In this paper I look at causality in the context of intervention research, and discuss some problems faced in the evaluation of causal hypotheses via interventions. I draw attention to a simple problem for evaluations that employ randomized controlled trials. The common alternative to randomized trials, the observational study, is shown to face problems of a similar nature. I then argue that these problems become especially acute in cases where the intervention is complex (i.e. that involves intervening in a complex (...) system). Finally, I consider and reject a possible resolution of the problem involving the simulation of complex interventions. The conclusion I draw from this is that we need to radically reframe the way we think about causal inference in complex intervention research. (shrink)

This paper outlines a new method for identifying folk intuitions to complement armchair intuiting and experimental philosophy, and thereby enrich the philosopher’s toolkit. This new approach—trope analysis—depends not on what people report their intuitions to be but rather on what they have made and engaged with; I propose that tropes in fiction reveal which theories, concepts and ideas we find intuitive, repeatedly and en masse. Imagination plays a dual role in both existing methods and this new approach: it enables us (...) to create the scenarios that elicit our intuitions, and also to mentally represent them. The method I propose allows us to leverage the imagination of the many rather than the few on both counts—scenarios are both created and consumed by the folk themselves. (shrink)

In recent years, a number of philosophers have attempted to fix paradoxes of the counterfactual account of causation by making causation contrastive. In this framework, causation is understood to be not a two-place relationship between a cause and an effect but a three or four-place relationship between a cause, an effect and a contrast on the side of the cause, the effect or both. I argue that contrasting helps resolving certain paradoxes only if an account of admissibility of the chosen (...) set of contrasts is given. I show by means of numerous examples that it is contextual features that determine admissibility. This way, context becomes part of the semantics of causation. I finally argue that once contextualised, explicit contrasting is redundant: causation is therefore a three-place relationship between a cause, an effect and a context. (shrink)

Reflective equilibrium —the idea that we have to justify our judgments and principles through a process of mutual adjustment—is taken to be a central method in philosophy. Nonetheless, conceptions of RE often stay sketchy, and there is a striking lack of explicit and traceable applications of it. This paper presents an explicit case study for the application of an elaborate RE conception. RE is used to reconstruct the arguments from Thomson’s paper “Turning the Trolley” for why a bystander must not (...) divert a runaway trolley from five workmen onto one. Analyzing Thomson’s resulting position with the RE-criteria has two main results: Firstly, the adjustment of one of her commitments can be defended. Secondly, no justified position in RE was reached. With respect to RE as a method, the main results from this application are: There is at least one conception of RE that is sufficiently specified to be applicable; the RE criteria put real constraints on the process of justification; and an explicit application of RE has benefits in terms of clarity while at the same time providing guidance for how the justificatory process could be continued. (shrink)

We think about and refer to things that we’ve never perceived or experienced. This paper bears on how this could be. Someone is testimonially acquainted with something just in case the explanation of one’s ability to think de re thoughts about it essentially appeals to communication with others who already have that ability. The main motivation for the claim that testimonial acquaintance is possible is that it best explains how we can think de re about and refer to things we’ve (...) never perceived or experienced: we gain these abilities in part by communicating with those who already have them. The main point of this paper is negative: to argue that testimonial acquaintance, as we currently understand it, is impossible. Although one might take this to show that we cannot be acquainted with things we haven’t perceived or experienced, I prefer to think of it as motivation for rethinking the nature of acquaintance. (shrink)

My aim is to rekindle interest in David Lewis's (1983) infamous but neglected Mixed Theory of mental states. The Mixed Theory is a mix of physicalism and functionalism designed to capture the intuitions that both Martians and abnormal human Madmen can be in pain. The Mixed Theory is widely derided. But I offer a new development of the Mixed Theory immune to its most prominent objections. In doing so, I uncover a new motivation for the Mixed Theory: its unique ability (...) to explain cases in which it is plausibly indeterminate whether something is in pain. The Mixed Theory, whether or not it is ultimately correct, at least deserves renewed attention. (shrink)

The connection between theories of truth and meaning is explored. Theories of truth and meaning are connected in a way such that differences in the conception of what it is for a sentence to be true are engendered by differences in the conception of how meanings depend on each other, and on a base of underlying facts. It is argued that this view is common ground between Davidson and Dummett, and that their dispute over realism is really a dispute in (...) the theory of meaning over holism and molecularism. The view offered is contrasted with influential pragmatist and modest views. (shrink)

In "Action without interaction" I showed that one might act on a physical system, without interacting with it, by the procedure of making it disappear. This paper presents further extensions and a critique of that result. These extensions show why physical actions without interaction are possible, while underscoring the philosophical fertility of a characteristic approach to the actual infinite inaugurated by Benardete.

Physicist Max Tegmark argues that if there are infinite universes or sub-universes, we will encounter our exact duplicates infinite times, the nearest within 10^10^115 m. Tegmark assumes Humean supervenience and a finite number of possible combinations of elementary quantum states. This paper argues on the contrary that Tegmark’s argument fails to hold if possible thoughts, persons, and life histories are all infinite in number. Are there infinite thoughts we could possibly think? This paper will show that there are. If so, (...) then it is not only Tegmark’s specific claim about our duplication that is called into question. We additionally acquire another strong argument against Humean supervenience. (shrink)

Traditionally, Physics has been dominated by the image of objects, that is, by the atomistic metaphysics of absolutely intrinsic properties of qualitatively unchangeable individual entities. The first major challenge to this metaphysics inside physics comes with quantum mechanics, specifically with the well-known phenomenon known as ‘quantum entanglement’. From quantum entanglement it seems that we can conclude that: quantum objects are not independent entities; wholes have an ontological priority over their parts. However, it is arguable that is too risky to infer (...) such conclusions directly from quantum mechanics. If epistemological consequences of quantum mechanics are inescapable, is seems unwise to take any direct ontological consequences from it. After all, quantum mechanics does not refer to the states of physical entities before measurements, but it is just a theory that allows us to calculate the possible outcomes from any given measurement. Still, this does not mean that, indirectly, quantum mechanics does not give some account of quantum reality that deeply challenges traditional objects metaphysics, namely, through the simple existence of the measurement problem; and through the experimental violations of Bell inequality. Even challenged, the object picture in physics can thus prevail. But we must also not forget that on the last decades, Quantum Physics has being evolved beyond the direct scope of quantum mechanics. I will try to argue that in at least some contemporary approaches in Quantum Physics we can see that there is a movement pointing towards a relational ontological view, according to which the ontological primacy is not to be given to individual entities, as self-sufficient elements with their own intrinsic and immutable identities—as in traditional object’s metaphysics—but to some sort of relational structures. I argue that although this relational metaphysics is still to be developed, it will have to be able to account for both the relational and object natures of quantum entities. (shrink)

Most practitioners of the metaphysics of science agree that it should be a naturalized metaphysics. But, just as in other areas of philosophy, there is no consensus on what constitutes naturalism. Here I will focus on just one aspect, viz., the idea that the metaphysics of science should be epistemically naturalized. In the first section I will characterize the kind of epistemic naturalism relevant to the metaphysics of science. The main idea, drawing on the work of Penelope Maddy, is that (...) metaphysical inquiry is to be conducted and metaphysical claims justified in the very same way that scientific inquiry is conducted and scientific claims are justified. I then examine two prominent examples of metaphysicians of science proposing to “naturalize” the metaphysics of science, and argue that they fail to be epistemically natural. These are approaches due to James Ladyman and Don Ross, and to Anjan Chakravartty. Considering their failures shows us a way forward for naturalized metaphysics of science that is both more metaphysical and more scientific. (shrink)

We show that truth conditions for counterfactuals need not always be given in terms of a vague notion of similarity. To this end, we single out the important class of historical counterfactuals and give formally rigorous truth conditions for these counterfactuals, employing a partial ordering relation called "comparative closeness" that is defined in the framework of branching space-times. Among other applications, we provide a detailed analysis of counterfactuals uttered in the context of lost bets. In an appendix we compare our (...) theory with the branching space-times based reading of counterfactuals recently proposed by Belnap. (shrink)

The truly philosophical issue in machine conscioiusness is whether machines can have 'hard consciounsess'. Criteria for hard consciousness are higher than for phenomenal consciousness, since the latter incorporates first-person functional consciousness.

In order to vindicate rational-choice theory as a mode of explaining social patterns in general - social patterns beyond the narrow range of economic behaviour - we have to recognize the legitimacy of explaining the resilience of certain patterns of behaviour: that is, explaining, not necessarily why they emerged or have been sustained, but why they are robust and reliable. And once we allow the legitimacy of explaining resilience, then we can see how functionalist theory may also serve us well (...) in social science; we lose the basis - the empty black box argumenton which the rational-choice critique of the theory has mostly been grounded. (shrink)

Philosophy invariably starts with the attempt to spell out ideas and beliefs that we already hold, whether on topics like time or causality, colour or value, consciousness or free will, democracy or justice or freedom. It may go well beyond such pre-philosophical assumptions in its further developments, regimenting them in unexpected ways, revising them on novel lines, even discarding them entirely in favour of other views. But philosophy always begins with the articulation of ordinary ideas and beliefs. This is where (...) its ladder starts. As the name suggests, articulation requires that the ideas and beliefs articulated were already there waiting to be spelled out, and that identifying them is a matter of analysing.. (shrink)

Michael Strevens develops kairetic account of causal explanations as a brand of explanatory reductionism. He argues that explanations in higher-level sciences are complete only because they can be potentially deepened—that is, added kernels of causal processes all the way down to the level of micro-physical relations. Thus, they are, in essence, the result of abstraction from deeper causal explanatory levels. I argue that Strevens’s discussion of the notion of depth in science is limited to a very narrow domain, the boundaries (...) of which are determined by a simplistic amalgam of science textbook and everyday cases analyzed by means of rational metaphysics. In contrast to his view, history of scientific practice shows that scientific explanations are typically bounded within a level and do not draw their viability from their potential for lower-level explanatory deepening. Moreover, a result of such deepening of higher-level explanations produces changes and refinements much more complex than Strevens’s account assumes. (shrink)

The criticisms levelled at the notion of truth by an anti-realist and an entity-realist are critically examined. The upshot of the discussion will be that whilst neither of the two anti-truth philosophers have succeeded in establishing their cases against truth, for entity-realists to reject the notion of truth is to throw out the baby with the bath water: entity-realism without the notion of correspondence truth will degenerate into anti-realism.

In this paper, I argue that the spatiotemporalist approach way of modeling the fundamental constituents, structure, and composition of the world has taken a wrong turn. Spatiotemporalist approaches to fundamental structure take the fundamental nature of the world to be spatiotemporal: they take the category of spatiotemporal to be fundamental. I argue that the debates over the nature of the fundamental space in the physics show us that (i) the fact that it is conceivable that the manifest world could be (...) exactly as it appears to us, even though spatiotemporal entities are not fundamental, means that a central premise of spatiotemporalism, that we may assume, given ordinary experience, that the world is fundamentally spatiotemporal, is false. (ii) Spatiotemporalism must be seen as a contingent, a posteriori physical truth. Finally, (iii) I argue that a metaphysically deeper conclusion follows from the debate over the nature of the fundamental space. The debate in physics over which sort of space is the fundamental space suggests that physicists have discovered that, even if a spacetime is an actual constituent of the world-space category, there is a world-space category that is more fundamental than the category of spacetime. (shrink)

Derek Parfit’s thesis that identity doesn’t matter in survival has been extensively discussed except for its metaphysical robustness. How can we justify the abandonment of identity in the way Parfit suggests? My argument is the following. Those who want to endorse the thesis that identity doesn’t matter (and, therefore, abandon identity across time) should adopt exdurantism, i.e. a metaphysics according to which the world is composed by temporal parts each existing at a time and according to which there is nothing (...) as a numerically same entity which exists at different times. I show that the metaphysics behind Parfit’s theory is neither compatible with endurantism, according to which object persists by being wholly present at different times, nor with perdurantism, according to which entities are aggregates of temporal and spatial parts. (shrink)

The standard argument against ordered tuples as propositions is that it is arbitrary what truth-conditions they should have. In this paper we generalize that argument. Firstly, we require that propositions have truth-conditions intrinsically. Secondly, we require strongly equivalent truth-conditions to be identical. Thirdly, we provide a formal framework, taken from Graph Theory, to characterize structure and structured objects in general. The argument in a nutshell is this: structured objects are too fine-grained to be identical to truth-conditions. Without identity, there is (...) no privileged mapping from structured objects to truth-conditions, and hence structured objects do not have truth-conditions intrinsically. Therefore, propositions are not structured objects. (shrink)

A substantial literature supports the attribution of intentional states such as beliefs and desires to groups. But within this literature, there is no substantial account of group concepts. Since on many views, one cannot have an intentional state without having concepts, such a gap undermines the cogency of accounts of group intentionality. In this paper I aim to provide an account of group concepts. First I argue that to fix the semantics of the sentences groups use to make their decisions (...) or express their beliefs, we need to appeal to a conventional semantics like that of Lewis. I then argue that the same reasons we have for taking group intentional states to be irreducible to the intentional states of their members apply also to the terms fixed by a conventional semantics. It follows that the meanings of terms in the sentences expressing a group's intentional states are also fixed by facts about the group, not its members. And recognizing this, I argue, amounts to attributing concepts to groups. Finally, I discuss a real-life example of a group concept—the meaning of ‘meter’ as fixed by the International Bureau of Weights and Measurements—and I discuss the upshot of these considerations for the question of social externalism about concepts. (shrink)

This paper defends the view that Newtonian forces are real, symmetrical and non-causal relations. First, I argue that Newtonian forces are real; second, that they are relations; third, that they are symmetrical relations; fourth, that they are not species of causation. The overall picture is anti-Humean to the extent that it defends the existence of forces as external relations irreducible to spatio-temporal ones, but is still compatible with Humean approaches to causation (and others) since it denies that forces are a (...) species of causation. (shrink)

According to Relational Quantum Mechanics the wave function \ is considered neither a concrete physical item evolving in spacetime, nor an object representing the absolute state of a certain quantum system. In this interpretative framework, \ is defined as a computational device encoding observers’ information; hence, RQM offers a somewhat epistemic view of the wave function. This perspective seems to be at odds with the PBR theorem, a formal result excluding that wave functions represent knowledge of an underlying reality described (...) by some ontic state. In this paper we argue that RQM is not affected by the conclusions of PBR’s argument; consequently, the alleged inconsistency can be dissolved. To do that, we will thoroughly discuss the very foundations of the PBR theorem, i.e. Harrigan and Spekkens’ categorization of ontological models, showing that their implicit assumptions about the nature of the ontic state are incompatible with the main tenets of RQM. Then, we will ask whether it is possible to derive a relational PBR-type result, answering in the negative. This conclusion shows some limitations of this theorem not yet discussed in the literature. (shrink)

Helen Beebee has recently argued that David Lewis’s account of compatibilism, so-called local miracle compatibilism, allows for the possibility that agents in deterministic worlds have the ability to break or cause the breaking of a law of nature. Because Lewis’s LMC allows for this consequence, Beebee claims that LMC is untenable and subsequently that Lewis’s criticism of van Inwagen’s Consequence Argument for incompatibilism is substantially weakened. I review Beebee’s argument against Lewis’s thesis and argue that Beebee has not refuted LMC (...) and concomitantly has not demonstrated that Lewis’s criticism of the Consequence Argument fails. (shrink)

A definition of causation as probability-raising is threatened by two kinds of counterexample: first, when a cause lowers the probability of its effect; and second, when the probability of an effect is raised by a non-cause. In this paper, I present an account that deals successfully with problem cases of both these kinds. In doing so, I also explore some novel implications of incorporating into the metaphysical investigation considerations of causal psychology.

What are the requirements on an adequate genuine modal realist analysis of modal discourse? One is material adequacy: the modal realist must provide for each candidate analysandum an analysans in the language of counterpart theory which by his lights has the same truth value as the candidate analysandum. Must the material biconditional joining these be necessarily true? This is the requirement of strict adequacy. It is not satisfied if Lewis’s 1968 scheme provides the analysis. John Divers puts forward a modification, (...) which identifies cases of ‘advanced modalizing’ in which the modal operator is semantically redundant. Even with this modification modal realist analyses of statements of modal discourse will be strictly inadequate. Strict adequacy can be achieved by extending the redundancy interpretation to all de dicto modal statements. The price is the denial of de dicto contingency. But perhaps material adequacy is enough. If the modal realist has a systematic means of replacing every sentence of quantified modal logic which he considers true by a sentence of counterpart theory that he considers true, perhaps he need do no more. Still, traditionally philosophical analysis aims at strict adequacy so it is as well to know that this is a test the modal realist analysis fails unless he abandons de dicto contingency. (shrink)

The paper defends Functionalism against the charge that it would make mental properties inefficacious. It outlines two ways of formulating the doctrine that mental properties are Functional properties and shows that both allow mental properties to be efficacious. The first (Lewis) approach takes functional properties to be the occupants of causal roles. Block [1990] has argued that mental properties should not be characterized in this way because it would make them properties of the ?implementing science?, e. g. neuroscience. I show (...) why this is not a problem. The second way of formulating the doctrine takes functional properties to be causal role properties. I claim that mental properties so understood would only be inefficacious if a law-centred rather than a property-centred approach is adopted to the introduction of efficacy into the world. I develop a property-centred account that explains how mental properties can be efficacacious without introducing systematic overdetermination. At the close, I provide a better characterization of the difference between these two approaches and offer an explanation as to why my way of resolving the problems has been missed. (shrink)

There’s an argument around from so-called “linguistic theories of vagueness”, plus some relatively uncontroversial considerations, to powerful metaphysical conclusions. David Lewis employs this argument to support the mereological principle of unrestricted composition, and Theodore Sider employs a similar argument not just for unrestricted composition but also for the doctrine of temporal parts. This sort of argument could be generalised, to produce a lot of other less palatable metaphysical conclusions. However, arguments to Lewis’s and Sider’s conclusions on the basis of considerations (...) about vagueness are uncompelling, even if we accept the crucial premises about vagueness. And a good thing too, since the generalised form of the argument would prove far too much. (shrink)

Foucault appears to challenge traditional views of truth, reason, and knowledge in the doctrine of power/knowledge developed in his post?1970 writings. This doctrine applies to all the sciences (and to non?scientific and non?discursive practices that are not discussed here). Foucault's notions of discourse (1) and power (3) are sufficiently discussed to set out his explanatory theory of the cause of our discourses and their change. In (4) three theses concerning the power/knowledge link are distinguished, of which the more important is (...) called the ?Primacy Thesis?, namely the causal priority of power over knowledge. Various formulations of this thesis are investigated, some of which are contrasted with Francis Bacon's more plausible account of the link between power and knowledge. In (5), six criticisms of the thesis are discussed. Foucault's notion of truth is explored on the way, more particularly in (6). The outcome of the discussion is that Foucault's doctrine of power/knowledge, apart from some insights for an empirically based sociological investigation into the causes of some of our scientific beliefs, is generally false; for philosophical theories of truth and knowledge it has nothing to say. In fact if Foucault is to be a critic of the beliefs which circulate as a result of power, he needs the traditional account of truth and knowledge. (shrink)

One very popular kind of semantics for subjunctive conditionals is aclosest-worlds account along the lines of theories given by David Lewisand Robert Stalnaker. If we could give the same sort of semantics forindicative conditionals, we would have a more unified account of themeaning of ``if ... then ...'' statements, one with manyadvantages for explaining the behaviour of conditional sentences. Such atreatment of indicative conditionals, however, has faced a battery ofobjections. This paper outlines a closest-worlds account of indicativeconditionals that does better (...) than some of its cousins in explaining thebehaviour of such conditionals. The paper then discusses objectionsoffered by Dorothy Edgington and Frank Jackson to closest-worldsaccounts of indicative conditionals, and shows that these objections canbe met by the account outlined. (shrink)

Iteration presents opposing puzzles for a theory of the imagination. The first puzzle, noted by David Lewis, is that when a person pretends to pretend, the iteration is often preserved. Let’s call this the puzzle of ‘pre- served iteration’. At the other pole, Gregory Currie has noted that very often when we pretend to pretend, the iteration does collapse. We might call this the puzzle of ‘collapsed iteration’. Somehow a theory of the imagination must be able to address these two (...) puzzles. I argue that an empirically inspired cognitive theory of the imagination (Nichols & Stich 2000) can accommodate both puzzles. (shrink)

The late nineteenth- and early twentieth-century debate over ‘internal’ and ‘external’ relations is well explored, as far as its course in Britain is concerned. F. H. Bradley’s idealistic internalism, on the one hand, and Bertrand Russell’s realistic externalism, on the other, were at the center of this debate. Less well known, however, is that there was also a discussion about relations in the United States at the time. The central figures in this discussion were Edward Gleason Spaulding and John Dewey. (...) Like Russell, Spaulding advocated a realist-inspired externalism, while Dewey criticized this viewpoint from a pragmatist perspective. The aim of the present paper is to reconstruct the exchange between Spaulding and Dewey and to elaborate the specifics of this exchange. In doing so it will emerge, among other things, that, in contrast to Bradley’s idealist rejection of externalism, Dewey’s pragmatist attempt at a refutation was more in line with common sense and science and that Spaulding’s version of externalism differed markedly from Russell’s in its strong empirical orientation. Overall, an undeservedly forgotten chapter in the history of early twentieth-century American philosophy is revisited and reassessed. (shrink)

Necessitism is the controversial thesis that necessarily everything is necessarily something, namely that everything, everywhere, necessarily exists. What is controversial about necessitism is that, at its core, it claims that things could not have failed to exist, while we have a pre-theoretical intuition that not everything necessarily exists. Contingentism, in accordance with common sense, denies necessitism: it claims that some things could have failed to exist. Timothy Williamson is a necessitist and claims that David Lewis is a necessitist too. The (...) paper argues that, granted the assumptions that lead to interpret the Lewisian as a necessitist, she can preserve contingentist intuitions, by genuinely agreeing with the folk that existence is contingent. This is not just the uncontroversial claim that the Lewisian, as a result of the prevalence of restricted quantification in counterpart theoretic regimentations of natural language, can agree with the folk while disagreeing with them in the metaphysical room. Rather, this is the claim that it is in the metaphysical room that the Lewisian can endorse the intuitions lying behind contingentism. (shrink)

In rejecting the Principle of AlternatePossibilities (PAP), Harry Frankfurt makes useof a special sort of counterfactual of thefollowing form: ``he wouldn''t have doneotherwise even if he could have''''. Recently,other philosophers (e.g., Susan Hurley (1999,2003) and Michael Zimmerman (2002)) haveappealed to a special class of counterfactualsof this same general form in defending thecompatibility of determinism andresponsibility. In particular, they claim thatit can be true of agents that even if they aredetermined, and so cannot do otherwise, theywouldn''t have done otherwise even if (...) they couldhave. Using as a central case an argument ofSusan Hurley''s, I point out that thecounterfactuals in question are both``interlegal'''' and ``indeterministic'''', and I raisedoubts about whether this special class ofcounterfactuals have clear truth conditions. Finally I suggest that acknowledging thesepoints leads to an appreciation of the realstrength of Frankfurt-style examples. (shrink)

When do two mental items belong to the same life? We could be content with the answer -just when they have certain volitional qualities in common. An affinity is noted between that theory and Berkeley's early doctrine of the self. Some rivals of the volitional theory invoke a spiritual or physical owner of mental items. They run a risk either of empty formality or of causal superstition. Other rivals postulate a non-transitive and symmetrical relation in the set of mental items. (...) They must allow in consequence either for joint ownership of one and the same mental item, or for incompatible simultaneous decisions by one and the same person, or for new forms of death. This makes them disquieting. Another rival invokes a transitive and symmetrical relation defined in terms of co-consciousness. Even that allows for incongruous simultaneities. The volitional theory is free from such disadvantages. (shrink)

While philosophers tend to consider a single type of causal history, biologists distinguish between two kinds of causal history: evolutionary history and developmental history. This essay studies the peculiarity of development as a criterion for the individuation of biological traits and its relation to form, function, and evolution. By focusing on examples involving serial homologies and genetic reprogramming, we argue that morphology (form) and function, even when supplemented with evolutionary history, are sometimes insufficient to individuate traits. Developmental mechanisms bring in (...) a novel aspect to the business of classification—identity of process-type—according to which entities are type-identical across individuals and natural kinds in virtue of the fact that they form and develop through similar processes. These considerations bear important metaphysical implications and have potential applications in several areas of philosophy. (shrink)

This article examines the implications of the holonomy interpretation of classical electromagnetism. As has been argued by Richard Healey and Gordon Belot, classical electromagnetism on this interpretation evinces a form of nonseparability, something that otherwise might have been thought of as confined to nonclassical physics. Consideration of the differences between this classical nonseparability and quantum nonseparability shows that the nonseparability exhibited by the classical electromagnetism on the holonomy interpretation is closer to separability than might at first appear.

Let us imagine an ideal ethical agent, i.e., an agent who (i) holds a certain ethical theory, (ii) has all factual knowledge needed for determining which action among those open to her is right and which is wrong, according to her theory, and who (iii) is ideally motivated to really do whatever her ethical theory demands her to do. If we grant that the notions of omniscience and ideal motivation both make sense, we may ask: Could there possibly be an (...) ideal utilitarian, that is, an ideal ethical agent whose ethical theory says that our only moral obligation consists in maximizing utility? I claim that an ideal agent cannot be utilitarian. My reasoning against ideal utilitarianism will parallel Putnam's famous argument against the brains in a vat. Putnam argues that an envatted brain cannot describe its own situation because its words do not refer to brains and vats; I argue that an ideal utilitarian cannot entertain or communicate the beliefs necessary to being a utilitarian. (shrink)