Equal probability model

The equal probability model by JC Harsanyi is a thought experiment for modeling a hypothetical starting situation for a rational and ethically justifiable decision.

In his theory, John Harsanyi advocates a moral philosophy that is not based on the foundations of institutions, but measures actions against a demonstrable social benefit. In doing so, Harsanyi tries to subordinate the moral decision-making situations in which people find themselves to consequentialist rationality postulates and thus develops his modern version of classical utilitarianism . Harsanyi sees ethics as a sub-theory of a general theory of rational action.

Classification of the theory

Harsanyi's theory is part of a utilitarian ethic and can therefore be ascribed to normative ethics. Harsanyi himself speaks out in favor of preferential utilitarianism . With regard to the debate between rule and action utilitarians, he considers rule utilitarianism to be more suitable for ensuring cooperation and reliable behavioral expectations among several actors.

Basis of the model

Several pillars form the basis of Harsanyi's theory. On the one hand, modern decision theory should be mentioned. It is a branch of probability theory and is used to evaluate the consequences of actions.

Harsanyi names game theory as the second pillar . It is about the modeling of situations in interaction systems that consist of at least two individuals.

Adam Smith's considerations represent a further basis . He designed the figure of an objective and compassionate observer (of society). At Harsanyi, every decision-maker is an observer. However - in contrast to Smith - Harsanyi's observer "lives" in society and is affected by his decisions, which of course makes it difficult to ignore personal interests during the (moral) decision-making process.

Harsanyi also bases his theory on Immanuel Kant . At Harsanyi, the Kantian universalisability also plays an important role. A decision that cannot be universalized cannot be a moral one.

Ultimately, as already seen, utilitarianism provides the moral-philosophical foundation. Like utilitarianism, Harsanyi also takes the maximization of social utility (here: average utility ) as a basic principle.

The model of the average utility principle

The thought experiment is based on the fact that a moral decision - in order to be qualified as such at all - must not depend on the personal preferences of the decision maker. Since such personal preferences are hardly suppress Harsanyi designed the " equal probability model " ( "equiprobability model"). This involves the following decision-making situation: A society consists of members. A decision should now be made between several alternatives. None of the individuals has any knowledge of the position in which they will find themselves later, when one has decided on an alternative. It could take the best place (place 1) as well as the worst place (place ). The individual must expect to achieve each of the positions in later society with the same probability. ${\displaystyle n}$ ${\displaystyle n}$ ${\displaystyle n}$ ${\displaystyle n}$

The decision of the individual must be based on his expected benefit ( ), the expected benefit . The expected utility ( ) resulting from the environmental conditions with the probability of occurrence (with , and ) is defined by . ${\displaystyle U}$ ${\displaystyle \operatorname {E} (U)}$ ${\displaystyle \operatorname {E} (U(s))}$ ${\displaystyle s_{i}}$ ${\displaystyle \pi _{i}}$ ${\displaystyle i\in \mathbf {N} }$ ${\displaystyle 0\leq \pi _{i}\leq 1}$ ${\displaystyle \sum \pi _{i}=1}$ ${\displaystyle \operatorname {E} (U(s))=\sum _{i=1}^{n}\pi _{i}U(s_{i})}$

However, Harsanyi is not concerned with maximizing the benefit of an individual, but with an impartial decision maker (the individual ) maximizing the overall welfare of society ( ). Society consists of individuals whose preferences must all be weighted equally. As mentioned above, "lives" of decision-makers in society, that is a part of . ${\displaystyle i}$ ${\displaystyle W_{i}}$ ${\displaystyle j}$ ${\displaystyle i}$ ${\displaystyle j}$

For DC probability of all positions, ie with and in the equal weighting of all preferences, arises as a social welfare function of an alternative: . ${\displaystyle \pi _{1}=\pi _{2}=\cdots =\pi _{n}}$ ${\displaystyle W_{i}={\frac {1}{n}}\sum _{j=1}^{n}U_{j}}$

This equation represents the welfare of society as a whole as the arithmetic mean of all individual utility functions. Since a mean can also be referred to as an average, the equation is called the "average utility principle ". If all environmental conditions are equally likely, the amount of expected utility of an individual actor in the equal-probability model is equal to the average utility.

Maximizing the average benefit as a decision rule

For all available alternatives, the benefit characteristics are first determined in each possible situation, each characteristic being given the same probability. Then the resulting average utility is calculated for all achievable alternatives. The alternative with the maximum average benefit is then selected.

Maximizing the average benefit

Harsanyi sees the maximization of the average utility as the only decision rule whose application is permissible in the situation of the equal-probability model. In the related concept of the " veil of ignorance " by John Rawls , a decision according to the Maximin rule is required. Harsanyi strongly opposes such an approach.

Another difference to Rawls is that Harsanyi assigns a weight of to each voice, while Rawls gives each voice an infinite weight. Majority decisions are therefore possible at Harsanyi. With Rawls, on the other hand, any individual can veto the decision. ${\displaystyle {\frac {1}{n}}}$

Individual evidence

1. ↑ HARSANYI, JOHN C. (1977): Morality and the Theory of Rational Behavior in Social Research 44, pp 625
2. ↑ HOMANN, KARL (1988): Rationality and Democracy (The Unit of Social Sciences 57), Tübingen: Mohr (Siebeck), p. 220
3. ↑ HARSANYI, JOHN C. (1977): Morality and the Theory of Rational Behavior in Social Research 44, pp 623-630.
4. ↑ See SMITH, ADAM (1790): Theory of ethical feelings, Felix Meiner Verlag, Hamburg 2004
5. ↑ HARSANYI, JOHN C. (1975): Can the Maximin Principle Serve as a Basis for Morality? A Critique of John Rawls's Theory, in: American Political Science Review 69, pp. 594-606