Abstract
We examine a distinctive kind of problem for decision theory, involving what we call discontinuity at infinity. Roughly, it arises when an infinite sequence of choices, each apparently sanctioned by plausible principles, converges to a ‘limit choice’ whose utility is much lower than the limit approached by the utilities of the choices in the sequence. We give examples of this phenomenon, focusing on Arntzenius et al.’s Satan’s apple, and give a general characterization of it. In these examples, repeated dominance reasoning (a paradigm of rationality) apparently gives rise to a situation closely analogous to having intransitive preferences (a paradigm of irrationality). Indeed, the agents in these examples are vulnerable to a money pump set-up despite having preferences that exhibit no obvious defect of rationality. We explore several putative solutions to such problems, particularly those that appeal to binding and to deliberative dynamics. We consider the prospects for these solutions, concluding that if they fail, the examples show that money pump arguments are invalid.
Similar content being viewed by others
Notes
The expression “discontinuity at infinity” is not common, although Atkinson and Johnson (2010) use it in much the way we intend in a discussion of the physics of large versus actually infinite ensembles. They write: “mathematically there is no paradox: there is simply a difference between the infinite limit and the value at infinity—we might call the phenomenon a discontinuity at infinity.” Similar phenomena are identified in Batterman’s discussion of asymptotic reasoning, critical points and infinite idealizations; see Batterman (2002, 2005). Norton (2012, p. 212) speaks of “the diverging of limit properties and limit systems.”
The following observation applies specifically to one of the examples discussed below: Satan’s apple.
In this puzzle, the outcome is determined entirely by your (ultimate) choice of strategy. So we identify your preference ordering on strategies with your preference ordering on outcomes.
In fact, each \(B_{n}\) has infinite expectation. However, for each prize of $2\(^{k}\) with \(k=n+1\), the strategy \(B_{n+1}\) guarantees twice the chance that \(B_{n}\) does of winning at least that much money. Thus, although both \(B_{n}\) and \(B_{n+1}\) have equal infinite expectation, the fact that the latter offers uniformly better chances seems to justify the preference ordering \(B_{0} < B_{1} < {\cdots }\). We acknowledge that this preference ordering rests on less secure ground than that in the other two examples; accordingly, in the remainder of our paper, we concentrate on Puzzle #1 and Puzzle #3.
While we previously used the ‘\(<\)’ symbol to denote the operative preference relation on strategies, its usage here is appropriate, since the preference ordering on strategies is also a dominance ordering (given that there is only one state).
We continue to focus on the deterministic case, so there is no need for chance nodes.
The term ‘strategy’ is sometimes used for a function that maps each choice point in the tree to some act available at that choice point, while ‘profile’ is reserved for a complete sequence of choices. Here, we use the terms ‘strategy’ and ‘profile’ interchangeably, in the latter sense.
In fact, all of them are dominated by any strategy Accept exactly the first N pieces, provided \(N > n(n-1)\).
Slight qualification: in the original version of Satan’s apple, taking all the pieces is a terrible outcome, but Eve can do slightly worse by omitting some pieces while still taking infinitely many.
Continuity of preferences can also be defined directly from the preference ordering, without invoking a utility function (see Appendix 1).
Whatever norm is used to find the profile Take all the pieces rationally impermissible must be restricted to agents capable of binding.
Meacham (2010) offers a sustained critique of the binding approach; here, we focus on objections distinct from his.
This is one simple way to achieve a distribution with the desired properties; note in particular that \(P(X \in \mathbb {N}) = 1\).
Defining Eve’s expected utility raises technical difficulties, so we put the point in terms of the number of slices expected.
This follows from \(P(ALL/S_n )=\frac{P(S_n /ALL)P(ALL)}{P(S_n /ALL)P(ALL)\;+\;P(S_n /\sim ALL)P(\sim ALL)}\), since \(P\left( {S_n / ALL} \right) = 1\) and \(P\left( {S_n /\,\sim ALL} \right) \,=\Sigma _{i \ge n} P\left( {T_n /\sim ALL} \right) \,=\Sigma _{i \ge n} p_i \).
There is also a separate problem of determining the values of the conditional probabilities \(P^{\prime } \left( {T_k / S_{n+1}} \right) \) and \(P^{\prime } \left( {ALL/S_{n+1} } \right) \) when \(P^{\prime } \left( {S_{n+1} } \right) \,=\,0\).
This argument fails if \(P(S_{n}) =\) 0: we could then have \(P\)(ALL) \(=\) 0 but \(P\left( {ALL/S_n } \right) \ne 0\). Rather than explore this possibility here, we consider a similar idea in the next section.
More generally, the agent revises her credences for states as well as acts (Skyrms 1990).
Note: this requires a utility function that is positive and bounded away from 0.
The conditional probabilities are primitive; they are defined even if the proposition conditioned upon has probability 0. Indeed, \(q_{n}\) may be interpreted as Eve’s credence that she would take piece \(n\) if she were to take pieces 1, ..., \(n-1\). These probabilities are not updated based on the evidence of Eve’s past choices; instead, as we shall see, they are updated on the basis of her deliberations.
Note that the replicator dynamics requires a utility function that is positive and bounded away from 0. This is easily obtained by adding \(+\)1001 to Eve’s utility function as it was defined in Sect. 2.
This definition is meant to correspond to the familiar notion of an evolutionarily stable strategy; see Skyrms (1996).
She takes piece \(i\) because the probability of taking the later piece \(j\) is zero, hence Pr(ALL / \(S_{i}) =\) 0, and therefore (from 2\(^{\prime }\)) the expected utility of taking piece \(i\) is higher than that of not taking it. She maintains \(q_{i} =\) 0 because, as we have seen, extreme assignments never change under the updating rule.
References
Arntzenius, F., Elga, A., & Hawthorne, J. (2004). Bayesianism, infinite decisions, and binding. Mind, 113, 251–283.
Atkinson, D., & Johnson, P. (2010). Nonconservation of energy and loss of determinism. II. Colliding with an open set. Foundations of Physics, 40, 179–189.
Barrett, J., & Arntzenius, F. (1999). An infinite decision puzzle. Theory and Decision, 46(1), 101–103.
Batterman, R. (2002). The devil in the details: Asymptotic reasoning in explanation, reduction and emergence. New York: Oxford University Press.
Batterman, R. (2005). Critical phenomena and breaking drops: Infinite idealizations in physics. Studies in History and Philosophy of Modern Physics, 36(2), 225–244.
Gilboa, I. (1994). Can free choice be known? In C. Bicchieri, R. Jeffrey, & B. Skyrms (Eds.), The logic of strategy. Oxford: Oxford University Press.
Hájek, A. (2008). Dutch book arguments. In A. Paul, P. Prasanta, & P. Clemens (Eds.), The Oxford handbook of rational and social choice (pp. 173–195). Oxford: Oxford University Press.
Jeffrey, R. (1983). The logic of decision (2nd ed.). Chicago: University of Chicago Press.
Levi, I. (1997). The covenant of reason: Rationality and the commitments of thought. Cambridge: Cambridge University Press.
Luce, D., & Raiffa, H. (1957). Games and decisions: Introduction and critical survey. New York: Wiley.
Meacham, C. (2010). Binding and its consequences. Philosophical Studies, 149(1), 49–71.
Norton, J. D. (2012). Approximation and idealization: Why the difference matters. Philosophy of Science, 79(2), 207–232.
Nover, H., & Hájek, A. (2004). Vexing expectations. Mind, 113, 305–317.
Pollock, J. (1983). How do you maximize expectation value? Noûs, 17, 409–421.
Resnik, M. (1987). Choices: An introduction to decision theory. Minneapolis, MN: University of Minnesota Press.
Rudin, W. (1974). Real and complex analysis (2nd ed.). New York: McGraw-Hill.
Skyrms, B. (1990). The dynamics of rational deliberation. Cambridge, MA: Harvard University Press.
Skyrms, B. (1996). Evolution of the social contract. Cambridge: Cambridge University Press.
Slote, M. (1985). Common-sense morality and consequentialism. London: Routledge and Kegan Paul.
Slote, M. (1989). Beyond optimizing: A study of rational choice. Cambridge, MA: Harvard University Press.
Spohn, W. (1977). Where Luce and Krantz do really generalize Savage’s decision model. Erkenntnis, 11, 113–134.
Acknowledgments
We thank especially Sharon Berry, Rachael Briggs, Kenny Easwaran, Christopher Hitchcock, Yoaav Isaacs, Ralph Miles, Daniel Nolan, Paolo Santorio, Wolfgang Schwarz, Julia Staffel, and Orri Stefánsson for very helpful discussion. One of us (Paul Bartha) also acknowledges support from Australian National University in the form of a Visiting Research Fellowship.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1: Continuity and transitivity
In this section, our main objective is to state and prove an interesting partial converse to Theorem 4.1. To state this converse, we need to define two notions of limit.
First, let \(U\) be a real-valued function on a topological space X, not necessarily continuous. Then the limit of \(U(x')\) as \(x' \rightarrow x\) (or for short, the limit of \(U\) at \(x\)) is defined to be the unique real number \(r\) with the following property: for any open interval (\(a, b\)) around \(r\), there is an open neighborhood V of \(x\) such that \(U\) maps V \(-\) {\(x\)} into (\(a, b\)). If no such \(r\) exists, the limit of \(U\) at \(x\) is undefined. The idea here is that we can make \(U(x')\) as close as we like to \(r\) by making \(x'\) sufficiently close to \(x\), provided \(x \ne x'\). Continuity is definable in terms of limits: \(U\) is continuous at \(x\) iff the limit of \(U\) at \(x\) is \(U(x)\).
Transitive intransitivity implies that the limit of \(U\) at \(A\) either does not exist or is less than \(U(A)\). When the limit of \(U\) at \(A\) exists, this necessary condition for transitive intransitivity is almost sufficient. To formulate a sufficient condition, we need one more definition. A topological space X is said to be first countable if for each point \(x\), there is a countable family {\(V_{i}: i =\) 1, 2, ...} of open neighborhoods of \(x\) such that for any open neighborhood \(V\) of \(x, V_{i} \subseteq V\) for some \(i\). Such a family is called a countable base at x. All of the spaces we are concerned with in this paper are first countable. We then have the following result.
Theorem 7.1
Let \(U\) be a real-valued function on a first countable topological space X, and for all \(x \in \) X, let \(U'(x)\) be the limit of \(U\) at \(x\). If \(U'\) is well-defined throughout X (i.e., limits exist everywhere), then the ordering \(\le \) generated by \(U\) is transfinitely intransitive iff there is an \(x \in \) X such that (a) \(U(x) < U'(x)\), and (b) every neighborhood of \(x\) contains a point \(x'\) such that \(U(x') < U'(x)\).
Proof
Left-to-right: Let \(x < x_{1} < x_{2} < \cdots \rightarrow x\). Then \(U(x) < U(x_{1}) < U(x_{2}) < \cdots \). Clearly, the sequence \(U(x_{1}), U(x_{2}), \ldots \) converges to \(U'(x)\), and this value is clearly greater than any of the \(U(x_{i})\), and thus greater than \(U(x)\). Thus, (a) holds. For (b), note that any open neighborhood \(V\) of \(x\) contains some of the points \(x_{i}\), and since \(U(x_{i}) < U'(x)\), (b) holds.
Right-to-left: Suppose \(x\) satisfies (a) and (b), and let {\(V_{i}: i =\) 1, 2, ...} be a countable base at \(x\). By (b), for each \(i\) we can pick \(y_{i} \in V_{i}\) such that \(U'(x) > U(y_{i})\). Any open neighborhood of X contains some \(V_{i}\), and hence contains all but finitely many of the \(y_{i}\), so the sequence \(y_{1}, y_{2}, \ldots \) converges to \(x\). Also, for every \(i\), there is a \(j > i\) such that \(U(y_{i}) < U(y_{j})\). To see this, simply pick an interval (\(a, b\)) around \(U'(x)\) that excludes \(U(y_{i})\). By the definition of the limit, there is an open neighborhood \(V\) of \(x\) such that \(U\) maps \(V-\){\(x\)} into (\(a, b\)), and there is some \(V_{j} \subseteq V\); this gives us the required \(j > i\). Thus, we can pick a subsequence \(x_{1}, x_{2}, \ldots \) of \(y_{1}, y_{2}, \ldots \) such that \(U(x_{1}) < U(x_{2}) < \cdots \). This subsequence clearly converges to \(x\), and the sequence \(U(x_{1}), U(x_{2}), \ldots \) obviously converges to \(U'(x)\). By (a), \(U(x) < U'(x)\), and thus, \(U(x) < U(x_{i})\) for some \(i\). Hence, \(U\) is transfinitely intransitive. \(\square \)
If we drop the condition that \(U\) has a well-defined limit everywhere, we are not aware of any simple necessary and sufficient condition for the transfinite intransitivity of the generated ordering \(\le \).
There is a second connection between transfinite intransitivity and continuity. Instead of assuming that the preference ordering \(\le \) comes from a utility function, let us simply assume that \(\le \) is complete, meaning that for all \(A, B \in \) X, either \(A \le B\) or \(B \le A\). There is then a standard definition of continuity for preference orderings. Specifically, for any given point \(A\), let U\(_{A} =\) {\(B\): \(A \le B\)} and L\(_{A} =\) {\(C\): \(C \le A\)}. Then \(\le \) is said to be continuous if for every \(A\), the sets U\(_\mathrm{A}\) and L\(_\mathrm{A}\) are topologically closed. (A set \(S\) is closed iff whenever a sequence of points in \(S\) converges, it converges to a point in \(S\).) Now suppose that \(\le \) is transfinitely intransitive, and let \(A < A_{1} < A_{2} \ldots \rightarrow A\). Then the set {\(B: A_{1} \le B\)} is not closed, because there is a sequence of points from that set whose limit, \(A\), is not in that set. Thus, transfinite intransitivity of \(\le \) implies discontinuity of \(\le \) in the standard sense, further justifying our talk of “discontinuity at infinity.”
Appendix 2: Extreme equilibria
The purpose of this section is to prove Theorem 5.1. We first prove a result based upon an elementary fact about infinite products.
Lemma
\(\hbox {lim}_{n \rightarrow \infty } P\left( {ALL/S_n } \right) = 1 \hbox { iff }\sum \nolimits _{n=1}^\infty {(1-q_n )} <\infty \), and \(P\)(ALL / \(S_{n})\) \(=\) 0 for all \(n\) iff \(\sum _{n=1}^\infty {(1-q_n )} =\infty \).
Proof
If \(0 \le u_{n} < 1\), then \(\prod _{n=1}^\infty {(1-u_n } )>0 \hbox { iff } \sum _{n=1}^\infty {u_n } <\infty \) (Rudin 1974, p. 322).
Recall that \(P(ALL / S_{n})=\prod _{i=n+1}^\infty {q_i }\). If \(q_{i}\) = 0 for infinitely many \(i\), then \(P(ALL / S_{n}) =\) 0 for all \(n\) and \(\sum _{n=1}^\infty {(1-q_n )} =\infty \). If \(q_{i} =\) 0 for finitely many \(i\), it follows from (Rudin 1974) that either \(P(ALL / S_{n}) =\) 0 and \(\sum _{i=n+1}^\infty {(1-q_i )} =\infty \), or \(P(ALL / S_{n}) >\) 0 and \(\sum _{i=n+1}^\infty {(1-q_i )} <\infty \). In the latter case, \(P(ALL / S_{n})\) must converge to 1 as \(n \rightarrow \infty \). \(\square \)
Theorem 7.2
Every equilibrium is an extreme assignment. (For all conditional probabilities \(q_{n}=P(S_{n} / S_{1} \cdot {\ldots } \cdot S_{n-1})\), either \(q_{n} =\) 0 or \(q_{n} =\) 1.)
Proof
Suppose that {\(q_{n}\)} is an equilibrium probability assignment. Either \(\sum _{n=1}^\infty {(1-q_n )} <\infty \) or \(\sum _{n=1}^\infty {(1-q_n )} =\infty \). If \(\sum _{n=1}^\infty {(1-q_n )} =\infty \), then \(P(ALL / S_{n}) = 0\) for all \(n\). It follows that for all \(n\), the expected utility calculations in 2’ favour Take piece n. The dynamics implies upward pressure (\(q_{n}' > q_{n}\)) unless \(q_{n} =\) 0 or \(q_{n} =\) 1. Since we have an equilibrium, it follows that \(q_{n} =\) 0 or \(q_{n} =\) 1 for all \(n\). Hence, we have an extreme assignment.
If instead \(\sum _{n=1}^\infty {(1-q_n )} <\infty \), then lim\(_{n\rightarrow \infty } P(ALL / S_{n}) =\) 1; also, lim\(_{n\rightarrow \infty } q_{n} =\) 1. Combining these facts with 2’, Eve reaches a point \(N\) beyond which each \(q_{n}\) is close to 1 and \(P(ALL / S_{n})\) is so large that her expected utility computations favour Reject piece n. For \(n > N\), the dynamics implies downward pressure (\(q_{n}' < q_{n}\)) unless \(q_{n} =\) 1. Since we have an equilibrium, it follows that \(q_{n}\) = 1 for all \(n > N\). So the ‘tail end’ of the probability assignment, from \(N+1\) onwards, consists entirely of 1s.
To finish up, we show that, for this case, each of \(q_{1}, {\ldots }, q_{N}\) must also be 0 or 1. If not, let \(k\) be the largest index such that 0 \(< q_{k} <\) 1. If \(q_{n} =\) 0 for some \(n > k\), then the dynamics implies upward pressure on \(q_{k}\): since \(P(ALL / S_{k}) =\) 0, EU\((S_{k}) >\) EU(\(\sim S_{k}\)) and \(q_{k}' > q_{k}\), contradicting the fact that we have an equilibrium. If \(q_{n} =\) 1 for all \(n > k\), then the dynamics implies downward pressure on \(q_{k}\): since \(P(ALL / S_{k}) =\) 1, EU\((S_{k}) <\) EU(\(\sim S_{k}\)) and \(q_{k}' < q_{k}\), again contradicting the fact that we have an equilibrium. Hence, \(q_{k} =\) 0 and each of \(q_{1}, {\ldots }, q_{N}\) must also be 0 or 1.
Hence, the equilibrium assignments for Eve are exactly the extreme assignments in which all of her conditional probabilities \(q_{n}=P(S_{n} / S_{1} \cdot {\ldots } \cdot S_{n-1})\) are either 0 or 1. \(\square \)
Rights and permissions
About this article
Cite this article
Bartha, P., Barker, J. & Hájek, A. Satan, Saint Peter and Saint Petersburg. Synthese 191, 629–660 (2014). https://doi.org/10.1007/s11229-013-0379-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11229-013-0379-9