Chuanjie Sheng

This paper clarifies and interprets some basic quantitative concepts of value, utility and utility function from a utilitarian point of view. First, I discuss the question as to whether value is objective or subjective. I hold that value is subjective and statistical in nature (although from the various subjective values of a certain object a norm can usually be obtained). Second, I emphasize the distinction between use value and exchange value in relation to utility. Third, I propose a law of diminishing incremental interest, which refers to the incremental (marginal) utility of money. Fourth, I identify the utility of money with the von Neumann-Morgenstern utility. Fifth, I question the necessity of the usual normalization of utility functions and the restricted linear transformation (and the consequent concept of “strategic equivalence”). Sixth, I discuss in detail the terminal values and utilities of a utility function from a philosophical rather than mathematical point of view, particularly the boundedness of a utility function and the magnitudes of V0 and U0. Finally, I conclude that, in order to be able to have interpersonal comparisons of utility, utility should have the same dimension as value rather than no dimension, and the normalization problem should be reconsidered in the light of terminal values and utilities.

This paper clarifies some basic concepts or assumptions of the prisoner's dilemma, asserts the independence between the two agentsA andB, and advocates the application of the dominance principle of decision theory to the prisoner's dilemma. It discusses several versions of the prisoner's dilemma, including the one-shot and repeated cases of a noncooperative game from a purely egoistic point of view. The main part of this paper, however, is a study of the problem from a moral point of view through a special decision-theoretic approach. Morality is taken into account by incorporating the utility of the feeling of moral satisfaction for the agent, as a part of the total utility for the agent, into the decision-theoretic model. In this way the problem will appear as a purely technical decision problem, and the conflicts between various assumptions, or the dilemma caused by the problem, will no longer exist. It is also pointed out that in a more general case, for some values of the coefficient of moralityk, dominance will not exist so that the dominance principle will not be applicable.

At Moore’s time, the main-stream ethical theory is the doctrine that pleasure alone is good as an end as held by the hedonistic utilitarianism. Moore, however, asserts that good, not composed of any parts, is a simple notion and indefinable, and naturalistic ethical theories, in particular hedonistic utilitarianism, interpret intrinsic good as a property of a single natural object---pleasure, which is also the sole end of life, thus violates naturalistic fallacy. Moore seems to believe that there exist things other than pleasure that are also intrinsically good and has searched for them. But Moore has not clearly stated what these things are, nor has he given any justification for why they are intrinsically good. This paper discusses Moore’s arguments and difficulties of utilitarianism. With the subjectivistic utilitarian theory of value, Unified Utilitarian Theory (UTT) discards the classification of value into intrinsic and instrumental and proves to be exempt of all theexisting difficulties with utilitarianism related to pleasure, including naturalistic fallacy, vagueness of pleasure and sole end of life, double counting, etc.