Cambridge University Press, 2022
A defence of money-pump arguments for the axioms of Expected Utility Theory.
A Simpler, More Compelling Money Pump with Foresight
with Wlodek Rabinowicz
You may think money pumps directed at agents with cyclic preferences can be avoided by foresight. But some money pumps also works against agents with foresight. A standard objection to these pumps is that they assume that agents would act rationally and retain trust in their future rationality even at choice nodes that can only be reached by irrational choices. We present a new money pump with foresight that does not need this assumption.
The Journal of Philosophy 117 (10): 578–589, 2020
The Sequential Dominance Argument for the Independence Axiom of Expected Utility Theory
Philosophy and Phenomenological Research 103 (1): 21–39, 2021
Independence is the idea that, if X is preferred to Y, then a lottery between X and Z is preferred to a lottery between Y and Z given the same probability of Z. The standard objection to Independence is based on the alleged rationality of Allais and Ellsberg preferences. This paper argues that Allais and Ellsberg preferences can be shown to irrational with fairly minimal assumptions. But, to defend the versions of Independence which are strong enough to serve in the standard characterization of Expected Utility Theory, we need stronger assumptions.
Money Pumps, Incompleteness, and Indeterminacy
Philosophy and Phenomenological Research 92 (1): 60–72, 2016
The money-pump argument is the standard argument that rational preferences are transitive. This argument, I argue, fails if preferences are rationally permitted to be incomplete but it works if preferences are rationally permitted to be indeterminate and rationally required to be complete.
Update: I now think there is a compelling money-pump argument for Completeness. See Money-Pump Arguments, pp. 22–39.
The Irrelevance of the Diachronic Money-Pump Argument for Acyclicity
The Journal of Philosophy 110 (8): 460–464, 2013
The money-pump argument is the standard argument that rational preferences are acyclic. It shows that, if one has cyclic preferences, one may accept a series of trades that results in a sure loss. Two stock objections are that one may get the drift and refuse the trades and that one may adopt a plan to only accept some of the trades. This paper argues that, if the diachronic money-pump argument is cogent, so is a more direct synchronic argument. The stock objections to the money-pump argument do not affect this synchronic argument.
A Money-Pump for Acyclic Intransitive Preferences
Dialectica 64 (2): 251–257, 2010
The standard argument that rational preferences are transitive is the money-pump argument. Still, the standard version of this argument only works against cyclic preferences. This paper presents a money-pump argument that also works against acyclic intransitive preferences.
A Universal Money Pump for the Myopic, Naive, and Minimally Sophisticated
Past money pumps against cyclic preferences have either worked for myopic and naive agents or for sophisticated agents, but they have not worked for all of these kinds of agents. I present a universal money pump that does so. It also works for minimally sophisticated agents who do not assume that they will choose rationally at nodes that can only be reached by irrational choices.
The Possibility of Undistinguishedness
Philosophical Studies, 180 (2): 609–613, 2023
It is natural to assume that every value bearer must be good, bad, or neutral. I argue that this assumption is false if value incommensurability is possible. More precisely, if value incommensurability is possible, then there is a fourth category of absolute value, in addition to the good, the bad, and the neutral.
Does the Collapsing Principle Rule Out Borderline Cases?
Utilitas 30 (4): 483–492, 2018
Luke Elson presents two examples that purport to show that John Broome’s Collapsing Principle rules out borderline cases. This paper argues that they do so only by making some plausible ancillary assumptions, which could be consistently rejected. The conclusion is that the Collapsing Principle does not rule out borderline cases but it is implausible.
Still Not ‘Good’ in Terms of ‘Better’
Noûs 50 (4): 854–864, 2016
Erik Carlson puts forward a new way of defining monadic value predicates, such as ‘good’, in terms of dyadic value relations, such as ‘better’. Earlier definitions of this kind have the unwanted feature that they rule out some reasonable axiologies by conceptual fiat. Carlson claims that his definitions do not have this drawback. This paper argues that they do.
Neither ‘Good’ in Terms of ‘Better’ nor ‘Better’ in Terms of ‘Good’
Noûs 48 (3): 466–473, 2014
This paper argues against defining either of ‘good’ and ‘better’ in terms of the other. Against defining ‘good’ in terms of ‘better’, I argue that it would rule out some reasonable axiologies. Against defining ‘better’ in terms of ‘good’, I argue that it either cannot allow for the incorruptibility of intrinsic goodness or it would break down in cases where both of the relata of ‘better’ are bad.
Value-Preference Symmetry and Fitting-Attitude Accounts of Value Relations
The Philosophical Quarterly 63 (252): 476–491, 2013
Value-preference symmetry is the claim that for every value relation, there is a corresponding preference relation, and vice versa. This paper argues that, if this symmetry holds, Joshua Gert’s and Wlodek Rabinowicz’s frameworks for value relations are either inadequate or without support. I present a simpler framework for value relations that allows for value-preference symmetry.
Indeterminacy and the Small‑Improvement Argument
Utilitas 25 (4): 433–445, 2013
The small-improvement argument is the standard argument against the view that, for any two things, either one is better than the other or they are equally good. This paper argues that the small-improvement argument fails since it relies on some comparisons that might be indeterminate. I defend this view from two objections by Ruth Chang, namely, the argument from phenomenology and the argument from perplexity. I also offer a new counter-example to John Broome’s collapsing principle.
Second Thoughts about My Favourite Theory
Pacific Philosophical Quarterly, 103 (3): 448–470, 2022
I argue that My Favourite Theory is vulnerable to a value pump. The argument generalizes to other approaches that avoid intertheoretic comparisons of value, such as My Favourite Option, the Borda Rule, and the Principle of Maximizing Expected Normalized Moral Value.
In Defence of My Favourite Theory
with Olle Torpman
This paper defends the My Favourite Theory approach to moral uncertainty. On this approach, a morally conscientious agent chooses an option that is permitted by the most credible moral theory. In defence of this approach, we argue that, unlike its main rivals, it prescribes dynamically consistent choices without relying on intertheoretic comparisons of value. We rebut the arguments that have been levelled against the approach and offer some arguments against intertheoretic comparisons of value.
Update: I’ve had second thoughts. See ‘Second Thoughts about My Favourite Theory’.
Pacific Philosophical Quarterly 95 (2): 159–174, 2014
An Extended Framework for Preference Relations
Economics and Philosophy 27 (3): 360–367, 2011
The traditional framework for preference relations analyses them in terms of a hypothetical choice between the relata. I extend this approach by analysing dyadic preference relations in terms of two hypothetical choices: the choice between keeping the first relata or swapping it for the second relata; and the choice between keeping the second or swapping it for the first.
Note: A misprinted version was published in Economics and Philosophy 27 (2): 101–108, 2011.
Conflicting Reasons in the Small‑Improvement Argument
with Nicolas Espinoza
The small-improvement argument is the standard argument that rational preferences need not be complete. This paper argues that while there might be reasons to believe each of the premises in the small-improvement argument, there is a conflict between these reasons such that they provide no reason to believe the conclusion.
Update: Erik Carlson has shown that a revised version of the Small-Improvement Argument can get around this challenge. See his ‘The Small-Improvement Argument Rescued’. Moreover, I now think there is a compelling money-pump argument for Completeness. See Money-Pump Arguments, pp. 22–39.
The Philosophical Quarterly 60 (241): 754–763, 2010
The Need for Merely Possible People
Journal of Ethics and Social Philosophy, forthcoming
W. V. Quine wished to restrict the interests that matter to those of actual people. This restriction can, given some minimal ethical assumptions, lead to choices that are worse than the opposite choices for everyone. Hence we need merely possible people in population ethics.
with Petra Kosonen
We defend longtermism on purely prudential grounds. We argue that the main determinants of our expected well-being lies in the far future. This makes even person-afffecting views, commonsense morality, and partial views able to ential longtermism.
Essays on Longtermism, Oxford University Press, forthcoming
Utilitas, 34 (4): 386–391, 2022
A dialogue, in three parts, on utilitarian vulnerability to exploitation.
Utilitarianism without Moral Aggregation
Canadian Journal of Philosophy, 51 (4): 256–269, 2021
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 utilitarianism, the former would be better because the sum total of well-being would be greater. This justification involves moral aggregation, since it is based on a comparison between aggregates of different people’s well-being. I provide an alternative justification that can justify any utilitarian evaluation without relying on moral aggregation.
What Should We Agree on about the Repugnant Conclusion?
with Stéphane Zuber and 27 co-authors
We should agree that implying the Repugnant Conclusion isn’t fatal for a moral theory.
Utilitas, 33 (4): 379–383, 2021
Permissibility Is the Only Feasible Deontic Primitive
Philosophical Perspectives 34 (1): 117–133, 2020
Moral obligation and permissibility are usually thought to be interdefinable. I argue that at most one direction of this alleged interdefinability works. If we do not require the conceptual possibility of moral dilemmas, then there is a way to plausibly define obligation in terms of permissibility.
Our Intuitive Grasp of the Repugnant Conclusion
The Oxford Handbook of Population Ethics, Oxford University Press, pp. 371–389, 2022
The Repugnant Conclusion is a counter-intuitive implication of Total Utilitarianism. A compelling defence of the latter is that our intuitions are unreliable in this case because the Repugnant Conclusion involves very large numbers. This chapter surveys earlier proposals of this kind and proposes a variation based on the idea that our intuitions’ unreliability is due to a slight insensitivity in our intuitive grasp of the relevant factors.
Population Axiology and the Possibility of a Fourth Category of Absolute Value
Economics and Philosophy 36 (1): 81–110, 2020
Critical-Range Utilitarianism is a variant of Total Utilitarianism which can avoid both the Repugnant Conclusion and the Sadistic Conclusion in population ethics. Yet Standard Critical-Range Utilitarianism entails the Weak Sadistic Conclusion. I defend a version of Critical-Range Utilitarianism which does not entail the Weak Sadistic Conclusion. This is made possible by what I call ‘undistinguishedness’, a fourth category of absolute value in addition to goodness, badness, and neutrality.
Is Objective Act Consequentialism Satisfiable?
Analysis, 79 (2): 193–202, 2019
The Principle of Satisfiability says that, in every possible choice situation with a finite number of alternatives, there is at least one performable act such that, if one were to perform that act, it would not be wrong. This paper argues that Objective Act Consequentialism violates the Principle of Satisfiability.
Consequentialism with Wrongness Depending on the Difficulty of Doing Better
Thought 5 (2): 108–118, 2016
Moral wrongness comes in degrees. On a consequentialist view of ethics, the wrongness of an act should depend, I argue, in part on how much worse the act’s consequences are compared to those of its alternatives and in part on how difficult it is to perform the alternatives with better consequences. I extend act consequentialism to take this into account.
Combinative Consequentialism and the Problem of Act Versions
Philosophical Studies 167 (3): 585–596, 2014
If both an act and a more specific version of that act are among an agent’s alternatives, act consequentialism yields deontic paradoxes. The standard response to this problem is to restrict the application of act consequentialism to certain relevant alternative sets. One such approach is maximalism, which says that act consequentialism should only be applied to maximally specific acts. This paper argues against maximalism and proposes instead that the relevant alternatives should be the exhaustive combinations of acts the agent can jointly perform without performing any other act in the situation.
Ex-Post Average Utilitarianism Can Be Worse for All Affected
with Dean Spears
Ex-Post Average Utilitarianism can be worse for all affected people. And we can show this without assuming any well-defined expectations of well-being for merely-possible people.
Against Negative Utilitarianism
Negative Utilitarianism is variant of classical utilitarianism. It says that you ought to minimize the sum total of pain. I present a new counter-example to Negative Utilitarianism and to other suffering-focused variations of classical utilitarianism.
Rule Consequentialism and the Problem of Ties
According to Rule Consequentialism, an act ought to be done if it is required by a rule that is better than any other rule. This formulation does not tell us what to do if there are multiple optimal rules. It’s surprisingly hard to find a plausible tie-breaker rule that is in keeping with consequentialism. I defend a new tie-breaker rule that is.
Dennett’s Prime-Mammal Objection to the Consequence Argument
Daniel C. Dennett argues that the Consequence Argument suffers from the same error as a clearly unconvincing argument that there are no mammals. I argue that the latter is unconvincing as it takes the form of a sorites-like infinite regresss, whereas the Consequence Argument does not. Hence Dennett’s objection misses its mark.
Dennett and Taylor’s Alleged Refutation of the Consequence Argument
Analysis, 80 (3): 426–433, 2020
Daniel C. Dennett and Christopher Taylor claim that the Consequence Argument relies on an inference rule that is invalid given a fairly plausible account of having the power to cause something. I show that Dennett and Taylor’s refutation does not work against a better, more standard version of the Consequence Argument.
A Strengthening of the Consequence Argument for Incompatibilism
Analysis 77 (4): 705–715, 2017
The aim of the Consequence Argument is to show that, if determinism is true, no one has, or ever had, any choice about anything. The argument has two premises: the fixity of the past and the fixity of the laws. The traditional version of the argument relies on fairly weak versions of these premises, but it requires an invalid inference rule. A standard response is to strengthen both premises. I show that we only need to strengthen one of them.
Decisions under Ignorance and the Individuation of States of Nature
Thought, 11 (2): 86–92, 2022
The most influential objection to the Laplace rule is that it is sensitive to the individuation of states of nature. This is problematic since the individuation of states seems arbitrary. I argue that this objection proves too much. I show that all plausible rules for decisions under ignorance will be sensitive to the individuation of states of nature.
A Note in Defence of Ratificationism
Erkenntnis 75 (1): 147–150, 2011
Andy Egan argues that neither evidential nor causal decision theory gives the intuitively right recommendation in the cases The Smoking Lesion, The Psychopath Button, and The Three-Option Smoking Lesion. Furthermore, Egan argues that we cannot avoid these problems by any kind of ratificationism. This paper develops a version of ratificationism that gives the intuitively right recommendations in each of these cases.
Preference and Choice
Doctoral Thesis, Royal Institute of Technology, 2011
Measurement of Freedom of Choice
Freedom of Choice and Expected Compromise
Social Choice and Welfare 35 (1): 65–79, 2010
This paper develops a measure of freedom of choice based on the proposal that an option set offers more freedom of choice than another if and only if the expected degree of dissimilarity between a random alternative from the universal set of all possible alternatives and the most similar offered alternative in the option set is smaller.
Update: I have changed my mind regarding this measure. See ‘A Paradox for the Intrinsic Value of Freedom of Choice’, p. 911n41.
The Ex-Ante Difference Principle Can Be Worse for Everyone Whatever Happens
Alexander Motchoulski and Phil Smolenski try to defend the Ex-Ante Difference Principle from my earlier challenge and claim that there are formal proofs that this principle would be chosen behind the Veil of Ignorance after all. I rebut their objections and show that these alleged proofs don't work. In doing so, I present a counter-example that does not rely on any access to probabilities and a case where the Ex-Ante Difference Principle is worse for everyone whatever happens (making worries about risk attitudes irrelevant).
Ex-Ante Prioritarianism Violates Sequential Ex‑Ante Pareto
Utilitas, 34 (2): 167–177, 2022
Ex-Post Prioritarianism adjusts for prioritizing the worse off in final outcomes, whereas Ex-Ante Prioritarianism adjusts for prioritizing the worse off on each person’s expectation. 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. I argue that Ex-Ante Prioritarianism suffers from much the same problem: It violates a sequential version of Ex-Ante Pareto.
The Levelling-Down Objection and the Additive Measure of the Badness of Inequality
Economics and Philosophy 36 (3): 401–406, 2020
The Levelling-Down Objection is a standard objection to egalitarian theories where equality is the only thing that has intrinsic value. But most egalitarians are value pluralists; in addition to equality, they hold that the egalitarian currency in which we are equal or unequal is intrinsically valuable. I argue that, given the Additive Individual-Complaints Measure of the badness of inequality, the Levelling-Down Objection still minimizes the weight that the badness of inequality may have in the overall evaluation of outcomes.
A Paradox for the Intrinsic Value of Freedom of Choice
Noûs 54 (4): 891–913, 2020
A standard liberal claim is that freedom of choice is not only instrumentally valuable but also intrinsically valuable, that is, valuable for its own sake. I argue that each one of five conditions is plausible if freedom of choice is intrinsically valuable. Yet there exists a counter-example to the conjunction of these conditions. Hence freedom of choice is not intrinsically valuable.
The Difference Principle Would Not Be Chosen behind the Veil of Ignorance
The Journal of Philosophy 115 (11): 588–604, 2018
John Rawls argues that the Difference Principle would be chosen by parties who deliberate behind the Veil of Ignorance, where they do not know who they are and are not even able to assign or estimate any probabilities to their turning out be any particular member of society. This paper argues that the parties behind the Veil of Ignorance would reject the Difference Principle.
Sequential Dominance and the Anti‑Aggregation Principle
Philosophical Studies 172 (6): 1593–1601, 2015
T. M. Scanlon’s anti-aggregation principle says that it is wrong to save a larger number of people from minor harms rather than a smaller number from much more serious harms. I show that this principle, in some cases, requires that one knowingly makes everyone worse off. This result holds regardless of how much worse a harm has to be in order to count as a much more serious harm and whether one accepts a counterfactual or temporal view of harm.
Do Lefty and Righty Matter More than Lefty Alone?
with Petra Kosonen
Derek Parfit argued that having your brain halves successfully transplanted to two separate bodies is better for you than death. This paper argues that, in addition, it is better for you to have both transplants succeed rather than just one. This suggests a total, rather than average, view of aggregation for prudential concern.
Is Psychology What Matters in Survival?
Australasian Journal of Philosophy, 99 (3): 504–516, 2021
On 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. I argue that this is implausible if the continuity is allowed be temporally unordered. I also argue that, if the continuity is required to be temporally ordered, it cannot plausibly be purely psychological. So psychological continuity is not what matters in survival.
Non-Branching Personal Persistence
Philosophical Studies, 176 (9): 2307–2329, 2019
Given reductionism about people, personal persistence must fundamentally consist in some kind of impersonal continuity relation. Typically, these continuity relations can hold from one to many. And, if they can, the analysis must include a non-branching clause to avoid non-transitive identities or multiple occupancy. This paper explores what logical form this clause should take.
Note: You can test the different accounts of the structure of personal persistence in the Personal-Persistence Lab.
The Unimportance of Being Any Future Person
Philosophical Studies 175 (3): 745–750, 2018
Derek Parfit’s argument against the platitude that identity is what matters in survival does not work given his intended reading of the platitude, namely, that what matters in survival to some future time is being identical with someone who is alive at that time. I develop Parfit’s argument so that it works against the platitude on this intended reading.
Phenomenal Continuity and the Bridge Problem
Philosophia 39 (2): 289–296, 2011
Any theory that analyses personal identity in terms of phenomenal continuity needs to deal with the ordinary interruptions of our consciousness that it is commonly thought that a person can survive. This is the bridge problem. This paper offers a new solution to the bridge problem based on the proposal that dreamless sleep need not interrupt phenomenal continuity.
History of Philosophy
Bentham’s Binary Form of Maximizing Utilitarianism
British Journal for the History of Philosophy
26 (1): 87–109, 2018
Jeremy Bentham is often read as defending a satisficing, rather than maximizing, version of utilitarianism, where an act is right as long as it produces more pleasure than pain. This is surprising given his maximizing slogan ‘the greatest happiness of the greatest number’. I argue that Bentham consistently defends a binary form of maximizing utilitarianism, where an act’s consequences are compared to those of not performing the act.
Did Locke Defend the Memory Continuity Criterion of Personal Identity?
Locke Studies 10:113–129, 2010
This paper argues that the traditional memory interpretation of John Locke’s account of personal identity is mistaken and defends a memory continuity view according to which a sequence of overlapping memories is necessary and sufficient for personal identity. On the new interpretation, Locke is not vulnerable to Thomas Reid’s Brave Officer argument.
The Ontological Proof
A Patch to the Possibility Part of Gödel’s Ontological Proof
Analysis 80 (2): 229–240, 2020
Kurt Gödel’s version of the Ontological Proof derives rather than assumes the crucial Possibility Claim: the claim that it is possible that something God-like exists. Gödel’s derivation starts off with a proof of the Possible Instantiation of the Positive: the principle that, if a property is positive, it is possible that there exists something that has that property. I argue that Gödel’s proof of this principle relies on some implausible axiological assumptions but it can be patched so that it only relies on plausible principles. But Gödel’s derivation of the Possibility Claim also needs a substantial axiological assumption, which is still open to doubt.
A Computer Simulation of the Argument from Disagreement
with Martin Peterson
According to the argument from disagreement, widespread and persistent disagreement on ethical issues indicates that our moral opinions are not influenced by moral facts, either because no such facts exist or because they are epistemically inaccessible or inefficacious for some other reason. This paper sheds new light on this argument by putting it to test in a computer simulation.
Note: You can run your own simulations in the Opinion-Dynamics Lab.
Synthese 184 (3): 387–405, 2012
Review of Louis Narens and Brian Skyrms, The Pursuit of Happiness: Philosophical and Psychological Foundations of Utility
with Krister Bykvist
Economics and Philosophy, forthcoming
Review of Iwao Hirose, Moral Aggregation
Mind 126 (503): 964–967, 2017
Review of Karin Enflo, Measures of Freedom of Choice
Theoria 81 (1): 87–92, 2015
Mycket sent om Hedenius och Mathlein
Filosofisk tidskrift, 42 (4): 35–36, 2021
An errata for my published works
Note: These errors have been fixed in the eprint versions on this page.