Johan E. Gustafsson

We argue that under general conditions, agents who exhibit Ellsberg-style ambiguity aversion (or ambiguity seekingness) are subject to money pumps: they will give up, for no gain, what they could have kept for free. The argument differs from existing results in three ways. First, it gains generality by depending on only minimal assumptions about the target agent. Second, unlike standard approaches that appeal to backward induction, the argument does not assume that rational agents expect that they will choose rationally after having made an irrational choice. Third, the argument is particularly simple.

Ideal Contractualism views principles of justice as corresponding to what rational, mutually disinterested persons would collectively choose behind a veil of ignorance. It is well-known that Ideal Contractualism faces profound challenges in accounting for justice between generations. We present a unified solution to these problems that involves rejecting the assumption that the parties conceive of their choices as causally efficacious and assumes instead that the parties choose in light of the news value of their decision. And we explore what concrete principles would be chosen by the parties as governing intergenerational justice against the backdrop of this assumption.

In this paper, I argue against defining either of ‘good’ and ‘better’ in terms of the other. According to definitions of ‘good’ in terms of ‘better’, something is good if and only if it is better than some indifference point. Against this approach, I argue that the indifference point cannot be defined in terms of ‘better’ without ruling out some reasonable axiologies. Against defining ‘better’ in terms of ‘good’, I argue that this approach either cannot allow for the incorruptibility of intrinsic goodness or it breaks down in cases where both of the relata of ‘better’ are bad.

According to Stochastic Dominance, it is rationally obligatory to prefer one gamble to another if it gives you the same chances of getting final outcomes you prefer. According to Statewise Maximality, it is rationally permissible not to disprefer a gamble if it is guaranteed not to result in a final outcome you disprefer. These principles conflict in cases involving incomplete preferences, known as Opaque Sweetening cases. In this paper, we argue for Stochastic Dominance and against Statewise Maximality in Opaque Sweetening cases. First, we rebut two standard arguments for statewise maximality, which we call the Argument from Full Information and the Argument from the Primacy of Final Outcomes. We then provide an argument for the verdict of Stochastic Dominance in Opaque Sweetening cases. This argument appeals to the principle of Transitivity, the Sure-Thing Principle, and the claim that stochastic reasoning is appropriate at least for choices which do not involve incomparability.

Critical-Level Utilitarianism entails one of the Repugnant Conclusion and the Sadistic Conclusion, depending on the critical level. Indeterminate Critical-Level Utilitarianism is a version of Critical-Level Utilitarianism where it is indeterminate which well-being level is the critical level. Undistinguished Critical-Range Utilitarianism is a variant of Critical-Level Utilitarianism where additions of lives in a range of well-being between the good and the bad lives make the resulting outcome incomparable to the original outcome. These views avoid the Repugnant Conclusion, they avoid the Sadistic Conclusion, and they agree on all outcome comparisons not involving indeterminacy or incomparability. So it may seem unclear whether we have any reason to favour one of these theories over the other. But I argue that Indeterminate Critical-Level Utilitarianism still entails the disjunction of the Repugnant Conclusion and the Sadistic Conclusion, which is also repugnant. By contrast, Undistinguished Critical-Range Utilitarianism does not entail this conclusion.

Completeness says that, for every pair of prospects, at least one of the prospects is at least as preferred as the other. I present a new money-pump argument that Completeness is a requirement of rationality. In comparison with earlier money-pump arguments for Completeness, this argument relies on a unidimensional form of stochastic dominance and the behavioural assumption that agents pick in a probabilistic manner when no option is optimal. Moreover, unlike some of the previous arguments, the new argument is based on a forcing money pump, that is, an exploitation scheme where the agent is rationally required at each step to go along with the scheme.

The money-pump argument aims to show that cyclic preferences are irrational. The argument can be based on a number of different exploitation schemes that vary in what needs to be assumed about the agent. The Standard Money Pump works for myopic and naive agents, but not for sophisticated agents who use backward induction. The Upfront Money Pump works for sophisticated agents, but not for myopic or naive agents. In this paper, I present a new money pump, the Universal Money Pump, that works for myopic, naive, and sophisticated agents. Moreover, the Universal Money Pump (just like the Upfront Money Pump) also works for minimally sophisticated agents who need not assume that they will choose rationally at nodes that can only be reached by irrational choices. This enables an argument that rational preferences are acyclic, which is based on weaker assumptions about dynamic rationality than existing money-pump arguments.

According to Act Consequentialism, an act is right if and only if its outcome is not worse than the outcome of any alternative to that act. This view, however, leads to deontic paradoxes if the alternatives to an act are all other acts that can be done in the situation. A typical response is to only apply this rightness criterion to maximally specific acts and to take the alternatives to a maximally specific act to be the other maximally specific acts that can be done in the situation. (This view can then be supplanted by a separate account for the rightness of acts that are not maximally specific.) This paper defends a rival view, Binary Act Consequentialism, where, for any voluntary act, that act is right if and only if its outcome is not worse than the outcome of not doing that act. Binary Act Consequentialism, which dates back to Jeremy Bentham, has few supporters. A number of seemingly powerful objections have been considered fatal. In this paper, I rebut these objections and put forward a positive argument for the view.

W. V. Quine wished to restrict the interests that matter to those of actual people. Actual-Population Utilitarianism is a version of utilitarianism where, following Quine, only the interests of actual people matter. It is well known that ethical theories of this kind, which depend on what is actual, typically lead to normative variance. In this paper, I put forward a new objection to Actual-Population Utilitarianism. I present a case in which Actual-Population Utilitarianism prescribes choices that are worse for everyone whose interests matter than the opposite choices. Moreover, I put forward a case in which total and average versions of Actual-Population Utilitarianism prescribe choices that are worse than the opposite choice for everyone (full stop). This objection also applies if the Actual-Population Restriction is combined with non-utilitarian views, given some minimal assumptions.

Anna Mahtani describes a puzzle meant to show that the Ex-Ante Pareto Principle is incomplete as it stands and, since it cannot be completed in a satisfactory manner, decades of debate in welfare economics and ethics are undermined. In this paper, we provide a better solution to the puzzle which saves the Ex-Ante Pareto Principle from this challenge. We also explain how the plausibility of our solution is reinforced by its similarity to a standard solution to an analogous puzzle in quantified epistemic logic. We also show that even if the puzzle were to remain unsolved, its impact on welfare economics and ethics would be limited.

The Consequence Argument is the classic argument for the incompatibility of determinism and our ability to do otherwise. Daniel C. Dennett objects that the Consequence Argument suffers from the same error as a clearly unconvincing argument that there are no mammals. In this paper, I argue that these arguments do not suffer from the same error. The argument that there are no mammals is unconvincing as it takes the form of a sorites, whereas the Consequence Argument does not. Accordingly, Dennett's objection misses its mark.

According to Longtermism, our acts’ expected influence on the expected value of the world is mainly determined by their effects in the far future. There is, given total utilitarianism, a straightforward argument for Longtermism due to the enormous number of people that might exist in the future, but this argument does not work on person-affecting views. In this paper, we will argue that these views might also lead to Longtermism if Prudential Longtermism is true. Prudential Longtermism holds for a person if and only if our acts’ overall influence on that person’s expected well-being is mainly determined by the acts’ effects in the far future. We argue that (due to a small chance of anti-ageing and uploading) there could be an enormous amount of prudential value for some contemporary person in the far future and that value may be so large that it dominates their overall expectation of lifetime well-being.

How do you make decisions under ignorance? That is, how do you decide when you lack subjective probabilities for some of your options’ possible outcomes? One answer is that you follow the Laplace Rule: you assign an equal probability to each state of nature for which you lack a subjective probability (that is, you use the Principle of Indifference) and then you maximize expected utility. The most influential objection to the Laplace Rule is that it is sensitive to the individuation of states of nature. This sensitivity is problematic because the individuation of states seems arbitrary. In this paper, however, I argue that this objection proves too much. I argue that all plausible rules for decisions under ignorance are sensitive to the individuation of states of nature.

Derek Parfit argues that fission is prudentially better for you than ordinary death. But is having more fission products with good lives prudentially better for you than having just one? In this paper, we argue that it is. We argue that, if your brain is split and the halves are transplanted into two recipients (who both have good lives), then it is prudentially better for you if both transplants succeed than if only one of them does (other things being equal). This upshot rules out, among other things, that the prudential value of standing in the relation that matters in survival to multiple people is equal to their average well-being.

A straightforward way to handle moral uncertainty is simply to follow the moral theory in which you have most credence. This approach is known as My Favourite Theory. In this paper, I argue that, in some cases, My Favourite Theory prescribes choices that are, sequentially, worse in expected moral value than the opposite choices according to each moral theory you have any credence in. In addition this, problem generalizes to other approaches that avoid intertheoretic comparisons of value, such as My Favourite Option, the Borda Rule, Variance Normalization, and the Principle of Maximizing Expected Normalized Moral Value.

Suppose that you prefer A to B, B to C, and C to A. Your preferences violate Expected Utility Theory by being cyclic. Money-pump arguments offer a way to show that such violations are irrational. Suppose that you start with A. Then you should be willing to trade A for C and then C for B. But then, once you have B, you are offered a trade back to A for a small cost. Since you prefer A to B, you pay the small sum to trade from B to A. But now you have been turned into a money pump. You are back to the alternative you started with but with less money. This Element shows how each of the axioms of Expected Utility Theory can be defended by money-pump arguments of this kind. The Element also defends money-pump arguments from the standard objections to this kind of approach. This title is also available as Open Access on Cambridge Core.

Prioritarianism is a variant of utilitarianism. It differs from utilitarianism in that benefiting individuals matters more the worse off these individuals are. On this view, there are two standard ways of handling risky prospects: Ex-Post Prioritarianism adjusts for prioritizing the worse off in final outcomes and then values prospects by the expectation of the sum total of those adjusted values, whereas Ex-Ante Prioritarianism adjusts for prioritizing the worse off on each individual's expectation and then values prospects by the sum total of those adjusted expectations. A standard objection to Ex-Post Prioritarianism is that it violates Ex-Ante Pareto, that is, it prescribes choices that worsen the expectations for everyone. In this article, I argue that Ex-Ante Prioritarianism suffers from much the same problem: it violates a sequential version of Ex-Ante Pareto, that is, it prescribes sequences of choices that worsen the expectations for everyone.

Is an outcome where many people are saved and one person dies better than an outcome where the one is saved and the many die? According to the standard utilitarian justification, the former is better because it has a greater sum total of well-being. This justification involves a controversial form of moral aggregation, because it is based on a comparison between aggregates of different people's well-being. Still, an alternative justification—the Argument for Best Outcomes—does not involve moral aggregation. I extend the Argument for Best Outcomes to show that any utilitarian evaluation can be justified without moral aggregation.

One might think that money pumps directed at agents with cyclic preferences can be avoided by foresight. This view was challenged two decades ago by the discovery of a money pump with foresight, which works against agents who use backward induction. But backward induction implausibly assumes that the agent would act rationally and retain her trust in her future rationality even at choice nodes that could only be reached if she were to act irrationally. This worry does not apply to BI-terminating decision problems, where at each choice node backward induction prescribes a move that terminates further action. For BI-terminating decision problems, it is enough to assume that rationality and trust in rationality are retained at choice nodes reachable by rational moves. The old money pump with foresight was not BI-terminating. In this paper, we present a new money pump with foresight, one that is both BI-terminating and considerably simpler.

According to the Psychological-Continuity Account of What Matters, you are justified in having special concern for the well-being of a person at a future time if and only if that person will be psychologically continuous with you as you are now. On some versions of the account, the psychological continuity is required be temporally ordered, whereas, on other versions, it is allowed to be temporally unordered. In this paper, I argue that the account is implausible if the psychological continuity is allowed to be temporally unordered. I also argue that, if the psychological continuity is required to be temporally ordered, it cannot plausibly be purely psychological (in the sense that the psychological continuity is not required to be caused through spatio-temporal continuity of a brain). The upshot is that no plausible version of the Psychological-Continuity Account of What Matters is purely psychological. So psychological continuity is not what matters in survival.

Moral obligation and permissibility are usually thought to be interdefinable. Following the pattern of the duality definitions of necessity and possibility, we have that something’s being permissible could be defined as its not being obligatory to not do it. And that something’s being obligatory could be defined as its not being permissible to not do it. In this paper, I argue that neither direction of this alleged interdefinability works. Roughly, the problem is that a claim that some act is obligatory or permissible entails that there is a moral law, whereas a negative claim that some act is not obligatory or not permissible does not. Nevertheless, one direction of the interdefinability can potentially be salvaged. I argue that, if we do not require the conceptual possibility of moral dilemmas, then there is a way to plausibly define obligation in terms of permissibility. I conclude that permissibility is the only feasible deontic primitive.

The Levelling-Down Objection is a standard objection to monistic egalitarian theories where equality is the only thing that has intrinsic value. Most egalitarians, however, are value pluralists; they hold that, in addition to equality being intrinsically valuable, the egalitarian currency in which we are equal or unequal is also intrinsically valuable. In this paper, I shall argue that the Levelling-Down Objection still minimizes the weight that the intrinsic badness of inequality could have in the overall intrinsic evaluation of outcomes, given a certain way of measuring the badness of inequality, namely, the Additive Individual-Complaints Measure.