any guidelines for getting through the introductory courses (Macro and Micro) as well as any reading suggestions would be great. Thanks!

[JMeganSnow]

The Prisoner's Dilemna

You and your partner rob a bank and you get caught.  You are seperated and given the following options (you can not collaborate and it is assumed that you did not determine how you would respond to this situation beforehand):

-If you both confess you each get 5 years in jail.

-If neither of you confess you each get 2 years in jail.

-If one confesses and the other does not then the one who confesses goes free and the one who does not gets 10 years in jail.

What would you do?

This is not the version of the Prisoner's dilemma that's familiar to me. The one I've heard goes something like this:

You and a friend are in prison for a crime and receive a certain median treatment. If one of you rats on the other, the rat receives better treatment and the ratee receives worse. If you both rat, you both receive worse treatment.

It's essentially similar, though. And equally stupid. Catch 22 works only if you drop context, forgetting that there are OTHER options. I love these idiot "ethics" problems, they're so dumb.