Skip to content
TheChatArchiveThe Chat Archive
Geminigemini-2.0-flash-grounding· GoogleHugging Face LMArena Search Arena 24K

Explain Harsanyi's utilitarian theorem.

From a Chatbot Arena battleThe same person sent this prompt to two anonymous models at once. Compare it with gpt-4o-mini-search-preview’s answer.
  1. User

    Explain Harsanyi's utilitarian theorem.

  2. Gemini Assistant

    Okay, here's an explanation of Harsanyi's utilitarian theorem:

    Harsanyi's utilitarian theorem, a cornerstone in moral philosophy and welfare economics, essentially argues that if both individual preferences and social preferences adhere to certain axioms of rationality, then social welfare must be a weighted average of individual utilities. It provides a rigorous justification for utilitarianism, the ethical theory that advocates for maximizing overall social welfare, defined as the sum of individual well-being.

    Here's a breakdown of the key components and implications:

    Core Idea:

    The theorem demonstrates that under specific conditions, the only way to create a social preference ordering that respects individual rational preferences is to weight and aggregate those individual preferences. This resulting social preference ordering is utilitarian in nature.

    Assumptions/Axioms:

    Harsanyi's theorem rests on several crucial assumptions:

    1. Individual Rationality (Axioms of Rational Choice): Each individual's preferences must be rational. This typically means that individual preferences are:

      • Complete: For any two options, an individual either prefers one to the other or is indifferent between them.
      • Transitive: If an individual prefers A to B and B to C, then they must prefer A to C.
      • Reflexive: An individual must be indifferent between A and A
    2. Social Rationality: Social preferences (the preferences of society as a whole) must also be rational, adhering to the same axioms of completeness, transitivity, and reflexivity.

    3. Pareto Principle (or Pareto Indifference): If everyone is indifferent between two options, then society must also be indifferent between them. In simpler terms, if no individual is made better or worse off by switching between two options, then society should also be indifferent. A stronger version, Pareto improvement, states if at least one person prefers option A to option B, and no one dislikes the switch to A, then society should prefer option A.

    4. von Neumann-Morgenstern Axioms (for choices involving risk): When individuals and society face choices with uncertain outcomes (lotteries), their preferences must satisfy the von Neumann-Morgenstern axioms. These axioms are about how people should behave when facing risk, and they're needed to represent preferences with utility functions. The main ones are:

      • Continuity: There exists a probability mixture such that an individual is indifferent between the mixture and some other option.
      • Independence: If an individual prefers A to B, then they must prefer a lottery that yields A with probability p to a lottery that yields B with probability p, even if both lotteries also have a common prize C with probability 1-p.

    The Theorem's Conclusion:

    If all the above axioms hold, then there exist utility functions representing individual and social preferences such that the social utility function (W) is a weighted sum of the individual utility functions (Ui):

    W = a₁U₁ + a₂U₂ + ... + aₙUₙ

    Where:

    • W is the social welfare function.
    • Uᵢ is the utility function of individual i.
    • aᵢ is a non-negative weight assigned to individual i's utility. These weights reflect the relative importance of each individual's well-being in determining social welfare.

    Implications and Significance:

    • Justification for Utilitarianism: The theorem provides a powerful, axiomatic basis for utilitarianism. It shows that if you accept certain fundamental principles of rational choice and social aggregation (like Pareto), then you are logically compelled to accept a form of utilitarianism.
    • Welfare Economics: It's a cornerstone in welfare economics, providing a framework for evaluating social policies and resource allocation.
    • Ethical Considerations: It raises important ethical questions about how to determine the weights (aᵢ) assigned to different individuals. Should everyone's utility count equally (as in classical utilitarianism), or should some individuals' preferences be given more weight than others?
    • Resolving Interpersonal Comparisons of Utility: One of the long-standing problems in economics is comparing utility across individuals. Harsanyi's theorem provides a way to make these comparisons meaningful within a specific axiomatic framework.
    • Criticisms: The theorem is not without its critics. Some question the realism or ethical implications of the axioms themselves. For example, the assumption of social rationality is sometimes challenged. Others question whether the von Neumann-Morgenstern axioms accurately describe human behavior under risk.

    In simpler terms:

    Imagine you're trying to decide what kind of pizza to order for a group of friends. Harsanyi's theorem suggests that if everyone has rational preferences for different toppings, and the group's overall preference for pizza toppings is also rational, then the best way to choose a pizza that makes the group happy is to figure out how much each person likes each topping and then combine those preferences in a weighted way. The weights could represent how hungry each person is, or how much you value their happiness. If everyone is equally important, then you simply add up everyone's preferences for each topping to get the best outcome for the group.

    To make sure I've covered everything, I'll run a quick search for any additional details or nuances about Harsanyi's utilitarian theorem that might be helpful. Okay, I've reviewed the search results and here's a summary of some key points and criticisms related to Harsanyi's Utilitarian Theorem, building on the previous explanation:

    Key Additions and Nuances:

    • The Impartial Observer: Harsanyi also framed his argument using the concept of an "impartial observer."[1] Imagine someone who doesn't know what position they will occupy in society. Harsanyi argued that such an observer would choose to maximize the expected utility, leading to the same utilitarian conclusion.
    • Von Neumann-Morgenstern Utilities as 'Welfares': Harsanyi claimed that von Neumann-Morgenstern utilities should be identified with welfares[2].
    • Alternative Axiomatic Foundation: An alternative axiomatic foundation for Vickrey-Harsanyi utilitarianism exists[3].

    Criticisms and Considerations:

    • Fairness and Risk Attitudes: Harsanyi's utilitarianism has faced criticisms regarding fairness and different attitudes toward risk[2].
    • Unacceptable Inequalities: Maximizing expected social welfare could lead to unacceptable inequalities of utility[3]. Some argue that the theorem might justify significant disparities in well-being if it leads to a higher overall sum of utility.
    • Interpersonal Comparisons of Utility: A core challenge lies in comparing utility across individuals. Critics argue that von Neumann-Morgenstern utility doesn't inherently provide a basis for interpersonal comparison needed for utilitarianism.
    • The Meaning of Utility: Some argue that Harsanyi confuses a mathematical sum of VNM utilities with the substantive view of utilitarianism[4].
    • Whose Preferences Count?: Even behind a veil of ignorance, individuals might have different views about others' risk attitudes and how to weight their utility functions[3].
    • Normalization of Utility Functions: Determining equal weights requires settling on a way to normalize utility functions[5].
    • Alternative Interpretations of Utilitarianism: The theorem doesn't dictate what is being added up (e.g., happiness, preferences, etc.) it just says that if you have a way of measuring individual well-being, social welfare is the sum of those measures[5].

    In conclusion, Harsanyi's Utilitarian Theorem is a significant result that provides a framework and justification for utilitarianism under specific conditions. However, it's important to be aware of the assumptions and the ongoing debates surrounding its interpretation and implications.

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.