Marcus Arvan

This chapter argues that the ideal of liberal neutrality entails that two economic dogmas—one descriptive and one normative—have led many theorists, policymakers, journalists, and members of the public to wrongly embrace an elitist and technocratic conception of normative economics. I then argue that both dogmas should be rejected in favor of a resolutely democratic conception of economics: one in which the members of a democratic society are entitled to collectively determine on an ongoing basis, through their expressed democratic will, which phenomena economics is concerned with and what constitutes good economics for their polity. Critically, the liberal-democratic conception of economics defended is anti-populist, holding not that common citizens “know better” than economic and political elites, but instead that there should be a particular division of labor between the general public and experts—one in which the proper role of experts is to educate the public about complex phenomena and guide democratic polities by ideals of rational choice theory, but otherwise pursue the public’s conception of "the economic" and how economic activity should be normatively evaluated.

Peer review is often taken to be the main form of quality control on academic research. Usually journals carry this out. However, parts of maths and physics appear to have a parallel, crowd-sourced model of peer review, where articles are posted on the arXiv to be publicly discussed. In this article we argue that crowd-sourced peer review is likely to do better than journal-solicited peer review at sorting articles by quality. Our argument rests on two key claims. First, crowd-sourced peer review will lead on average to more reviewers per article than journal-solicited peer review. Second, due to the wisdom of the crowds, more reviewers will tend to make better judgements than fewer reviewers will. We make the second claim precise by looking at the Condorcet jury theorem as well as two related jury theorems developed specifically to apply to peer review.

Peer review is often taken to be the main form of quality control on academic research. Usually journals carry this out. However, parts of maths and physics appear to have a parallel, crowd-sourced model of peer review, where articles are posted on the arXiv to be publicly discussed. In this article we argue that crowd-sourced peer review is likely to do better than journal-solicited peer review at sorting articles by quality. Our argument rests on two key claims. First, crowd-sourced peer review will lead on average to more reviewers per article than journal-solicited peer review. Second, due to the wisdom of the crowds, more reviewers will tend to make better judgements than fewer reviewers will. We make the second claim precise by looking at the Condorcet jury theorem as well as two related jury theorems developed specifically to apply to peer review.

This paper argues that the existence and nature of consciousness are best explained by manifest asymmetries between space and time. I show that because the present manifests as immeasurably short in duration, to reason effectively about its environment in real-time a cognitive system must (i) construct integrated qualitative world-models, (ii) distributed across space, (iii) at every present instant, (iv) updating them accurately and coherently across time. I then show that this explains why brains systematically utilize analog computation, why consciousness appears so confounding, and why digital computers probably cannot realize consciousness or intentional mental states (beliefs, desires, hopes, fears, etc.) that causally impact their behavior. First, given the ephemeral nature of the present, to reason effectively about its environment a system must model the world as it is now, representing various qualities of the present state of the world (colors, sounds, shapes, etc.) simultaneously as an integrated whole. Second, only analog computation can achieve this. Whereas digital computers consist only of unintegrated binary (“on/off”) representations at any present instant, analog computation can utilize graded magnitudes of diverse phenomena (such as different levels of electrical activation) to simultaneously represent different qualities in the world at a single moment, instantaneously integrating them into a unified world-model distributed across space. Third, consciousness appears to be so philosophically and scientifically mysterious because, to fit such qualitative representations into an irreducibly short present, brains must compress them into maximally simple units (i.e. individual color-qualities, sounds, etc.), which thus appear to any system which has them to resist any further (scientific or functional) explanation as an apparently-irreducible “subjective point-of-view.” Finally, insofar as intentional mental states manifest as depending upon conscious experience, insofar as causation manifests as a relationship between occurrent events, and insofar as events manifest as occurring at the present, it follows that only analog computing devices (e.g., brains) can realize consciousness, intentional mental states, and use such states to cause their future behavior.

This chapter reviews John Rawls’s argument that justice is comprised by principles that rational individuals would agree upon in a situation of radical ignorance. Unjust inequalities are shown to be grounded in people’s knowledge of how to unfairly take advantage of their situation. Thus, Rawls argues, principles of justice can be identified by imagining people in a hypothetical situation where they are made ignorant of their own identity by a “veil of ignorance.” If no one knew who they were—and hence, couldn’t unfairly privilege themselves—which principles of justice would rational people agree upon? While some argue that they would agree upon a utilitarian principle of maximizing happiness—which can justify sacrificing the few for the good of the many—Rawls instead argues that it would be more rational to agree upon an egalitarian distribution of rights, opportunities, and wealth that advantages everyone, particularly society’s worst-off group. Societies that conform to Rawls’s principles relatively well are identified as comparatively happy, stable, and non-corrupt. However, criticisms of Rawls’s theory are identified, as are debates on deeper moral and prudential justifications of the theory, and how to extend it to international/global justice, gender and family, and the rectification of injustice.

This paper outlines how a potential new interpretation of quantum mechanics—the Many-Interacting Simulations (MIS) interpretation—may explain why our world has many of the bizarre quantum features it does. §1 provides an overview of what has been termed the Peer-to-Peer (P2P) Simulation Hypothesis, showing how it explains general features of quantum phenomena. §2 then examines dead reckoning, a framework utilized to ensure ‘believable’ simulated worlds for P2P networked games. §3 details how dead reckoning in P2P games reproduces—and hence, computationally explains—properties broadly analogous to (§3.1) quantum mechanical standing waves, (§3.2) Gaussian wave packets, (§3.3) the double-slit experiment, and (§3.4) the ‘Wigner’s friend’ thought experiment. §4 then details how, if this explanation is correct, then traditional interpretations of quantum mechanics each contain elements of truth. §5 concludes that quantum phenomena may be the ‘glitches in the Matrix’ that many have suggested would be evidence that we live in a computer simulation.

Conservatives and progressives rarely agree on much—but one thing many agree upon is that it’s not OK to be a moderate. This book shows they are wrong. In Why It’s OK to be a Moderate, Marcus Arvan shows how many of history’s worst evils have resulted from far-right and far-left radicalism, how escalating conflicts between conservatives and progressives are undermining democracy, and how many widely hailed social and political achievements have been achieved by moderates and radicals working in constructive tension with each other. Using philosophy, science, and historical analysis, Arvan shows that critics of moderates tend to equate them with spineless centrists, but that most moderates aren’t centrists, falling into diverse categories across the political spectrum. Arvan then shows that although radicals tend to be popular in their era, many of them have gone down in infamy, while many moderates, like Abraham Lincoln or Clement Attlee, have endured short-term unpopularity to “make history.” Arvan shows that it’s OK to be a moderate precisely because not everyone should be one. He makes this case to you, showing that whatever your reasonable political ideology may be, things tend to go best politically when radicals and moderates effectively complement each other’s virtues while counterbalancing the other’s vices.

Michael Huemer claims to give an ontological proof of robust moral realism, the influential view that we have non-selfish, categorical, observer-independent reasons for action. This paper argues that one of Huemer’s premises – that knowing that baby torture is not objectively wrong would provide us with no first-person reasons to torture babies – is false of agents with sadistic desires. This in turn falsifies Huemer’s further premise that the premises of his “Antitorture Argument” are true independent of interests, desires, or other attitudes. My refutation is then shown to not only deepen extant doubts about whether robust realists can adequately account for the first-person normative authority of morality. It also suggests that realists about categorical moral reasons must either complete a project – moral constructivism – which has never clearly succeeded, or else defend views on the nature of categorical reasons that strain credulity and beg the question against their opponents.

The dominant theory of the evolution of moral cognition across a variety of fields is that moral cognition is a biological adaptation to foster social cooperation. This chapter argues, to the contrary, that moral cognition is likely an evolutionary exaptation: a form of cognition where neurobiological capacities selected for in our evolutionary history for a variety of different reasons—many unrelated to social cooperation—were put to a new, prosocial use after the fact through individual rationality, learning, and the development and transmission of social norms. This chapter begins with a brief overview of the emerging behavioral neuroscience of moral cognition. It then outlines a novel theory of moral cognition that I have previously argued explains these findings better than alternatives. Finally, it shows how the evidence for this theory of moral cognition and human evolutionary history together suggest that moral cognition is likely not a biological adaptation. Instead, like reading sheet music or riding a bicycle, moral cognition is something that individuals learn to do—in this case, in response to sociocultural norms created in our ancestral history and passed down through the ages to enable cooperative living.

This chapter details three sources of normative moral disagreement and surveys 11 approaches to understanding its implications for normative ethics. Section 2 explains how normative moral disagreement can emerge from first-order commonsense moral disagreement, second-order metaethical disagreement over moral concepts and methods of ethics, and third-order metaphilosophical disagreement over the merits of different philosophical methods. Section 3 then details how moral disagreement has been argued to support either moral error theory (Section 3.1), moral skepticism (Section 3.2), moral relativism (Section 3.3), common moral foundations beneath areas of disagreement (Section 3.4), empirical approaches to ethics (Section 3.5), considerations of peer-disagreement (Section 3.6), convergence of moral frameworks (Section 3.7), moral compromise (Section 3.8), moral pluralism (Section 3.9), moral pragmatism (Section 3.10), or metaphilosophical examination of the epistemic merits of different philosophical methods (Section 3.11).

This paper uses famous problems from philosophy of science and philosophical psychology—underdetermination of theory by evidence, Nelson Goodman’s new riddle of induction, theory-ladenness of observation, and “Kripkenstein’s” rule-following paradox—to show that it is empirically impossible to reliably interpret which functions a large language model (LLM) AI has learned, and thus, that reliably aligning LLM behavior with human values is provably impossible. Sections 2 and 3 show that because of how complex LLMs are, researchers must interpret their learned functions largely in terms of empirical observations of their outputs and network behavior. Sections 4–7 then show that for every “aligned” function that might appear to be confirmed by empirical observation, there is always an infinitely larger number of “misaligned”, arbitrarily time-limited functions equally consistent with the same data. Section 8 shows that, from an empirical perspective, we can thus never reliably infer that an LLM or subcomponent of one has learned any particular function at all before any of an uncountably large number of unpredictable future conditions obtain. Finally, Section 9 concludes that the probability of LLM “misalignment” is—at every point in time, given any arbitrarily large body of empirical evidence—always vastly greater than the probability of “alignment.”

This article argues that the possibility that we live in a computer simulation has important implications for the philosophy of time and time travel. Section 2 distinguishes real time from simulated time. Section 3 argues that whatever is true of real time, simulated time realises a functional analogue within real time to versions of presentism in which features of the present play the role of an uninstantiated past and future. Section 4 then argues that whereas real time travel paradoxes depend upon the ultimate nature of ‘base-reality’ and consistency with observed physics, the possibility of simulated time travel depends instead on a simulation's code, which may allow seemingly ‘miraculous’ exceptions to ‘physical laws’. Finally, Section 5 concludes that if miraculous forms of time travel are discovered, this may provide evidence by inference to the best explanation that we live in a simulation.

This chapter explores the hidden depths beneath the vibrant veneer of AppleTV's breakout, award-winning sitcom – Ted Lasso. Ted Lasso depicts several flavors of toxic masculinity. Toxic masculinity is the wrong path, clearly a moral vice. It encourages harmful behavior, such as sexual assault and domestic violence. Toxic masculinity has also been found to harm men, increasing rates of depression, stress, and substance abuse, as well as alcoholism, cancer, and sexually transmitted infections. In contrast, Ted Lasso consistently depicts positive masculinity as balancing traditionally masculine and feminine traits.

According to the standard interpretation of Einstein’s field equations, gravity consists of mass-energy curving spacetime, and an additional physical force or entity—denoted by Λ (the ‘cosmological constant’)—is responsible for the Universe’s metric-expansion. Although General Relativity’s direct predictions have been systematically confirmed, the dominant cosmological model thought to follow from it—the ΛCDM (Lambda cold dark matter) model of the Universe’s history and composition—faces considerable challenges, including various observational anomalies and experimental failures to detect dark matter, dark energy, or inflation-field candidates. This paper shows that Einstein’s Equivalence Principle entails two possible physical interpretations of General Relativity’s field equations. Although the field equations facially appear to support the standard interpretation—that gravity consists of mass-energy curving spacetime—the field equations can be equivalently understood as holding that gravitational effects instead result from mass-energy accelerating the metric-expansion of a second-order spacetime fabric superimposed upon an absolute, first-order Euclidean space, resulting in the observational appearance of spacetime curvature. This alternative interpretation of relativity is shown to be empirically equivalent to the standard interpretation of relativity, albeit with a changing value for Λ (which is similar to how Λ is understood in the conception of Λ as ‘quintessence’, but in this case takes Λ to be gravity). The reconceptualization is then shown to potentially resolve major observational anomalies for the ΛCDM model, including recent observations conflicting with ΛCDM predictions, as well as failures to directly detect dark matter, dark energy, and inflation field/particle candidates.

This paper presents a new trilemma with respect to resolving the control and alignment problems in machine ethics. Section 1 outlines three possible types of artificial moral agents (AMAs): (1) 'Inhuman AMAs' programmed to learn or execute moral rules or principles without understanding them in anything like the way that we do; (2) 'Better-Human AMAs' programmed to learn, execute, and understand moral rules or principles somewhat like we do, but correcting for various sources of human moral error; and (3) 'Human-Like AMAs' programmed to understand and apply moral values in broadly the same way that we do, with a human-like moral psychology. Sections 2–4 then argue that each type of AMA generates unique control and alignment problems that have not been fully appreciated. Section 2 argues that Inhuman AMAs are likely to behave in inhumane ways that pose serious existential risks. Section 3 then contends that Better-Human AMAs run a serious risk of magnifying some sources of human moral error by reducing or eliminating others. Section 4 then argues that Human-Like AMAs would not only likely reproduce human moral failures, but also plausibly be highly intelligent, conscious beings with interests and wills of their own who should therefore be entitled to similar moral rights and freedoms as us. This generates what I call the New Control Problem: ensuring that humans and Human-Like AMAs exert a morally appropriate amount of control over each other. Finally, Section 5 argues that resolving the New Control Problem would, at a minimum, plausibly require ensuring what Hume and Rawls term ‘circumstances of justice’ between humans and Human-Like AMAs. But, I argue, there are grounds for thinking this will be profoundly difficult to achieve. I thus conclude on a skeptical note. Different approaches to developing ‘safe, ethical AI’ generate subtly different control and alignment problems that we do not currently know how to adequately resolve, and which may or may not be ultimately surmountable.

To what extent are authors morally culpable for harms caused by their published work? Can authors be culpable even if their ideas are misused, perhaps because they failed to take precautions to prevent harmful misinterpretations? Might authors be culpable even if they do take precautions—if, for example, they publish ideas that others can be reasonably expected to put to harmful uses, precautions notwithstanding? Although complete answers to these questions depend upon controversial views about the right to free speech, this paper argues that five notions from philosophy of law and legal practice—liability, burden of proof, legal causation, mens rea, and reasoning by precedent—can be adapted to provide an attractive moral framework for determining whether an author’s work causes harm, whether and how culpable the author is for causing such harm, steps authors may take to immunize themselves from culpability, and how to responsibly develop new rules for publishing ethics.

Liberalism is often claimed to be at odds with feminism and critical race theory (CRT). This article argues, to the contrary, that Rawlsian liberalism supports the central commitments of both. Section 1 argues that Rawlsian liberalism supports intersectional feminism. Section 2 argues that the same is true of CRT. Section 3 then uses Young’s ‘Five Faces of Oppression’—a classic work widely utilized in feminism and CRT to understand and contest many varieties of oppression—to illustrate how Rawlsian liberalism supports diverse feminist and CRT projects, and why it may be critical to achieve solidarity between feminism, CRT, and Rawlsian liberalism. Finally, Section 4 responds to five objections.

Alex Byrne contends that women are (simply) adult human females, claiming that this thesis has considerably greater initial appeal than the justified true belief (JTB) theory of knowledge. This paper refutes Byrne’s thesis in the same way the JTB theory of knowledge is widely thought to have been refuted: through simple counterexamples. Lessons are drawn. One lesson is that women need not be human. A second lesson is that biology and physical phenotypes are both irrelevant to whether someone is a woman, and indeed, female in a gendered sense. A third lesson is that trans women, cis women, alien women, and robot women are all women because to be a woman is to be an adult gendered female. This paper does not purport to settle complex normative questions of ethics or justice, including whether the ordinary meaning of “woman” ought to be retained or changed—though I do note plausible implications for these debates. This paper does purport to settle what the ordinary meaning of “woman” is, and in that regard contribute to important conceptual ground-clearing regarding what constitutes an ameliorative or revisionary definition of “woman.”

School boosters are tax-exempt organizations that engage in fundraising efforts to provide public schools with supplementary resources. This paper argues that prevailing forms of school boosting are defeasibly unjust. Section 1 shows that inequalities in public education funding in the United States violate John Rawls’s two principles of domestic justice. Section 2 argues that prevailing forms of school boosting exacerbate and plausibly perpetuate these injustices. Section 3 then contends that boosting thereby defeasibly violates Rawlsian principles of nonideal theory for rectifying injustice. Thus, boosting should be presumptively either made illegal or substantially reformed. Finally, Section 4 responds to potential objections.

This article argues that if panpsychism is true, then there are grounds for thinking that digitally-based artificial intelligence may be incapable of having coherent macrophenomenal conscious experiences. Section 1 briefly surveys research indicating that neural function and phenomenal consciousness may be both analog in nature. We show that physical and phenomenal magnitudes—such as rates of neural firing and the phenomenally experienced loudness of sounds—appear to covary monotonically with the physical stimuli they represent, forming the basis for an analog relationship between the three. Section 2 then argues that if this is true and micropsychism—the panpsychist view that phenomenal consciousness or its precursors exist at a microphysical level of reality—is also true, then human brains must somehow manipulate fundamental microphysical-phenomenal magnitudes in an analog manner that renders them phenomenally coherent at a macro level. However, Sect. 3 argues that because digital computation abstracts away from microphysical-phenomenal magnitudes—representing cognitive functions non-monotonically in terms of digits—digital computation may be inherently incapable of realizing coherent macroconscious experience. Thus, if panpsychism is true, digital AI may be incapable of achieving phenomenal coherence. Finally, Sect. 4 briefly examines our argument’s implications for Tononi’s Integrated Information Theory theory of consciousness, which we contend may need to be supplanted by a theory of macroconsciousness as analog microphysical-phenomenal information integration.

Michael Huemer argues that cross-cultural convergence toward liberal moral values is evidence of objective moral progress, and by extension, evidence for moral realism. Nathan Cofnas claims to debunk Huemer’s argument by contending that convergence toward liberal moral values can be better explained by ‘two related non-truth-tracking processes’: self-interest and its long-term tendency to result in social conditions conducive to greater empathy. This article argues that although Cofnas successfully debunks Huemer’s convergence argument for one influential form of moral realism – Robust Moral Realism, which holds that moral facts are non-natural, stance-independent normative facts – Cofnas’s debunking argument broadly supports a second type of moral realism: Enlightened Self-Interest Realism, the view that moral facts are reducible to stance-dependent requirements of instrumental (‘means-end’) rationality. Finally, this article argues that insofar as different Enlightened Self-Interest Realist theories make specific predictions about the intra- and inter-personal mechanisms behind moral convergence toward liberalism, empirical observations of cross-cultural convergence can provide independent support for Enlightened Self-Interest Realism. I conclude that this is an important mark in favor of Enlightened Self-Interest Realism over Robust Moral Realism.

Various theorists contend that we may live in a computer simulation. David Chalmers in turn argues that the simulation hypothesis is a metaphysical hypothesis about the nature of our reality, rather than a sceptical scenario. We use recent work on consciousness to motivate new doubts about both sets of arguments. First, we argue that if either panpsychism or panqualityism is true, then the only way to live in a simulation may be as brains-in-vats, in which case it is unlikely that we live in a simulation. We then argue that if panpsychism or panqualityism is true, then viable simulation hypotheses are substantially sceptical scenarios. We conclude that the nature of consciousness has wide-ranging implications for simulation arguments.

The Duhem-Quine thesis famously holds that a single hypothesis cannot be confirmed or disconfirmed in isolation, but instead only in conjunction with other background hypotheses. This article argues that this has important and underappreciated implications for metaethics. Section 1 argues that if one begins metaethics firmly wedded to a naturalistic worldview—due (e.g.) to methodological/epistemic considerations—then normativity will appear to be reducible to a set of social-psycho-semantic behaviors that I call the ‘normative stance.’ Contra Hume and Bedke (2012), I argue that the normative stance provides semantically-grounded entailments from natural truths to normative truths, reducing the latter to the former. Specifically, the normative stance explains the truth-conditions, truth-values, and truth-makers of normative propositions in terms of socially grounded cognitive-behavioral rules and other natural facts, thus explaining how there can be bona fide normative facts and properties in a wholly naturalistic world. I then show that the normative stance explains the apparent stance-independence and non-naturalness of normative reasons, intrinsic value, and categoricity of moral reasons as ‘user-illusions’ generated by people having strong psycho-social propensities—rooted in evolution and social cooperation—to take these normative stances. Section 2 then argues that while the normative stance may appear to naturalists to successfully explain normativity, it will not appeal to those who come to metaethics with different background commitments. I conclude that naturalists should take the normative stance to be a promising metaethical theory of normativity, and that whether it is a true theory of normativity is something that can only be ascertained by determining which background hypotheses—naturalistic or otherwise—we should have when doing metaethics.

Chapter 1 of this book argued that moral philosophy should be based on seven principles of theory selection adapted from the sciences. Chapter 2 argued that these principles support basing normative moral philosophy on a particular problem of diachronic instrumental rationality: the ‘problem of possible future selves.’ Chapter 3 argued that a new moral principle, the Categorical-Instrumental Imperative, is the rational solution to this problem. Chapter 4 argued that the Categorical-Instrumental Imperative has three equivalent formulations akin to but superior to Kant’s formulations of the Categorical Imperative. Chapter 5 argued that my principle’s three formulations make it rational to adopt a Moral Original Position to derive moral principles. The present chapter derives Four Principles of Fairness from the Moral Original Position--principles of coercion minimization, mutual assistance, fair negotiation, and virtue—and unifies them into a single principle of rightness: Rightness as Fairness. Finally, this chapter argues that Rightness as Fairness entails a novel approach to applied ethics called ‘principled fair negotiation’, illustrating how the theory provides a plausible new framework for addressing applied cases including lying, suicide, trolleys, torture, distribution of scarce resources, poverty, and the ethical treatment of animals.