Abstract
There are probably no unconditional preferences. Nevertheless it is useful to discriminate conceptually between conditional and unconditional, or categorical, preferences. A person’s preference for x over y is unconditional if this preference is independent of the person’s belief about the truth or falsity of any proposition p. If a person perferred x to y — prefers, e. g., eating vegetarian food to eating meat — but changed this preference after having been told that eating meat would improve her health, then she had a conditional preference. In this case the preference was not independent of any proposition.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
In Savage (1954), the subjectivist standpoint is even made explicit insofar as he identifies consequences with subjective states of the acting person (see Savage (1954), p. 14). For a philosophical exposition of a subjectivist cum consequentialist notion of practical rationality see Griffin (1986) . According to the structuralist account in the philosophy of science (see e. g. Sneed (1971), Moulines (1976) and Stegmüller (1979b)) a scientific paradigm is characterized fundamentally by two components: a (mathematical) structure and a (paradigmatic) range of applications. Basic concepts are (partly) interpreted by specific examples of application. The paradigm of standard decision theory common to all models is subjectivist cum consequentialist in terms of standard applications (on ‘paradigmatic’ case-studies cf. e. g. Becker (1976)). For an analysis of different models of decision theory in terms of expliciteness and application see Spohn (1978); supplementary material is contained in Nida-Rümelin (1987a) and in Schmidt (1995).
Jeffrey (1965a).
To my knowledge, Frederic Schick first brought up the problem of interdependent preferences in the context of collective choice theory, but his critique did not initiate a fundamental revision of preference theory. In his paper ‘Democracy and Interdependent Preferences’ (Schick (1972)), he proposed two new procedures for aggregating preferences, one is a version of the ranking method, the other relies on the identification of an optimal attainable equilibrium (cf. also Schick (1971)). Certainly, both methods allow for a much more refined analysis of collective choice procedures compared to the usual ones in Arrow’s conceptual frame (cf. Arrow (1951)), but if my argument is sound, new procedures of aggregation based on a uniform concept of preference do not really solve the problem. I shall not discuss how the concept of preference has to be modified in order to allow for an adequate treatment of interdependent preferences. One possibility is to use the concept of a social position, which, e. g., allows to make the concept of categorical preferences more explicit, cf. Kern/Nida-Rümelin (1994) ch. 8 and 11.
Cf. Frankfurt (1971); Dennett, ‘Conditions of Personhood’, in Dennett (1978).
For a critical discussion and game-theoretic criteria of rational choice, esp. the notion of equilibrium cf. W. Spohn, ‘Wie läßt sich Spieltheorie verstehen?’, in Nida-Rümelin (1 994c).
Cf. Nida-Rümelin (1987a), ch. 7.
Facing the general practise of using cardinal utility functions even when the utility is not arrived at by constrained preferences, this does not seem to be too strong a heuristic assumption in this context. See, e. g., the discussion of different criteria for decision situations under uncertainty in Luce/Raiffa (1957), pp. 275 ff. However, although I cannot give a full formal treatment in this chapter, I believe it to be possible to cardinalize interdependent preferences using a modification of the classical axiomatics, e. g. the one in Luce/Raiffa (1957), pp. 24 ff, along the following lines. Take as alternatives first of all a set of states, where two different states are regarded as equal if they merely differ in a person’s preference. Then differenciate this set of states by adding to the descriptions phrases such as ‘... and the value P assigns to it is x.’ With this set, form gambles in the usual fashion and constrain a person’s preferences over this extended set by the standard coherence axioms. Of course, this only leads to a class of utility functions which can be narrowed down, for our purposes, by linear transformation to functions bounded by 0 and 1.
For a mathematical treatment see Nida-Rümelin/Schmidt/Munk (1996).
The model used here is extremely simple. But it suffices to see that the answer which one might try, namely to save the uniform value function based on revealed subjective preferences, which is to regard equilibrium states of interdependent valuations as the real utilities, would be inadequate. There are relations to the theory of dynamical games, see e. g. Holler/Illing (1993, pp. 113 ff.) and Skyrms (1992), which can not be discussed here.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1997 Springer Science+Business Media Dordrecht
About this chapter
Cite this chapter
Nida-Rümelin, J. (1997). Interpersonal Dependency of Preferences. In: Economic Rationality and Practical Reason. Theory and Decision Library, vol 24. Springer, Dordrecht. https://doi.org/10.1007/978-94-015-8814-0_7
Download citation
DOI: https://doi.org/10.1007/978-94-015-8814-0_7
Publisher Name: Springer, Dordrecht
Print ISBN: 978-90-481-4828-8
Online ISBN: 978-94-015-8814-0
eBook Packages: Springer Book Archive