Might theory X be a theory of diminishing marginal value?

Act Utilitarianisms divide into Total and Average versions. Total ver-sions seem to imply Part’s “Repugnant Conclusion”. Average versions are proposed in part to avoid the Repugnant Conclusion, but these are subject to “Mere Addition ” arguments as detailed by Hudson in “The Di-minishing Marginal Value of Happy People”. Thus, various intermediate versions of utilitarianism, such as the one investigated by Hurka in “Value and Population Size”, take on interest. But Hudson argues that such compromise theories are subject to the mere addition arguments, and are therefore no improvement over Average Utilitarianism. I disagree: some such compromise solutions escape Hudson’s Mere Addition arguments. 1. Totalism Utilitarianisms require us to maximize value. We can develop this basic idea in the following way.1 At any time, some possible worlds are accessible to me. These worlds may be good or bad to various degrees. Intuitively, act utilitarianism claims that I ought to behave as I do in the best of these worlds. Equivalently: AU: it is right at some time t for S to see to p iff p is true in some world w such that i) w is accessible to S at t, and ii) there is no world w ′ which is accessible to S as of t, and better than w. AU is incomplete, for we have said nothing about which worlds are better than others. Only when coupled with a theory of the better than relation for possible worlds does AU generate results. The true theory of this relation, “Theory X”, has proved to be elusive.2 One family of such theories is familiar. Assume that for every person x existing at world w, there is a number, Ux,w, which represents her life utility at ∗I would like to thank Fred Feldman for his kind and invaluable help with this paper, and also Phillip Bricker for his careful comments.