Ayn Rand/Objectivism Sightings
Free Radical Updates
Local Club Meeting Plans
News & Interesting Links
Rights, Games, and Self-Realization Part 1
Rights against Personal Injury for Two in Isolation
The broad yet personal moral ideal on which I shall later rely is this: to live a life self-chosen, to be a self unified with its life (Boydstun 1984). The moral personality will revere the moral ideal realized in himself and in others, but the first layer of the rights between two people, considered as absolutely isolated from wider society, can be laid out assuming more impoverished personalities. Such people are rightly called players by game theorists. A player recognizes (superficially) his own end-in-itself value, but only his own, and all players realize that that is the situation. Players awaken. They come to moral life. Contouring their behavior to selves seeking value and meaning, to the end-in-itself value of those beings, utility functions are transformed, and the players become persons. Transformation can proceed only so far without reciprocity, but at any stage, unilateral respect for rights appropriate to the circumstances will be possible (cf. Nozick 1981, 474–535; Mack 1984, 158–59).*
Consider two players, A and B, in isolation from wider society. From the game-theoretic perspective, with each player regarding the other as having at most instrumental value, when could the deliberate use of force be justified? Self-defense easily qualifies. If A is a threat to B, then B could reasonably respond with force destructive of A on the supposition that only B is end-in-itself valuable. That is the case whether A poses a threat to B intentionally or only incidentally in the course of another pursuit. Nor would it matter that A could not have foreseen that his action could bring harm to B.
The fact that defensive force is a sound move in the game means that it must be taken into account in resorting to force. Given formidable responsive powers of an opponent and substituting game-theoretic imperatives for the moral imperatives in the formal definition of a right, we obtain a first approximation to a right against personal injury: it will be game-theoretically right to use force in self-defense, and it will be wrong (a blunder) to initiate the use of force. That is not to say that A and B could not be found in violent conflict even though both have respected the rule. The chain of violence could begin with an accident.
The amount of force that A and B may rightfully apply in response to a threat is limited by game strategy alone where neither can be sure that his attack would disable his opponent. A and B here face a natural game—call it the Engagement game—having the following four payoff possibilities, where 1 is preferred over 2 over 3.
A stops and B stops: (1,1)
A fights and B stops: (1,3)
A stops and B fights: (3,1)
A fights and B fights: (2,2)
Each pair of numbers (a,b) represents the ordinal utilities to player A and player B, respectively, for a particular combination of choices, fight or not, made by A and B.
The game known as Prisoner’s Dilemma has the following four payoff possibilities, where 1 is preferred over 2 over 3 over 4.
A cooperates and B cooperates: (2,2)
A defects and B cooperates: (1,4)
A cooperates and B defects: (4,1)
A defects and B defects: (3,3)
In Prisoner’s Dilemma (PD), each player has four distinct possible payoffs, whereas in Engagement (E), each player has only three. Nevertheless, E has in common with PD a number of strategically significant properties. Like PD, E is a game of partial conflict, rather than a game of complete opposition; there is at least one outcome (first possible choice combination) that is preferred by both players over at least one other outcome (fourth possible choice combination). Like PD, E presents an option, fight, that minimizes the damage to a player assuming the worst about the opponent; for each player in E, fight dominates stop. The implied natural outcome of E, the one that would be expected most frequently when the game is played by rational players, is continued fighting. As in PD, the natural outcome of E is an equilibrium in the sense that either player would suffer if only he were to change strategy. Finally, as in PD, the natural outcome of E is Pareto-deficient; there is another possible outcome (first possible choice combination) that both players would prefer (Rapoport, Guyer, and Gordon 1976).
Fortunately, the natural outcome of Engagement (both fight) is less stable than the natural outcome of Prisoner’s Dilemma. The payoff found in PD, but not in E, is the one that can obtain because of what is known as temptation. If either player in E can and does retreat, there is no immediate incentive for the other to inflict avoidable harm beyond that required for self-defense; if either player in PD were to cooperate, the other would have added incentive to defect. When we add into account the fact that real engagements often occur over an appreciable time interval and are naturally decomposable into subsidiary successive steps, peace becomes even easier; we then have an iterated game of Engagement. Robert Axelrod has demonstrated the strong possibilities of cooperation (assuming players do not have inordinately high time preferences) in the iterated Prisoner’ Dilemma (1984). Surely iteration facilitates cooperation in Engagement as well.
There is here no sense to any concept of corrective justice even when we bring moral concerns into consideration. It is true that if A accidentally injured B, then A morally ought to do what he can to make B whole again. It is true as well that if A aggressed against B—if A, under no threat from B, maliciously or negligently generated the force that brought harm to B—then A morally ought to make B whole and may deserve to be punished in addition (Nozick 1981, 363–97; 1974, 59–63, 137–46). However, under the circumstances we have assumed for A and B, there is no one available to forcibly punish A with impunity. In the context of two people roughly equal in power and isolated from wider society, we find that each person has, game-theoretically and morally, a right to self-defense and a right of retaliation for the sake of deterrence, but there is here no point in speaking of rights to recompense nor of rights to punish (cf. Nozick 1974, 137–42).
A right against aggression or against the use of responsive force in excess of that necessary for defense evidently does not follow from game-theoretic considerations alone in those cases where A and B, in isolation from wider society, are very unequally matched. If only B has a gun, for example, and A is a threat (innocently or not-so-innocently) to B, then B’s choice between a defense that entails killing A and a defense that does not is arbitrary from the value-impoverished, game-theoretic point of view. True, if A were to live, he might become instrumentally valuable to B in the future, but for some A’s and B’s, that is not realistic. Where A is of no plausible value to B, it would be quite all “right” for B to kill A, whether or not that is required for self-defense. Taking into account considerations of what Nozick called value-theoretic aspects (moral aspects), B ought always to respond to A as a person, to that person’s end-in-itself value in addition to any instrumental value that B might possess. B should avoid killing A if possible (Nozick 1981, 462–73; Mack 1977, 289–91).
Rights to Liberty for Two in Isolation
The labor of A might be of instrumental value to B. If B were sufficiently powerful, he could make A his slave. There would remain some sense to saying that A has a right to be free, even supposing that A has no possibility of freeing herself by force. In the course of nature, B might one day need the assistance of A to survive. A might then take her freedom through a morally right act of omission.
If A and B were of roughly equal power and if the labor of each could be of instrumental value to the other, impressment is not a game-theoretically sensible option for either player. The payoff possibilities take the form of the game I shall call Equal Power.
A no force and B no force: (2,2)
A force and B no force: (1,4)
A no force and B force: (4,1)
A force and B force: (3,3)
Here I assume that the disutility of losing his freedom is to each worse than the damage expected in defending his freedom. We see that the payoff possibilities for Equal Power are those of a Prisoner’s Dilemma. A and B here face an iterated Prisoner’s Dilemma. From game-theoretic considerations alone, each should refrain from trying to enslave the other, and each should forcibly resist attempts by the other (Axelrod 1984).
A master-slave relation is, moreover, morally wrong for the master on at least two counts. Firstly, the master is not responsive to the end-in-itself value of the person he enslaves (Nozick 1981, 470). Secondly, the interposition of a slave in the life of a master disrupts the unity of the master with his own life. Unity of the self with its life requires engagement to the process of living, to the processes that make its life. That loss of unity in the master seems to me akin to Nozick’s broader suggestion that the life of a violator of rights loses meaning in the violations. In contrast to the master and slave, persons making willing transactions (at least persons who retain a fundamentally independent existence) have their individual unities preserved (Nozick 1974, 48–51).**
[In the original 1987 text of the present essay, there followed a treatment of rights surrounding paternalistic acts for two persons in isolation. That is here omitted.]
*On this paragraph, see also Nozick 1993, 50–59, and Mack 1998.
**See also Khawaja 1997, 111–30.
Axelrod, R. 1984. The Evolution of Cooperation. Basic Books.
Khawaja, I. 1997. A Perfectionist-Egoist Theory of the Good. Objectivity 2(5):95–147.
Mack, E. 1977. How to Derive Libertarian Rights. Reprinted in Reading Nozick. 1981. J. Paul, editor. Rowman and Littlefield.
——. 1984. The Fundamental Moral Elements in Rand’s Theory of Rights. In The Philosophical Thought of Ayn Rand. D. Den Uyl and D. Rasmussen, editors. University of Illinois Press.
——. 1998. On the Fit of Egoism and Rights. Reason Papers 23:1–21.
Nozick, R. 1974. Anarchy, State, and Utopia. Basic Books.
——. 1981. Philosophical Explanations. Harvard University Press.
——. 1993. The Nature of Rationality. Princeton University Press.
Rapoport, A., M.J. Guyer, and D.G. Gordon 1976. The 2x2 Game. University of Michigan Press.
Discuss this Article (4 messages)