In this position paper for the 2019 CSCW workshop Good Systems: Ethical AI for CSCW I propose one tool and one idea for navigating the complex ethical problem space that results from the interaction of human and/or AI agents in shared, hopefully cooperative, computing environments.

This paper raises two problems for plan-expressivism concerning normative judgments about non-corealizable actions: actions which cannot both be performed. First, plan-expressivists associate normative judgment with an attitude which satisfies a corealizability constraint, but this constraint is (in the interpersonal case) unwarranted, and (in the intrapersonal case) warranted only at the price of a contentious normative premise. Ayars (2022) holds that the pair of judgments ‘A should φ’ and ‘B should ψ’ is coherent only if one believes that A can φ (...) while B ψ’s. But this is false. Both Gibbard (2003) and Ayars hold that the pair of judgments ‘A should φ’ and ‘A should ψ’ is coherent only if one believes that A can φ and ψ. But this assumes possibilism. Second, the paper demonstrates, cases involving interpersonal non-corealizability prompt judgments about what multiple agents should do which – contra Gibbard – are not plausibly associated with any planning subject. (shrink)

Recent advances in natural language processing have given rise to a new kind of AI architecture: the language agent. By repeatedly calling an LLM to perform a variety of cognitive tasks, language agents are able to function autonomously to pursue goals specified in natural language and stored in a human-readable format. Because of their architecture, language agents exhibit behavior that is predictable according to the laws of folk psychology: they function as though they have desires and beliefs, and then (...) make and update plans to pursue their desires given their beliefs. We argue that the rise of language agents significantly reduces the probability of an existential catastrophe due to loss of control over an AGI. This is because the probability of such an existential catastrophe is proportional to the difficulty of aligning AGI systems, and language agents significantly reduce that difficulty. In particular, language agents help to resolve three important issues related to aligning AIs: reward misspecification, goal misgeneralization, and uninterpretability. (shrink)

How should you decide what to do when you’re uncertain about basic normative principles? A natural suggestion is to follow some "second-order:" norm: e.g., obey the most probable norm or maximize expected choiceworthiness. But what if you’re uncertain about second-order norms too—must you then invoke some third-order norm? If so, any norm-guided response to normative uncertainty appears doomed to a vicious regress. This paper aims to rescue second-order norms from the threat of regress. I first elaborate and defend the claim (...) some philosophers have made that the regress problem forces us to accept normative externalism, the view that at least one norm is incumbent on all agents regardless of their normative beliefs. But, I then argue, we need not accept externalism about first-order norms, thus closing off any question of how agents should respond to normative uncertainty. Rather, we can head off the threat of regress by ascribing external force to a single second-order norm: the enkratic principle. (shrink)

How should you decide what to do when you're uncertain about basic normative principles (e.g., Kantianism vs. utilitarianism)? A natural suggestion is to follow some "second-order" norm: e.g., "comply with the first-order norm you regard as most probable" or "maximize expected choiceworthiness". But what if you're uncertain about second-order norms too -- must you then invoke some third-order norm? If so, it seems that any norm-guided response to normative uncertainty is doomed to a vicious regress. In this paper, I aim (...) to rescue second-order norms from this threat of regress. I first elaborate and defend the suggestion some philosophers have entertained that the regress problem forces us to accept normative externalism, the view that at least one norm is incumbent on agents regardless of their beliefs or evidence concerning that norm. But, I then argue, we need not accept externalism about first-order (e.g., moral) norms, thus closing off any question of what an agent should do in light of her normative beliefs. Rather, it is more plausible to ascribe external force to a single, second-order rational norm: the enkratic principle, correctly formulated. This modest form of externalism, I argue, is both intrinsically well-motivated and sufficient to head off the threat of regress. (shrink)

In philosophical action theory there is a wide agreement that intentions, often understood in terms of plans, play a major role in the deliberation of rational agents. Planning accounts of rational agency challenge game- and decision-theoretical accounts in that they allow for rationality of actions that do not necessarily maximize expected utility but instead aim at satisfying long-term goals. Another challenge for game-theoretical understanding of rational agency has recently been put forth by the theory of team reasoning in which (...) the agents select their actions by doing their parts in the collective action that is best for the group. Both planning and team reasoning can be seen as instances of a similar type of reasoning in which actions are selected on the basis of an evaluation of a larger unit than an individual’s momentary act. In recent theories of collective agency, both planning and team reasoning have been defended against orthodox game theory, but, interestingly, by different authors: Raimo Tuomela has defended team reasoning in his theory of group agency, but he ignores temporally extended planning in this context. Michael Bratman has extended his theory of planning to the case of shared agency, but he does not seem to see a role for team reasoning in understanding shared intentional activities. In this paper, we argue that both accounts suffer from this one-sidedness. We aim to combine the main insights of Tuomela’s we-mode approach and Bratman’s planning approach into a fruitful synthesis that we think is necessary for understanding the nature of group agency. (shrink)

In this paper I develop an objection to the version of expressivism found in Allan Gibbard’s book Thinking How to Live, and I suggest that the difficulty faced by Gibbard’s analysis is symptomatic of a problem for expressivism more generally. The central claim is that Gibbard’s expressivism is unable to account for certain normative judgments that arise in the process of evaluating cases of innocent mistakes. I begin by considering a type of innocent mistake that Gibbard’s view is able to (...) capture, one that can occur in situations in which our judgments of what it makes sense to do come apart from our judgments of what it makes sense to plan to do. Whether or not such mistakes are possible is a normative question, and I argue against Gibbard that we should adopt a normative stance that rules out such mistakes. This leads me to consider a second type of innocent mistake, one that can arise when an agent is constituted in such a way as to be incapable of recognizing the appropriate course of action. I argue here that our full normative assessment of the situation incorporates judgments that cannot be captured by Gibbard’s expressivism. I conclude by suggesting that any form of expressivism that shares Gibbard’s commitment to account for an intimate tie between normative judgments and action will face a similar problem. (shrink)

Clinical guidelines are special types of plans realized by collective agents. We provide an ontological theory of such plans that is designed to support the construction of a framework in which guideline-based information systems can be employed in the management of workflow in health care organizations. The framework we propose allows us to represent in formal terms how clinical guidelines are realized through the actions of are realized through the actions of individuals organized into teams. We provide various levels of (...) implementation representing different levels of conformity on the part of health care organizations. Implementations built in conformity with our framework are marked by two dimensions of flexibility that are designed to make them more likely to be accepted by health care professionals than standard guideline-based management systems. They do justice to the fact 1) that responsibilities within a health care organization are widely shared, and 2) that health care professionals may on different occasions be non-compliant with guidelines for a variety of well justified reasons. The advantage of the framework lies in its built-in flexibility, its sensitivity to clinical context, and its ability to use inference tools based on a robust ontology. One disadvantage lies in its complicated implementation. (shrink)

Cet article cherche à approfondir la conclusion de Hume selon laquelle l’induction, malgré son importance comme méthode inférentielle, est injustifiable. J’argumente en particulier que l’induction devrait être comprise comme un processus psychologique fondamental. Mon plan est le suivant : après un résumé du problème de l’induction humien, j’offre une rapide argumentation en faveur de la conceptualisation de l’induction comme une inférence à la meilleure explication (IME). Ensuite, je propose un argument original visant à montrer que l’IME est un processus psychologique (...) fondamental. Cet argument s’appuie sur une analyse du concept d’explication prouvant qu’une explication nécessite toujours l’existence d’un agent épistémique. J’utilise ce résultat afin de montrer que l’IME est un processus psychologique fondamental. Si l’induction est bel et bien une forme d’IME, l’induction est donc elle aussi un processus psychologique fondamental. (shrink)

Traditionally, policy-making at the national, state and local levels has been associated with politicians, while planning and implementation are handled mainly by planners and administrators, respectively. Education is, therefore, not different; hence future context will continue to be in the hands of politicians. However, the role of politicians in this perspective can be directed to the expectation of society if planners perform their specialist functions effectively. In this capacity, the technical and professional advice of professional planners is necessary to (...) chart a realistic or rational course for the future. The planner, therefore, is the catalyst in the politics of education. The planner, therefore, is the catalyst in the politics of education. The planners are always in the background, yet they remain the most crucial agents in growth and development in education. As is already well known, politics is the game of power, control and authority, and the influence of these forces in education determines the pattern and nature of politics in education in a given society. Because of the need to control these forces (politicians), there is always the struggle for the concentration of decisions or policy-making at certain levels of government. This is partly related to the scope of centralization and decentralization in the general policy and education in particular. Suffice it to say that centralization and power aggrandizement has become the most typical characteristics of policy-making machinery in developing countries. (shrink)

This paper offers an overview of the ways agents might extend over time and the characteristic structure of extended human agency. Agency can extend in two distinct but combinable modes: the ontological, which gives rise to simple continuous agents; and the conceptual, which gives rise to agents who conceive of and care about distal times, and have minimal planning abilities. Our extended form of agency combines both. But we are still limited by the temporal locality in the operation of (...) our psychological and executive powers. To account for this locality, I introduce the notion of 'temporal selves,' as the loci of immediacy in the agent's determination of their psychology, conduct, and practical standpoint. I argue that the passage of time generates, by itself, the threat of temporal alienation from distant temporal selves. A genuinely extended agency requires temporal identification: the sharing, by separate temporal selves, of a temporally extended and integrated practical standpoint. This temporal identification cannot be produced simply by temporal identity as continuity. What is required is temporal identity in the mode of unity and integration. This identity does not precede temporal identification but is co-constituted with it. I offer a preliminary account of the structure of the units of integration for agents who aspire to persist in the mode of unity and integration. I close with a cautionary note: the complex structure of integration, although familiar in everyday life, is often missed by standard philosophical accounts, which tend to focus on simple models of extended agency. (shrink)

Over the last decade, multi-agent systems have come to form one of the key tech- nologies for software development. The Formal Approaches to Multi-Agent Systems (FAMAS) workshop series brings together researchers from the fields of logic, theoreti- cal computer science and multi-agent systems in order to discuss formal techniques for specifying and verifying multi-agent systems. FAMAS addresses the issues of logics for multi-agent systems, formal methods for verification, for example model check- ing, and formal approaches (...) to cooperation, multi-agent planning, communication, coordination, negotiation, games, and reasoning under uncertainty in a distributed environment. In 2007, the third FAMAS workshop, FAMAS'007, was one of the agent workshops gathered together under the umbrella of Multi-Agent Logics, Languages, and Organ- isations - Federated Workshops, MALLOW'007, taking place from 3 to 7 September 2007 in Durham. This current special issue of the Logic Journal of the IGPL gathers together the revised and updated versions of the five best FAMAS'007 contributions. (shrink)

An important module in the Belief-Desire-Intention architecture for artificial agents (which builds on Michael Bratman's work in the philosophy of action) focuses on the task of intention reconsideration. The theoretical task is to formulate principles governing when an agent ought to undo a prior committed intention and reopen deliberation. Extant proposals for such a principle, if sufficiently detailed, are either too task-specific or too computationally demanding. I propose that an agent ought to reconsider an intention whenever some incompatible (...) prospect is sufficiently valuable along some dimension that can be assessed at zero or near-zero computational cost. (shrink)

By introducing elements of phenomenological philosophy to the analysis of human needs in economics; from Sartrean postulates as well as the nature and essence of individual’s needs, has been revealed a theorethical framework that serves to ponder human being’s existential behavior by means of their phenomenologic social choices and welfare. Defining a planning agent under strong assumptions of rationality and projective efficacious capabilities, the Arrow’s theorem has been proved for the economic agent aware of its finitude in (...) this world. (shrink)

The leading reductive approaches to shared agency model that phenomenon in terms of complexes of individual intentions, understood as plan-laden commitments. Yet not all agents have such intentions, and non-planning agents such as small children and some non-human animals are clearly capable of sophisticated social interactions. But just how robust are their social capacities? Are non-planning agents capable of shared agency? Existing theories of shared agency have little to say about these important questions. I address this lacuna by (...) developing a reductive account of the social capacities of non-planning agents, which I argue supports the conclusion that they can enjoy shared agency. The resulting discussion offers a fine-grained account of the psychological capacities that can underlie shared agency, and produces a recipe for generating novel hypotheses concerning why some agents do not engage in shared agency. (shrink)

According to a widely accepted constraint on the content of intentions, here called the exclusivity constraint, one cannot intend to perform another agent’s action, even if one might be able to intend that she performs it. For example, while one can intend that one’s guest leaves before midnight, one cannot intend to perform her act of leaving. However, Deborah Tollefsen’s (2005) account of joint activity requires participants to have intentions-in-action (in John Searle’s (1983) sense) that violate this constraint. I (...) argue that the exclusivity constraint should not be accepted as an unconditional constraint on the contents of intentions-in-action: one may intend to perform a basic action that belongs both to oneself and to another agent. Based on the phenomenology of tool use, I first argue that intentions-in-action of one’s basic actions may be technologically extended, meaning that their contents are not restricted to concern the agent’s bodily movements. In analogy with this, I then argue that the phenomenology of some skillful joint activities supports the idea that one’s basic intentions-in-action may be socially extended, in violation of the widely accepted exclusivity constraint. Tollefsen’s account is specifically constructed to account for the joint activities of infants and toddlers who lack the capacity to think of others as planning agents and grasp their plan-like intentions (a capacity required by Michael Bratman’s (1992, 1993, 2009a, b) influential account of joint activity). At the end of the paper, I raise some doubts regarding the extent to which infants and toddlers have socially extended intentions-in-action. (shrink)

Persons are widely believed to be rational, planning agents that are both author and main character of their life stories. A major goal is to keep these narratives coherent as they unfold, and part of a fulfilled life allegedly stems from this coherence. My aim is to challenge these convictions by considering two related claims about persons and their lives. Contrary to the widespread theoretical conviction in philosophy of mind and action, persons are fundamentally emotional and affective rather than (...) rational and deliberative beings. And so, on a practical level, persons need not constantly aspire to integrate their past, present, and future into a coherent whole in order to live fulfilled lives. Needless to say, I cannot hope to defend these claims and their relation in great detail with a few brief strokes. In addition to theoretical reflections, I discuss some practical implications and potential benefits that come with discarding the daunting task of continuously keeping track of one's life story. Drawing on insights from a contemplative Buddhist tale, I venture that the practice of letting go can break the spell, and give rise to an alleviating source of liberation from life's troubles.Export citation. (shrink)

This paper explores the impact of furniture and machines on spatial organization and circulation systems in industrial buildings using space syntax analysis. While space syntax research covers various settings, industrial buildings have received limited attention. This study aims to address this gap by examining how machines and furniture influence spatial organization and circulation in two industrial buildings: A Nitrile Glove manufacturing facility and a textile manufacturing factory. The furnished and unfurnished floor plans were analyzed using space syntax software, DepthMapX, with (...) connectivity and agent-based analyses. The results indicate that unfurnished plans have centrally located connectivity values, whereas furnished plans create subspaces with varying connectivity. Agent-based analysis reveals that unfurnished spaces have high density in the center, while furnished spaces distribute density more evenly, resulting in more uniform circulation. This study concludes that industrial building spatial configurations result from a combination of architectural design and the placement of machines and furniture. (shrink)

What is involved in acting together with others? Most shared agency theorists endorse the Shared Intention Thesis, i.e., the claim that shared agency necessarily involves shared intentions. This article dissents from this orthodoxy and offers a minimalist account of shared agency—one where parties to shared activities need not form rich webs of interrelated psychological states. My account has two main components: a conceptual analysis of shared agency in terms of the notion of plan, and an explanation of undertheorized agency‐sharing mechanisms. (...) My analysis states that we act together just in case our activities conform to a plan and that plan figures in an explanation of our activities’ joint conformity to it. To sloganize:shared activity is plan‐coordinated activity. Sometimes, plan‐coordination goes by way of shared intentions. However, besides shared intentions, there are at least two additional families of agency‐sharing mechanisms. The first features a central planner who determines the content of a plan and attributes the different parts of that plan to a collection of agents. The second does away with the planner and involves a roughly Darwinian selection of patterns of activity. Both families of mechanisms enable us to act together even in the absence of shared intentions. (shrink)

I investigate the structure of pro-tempore disjunctive intentions: intentions directed at two or more eventually incompatible goals that are nonetheless kept open for the time being, while the agent is waiting to acquire more information to determine which option is better. These intentions are the basic tool for balancing, in our planning agency, rigidity and flexibility, stability and responsiveness to changing circumstances. They are a pervasive feature of intentional diachronic agency and contribute to secure dynamic consistency in our (...) plans. I show how they differ from simple disjunctive intentions, where the agent is indifferent between the options. I argue that pro-tempore disjunctive intentions meet the distinctive pressures of intentions and that, contrary to the initial impressions, they are both stable and successful at settling practical matters. In closing, I argue that pro-tempore disjunctive intentions are all-out intentions that are more explanatorily powerful than Holton's "partial" intentions. (shrink)

I here present some doubts about whether Mandik’s (2010) proposed intermediacy and recurrence constraints are necessary and sufficient for agentive experience. I also argue that in order to vindicate the conclusion that agentive experience is an exclusively perceptual phenomenon (Prinz, 2007), it is not enough to show that the predictions produced by forward models of planned motor actions are conveyed by mock sensory signals. Rather, it must also be shown that the outputs of “comparator” mechanisms that compare these predictions against (...) actual sensory feedback are also coded in a perceptual representational format. (shrink)

This chapter aims to distinguish between pathologies of agency in the strict sense and mere sources of impediments or distortion. Expanding on a recent notion of necessarily less-than-successful agency, it complements a mainstream approach to mental disorders and anomalous psychological conditions in the philosophy of mind and action. According this approach, the interest of such clinical case studies is heuristic, to differentiate between facets of agency that are functionally and conceptually separate even though they typically come together. Yet, in the (...) absence of independent criterion for a pathology of as opposed to inner obstacle to agency, this heuristic is at risk of becoming circular or uninformative, falling back on a clinical diagnosis it is meant to take as a starting point only. The chapter develops such a criterion and shows how it could work tracking agential achievement across two core dimensions of agency: planning and responsiveness to reasons. The discussion concludes with some implications on assessing decisional capacity and safeguarding agent autonomy in psychiatric settings. (shrink)

Joint actions often require agents to track others’ actions while planning and executing physically incongruent actions of their own. Previous research has indicated that this can lead to visuomotor interference effects when it occurs outside of joint action. How is this avoided or overcome in joint actions? We hypothesized that when joint action partners represent their actions as interrelated components of a plan to bring about a joint action goal, each partner’s movements need not be represented in relation to (...) distinct, incongruent proximal goals. Instead they can be represented in relation to a single proximal goal – especially if the movements are, or appear to be, mechanically linked to a more distal joint action goal. To test this, we implemented a paradigm in which participants produced finger movements that were either congruent or incongruent with those of a virtual partner, and either with or without a joint action goal (the joint flipping of a switch, which turned on two light bulbs). Our findings provide partial support for the hypothesis that visuomotor interference effects can be reduced when two physically incongruent actions are represented as mechanically interdependent contributions to a joint action goal. (shrink)

The notion of an agent and the notion of a self are connected, for agency is one role played by the self. Millgram argues for a disunity thesis of agency on the basis of extreme incommensurability across some major life events. We propose a similar negative thesis about the self, that it is composed of relatively independent threads reflecting the different roles and different mind-sets of the person's life. Our understanding of those threads is based on theories of the (...) narrative construction of the self. Our disunity thesis is that there need be no overarching narrative that unifies those narrative threads. To explain how the threads hang together to produce coherent action, we make these positive claims: control normally switches smoothly and unconsciously between threads as circumstances require, within one thread there is likely to be acknowledgment of other threads, some situations require a temporary blending of threads, and some plans and policies reach across different threads and contribute to some coordination among them. Our account of a self provides an account of agency that has merits in comparison to Millgram's. Our narrative approach allows explanations of actions beyond rational deliberation. (shrink)

The Anscombean conception of intentional action endorses that an action's being intentional is in virtue of its internal structure, rather than any extra ingredient. Anscombe (2000) calls it the calculative order "that is there whenever actions are done with intentions," which features its teleological aspect. In his work reviving this tradition, Thompson (2008) highlights its temporal aspect. With the help of the concept "compound action," we shall see that typical intentional action attends to both aspects. The synthesis is met by (...) shifting from the aspect of action to agency. To act intentionally, an agent needs not only the capacity for calculating from a general end to a specific means, but also the capacity for planning: coordinate stages of different actions over time hierarchically, by occupying a certain kind of slots. (shrink)

The paper is an introductory essay to the Polish translation of M.E. Bratman’s paper The Fecundity of Planning Agency. Instead of summarizing the main drifts of Bratman’s work, the author tries to show a few important parallels between his approach to action theory and the so-called praxiology (or ‘theory of efficient action’) proposed by Tadeusz Kotarbinski. It occurs that there are important similarities between their approaches both to specific problems in action theory (as: concept of an agent, the (...) role of Buridan’s Ass, or mind-infusion of actions) and to the general question how to understood agency (norms of planning). A brief presentation of these similarities gives two central results. Firstly, it allows to interpret Bratman as a kind of praxiologist, and, secondly, it shows that praxiological thinking proposed by Kotarbinski is still a vital perspective in contemporary action theory. (shrink)

In a reflective and richly entertaining piece from 1979, Doug Hofstadter playfully imagined a conversation between ‘Achilles’ and an anthill (the eponymous ‘Aunt Hillary’), in which he famously explored many ideas and themes related to cognition and consciousness. For Hofstadter, the anthill is able to carry on a conversation because the ants that compose it play roughly the same role that neurons play in human languaging; unfortunately, Hofstadter’s work is notably short on detail suggesting how this magic might be achieved1. (...) Conversely in this paper - finally reifying Hofstadter’s imagination - we demonstrate how populations of simple ant-like creatures can be organised to solve complex problems; problems that involve the use of forward planning and strategy. Specifically we will demonstrate that populations of such creatures can be configured to play a strategically strong - though tactically weak - game of HeX (a complex strategic game).We subsequently demonstrate how tactical play can be improved by introducing a form of forward planning instantiated via multiple populations of agents; a technique that can be compared to the dynamics of interacting populations of social insects via the concept of meta-population. In this way although, pace Hofstadter, we do not establish that a meta-population of ants could actually hold a conversation with Achilles, we do successfully introduce Aunt Hillary to the complex, seductive charms of HeX. (shrink)

Which artificial intelligence (AI) systems are agents? To answer this question, I propose a multidimensional account of agency. According to this account, a system's agency profile is jointly determined by its level of goal-directedness and autonomy as well as is abilities for directly impacting the surrounding world, long-term planning and acting for reasons. Rooted in extant theories of agency, this account enables fine-grained, nuanced comparative characterizations of artificial agency. I show that this account has multiple important virtues and is (...) more informative than alternatives. More speculatively, it may help to illuminate two important emerging questions in AI ethics: 1. Can agency contribute to the moral status of non-human beings, and how? 2. When and why might AI systems exhibit power-seeking behaviour and does this pose an existential risk to humanity? (shrink)

While much has been written about the functional profile of intentions, and about their normative or rational status, comparatively little has been said about the subjective authority of intention. What is it about intending that explains the ‘hold’ that an intention has on an agent—a hold that is palpable from her first-person perspective? I argue that several prima facie appealing explanations are not promising. Instead, I maintain that the subjective authority of intention can be explained in terms of the (...) inner structure of intention. In adopting an intention the agent comes to see herself as criticizable depending on whether she executes the intention. This allows us to explain in first-personal terms why the agent becomes disposed to act and deliberate in ways that are characteristic of intention. As intention-formation involves profound changes to reflexive evaluative attitudes, this is the ‘Self-Evaluation’ view of the subjective authority of intention. (shrink)

Shared agency is a distinctive kind of sociality that involves interdependent planning, practical reasoning, and action between participants. Philosophical reflection suggests that agents engage in this form of sociality when a special structure of interrelated psychological attitudes exists between them, a set of attitudes that constitutes a collective intention. I defend a new way to understand collective intention as a combination of individual conditional intentions. Revising an initial statement of the conditional intention account in response to several challenges leads (...) to a specification of the properties these intentions need to have in order to be genuine commitments. I then show how a structure of conditional intentions with these properties settles a collection of agents on engaging in social interactions that display all the features typically associated with shared agency. (shrink)

The objective of this essay is to provide the beginning of a principled classification of some of the ways space is intelligently used. Studies of planning have typically focused on the temporal ordering of action, leaving as unaddressed questions of where to lay down instruments, ingredients, work-in-progress, and the like. But, in having a body, we are spatially located creatures: we must always be facing some direction, have only certain objects in view, be within reach of certain others. How (...) we manage the spatial arrangement of items around us is not an afterthought: it is an integral part of the way we think, plan, and behave. The proposed classification has three main categories: spatial arrangements that simplify choice; spatial arrangements that simplify perception; and spatial dynamics that simplify internal computation. The data for such a classification is drawn from videos of cooking, assembly and packing, everyday observations in supermarkets, workshops and playrooms, and experimental studies of subjects playing Tetris, the computer game. This study, therefore, focuses on interactive processes in the medium and short term: on how agents set up their workplace for particular tasks, and how they continuously manage that workplace. (shrink)

Why do some addicted people chronically fail in their goal to recover, while others succeed? On one established view, recovery depends, in part, on efforts of intentional planning agency. This seems right, however, firsthand accounts of addiction suggest that the agent’s self-narrative also has an influence. This paper presents arguments for the view that self-narratives have independent, self-fulfilling momentum that can support or undermine self-governance. The self-narrative structures of addicted persons can entrench addiction and alienate the agent (...) from practically feasible recovery plans. Strategic re-narration can redirect narrative momentum and therefore support recovery in ways that intentional planning alone cannot. (shrink)

This paper presents a decision problem called the holiday puzzle. The decision problem is one that involves incommensurable goods and sequences of choices. This puzzle points to a tension between three prima facie plausible, but jointly incompatible claims. I present a way out of the trilemma which demonstrates that it is possible for agents to have incomplete preferences and to be dynamically rational. The solution also suggests that the relationship between preference and rational permission is more subtle than standardly assumed.

It seems to be a humdrum fact of human agency that we act on intentions or decisions that we have made at an earlier time. At breakfast, you look at the Taco Hut menu online and decide that later today you’ll have one of their avocado burritos for lunch. You’re at your desk and you hear the church bells ring the noon hour. You get up, walk to Taco Hut, and order the burrito as planned. As mundane as this sort (...) of scenario might seem to be, philosophers have raised a problem in understanding it. If you are simply abiding by this morning’s decision, how are you acting autonomously? Your earlier self seems to be calling the shots; if you are just acting accordingly, without thinking through it or in some other way trying to ensure that the past decision conforms to your present standpoint, it is not clear how this amounts to an exercise of your present autonomous agency. It seems, rather, that your earlier self has succeeded in slaving you to her own purposes. She was the one who wanted (intended, judged it to be good, etc.) to have an avocado burrito. In simply following through, your current self seems to be just an automaton performing the commands left behind by your former self. Of course, you might not allow yourself to be shackled by your earlier self. You might refuse to follow anything but your own present judgments: you will only go to Taco Hut if this is what you judge you should do right now, and once at Taco Hut you will only eat the avocado burrito if that is what you want to eat once there. But if this is the way you generally operate, this seems to block your ability to make effective future-directed decisions. The puzzle, then, is one of explaining how the future self can do the bidding of her past self without losing her autonomy. We call this “the Problem of Diachronic Autonomy.” Philosophers raising this problem take it to show that there must be reasons or rational requirements to follow-through with our past decisions. According to these philosophers, we can only make sense of our diachronic autonomy if our past decisions put rational pressure on us to follow through. We argue that there is no Problem of Diachronic Autonomy. There is, in other words, no puzzling situation that needs explaining. Consequently, there is no need coming from this purported puzzle to think that our future-directed decisions generate reasons or rational requirements to follow through. The correct view of our diachronic autonomy is the “naïve” one: the “future self” can do the bidding of the “past self” without giving up its autonomy because, very simply, the past self is the same agent as the future self. I am acting autonomously when I get the avocado burrito, because I was the one who decided to get the burrito. I am acting on my own freely-formed decision. (shrink)

Adaptive preference formation is the unconscious altering of our preferences in light of the options we have available. Jon Elster has argued that this is bad because it undermines our autonomy. I agree, but think that Elster's explanation of why is lacking. So, I draw on a richer account of autonomy to give the following answer. Preferences formed through adaptation are characterized by covert influence (that is, explanations of which an agent herself is necessarily unaware), and covert influence undermines (...) our autonomy because it undermines the extent to which an agent's preferences are ones that she has decided upon for herself. This answer fills the lacuna in Elster's argument. It also allows us to draw a principled distinction between adaptive preference formation and the closely related phenomenon of character planning. (shrink)

This chapter discusses the structure of our temporally extended agency. We do not have the power to act directly at a distance, so any of our temporally extended projects must be sustained over its temporal unfolding by momentary actions. We need both the capacity to organize these momentary steps in light of a synoptic overview of the extended activity as a whole and to sustain our motivation to continue to pursue the extended activity. Hence, the distinctive mode in which we (...) act over time is that of ’planning agency,’ which requires a form of temporal identification that goes beyond mere psychological continuity. I then discuss the differences in internal structure between merely continuous and integrated activities, and between telic and atelic ones. I provide a taxonomy of the different kinds of temporal goods and values and how these structures might bear on the risk of crisis of meaning in our temporal existence. I then touch briefly on the issue of how our mortality might make a difference to the structure of what we find valuable and how it relates to the temporal features of our agency. In the closing section, I discuss the effects of the moving temporal location of the agent (with her local executive powers) relative to her extended pursuits, including the asymmetries generated by the distinction between past and future stages of the activity and the moving horizon of what the agent can anticipate and (indirectly) control. (shrink)

Given the endowment effect, the role of attention in decision-making, and the framing effect, most behavioral economists agree that it would be a mistake to accept the satisfaction of revealed preferences as the normative criterion of choice. Some have suggested that what makes agents better off is not the satisfaction of revealed preferences, but ‘true’ preferences, which may not always be observed through choice. While such preferences may appear to be an improvement over revealed preferences, some philosophers of economics have (...) argued that they face insurmountable epistemological, normative, and methodological challenges. This article introduces a new kind of true preference – values-based preferences – that blunts these challenges. Agents express values-based preferences when they choose in a manner that is compatible with a consumption plan grounded in a value commitment that is normative, affective, and stable for the agent who has one. Agents who choose according to their plans are resolute choosers. My claim is that while values-based preferences do not apply to every choice situation, this kind of preference provides a rigorous way for thinking about classic choice situations that have long interested behavioral economists and philosophers of economics, such as ‘Joe-in-the-cafeteria.’. (shrink)

I demonstrate that a "speech act" theory of meaning for imperatives is—contra a dominant position in philosophy and linguistics—theoretically desirable. A speech act-theoretic account of the meaning of an imperative !φ is characterized, broadly, by the following claims. -/- LINGUISTIC MEANING AS USE !φ’s meaning is a matter of the speech act an utterance of it conventionally functions to express—what a speaker conventionally uses it to do (its conventional discourse function, CDF). -/- IMPERATIVE USE AS PRACTICAL !φ's CDF is to (...) express a practical (non-representational) state of mind—one concerning an agent's preferences and plans, rather than her beliefs. -/- Opposed to speech act accounts is a preponderance of views which deny that a sentence's linguistic meaning is a matter of what speech act it is used to perform, or its CDF. On such accounts, meaning is, instead, a matter of "static" properties of the sentence—e.g., how it depicts the world as being (or, more neutrally, the properties of a model-theoretic object with which the semantic value of the sentence co-varies). On one version of a static account, an imperative 'shut the window!' might, for instance, depict the world as being such that the window must be shut. -/- Static accounts are traditionally motivated against speech act-theoretic accounts by appeal to supposedly irremediable explanatory deficiencies in the latter. Whatever a static account loses in saying (prima facie counterintuitively) that an imperative conventionally represents, or expresses a picture of the world, is said to be offset by its ability to explain a variety of phenomena for which speech act-theoretic accounts are said to lack good explanations (even, in many cases, the bare ability to offer something that might meet basic criteria on what a good explanation should be like). -/- I aim to turn the tables on static accounts. I do this by showing that speech act accounts are capable of giving explanations of phenomena which fans of static accounts have alleged them unable to give. Indeed, for a variety of absolutely fundamental phenomena having to do with the conventional meaning of imperatives (and other types of practical language), speech act accounts provide natural and theoretically satisfying explanations, where a representational account provides none. (shrink)

Alfred Mele’s zygote argument for incompatibilism is based on a case involving an agent in a deterministic world whose entire life is planned by someone else. Mele’s contention is that Ernie (the agent) is unfree and that normal determined agents are relevantly similar to him with regards to free will. In this paper, I examine four different ways of understanding this argument and then criticize each interpretation. I then extend my criticism to manipulation arguments in general. I conclude (...) that the zygote argument is no threat to compatibilism. (shrink)

Abstract: The standard picture of rationality requires that the agent acts so as to realize her most preferred alternative in the light of her own desires and beliefs. However, there are circumstances where such an agent can predict that she will act against her preferences. The story of Ulysses and the Sirens is the paradigmatic example of such cases. In those circumstances the orthodoxy requires the agent to be ‘sophisticated’. That is to say, she should take into (...) account her expected future choices and prevent her future self to act in certain ways. She should ‘bind’ herself to a certain course of action. This is a form of causal commitment. It is generally recognized that this form of self-commitment is the only one that is available to a rational agent. Rational commitment, where the agent gives herself a reason to act in a certain way rather than making herself act in that way, is considered not feasible. In this paper, I question this verdict. I sketch the broad outlines of a model of rational commitment, which takes as its starting point Michael Bratman’s ‘planning theory’ of intention. There are two important objections against this theory (one by John Broome and one by the Dutch philosopher Govert den Hartogh.) Both criticisms claim that such a theory is a form of ‘bootstrapping’ reasons for action into existence. In the remainder of the paper, I will defend the theory against these objections. This way, I hope to establish that defending the feasibility of rational commitment is not an obvious mistake. (shrink)

It is a commonplace in economics that we should disregard sunk costs. The sunk cost effect might be widespread, goes the conventional wisdom, but we would be better off if we could rid ourselves of it. In this paper, I argue against the orthodoxy by showing that the sunk cost effect is often beneficial. Drawing on discussions of related topics in dynamic choice theory, I show that, in a range of cases, being disposed to honor sunk costs allows an (...) class='Hi'>agent to mimic a resolute chooser, someone who adopts the best plan at the outset of a decision problem and sticks with it, even when resoluteness is unfeasible. I discuss several kinds of cases in which honoring sunk costs coincides with resolute choice. (shrink)

I argue that taking the Practical Conditionals Thesis seriously demands a new understanding of the semantics of such conditionals. Practical Conditionals Thesis: A practical conditional [if A][ought] expresses B’s conditional preferability given A Paul Weirich has argued that the conditional utility of a state of affairs B on A is to be identified as the degree to which it is desired under indicative supposition that A. Similarly, exploiting the PCT, I will argue that the proper analysis of indicative practical conditionals (...) is in terms of what is planned, desired, or preferred, given suppositional changes to an agent’s information. Implementing such a conception of conditional preference in a semantic analysis of indicative practical conditionals turns out to be incompatible with any approach which treats the indicative conditional as expressing non-vacuous universal quantification over some domain of relevant antecedent-possibilities. Such analyses, I argue, encode a fundamental misunderstanding of what it is to be best, given some condition. The analysis that does the best vis-à-vis the PCT is, instead, one that blends a Context-Shifty account of indicative antecedents with an Expressivistic, or non-propositional, treatment of their practical consequents. (shrink)

For ambitious metaphysical neo-sentimentalists, all normative facts are grounded in fitting attitudes, where fittingness is understood in naturalistic terms. In this paper, I offer a neo-sentimentalist account of blameworthiness in terms of the reactive attitudes of a morally authoritative subject I label a Nagelian Imp. I also argue that moral impermissibility is indirectly linked to blameworthiness: roughly, an act is morally impermissible if and only if and because it is not *possible* in the circumstances to adopt a plan of performing (...) it without meriting blame, assuming the agent is rational, informed, and meets the conditions of accountability. (shrink)

Attributions of what we shall call normative responsibilities play a central role in everyday moral thinking. It is commonly thought, for example, that parents are responsible for the wellbeing of their children, and that this has important normative consequences. Depending on context, it might mean that parents are morally required to bring their children to the doctor, feed them well, attend to their emotional needs, or to see to it that someone else does. Similarly, it is sometimes argued that countries (...) that emit most greenhouse gases are responsible for preventing catastrophic climate change. This responsibility might imply that these countries are morally required to take necessary steps individually and jointly to come to an agreement on and implement a workable plan, and to avoid steps that worsen the situation. More trivially, the grading of your student’s essays might be your responsibility, as might making sure there is wine at tomorrow’s picnic, and you might thus be required to see to it that essays are competently graded and suitable wine brought to the picnic. -/- Though attributions of normative responsibilities are legion, such responsibilities have received surprisingly little philosophical attention compared to its normative relatives, obligations and reasons, and compared to retrospective responsibility. This chapter hopes to improve on this situation by taking on two main tasks. The first, attempted in section 1, is to spell out the general structure of normative responsibility, in particular the relation between normative responsibilities and corresponding obligations and demands. We suggest that normative responsibilities are constituted by normative requirements that the responsible agents care appropriately about how well things go in certain regards, and that obligations generally can be seen as straightforward upshots of requirements to care. -/- The second task, taken on in section 2, is to provide an overview of prominent sources of normative responsibility and its distribution among agents. Why would the children’s wellbeing be the parents’ responsibility? Why not the neighbor’s, or the state’s, or everyone’s? Here we discuss a range of possible sources, including agents’ abilities, costs involved in taking on the responsibility in question, retrospective responsibility for the situation, promises or contracts, and certain social relationships. (shrink)

According to the popular Whole Life Satisfaction theories of happiness, an agent is happy when she judges that her life fulfils her ideal life-plan. Fred Feldman has recently argued that such views cannot accommodate the happiness of spontaneous or pre-occupied agents who do not consider how well their lives are going. In this paper, I formulate a new Whole Life Satisfaction theory which can deal with this problem. My proposal is inspired by Michael Smith’s advice-model of desirability. According to (...) it, an agent is happy when a more informed and rational hypothetical version of her would judge that the agent’s actual life matches the best life-plan for her. This view turns out to be a flexible model which can avoid many problems of the previous theories of happiness. (shrink)

Richard Holton has argued that the traditional account of intentions—which only posits the existence of all-out intentions—is inadequate because it fails to accommodate dual-plan cases; ones in which it is rationally permissible for an agent to adopt two competing plans to bring about the same end. Since the consistency norms governing all-out intentions prohibit the adoption of competing intentions, we can only preserve the idea that the agent in a dual-plan case is not being irrational if we attribute (...) to them a pair of partial intentions. I argue that, contrary to initial appearances, (i) Holton has yet to offer us an actual account of partial intentions, and (ii) that the traditional account of intentions already has the resources necessary to accommodate dual-plan cases. (shrink)

Sometimes we make a decision about an action we will undertake later and form an intention, but our judgment of what it is best to do undergoes a temporary shift when the time for action comes round. What makes it rational not to give in to temptation? Many contemporary solutions privilege diachronic rationality; in some “rational non-reconsideration” (RNR) accounts once the agent forms an intention, it is rational not to reconsider. This leads to other puzzles: how can someone be (...) motivated to follow a plan that is contrary to their current judgment? How can it be rational to form a plan to resist if we can predict that our judgment will shift? I show how these puzzles can be solved in a framework where there are multiple units of agency, distinguishing between the judgments of the timeslice and those of the person over time, and allowing that the timeslice can “self identify”, taking the person over time as the relevant unit of agency and doing intrapersonal team reasoning (with a different causal role for intentions than RNR accounts). On my account, resisting temptation is compatible with synchronic rationality, so synchronic and diachronic rationality are aligned. However, either resisting or succumbing to temptation can be instrumentally rational, depending on the unit of agency that is identified with. In order to show why we ought to resist temptation, we need to draw on a non-instrumental rationale. I sketch possible routes for doing this. (shrink)

This paper focuses on the Enkratic principle of rationality, according to which rationality requires that if an agent sincerely and with conviction believes she ought to X, then X-ing is a goal in her plan. We analyze the logical structure of Enkrasia and its implications for deontic logic. To do so, we elaborate on the distinction between basic and derived oughts, and provide a multi-modal neighborhood logic with three characteristic operators: a non-normal operator for basic oughts, a non-normal operator (...) for goals in plans, and a normal operator for derived oughts. We prove two completeness theorems for the resulting logic, and provide a dynamic extension of the logic by means of product updates. We illustrate how this setting informs deontic logic by considering issues related to the filtering of inconsistent oughts, the restricted validity of deontic closure, and the stability of oughts and goals under dynamics. (shrink)

The process of cooperative problem solving can be divided into four stages. First, finding potential team members, then forming a team followed by constructing a plan for that team. Finally, the plan is executed by the team. Traditionally, very simple protocols like the Contract Net protocol are used for performing the first two stages of the process. In an open environment however, there can be discussion among the agents in order to form a team that can achieve the collective intention (...) of solving the problem. For these cases fixed protocols like contract net do not suffice. In this paper we present an alternative solution, using structured dialogues, with an emphasis on persuasion, that can be shown to lead to the required team formation. The dialogues are described formally using modal logics and speech acts. (shrink)

This article addresses three main questions: What causes cognitive overload in the workplace? What analytical framework should be used to understand how agents interact with their work environments? How can environments be restructured to improve the cognitive workflow of agents? Four primary causes of overload are identified: too much tasking and interruption, and inadequate workplace infrastructure to help reduce the need for planning, monitoring, reminding, reclassifying information, etc… The first step in reducing the cognitive impact of these causes is (...) to enrich classical frameworks for understanding work environments, such as Newell and Simon’s notion of a task environment, by recognizing that our actual workplace is a superposition of many specific environments – activity spaces – which we slip between. Each has its own cost structure arising from the tools and resources available, including the cognitive strategies and interpretational frameworks of individual agents. These cognitive factors are significant, affecting how easy or difficult it is to perform an action, such as finding a specific paper in a “mess” desk. A few simple examples show how work environments can be redesigned and how restructuring can alter the cost structure of activity spaces. (shrink)