Abstract
This paper develops an account of what it is that rational agents choose and what it is that rational agents prefer. There are three desiderata to satisfy when offering such an account. First, the account should maintain canonical axioms of rational choice theory as intuitively plausible. Here I focus on contraction and expansion consistency properties. Second, the account should prevent canonical axioms of rational choice theory from becoming trivial—it should be possible to actually violate these axioms, less rational choice theory becomes useless for many purposes. Third, the account should allow rational choice theory to be put to several different philosophical projects. I show that existing accounts of what we choose and prefer fail along at least one of these metrics. The account I develop does not fail across any of these metrics.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
These are just two very common consistency properties, and only two common ways of defining these two properties. For extensive overview of different consistency properties and different ways of formulating them, see Sen (1977/1982: §4).
Some, particularly in the revealed preference literature, might find this approach unobjectionable. Adopting this kind of response to fleshing out what it is we choose and prefer, however, significantly circumscribes the kinds of things rational choice theory can be employed to do, and thus stakes out a highly controversial account of what the purpose of rational choice theory is. These issues are explored in more detail in Sect. 4 below, when we examine the possible worlds account of what we choose and prefer.
Rulli and Worsnip (2016: §3) similarly criticize certain ways of individuating options (“the de dicto response”) by arguing that individuating options in such a way makes rational choice theory unable to perform certain functions we want it to perform.
There are several in the literature who propose something along these lines. One example is Caplin and Leahy (2001: p. 60) who “replace the standard prize space with a space of ‘psychological states,’ comprising a complete (for model purposes) description of the individual’s state of mind.” Broome (1991: p. 103)’s principle of individuation by justifiers also includes the perception of choosing agents, as does Neumann (2007: p. 82)’s emphasis on “decision-sensitive features of the choice act.” Finally, Hausman (2011: pp. 3–4)’s understanding of preferences as total comparative evaluations defines preferences as comparative evaluations over states of affairs based on everything that matters to the chooser. This too makes essential reference to the perception of the choosing agent.
Technically, it is not choice options that have properties on the Dietrich–List model, but choice options in contexts that have properties (what Dietrich and List call option-context pairs) (Dietrich and List 2016: p. 186; pp. 188–189). So a mango simpliciter does not have properties, but rather a mango in the context of menu \(\hbox {S}_{1}\) has properties, a mango in the context of menu \(\hbox {S}_{2}\) has properties, and so on and so forth. (If a mango has the property p in every possible menu then we say that p is an option property of the mango. We can thus think of option properties as intrinsic properties of choice options x, y, and z.).
Continuing our discussion from the footnote above, on the Dietrich–List model, there are three kinds of properties option-context pairs may have: option properties, which depend solely on the nature of the choice option; context properties, which depend solely on the context the choice option is confronted in; and relational properties, which depend on both the option and the context (Dietrich and List 2016: p. 186). This richer account of properties—particularly context properties and relational properties—allows the same choice option (e.g., a mango) to have different properties in different contexts.
Here, the Dietrich–List model would somehow have to account for the fact that if the choosing agent chooses both mangos then they would be taking the last bit of everyone’s favorite fruit. The model could do this by understanding the choice options here not as pieces of fruit, but as distributions of fruit as the social states account does, such that each distribution where both mangos are selected would then have the property last of everyone’s favorite. This suggested revision changes nothing of the forthcoming analysis.
Instead of defining a context as merely a subset on X, one could here offer more detail by following the Dietrich–List model in defining a context K as an ordered pair (Y, \(\lambda )\), where Y is a subset on X, and \(\lambda \) a parameter that specifies further features of the choice environment (Dietrich and List 2016: p. 185). This latter object \(\lambda \) allows us to account for the fact that individuals might face the same menus (subsets on X) at times \(t_{1}\) and \(t_{2}\) but perceive very different choice problems if, say, they are sober at \(t_{1}\) and drunk at \(t_{2}\).
Thus far we have been postulating the existence of a perspective function to explain choices an agent makes, but we have not yet explained why it is we should think persons have perspective functions in the first place. A fuller account of the perspectival model will need to address why it is we should think persons have perspectives as we have defined them. Work on perspectives and evaluative diversity more generally could possibly fill such gaps. See here Page (2007), Gaus (2016) and Muldoon (2016). Moreover, for the strictly positivists we have not yet specified how an external observer can know the nature of a choosing agent’s perspective function. Note that given the complex nature of perspectives—as mappings from choice options in contexts to elements in \(X^\wedge \)—it is far more difficult to determine the nature of an agent’s perspective than, say, an agent’s preferences over brute choice options as defined by the commodity baskets account.
Returning to footnote eight above, the perspectival model will have to account for the fact that choosing both mangos from \(\hbox {S}_{2}\) would be deemed rude by Bertha. The suggested revision in the above footnote for the Dietrich–List model could be applied here as well, mutatis mutandis.
We here suppose that Esau and Franklin assign 0 to every other feature of the respective pieces of fruit besides their sweetness properties and their bruised-ness properties.
References
Arrow, K. J. (1951/2012). Social choice and individual values. New Haven: Yale University Press.
Broome, J. (1991). Weighing goods: Equality, uncertainty, and time. Oxford: Basil Blackwell.
Caplin, A., & Leahy, J. (2001). Psychological expected utility theory and anticipatory feelings. The Quarterly Journal of Economics, 116, 55–79.
Cresswell, M. J. (1972). The world is everything that is the case. Australasian Journal of Philosophy, 50, 1–13.
Dietrich, F., & List, C. (2011). A model of non-informational preference change. Journal of Theoretical Politics, 23, 145–164.
Dietrich, F., & List, C. (2013a). A reason-based theory of rational choice. Noûs, 47, 104–134.
Dietrich, F., & List, C. (2013b). Where do preferences come from? International Journal of Game Theory, 42, 613–637.
Dietrich, F., & List, C. (2016). Reason-based choice and context-dependence: an explanatory framework. Economics and Philosophy, 32, 175–229.
Dowding, K. (2002). Revealed preference and external reference. Rationality and Society, 14, 259–284.
Gaertner, W., & Yongsheng, X. (1999). Rationality and external reference. Rationality and Society, 11, 169–185.
Gaus, G. (2016). The tyranny of the ideal: Justice in a diverse society. Princeton: Princeton University Press.
Hausman, D. M. (2011). Preference, values, choice, and welfare. Cambridge: Cambridge University Press.
Hicks, J. R. (1939). Value and capital: An inquiry into some fundamental principles of economic theory. Oxford: Oxford University Press.
Muldoon, R. (2016). Social contract theory for a diverse world: Beyond tolerance. New York: Routledge.
Neumann, M. (2007). Choosing and describing: Sen and the irrelevance of independence alternatives. Theory and Decision, 63, 79–94.
Page, S. E. (2007). The difference: How the power of diversity creates better groups, firms, schools, and societies. Princeton: Princeton University Press.
Rulli, T., & Worsnip, A. (2016). IIA, rationality, and the individuation of options. Philosophical Studies, 173, 205–221.
Samuelson, P. (1947). Foundations of economic analysis. Cambridge: Harvard University Press.
Sen, A. (1977/1982). Social choice theory: A re-examination. In Choice, welfare, and measurement (pp. 158–200). Cambridge: Harvard University Press.
Sen, A. (1993/2004). Internal consistency and the act of choice. In Rationality and freedom (pp. 121–157). Cambridge: Harvard University Press.
Sen, A. (1995). Is the idea of purely internal consistency of choice Bizarre? In J. E. J. Althman & R. Harrison (Eds.), World, mind, and ethics: Essays on the ethical philosophy of Bernard Williams (pp. 19–31). Cambridge: Cambridge University Press.
Sen, A. (1997/2004). Maximization and the act of choice. In Rationality and freedom (pp. 158–205). Cambridge: Harvard University Press.
Author information
Authors and Affiliations
Corresponding author
Additional information
The author would like to thank Joel Chow, Jerry Gaus, Adam Gjesdal, Ryan Muldoon, Sarah Raskoff, Stephen G.W. Stich, and Robert “Bobby” Wallace, Jr., as well as those who attended his talk at the University of Maryland, for comments on earlier drafts of this paper. The author would also like to thank two anonymous referees for their immensely helpful feedback.
Rights and permissions
About this article
Cite this article
Kogelmann, B. What we choose, what we prefer. Synthese 195, 3221–3240 (2018). https://doi.org/10.1007/s11229-017-1369-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11229-017-1369-0