Preference

The notion of preference is central to many social science disciplines. It also raises a number of philosophical issues central to the philosophy of social sciences. However, there is considerable disagreement about how to interpret this concept, with various incompatible accounts in use. This entry starts by sketching the main formal properties of the preference notion and their justification. It then addresses the relation of preference to choice, and the various interpretations offered.

The preference concept is often characterized as a binary relation defined over a set X of alternatives. In particular, there are two basic relations, strict preference (≻) and indifference (∽), and one derived relation, “at least as good as” (≽), defined as follows:

In most applications, preferences are assumed to be complete. That is,

Completeness is sometimes difficult to justify. Indeed, we often do not have, and do not need, complete preferences. Consider a person who has to choose between three alternatives A,B, and C. If she knows that she prefers A to the others, she does not have to make up her mind about the relative ranking between B and C.

Incompleteness may arise for different reasons. Agents may not have a preference between two alternatives because they have not sufficiently thought about the alternatives (or lack sufficient information to do so). Thinking more about the issue will thus resolve incompleteness. Yet not all cases of incompleteness are resolvable. Even after maximum information intake and reflection, an agent may not be able to rank two alternatives. Such unresolvable incompleteness problems often arise when people consider alternatives to be incommensurable.

The other main assumption concerns the transitivity of preferences, which says that

People often violate transitivity. The controversial question is whether transitivity should be a normative-rationality requirement.

The money pump argument states that those who violate transitivity stand to be exploited. Take the case where an agent has the preferences A ≻ B, B ≻ C, and C ≻ A, (i.e., transitivity is violated), and she currently possesses C. Then, a cunning trader could offer to exchange (for a small fee) C for B, which the agent will accept given her preference. The trader could then do the same by offering first A and then C. In the end, the agent is left with the same alternative, C, as before the trade but has paid three small fees to the trader. Violating transitivity thus is considered irrational because it makes the violator subject to exploitation.

Another argument for the normative appropriateness of preference transitivity suggests that transitivity is constitutive of the meaning of preference. For example, Donald Davidson has claimed that violating transitivity undermines the very meaning of preferring one alternative to others.

All the arguments for preference transitivity are philosophically controversial. The money pump argument applies straightforwardly only to transitivity violations of strict, but not weak, preferences. Even then, additional conditions must be satisfied for the cunning trader to be able to arrange the money pump. A critic of the meaning argument can point out possible connections between intransitive preferences and choice.

...