PHILOSOPHICAL STUDIES最新文献

{"title":"The linguistic dead zone of value-aligned agency, natural and artificial","authors":"Travis LaCroix","doi":"10.1007/s11098-024-02257-w","DOIUrl":"https://doi.org/10.1007/s11098-024-02257-w","url":null,"abstract":"<p>The value alignment problem for artificial intelligence (AI) asks how we can ensure that the “values”—i.e., objective functions—of artificial systems are aligned with the values of humanity. In this paper, I argue that linguistic communication is a necessary condition for robust value alignment. I discuss the consequences that the truth of this claim would have for research programmes that attempt to ensure value alignment for AI systems—or, more loftily, those programmes that seek to design robustly beneficial or ethical artificial <i>agents</i>.\u0000</p>","PeriodicalId":48305,"journal":{"name":"PHILOSOPHICAL STUDIES","volume":"14 1","pages":""},"PeriodicalIF":1.3,"publicationDate":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142763082","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"哲学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"How to be a postmodal directionalist","authors":"Scott Dixon","doi":"10.1007/s11098-024-02237-0","DOIUrl":"https://doi.org/10.1007/s11098-024-02237-0","url":null,"abstract":"<p>According to directionalism, non-symmetric relations are distinct from their converses. Kit Fine (2000a) argues that the directionalist faces a dilemma; they must either (i) reject the principle Uniqueness, which states that no completion (fact, state of affairs, or proposition) is a completion of more than one relation, or (ii) reject the principle Identity, which states that each completion of a relation is identical to a completion of its converse (e.g., Dante’s loving Bice is identical to Bice’s being loved by Dante). Fine’s argument has been regarded as a decisive blow to directionalism. But new strategies for replying to it can be developed with the tools of the postmodal metaphysician, who is comfortable individuating relations and their completions hyperintensionally, allowing for necessary connections between distinct entities, and making use of hyperintensional notions like essence and grounding. In what follows, I develop postmodal strategies for denying both horns of Fine’s dilemma, concluding that the postmodal directionalist need not be concerned with Fine’s argument.</p>","PeriodicalId":48305,"journal":{"name":"PHILOSOPHICAL STUDIES","volume":"261 1","pages":""},"PeriodicalIF":1.3,"publicationDate":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142763349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"哲学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Closure and the structure of justification","authors":"Christoph Kelp, Matthew Jope","doi":"10.1007/s11098-024-02245-0","DOIUrl":"https://doi.org/10.1007/s11098-024-02245-0","url":null,"abstract":"<p>This paper considers two recent views on the structure of justification and closure of knowledge by Ernest Sosa. It provides reason to believe that neither view is ultimately viable and sketches a better alternative.</p>","PeriodicalId":48305,"journal":{"name":"PHILOSOPHICAL STUDIES","volume":"116 1","pages":""},"PeriodicalIF":1.3,"publicationDate":"2024-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142758198","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"哲学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A simpler model of judgment: on Sosa’s Epistemic Explanations","authors":"Antonia Peacocke","doi":"10.1007/s11098-024-02232-5","DOIUrl":"https://doi.org/10.1007/s11098-024-02232-5","url":null,"abstract":"<p>In <i>Epistemic Explanations</i>, Sosa continues to defend a model of judgment he has long endorsed. On this complex model of judgment, judgment aims not only at correctness but also at aptness of a kind of alethic affirmation. He offers three arguments for the claim that we need this model of judgment instead of a simpler model, on which judgment aims only at correctness. The first argument cites the need to exclude knowledge-spoiling luck from apt judgment. The second argument uses the complex model to distinguish judgment from mere guessing. The third argument involves the assessment of suspension of judgment as a performance. This paper shows why none of these arguments succeeds, and so recommends adopting the simpler model of judgment.</p>","PeriodicalId":48305,"journal":{"name":"PHILOSOPHICAL STUDIES","volume":"260 1","pages":""},"PeriodicalIF":1.3,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142753627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"哲学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Silence as complicity and action as silence","authors":"J. L. A. Donohue","doi":"10.1007/s11098-024-02246-z","DOIUrl":"https://doi.org/10.1007/s11098-024-02246-z","url":null,"abstract":"<p>Silence sometimes constitutes moral complicity. We see this when protestors take to the streets against racial injustice. Think of signs with the words: “Silence is complicity.” We see this in instances of sexual harassment, when we learn that many knew and said nothing. We see this in cases of wrongdoing within a company or organization, when it becomes clear that many were aware of the negligent or criminal activity and stayed silent. In cases like this we consider agents morally complicit in virtue of their silence. Flagrant injustices cry out for action, and sometimes remaining silent amounts to complicity in those injustices. What philosophy owes us is an account of how it could be that silence constitutes complicity. In this paper I argue that one possibility is an account grounded in problematic deliberative contribution. The core idea of “deliberative complicity,” as I call it, is that agents have moral duties concerning the moral deliberation of other agents, and failures in these duties can amount to moral complicity. For example, an agent aware that a colleague is sexually harassing his students has a deliberative obligation to report the misconduct, and their silence in failing to report constitutes a failure to fulfill their deliberative obligation, a failure that grounds their moral complicity in the harassment. If my argument is successful, it provides a distinctive reason to prefer a deliberative account of moral complicity: it can capture cases of silent complicity that other views of moral complicity cannot. And further, by turning our attention toward our interpersonal deliberative obligations, a deliberative account of complicity can incorporate helpful resources from recent work in social epistemology and speech act theory as we set out to determine when and why silence amounts to complicity. And when it does, we cannot stay silent. We must speak.</p>","PeriodicalId":48305,"journal":{"name":"PHILOSOPHICAL STUDIES","volume":"26 1","pages":""},"PeriodicalIF":1.3,"publicationDate":"2024-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142697093","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"哲学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Responses to Speaks, Stojnić and Szabó","authors":"Jeffrey C. King","doi":"10.1007/s11098-024-02219-2","DOIUrl":"https://doi.org/10.1007/s11098-024-02219-2","url":null,"abstract":"&lt;p&gt;Consider the class of contextually sensitive expressions whose context invariant meanings arguably do not suffice to secure semantic values in context. Demonstratives and demonstrative pronouns are the examples of such expressions that have received the most attention from philosophers. However, arguably this class of contextually sensitive expressions includes among other expressions modals, conditionals, tense, gradable adjectives, possessives, ‘only’, quantifiers, and expressions that take implicit arguments (e.g. ‘ready’ in sentences like ‘Molly is ready.’). Most theorists, including me, think that since the context invariant meanings of such expressions do not by themselves secure semantic values in context for these expressions, they must be supplemented in some way in context in order to secure semantic values in context. For this reason, I call these expressions &lt;i&gt;supplementives&lt;/i&gt;. I just said that supplementives need some sort of supplementation to secure semantic values in context. Of course, the question of what &lt;i&gt;form&lt;/i&gt; the supplementation in context takes is controversial. For example, ever since Kaplan claimed that the semantic value of a demonstrative or demonstrative pronoun in context is the &lt;i&gt;demonstratum&lt;/i&gt; of its associated &lt;i&gt;demonstration&lt;/i&gt;, there has been a lively controversy over whether that or some other account is the correct one. Call an account of how a given supplementive secures a semantic value in context a &lt;i&gt;metasemantics&lt;/i&gt; for the supplementive. In King [2018] I argue that all supplementives have felicitous uses in which they haven’t been assigned unique semantic values in context. This conclusion is somewhat surprising, since many uses of supplementives in which they have not been assigned unique semantic values in context are quite infelicitous. I call felicitous uses of supplementives in which they haven’t been assigned unique semantic values in context instances of &lt;i&gt;felicitous underspecification&lt;/i&gt;. The central idea is that in cases of felicitous underspecification, supplementives get assigned a &lt;i&gt;range of candidates for being their semantic values in contexts&lt;/i&gt; rather than being assigned unique semantic values in contexts. Consider an example. Glenn and I are out surfing at Lost Winds beach. There are some surfers to our south stretching a quarter mile or so down the beach. I notice that some surfers in an ill-defined group to our immediate south are getting incredible rides. I say to Glenn looking south toward them ‘Those guys are good.’ It seems easy to imagine that nothing in the context of utterance determines a unique group of surfers as the semantic value in context of ‘Those guys’. For example, it is easy to imagine that I didn’t intend any &lt;i&gt;specific, unique&lt;/i&gt; group to be the semantic value in context. Instead, there is a range of overlapping groups that are legitimate candidates for being the semantic value in context of ‘Those guys’. Nonetheless, my utterance is felicitous:","PeriodicalId":48305,"journal":{"name":"PHILOSOPHICAL STUDIES","volume":"41 1","pages":""},"PeriodicalIF":1.3,"publicationDate":"2024-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142678464","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"哲学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"In defense of virtual veridicalism","authors":"Yen-Tung Lee","doi":"10.1007/s11098-024-02256-x","DOIUrl":"https://doi.org/10.1007/s11098-024-02256-x","url":null,"abstract":"<p>This paper defends <i>virtual veridicalism</i>, according to which many perceptual experiences in virtual reality are veridical. My argument centers on perceptual variation, the phenomenon in which perceptual experience appears all the same while being reliably generated by different properties under different circumstances. It consists of three stages. The first stage argues that perceptual variation can occur in color perception without involving misperception. The second stage extends the argument to perceptual variation of space, arguing that it is possible for individuals to perceive distinct physical spaces as having the same experiential space without suffering from systematic misperception. The final stage proceeds to argue that perceptual variation without misperception in color and spatial perception can occur across virtual and ordinary environments. In that sense, given that ordinary experiences are presumably veridical, experiences in virtual reality are also veridical.</p>","PeriodicalId":48305,"journal":{"name":"PHILOSOPHICAL STUDIES","volume":"8 1","pages":""},"PeriodicalIF":1.3,"publicationDate":"2024-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142673893","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"哲学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"What are we to do? Making sense of ‘joint ought’ talk","authors":"Rowan Mellor, Margaret Shea","doi":"10.1007/s11098-024-02222-7","DOIUrl":"https://doi.org/10.1007/s11098-024-02222-7","url":null,"abstract":"<p>We argue for three main claims. First, the sentence ‘A and B ought to φ and ψ’ can express what we a call a joint-ought claim: the claim that the plurality A and B ought to φ and ψ respectively. Second, the truth-value of this joint-ought claim can differ from the truth-value of the pair of claims ‘A ought to φ’ and ‘B ought to ψ.’ This is because what A and B jointly ought to do can diverge from what they individually ought to do: it may be true that A and B jointly ought to φ and ψ respectively, yet false that A ought to φ and false that B ought to ψ; and vice-versa. Third, either of two prominent semantic analyses of ‘ought’—Mark Schroeder’s relational semantics, and Angelika Kratzer’s modal semantics—can model joint-ought claims and this difference in truth-value.</p>","PeriodicalId":48305,"journal":{"name":"PHILOSOPHICAL STUDIES","volume":"2 1","pages":""},"PeriodicalIF":1.3,"publicationDate":"2024-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142670267","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"哲学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Withhold by default: a difference between epistemic and practical rationality","authors":"Chris Tucker","doi":"10.1007/s11098-024-02233-4","DOIUrl":"https://doi.org/10.1007/s11098-024-02233-4","url":null,"abstract":"<p>In practical rationality, if two reasons for alternative actions are tied, then either action is *permissible*. In epistemic rationality, we get the Epistemic Ties Datum: if the reasons for belief and disbelief are tied, then withholding judgment is *required*. I argue that this difference is explained by a difference in default biases. Practical rationality is biased toward permissibility. An action is permissible unless the specific features of the situation (e.g., the costs and benefits that apply) make it prohibited. In contrast, epistemic rationality is biased toward withholding judgment. Withholding judgment is required unless the specific features of the situation (e.g., the evidence) makes belief or disbelief permissible. This difference explains the Epistemic Ties Datum. When the reasons for belief and disbelief are equally weighty, they cancel each other out. But then the only remaining reason is the default reason, or default bias, to withhold judgment. Since it is the only remaining reason, it requires us to withhold judgment.</p>","PeriodicalId":48305,"journal":{"name":"PHILOSOPHICAL STUDIES","volume":"128 1","pages":""},"PeriodicalIF":1.3,"publicationDate":"2024-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142670273","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"哲学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Disagreement, AI alignment, and bargaining","authors":"Harry R. Lloyd","doi":"10.1007/s11098-024-02224-5","DOIUrl":"https://doi.org/10.1007/s11098-024-02224-5","url":null,"abstract":"<p>New AI technologies have the potential to cause unintended harms in diverse domains including warfare, judicial sentencing, medicine and governance. One strategy for realising the benefits of AI whilst avoiding its potential dangers is to ensure that new AIs are properly ‘aligned’ with some form of ‘alignment target.’ One danger of this strategy is that–dependent on the alignment target chosen–our AIs might optimise for objectives that reflect the values only of a certain subset of society, and that do not take into account alternative views about what constitutes desirable and safe behaviour for AI agents. In response to this problem, several AI ethicists have suggested alignment targets that are designed to be sensitive to widespread normative disagreement amongst the relevant stakeholders. Authors inspired by voting theory have suggested that AIs should be aligned with the verdicts of actual or simulated ‘<i>moral parliaments</i>’ whose members represent the normative views of the relevant stakeholders. Other authors inspired by decision theory and the philosophical literature on moral uncertainty have suggested that AIs should maximise socially expected choiceworthiness. In this paper, I argue that both of these proposals face several important problems. In particular, they fail to select attractive ‘compromise options’ in cases where such options are available. I go on to propose and defend an alternative, bargaining-theoretic alignment target, which avoids the problems associated with the voting- and decision-theoretic approaches.</p>","PeriodicalId":48305,"journal":{"name":"PHILOSOPHICAL STUDIES","volume":"76 1","pages":""},"PeriodicalIF":1.3,"publicationDate":"2024-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142670269","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"哲学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}