A shared narrative in the literature on the evolution of cooperation maintains that social _learning_ evolves early to allow for the transmission of cumulative culture. Social _norms_, whilst present at the outset, only rise to prominence later on, mainly to stabilise cooperation against the threat of defection. In contrast, I argue that once we consider insights from social epistemology, an expansion of this narrative presents itself: An interesting kind of social norm — an epistemic coordination norm — was operative in (...) early and important instances of specialised social learning. I show how there’s a need for such norms in two key social learning strategies and explain how this need is constituted. In assessor-teaching (e.g. Castro et al., 2019b, 2021), epistemic coordination norms allow agents to coordinate around the _content_ of social learning, i.e., what is to be known and how this is to be done. These norms also allow agents to coordinate around the form of cultural learning in what’s sometimes called strategic social learning (Laland, 2004; Hoppitt & Laland, 2013; Heyes, 2018, Chap. 5) and elsewhere. Broadly speaking, this concerns how cultural learning is organised within the social group. The upshot is that the evolution of social learning and social norms are intertwined in important and underappreciated ways from early on. The above matters as it informs our views about the evolution of social norms more generally. Truly _social_ norms emerged to coordinate a plurality of complex behaviours and interactions, amongst them specialised social learning. I substantiate this view by contrasting it with Jonathan Birch’s views on the evolution of norms. What results is a general but cohesive narrative on the early evolution of social norms. (shrink)

Automated Influence is the use of Artificial Intelligence to collect, integrate, and analyse people’s data in order to deliver targeted interventions that shape their behaviour. We consider three central objections against Automated Influence, focusing on privacy, exploitation, and manipulation, showing in each case how a structural version of that objection has more purchase than its interactional counterpart. By rejecting the interactional focus of “AI Ethics” in favour of a more structural, political philosophy of AI, we show that the real problem (...) with Automated Influence is the crisis of legitimacy that it precipitates. (shrink)

Amy saves a man from drowning despite the risk to herself, because she is moved by his plight. This is a quintessentially supererogatory act: an act that goes above and beyond the call of duty. Beth, on the other hand, saves a man from drowning because she wants to get her name in the paper. On this second example, opinions differ. One view of supererogation holds that, despite being optional and good, Beth’s act is not supererogatory because she is not (...) praiseworthy; the other agrees that Beth is not praiseworthy but holds that her act is nevertheless supererogatory because, while supererogatory actions are generally performed by praiseworthy agents, an individual need not be praiseworthy for their act to be supererogatory. In this paper, I raise a problem for this latter position, which I shall call the Anti-Motivation View of supererogation. While the Anti-Motivation View rejects the claim that the agent’s motivations are important, it accepts that the agent’s intentions are. Thus, this view assumes that a clear, coherent distinction can be drawn between intentions and motivations in the context of supererogation. This distinction is often attributed to Mill; however, his original discussion reveals an inconsistency. Consider Clara, who saves a man from drowning because she wants to torture him later. According to Mill, Clara’s act is not morally good; those who follow Mill will have to accept that it is therefore not supererogatory. Mill asserts that Clara’s act differs from Amy’s and Beth’s not just in motivation but also in intention. I question this explanation and demonstrate that the distinction between motivation and intention is not as clear as the Anti-Motivation View has supposed and that the gap between the Anti-Motivation View and the alternative Praise-Based View is either much greater or much smaller than previously thought. (shrink)

A familiar part of debates about supererogatory actions concerns the role that cost should play. Two camps have emerged: one claiming that extreme cost is a necessary condition for when an action is supererogatory, while the other denies that it should be part of our definition of supererogation. In this paper, I propose an alternative position. I argue that it is comparative cost that is central to the supererogatory and that it is needed to explain a feature that all accounts (...) agree is central to the very notion of supererogation: optionality. Perhaps because of this agreement on its importance, few attempts have been made to clarify and explain the notion of optionality. I argue that giving an account of the optionality of supererogatory requires drawing a line between doing the bare minimum permissible and going beyond the bare minimum and that this line ought to be drawn based on comparative cost of alternative permissible acts. Having outlined my account and motivated it, I discuss and reject two concerns that might be raised: firstly, that it is extreme cost, not comparative cost, that matters and, secondly, that in fact no cost is needed for an act to be supererogatory. (shrink)

Animal ethics has often been criticized for an overreliance on “ideal” or even “utopian” theorizing. In this article, I recognize this problem, but argue that the “nonideal theory” which critics have offered in response is still insufficient to make animal ethics action-guiding. I argue that in order for animal ethics to be action-guiding, it must consider agent-centered theories of change detailing how an ideally just human-animal coexistence can and should be brought about. I lay out desiderata that such a theory (...) of change should suffice so as to be helpful in guiding action. Specifically, a theory of change should determine (1) who needs to do what in order for ideal justice to be achieved in the long run, (2) who should be expected to refuse compliance and how they should be moved to comply, and (3) why specific intermediate steps are necessary. I show how previous “nonideal” contributions, though helpful in other ways, are insufficiently determinate on these points and I sketch a (still somewhat utopian) theory of change for one specific context. This brings animal ethics a crucial step closer to being action-guiding in the real world. (shrink)

De Re Significance accounts of moral appraisal consider an agent’s responsiveness to a particular kind of reason, normative moral reasons de re, to be of central significance for moral appraisal. Here, I argue that such accounts find it difficult to accommodate some neuroatypical agents. I offer an alternative account of how an agent’s responsiveness to normative moral reasons affects moral appraisal – the Reasonable Expectations Account. According to this account, what is significant for appraisal is not the content of the (...) reasons an agent is responsive to, but rather whether she is responsive to the reasons it is reasonable to expect her to be responsive to, irrespective of their content. I argue that this account does a better job of dealing with neuroatypical agents, while agreeing with the De Re Significance accounts on more ordinary cases. (shrink)

Is logic normative for belief? A standard approach to answering this question has been to investigate bridge principles relating claims of logical consequence to norms for belief. Although the question is naturally an epistemic one, bridge principles have typically been investigated in isolation from epistemic debates over the correct norms for belief. In this paper we tackle the question of whether logic is normative for belief by proposing a Kripkean model theory accounting for the interaction between logical, doxastic, epistemic and (...) deontic notions and using this model theory to show which bridge principles are implied by epistemic norms that we have independent reason to accept, for example, theknowledge normand thetruth norm. We propose a preliminary theory of the interaction between logical, doxastic, epistemic and deontic notions that has among its commitments bridge principles expressing how logic is normative for belief. We also show how our framework suggests that logic is exceptionally normative. (shrink)

The paper addresses the following question: why do human beings, on Aristotle’s view, have an innate tendency to badness, that is, to developing desires that go beyond, and often against, their natural needs? Given Aristotle’s teleological assumptions (including the thesis that nature does nothing in vain), such tendency should not be present. I argue that the culprit is to be found in the workings of rationality. In particular, it is the presence of theoretical reason that necessitates the limitless nature of (...) human non-rational desires. (shrink)

This open access book revises Kant’s ethical thought in one of its most notorious respects: its exclusion of animals from moral consideration. The book gives readers in animal ethics an accessible introduction to Kant’s views on our duties to others, and his view that we have only ‘indirect’ duties regarding animals. It then investigates how one would have to depart from Kant in order to recognise that animals matter morally for their own sake. Particular attention is paid to Kant’s ‘Formula (...) of Humanity,' the role of autonomy and the moral law, as well as Kant’s notions of practical reason and animal instinct. The result is a deliberately amended version of Kantianism which nevertheless remains faithful to central aspects of Kant’s thought. The book’s final part illustrates the framework’s use in applied contexts, addressing the issues of using animals as mere means, the ethics of veganism and vegetarianism, and environmental protection. Nico Dario Müller shows how, when furnished with duties to animals, Kant's moral philosophy can be a powerful resource for animal ethicists. (shrink)

Cumulative cultural knowledge [CCK], the knowledge we acquire via social learning and has been refined by previous generations, is of central importance to our species’ flourishing. Considering its importance, we should expect that our best epistemological theories can account for how this happens. Perhaps surprisingly, CCK and how we acquire it via cultural learning has only received little attention from social epistemologists. Here, I focus on how we should epistemically evaluate how agents acquire CCK. After sampling some reasons why extant (...) theories cannot account for CCK, I suggest that things aren’t as bleak as they might look. I explain how agents deserve epistemic credit for how CCK is transmitted in cultural learning by promoting a central need of their social group: The efficient and safe transmission of CCK. A good initial fit exists between this observation and Greco’s knowledge-economy framework. Ultimately, however, Greco’s account doesn’t straightforwardly account for CCK because of its strict focus on testimony. I point out two issues in the framework due to this focus. The resulting view advocates giving epistemic credit to agents when they act to promote their communities’ epistemic needs in the right way and highlights the various ways in which agents come to do this. (shrink)

The Enkratic Principle enjoys something of a protected status as a requirement of rationality. I argue that this status is undeserved, at least in the epistemic domain. Compliance with the principle should not be thought of as a requirement of epistemic rationality, but rather as defeasible indication of epistemic blamelessness. To show this, I present the Puzzle of Inconsistent Requirements, and argue that the best way to solve it is to distinguish two kinds of epistemic evaluation – requirement evaluations and (...) appraisal evaluations. This allows us to solve the puzzle while accommodating traditional motivations for thinking of the Enkratic Principle as a requirement of rationality. (shrink)

Standing in San Marco Cathedral in Venice, you immediately notice the exquisitely decorated spandrels: the triangular spaces bounded on either side by adjoining arches and by the dome above. You would be forgiven for seeing them as the starting point from which to understand the surrounding architecture. To do so would, however, be a mistake. It is a similar mistaken inference that evolutionary biologists have been accused of making in assuming a special adaptive purpose for such biological features as fingerprints (...) and chins. I argue that a mistake of just this sort is being made by ethicists who appeal to the intrinsic value of supererogatory acts in their efforts to make space for supererogation in ethical theory. Many cases of supererogatory action are simply spandrels: by-products of uncontroversial commitments elsewhere in our moral thought. This is not to downplay their value but rather to show that their value need not be the justification for making room for the supererogatory. I demonstrate this by examining two areas: rights and the distribution of burdens among a group. My argument has significance for those who take themselves to be defends of the possibility of supererogatory actions, as well as those who are committed to the contrary and those who believe themselves to be indifferent on the matter. (shrink)

Can we make mistakes about what rationality requires? A natural answer is that we can, since it is a platitude that rational belief does not require truth; it is possible for a belief to be rational and mistaken, and this holds for any subject matter at all. However, the platitude causes trouble when applied to rationality itself. The possibility of rational mistakes about what rationality requires generates a puzzle. When combined with two further plausible claims – the enkratic principle, and (...) the claim that rational requirements apply universally – we get the result that rationality generates inconsistent requirements. One popular and attractive solution to the puzzle denies that it is possible to make rational mistakes about what rationality requires. I show why (contra Titelbaum (2015b), and Littlejohn (2015)) this solution is doomed to fail. (shrink)

I argue that there are some situations in which it is praiseworthy to be motivated only by moral rightness de dicto, even if this results in wrongdoing. I consider a set of cases that are challenging for views that dispute this, prioritising concern for what is morally important in moral evaluation. In these cases, the agent is not concerned about what is morally important, does the wrong thing, but nevertheless seems praiseworthy rather than blameworthy. I argue that the views under (...) discussion cannot accommodate this, and should be amended to recognise that it is often praiseworthy to be motivated to do what is right. (shrink)

There has been some debate as to whether or not it is possible to keep a promise, and thus fulfil a duty, to supererogate. In this paper, I argue, in agreement with Jason Kawall, that such promises cannot be kept. However, I disagree with Kawall’s diagnosis of the problem and provide an alternative account. In the first section, I examine the debate between Kawall and David Heyd, who rejects Kawall’s claim that promises to supererogate cannot be kept. I disagree with (...) Heyd’s argument, as it fails to get to the heart of the problem Kawall articulates. Kawall’s argument however fails to make clear the problem with promising to supererogate because his discussion relies on the plausibility of the following claim: that supererogatory actions cannot also fulfil obligations. I argue that this view is mistaken because there are clear examples of supererogatory actions that also fulfil obligations. In the final section, I give my alternative account of the problem, identifying exactly what is wrong with fulfilling a duty, and thus keeping a promise, to supererogate. My diagnosis emphasises the importance of identifying non-supererogatory actions when it comes to understanding the way in which supererogatory actions go above and beyond the call of duty. (shrink)

Is normative uncertainty like factual uncertainty? Should it have the same effects on our actions? Some have thought not. Those who defend an asymmetry between normative and factual uncertainty typically do so as part of the claim that our moral beliefs in general are irrelevant to both the moral value and the moral worth of our actions. Here I use the consideration of Jackson cases to challenge this view, arguing that we can explain away the apparent asymmetries between normative and (...) factual uncertainty by considering the particular features of the cases in greater detail. Such consideration shows that, in fact, normative and factual uncertainty are equally relevant to moral assessment. (shrink)

Incoherence is usually regarded as a bad thing. Incoherence suggests irrationality, confusion, paradox. Incoherentism disagrees: incoherence is not always a bad thing, sometimes we ought to be incoherent. If correct, Incoherentism has important and controversial implications. It implies that rationality does not always require coherence. Dilemmism and Incoherentism both embrace conflict in epistemology. After identifying some important differences between these two ways of embracing conflict, I offer some reasons to prefer Incoherentism over Dilemmism. Namely, that Incoherentism allows us to deliberate (...) about what we ought to believe using ordinary epistemology, and it does a better job of accommodating the positive features of incoherence. (shrink)

I argue for the unexceptionality of evidence about what rationality requires. Specifically, I argue that, as for other topics, one’s total evidence can sometimes support false beliefs about this. Despite being prima facie innocuous, a number of philosophers have recently denied this. Some have argued that the facts about what rationality requires are highly dependent on the agent’s situation and change depending on what that situation is like. (Bradley 2019). Others have argued that a particular subset of normative truths, those (...) concerning what epistemic rationality requires, have the special property of being ‘fixed points’—it is impossible to have total evidence that supports false belief about them. (Smithies 2012; Titelbaum 2015). Each of these kinds of exceptionality permits a solution to downstream theoretical problems that arise from the possibility of evidence supporting false belief about requirements of rationality. However, as I argue here, they incur heavy explanatory burdens that we should avoid. (shrink)

It is sometimes alleged that the study of emotion and the study of value are currently pursued as relatively autonomous disciplines. As Kevin Mulligan notes, “the philosophy and psychology of emotions pays little attention to the philosophy of value and the latter pays only a little more attention to the former.” (2010b, 475). Arguably, the last decade has seen more of a rapprochement between these two domains than used to be the norm (cf. e.g. Roeser & Todd 2014). But there (...) still seems to be considerable potential for exchange and dialogue if the situation is compared with their intimate relationship in central strands of early realist phenomenology. The philosopher perhaps most representative of this ecumenical approach is Husserl’s early student Dietrich von Hildebrand (1889-1977). From the very early stages of his philosophical career, Hildebrand has developed one of the most original, comprehensive and nuanced accounts of emotions at whose core is a detailed examination of their connection to value. While his central concern with the ethical significance of our affective life is in many ways continuous with Scheler’s work and draws crucially on Reinach’s philosophy of mind, Hildebrand’s own reflections considerably expand on and substantially modify the picture of the ontology and normative role of emotions defended by these authors. In this article, I reconstruct Hildebrand’s view of emotions with a particular focus on those aspects which represent his most distinctive contribution to this subject. (shrink)

In this paper, I argue that the widely held view that Aristotle's vicious agent is a principled follower of a wrong conception of the good whose soul, just like the soul of the virtuous agent, is marked by harmony between his reason and non-rational desires is an exegetical mistake. Rather, Aristotle holds – consistently and throughout the Nicomachean Ethics – that the vicious agent lacks any real principles of action and that his soul lacks unity and harmony even more than (...) the soul of the uncontrolled agent. (shrink)

Goethe carried out an enormous number of experiments before criticizing Newton's theory of light and colours in the Farbenlehre (1810). He managed to show that Newton's reasoning is based on a rather narrow choice of experiments, in which parameters such as the distance between the prism and the screen are fixed arbitrarily: Newton's famous spectrum (with its green centre) occurs only at a specific distance. Once you reduce the distance, the green centre disappears, and you see the two border spectra (...) instead. Newton can of course explain these border spectra; but things can be explained just as well the other way around: Then you start with a theory based on the two border spectra in order to derive the Newtonian spectrum. Both accounts fit many data (available in Newton and Goethe's days) equally well. Thus by insisting that the step to theory is not forced upon us by "reason and experiments" alone (as Newton would have it), Goethe revealed our own free, creative contribution to theory construction. And surprisingly, his theory of the prismatic colours seems no worse off than Newton's, even taking into account additional criteria of theory choice, such as simplicity or elegance. (shrink)

Moral theories that demand that we do what is morally best leave no room for the supererogatory. One argument against such theories is that they fail to realize the value of autonomy: supererogatory acts allow for the exercise of autonomy because their omissions are not accompanied by any threats of sanctions, unlike obligatory ones. While this argument fails, I use the distinction it draws – between omissions of obligatory and supererogatory acts in terms of appropriate sanctions – to draw a (...) parallel with psychological perfectionism. Through this parallel, I demonstrate that requiring what is morally best is in fact counter-productive. Thus, by its own lights, a theory that wants us to do what is best ought at the very least to tell us to believe that some actions are supererogatory. As the old adage goes, the best is the enemy of the good; I argue here that the supererogatory is the solution. (shrink)

I argue that on Aristotle’s account practical thinking is thinking whose origin (archē) is a desire that has as its object the very thing that one reasons about how to promote. This feature distinguishes practical from productive reasoning since in the latter the desire that initiates it is not (unless incidentally) a desire for the object that one productively reasons about. The feature has several interesting consequences: (a) there is only a contingent relationship between the desire that one practically reasons (...) about how to satisfy and the action one decides on; (b) practical thinking and action cannot be separated from the agent, whereas productive thinking and production can be outsourced to someone else. The view has consequences also for the distinction between action and production. Finally, I illustrate the usefulness and correctness of my account of practical thinking by using it to shed new light on Aristotle’s claim that the virtuous agent must decide on her virtuous actions ‘for themselves’. (shrink)

The theory and philosophy of artificial intelligence has come to a crucial point where the agenda for the forthcoming years is in the air. This special volume of Minds and Machines presents leading invited papers from a conference on the “Philosophy and Theory of Artificial Intelligence” that was held in October 2011 in Thessaloniki. Artificial Intelligence is perhaps unique among engineering subjects in that it has raised very basic questions about the nature of computing, perception, reasoning, learning, language, action, interaction, (...) consciousness, humankind, life etc. etc. – and at the same time it has contributed substantially to answering these questions. There is thus a substantial tradition of work, both on AI by philosophers and of theory within AI itself. - The volume contains papers by Bostrom, Dreyfus, Gomila, O'Regan and Shagrir. (shrink)

Die Entwicklungen in der Künstlichen Intelligenz (KI) sind spannend. Aber wohin geht die Reise? Ich stelle eine Analyse vor, der zufolge exponentielles Wachstum von Rechengeschwindigkeit und Daten die entscheidenden Faktoren im bisherigen Fortschritt waren. Im Folgenden erläutere ich, unter welchen Annahmen dieses Wachstum auch weiterhin Fortschritt ermöglichen wird: 1) Intelligenz ist eindimensional und messbar, 2) Kognitionswissenschaft wird für KI nicht benötigt, 3) Berechnung (computation) ist hinreichend für Kognition, 4) Gegenwärtige Techniken und Architektur sind ausreichend skalierbar, 5) Technological Readiness Levels (TRL) (...) erweisen sich als machbar. Diese Annahmen werden sich als dubios erweisen. (shrink)

Crispin Wright hat die bislang beste Rekonstruktion von Putnams Beweis gegen die skeptische Hypothese vom Gehirn im Tank vorgelegt. Aber selbst in Wrights Fassung hat der Beweis einen Mangel: Er wird mithilfe eines Prädikates wie z.B. "Tiger" geführt und funktioniert nur, wenn man sich darauf verlassen kann, dass es Tiger wirklich gibt. Aber die Skeptikerin bestreitet, über die Existenz von Tigern bescheid zu wissen. Das Problem lässt sich dadurch beheben, dass man den Beweis – statt mit dem extensionalen Begriff der (...) Referenz (wie bei Wright) – mit Intensionen führt. Zudem sollte der Beweis transzendental laufen, also einzig und allein auf den sprachlichen Voraussetzungen beruhen, die auch die Skeptikerin zur Formulierung ihrer Tank-Hypothese benötigt. -/- The best reconstruction of Putnam's brain-in-a-vat argument is due to Crispin Wright. But even Wright's version of the argument is not convincing. It employs predicates such as 'tiger' and reaches the conclusion only if we can rely upon the existence of tigers; and the skeptic disputes our knowledge regarding tigers. The problem is not insuperable: The argument must be run, not with the extensionalistic notion of reference (as in Wright's version), but with intensions. Furthermore, the argument should have a transcendental form; it should exclusively rely on conditions that are needed by the skeptic for formulating her own position. (shrink)

Quine's early arguments in favor of "Ontological Relativity" depend on wild but interesting alternatives to standard ontology,most apparent when viewing his unusual idea of undetached object parts. In Quine's later philosophy, by contrast, he invokes trivial proxy functions and simple permutations to standard ontology, and in so doing paves the way for his claim of "Inscrutability of Reference". Nevertheless, Quine's more recent alternatives to standard ontology have thus far remained uninteresting for ontological questions, as his later arguments fail to offer (...) any real alternative to our standard identity relation. For this reason, I have chosen to return to Quine's original idea of "Ontological Relativity" and put it to the test. In my investigation, development and subsequent defense of Quine's original presentation, I attempt to show that it is in fact possible to read Quine's ontology of undetached object parts into a large sphere of our language. (shrink)

By drawing attention to these facts and to the relationship between Cantor’s and Husserl's ideas, I have tried to contribute to putting Frege's attack on Husserl "in the proper light" by providing some insight into some of the issues underling criticisms which Frege himself suggested were not purely aimed at Husserl's book. I have tried to undermine the popular idea that Frege's review of the Philosophy of Arithmetic is a straightforward, objective assessment of Husserl’s book, and to give some specific (...) reasons for thinking that the uncritical reading of Frege's review has unfairly distorted philosophers' perception of a work they do not know very well. (shrink)

The «extended mind thesis» sounds like a substantive thesis, the truth of which we should investigate. But actually the thesis a) turns about to be just a statement on where the demarcations for the «mental» are to be set (internal, external,…), i.e. it is about the «mark of the mental»; and b) the choice about the mark of the mental is a verbal choice, not a matter of scientific discovery. So, the «extended mind thesis » is a remark on how (...) its supporters or opponents want to use the word ‘mind’, not a thesis of cognitive science or philosophy. The upshot of the extended mind discussion should not be to draw the line further out, but to drop the demarcation project. (shrink)

The theory that all processes in the universe are computational is attractive in its promise to provide an understandable theory of everything. I want to suggest here that this pancomputationalism is not sufficiently clear on which problem it is trying to solve, and how. I propose two interpretations of pancomputationalism as a theory: I) the world is a computer and II) the world can be described as a computer. The first implies a thesis of supervenience of the physical over computation (...) and is thus reduced ad absurdum. The second is underdetermined by the world, and thus equally unsuccessful as theory. Finally, I suggest that pancomputationalism as metaphor can be useful. – At the Paderborn workshop in 2008, this paper was presented as a commentary to the relevant paper by Gordana Dodig-Crnkovic " Info-Computationalism and Philosophical Aspects of Research in Information Sciences". (shrink)

Special Issue “Risks of artificial general intelligence”, Journal of Experimental and Theoretical Artificial Intelligence, 26/3 (2014), ed. Vincent C. Müller. http://www.tandfonline.com/toc/teta20/26/3# - Risks of general artificial intelligence, Vincent C. Müller, pages 297-301 - Autonomous technology and the greater human good - Steve Omohundro - pages 303-315 - - - The errors, insights and lessons of famous AI predictions – and what they mean for the future - Stuart Armstrong, Kaj Sotala & Seán S. Ó hÉigeartaigh - pages 317-342 - - (...) - The path to more general artificial intelligence - Ted Goertzel - pages 343-354 - - - Limitations and risks of machine ethics - Miles Brundage - pages 355-372 - - - Utility function security in artificially intelligent agents - Roman V. Yampolskiy - pages 373-389 - - - GOLEM: towards an AGI meta-architecture enabling both goal preservation and radical self-improvement - Ben Goertzel - pages 391-403 - - - Universal empathy and ethical bias for artificial general intelligence - Alexey Potapov & Sergey Rodionov - pages 405-416 - - - Bounding the impact of AGI - András Kornai - pages 417-438 - - - Ethics of brain emulations - Anders Sandberg - pages 439-457. (shrink)

While the consequences of mathematically-based software, algorithms and strategies have become ever wider and better appreciated, ethical reflection on mathematics has remained primitive. We review the somewhat disconnected suggestions of commentators in recent decades with a view to piecing together a coherent approach to ethics in mathematics. Calls for a Hippocratic Oath for mathematicians are examined and it is concluded that while lessons can be learned from the medical profession, the relation of mathematicians to those affected by their work is (...) significantly different. There is something to be learned also from the codes of conduct of cognate but professionalised quantitative disciplines such as engineering and accountancy, as well as from legal principles bearing on professional work. We conclude with recommendations that professional societies in mathematics should sponsor an (international) code of ethics, institutional mission statements for mathematicians and syllabuses of ethics courses for incorporation into mathematics degrees. (shrink)

In der Literatur zur Wirkungsgeschichte der Farbenlehre Goethes aus dem Jahr 1810 grassieren zwei Vorurteile: (1) Nur ein einziger Physiker von Rang (Seebeck) habe sich auf Goethes Projekt wissenschaftlich eingelassen. (2) Schon zu Goethes Lebzeiten habe sich die Fachwissenschaft mit überwältigender Mehrheit gegen den Dichter ausgesprochen. Beide Behauptungen sind falsch. ad (1): Der bedeutende Physiker und Chemiker Johann Ritter hat zwischen 1800 und 1801 eng mit Goethe kooperiert, dieselbe Forschungsmethode eingesetzt wie Goethe und aufgrund dieser Kooperation das UV-Licht entdeckt. Bis (...) zu seinem Lebensende war er der Ansicht, dass die newtonische Theorie durch Fortführung der Experimente Goethes unterminiert werden könne. ad (2): Wie sich bei einer Kampfabstimmung unter den MINT-Forschern der Goethe-Zeit ergibt (bei der sämtliche veröffentlichte Voten von 1810 bis 1832 gezählt werden), stimmten gegen Goethe die Hälfte der Naturwissenschaftler, für ihn ein Drittel, bei 15 Prozent Stimmenthaltungen. Goethe hat die Abstimmung in der Tat verloren, doch eine verheerende Niederlage sähe anders aus. (shrink)

Artificial intelligence (AI) and robotics are digital technologies that will have significant impact on the development of humanity in the near future. They have raised fundamental questions about what we should do with these systems, what the systems themselves should do, what risks they involve, and how we can control these. - After the Introduction to the field (§1), the main themes (§2) of this article are: Ethical issues that arise with AI systems as objects, i.e., tools made and used (...) by humans. This includes issues of privacy (§2.1) and manipulation (§2.2), opacity (§2.3) and bias (§2.4), human-robot interaction (§2.5), employment (§2.6), and the effects of autonomy (§2.7). Then AI systems as subjects, i.e., ethics for the AI systems themselves in machine ethics (§2.8) and artificial moral agency (§2.9). Finally, the problem of a possible future AI superintelligence leading to a “singularity” (§2.10). We close with a remark on the vision of AI (§3). - For each section within these themes, we provide a general explanation of the ethical issues, outline existing positions and arguments, then analyse how these play out with current technologies and finally, what policy consequences may be drawn. (shrink)

Rarely does research in the history and philosophy of science lead to new empirical results, but that is exactly what has happened in one of the essays of this special issue: Rang and Grebe-Ellis have developed new experimental techniques to perform measurements Goethe proposed 217 years ago. These measurements fit neatly with Goethe’s idea of polarity—his complementary spectrum is not only an optical, but also a thermodynamical counterpart of Newton’s spectrum. I use the new measurements, firstly, to argue against the (...) asymmetries between light and darkness posited by Lyre and Schreiber; and, secondly, to explicate the alternative theory that Goethe had introduced to urge scientific pluralism. In my replies to exegetical criticism by Böhler, Hampe and Zemplén, I show that the main goal of Goethe’s Farbenlehre was indeed to expose symmetries between light and darkness. Furthermore, I argue that it is worthwhile to focus on the experiments, arguments and hypotheses of the Farbenlehre, and not merely on rhetorical, narrative or stylistical aspects, as Böhler and Hampe would have it. Goethe’s criticism of Newton is often dismissed, but it is in fact surprisingly relevant today. (shrink)

Goethes Protest gegen Newtons Theorie des Lichts und der Farben ist besser, als man gemeinhin denkt. Man kann diesem Protest in den wichtigsten Elementen folgen, ohne Newton in der physikalischen Sache unrecht zu geben. Laut meiner Interpretation hat Goethe in Newtons wissenschaftsphilosophischer Selbsteinschätzung eine entscheidende Schwäche aufgedeckt: Newton glaubte, mithilfe prismatischer Experimente beweisen zu können, dass das Licht der Sonne aus Lichtstrahlen verschiedener Farben zusammengesetzt sei. Goethe zeigt, dass dieser Übergang vom Beobachtbaren zur Theorie problematischer ist, als Newton wahrhaben wollte. (...) Wenn Goethe darauf beharrt, dass uns der Übergang zur Theorie nicht von den Phänomenen aufgezwungen wird, dann kommt dadurch unser eigener, freier und kreativer Beitrag zur Theorie-Bildung ans Tageslicht. Und diese Einsicht Goethes gewinnt eine überraschende Schärfe, weil Goethe plausibel machen kann, dass sich alle entscheidenden prismatischen Experimente Newtons ebenso gut mit einer alternativen Theorie vereinbaren lassen. Wenn ich recht sehe, war Goethe der erste Wissenschaftsphilosoph, der mindestens eine empirisch äquivalente Alternative zu einer wohletablierten physikalischen Theorie gesehen hat: Damit war Goethe seiner Zeit um ein gutes Jahrhundert voraus. (shrink)

Cet article se propose d’étudier la question de la corruption démocratique à partir d’un cas précis, celui de la crise du traitement des déchets à Naples, communément nommée « crise des ordures ». En analysant trois formes ou niveaux de corruption démocratique lors de cette crise, l’article souhaite souligner que le terme de corruption démocratique, loin de désigner un mécanisme précis, qualifie, au contraire, des actes, des pratiques et des phénomènes très divers.La crise napolitaine est marquée, d’une part, par l’implication (...) du réseau mafieux local – la Camorra – et, d’autre part, par de graves problèmes environnementaux. L’article analyse donc, dans un premier temps, le phénomène de corruption démocratique lié à la présence d’un réseau mafieux et s’attache ensuite à montrer que ce phénomène n’est pas uniquement lié à la présence de la Camorra. En montrant que la mise en place des commissariats extraordinaires est une réponse gouvernementale technocratique à la crise napolitaine, l’article met ensuite en exergue le lien peu évident existant entre corruption démocratique et environnement. Il démontre la difficulté pour nos démocraties libérales et représentatives de répondre démocratiquement à l’enjeu environnemental contemporain et soutient que cette difficulté relève d’une forme de corruption démocratique. (shrink)

While it is often said that robotics should aspire to reproducible and measurable results that allow benchmarking, I argue that a focus on benchmarking can be a hindrance for progress in robotics. The reason is what I call the ‘measure-target confusion’, the confusion between a measure of progress and the target of progress. Progress on a benchmark (the measure) is not identical to scientific or technological progress (the target). In the past, several academic disciplines have been led into pursuing only (...) reproducible and measurable ‘scientific’ results – robotics should be careful to follow that line because results that can be benchmarked must be specific and context-dependent, but robotics targets whole complex systems for a broad variety of contexts. While it is extremely valuable to improve benchmarks to reduce the distance be- tween measure and target, the general problem to measure progress towards more intelligent machines (the target) will not be solved by benchmarks alone; we need a balanced approach with sophisticated benchmarks, plus real-life testing, plus qualitative judgment. (shrink)

Millikan's account of substance concepts is based on a notion of “substance” expanded from realist notions of individuals and natural kinds. Her metaphysical notion, based on “inductive potential,” is shown to be too puristic and needs to incorporate cognizing subjects. This could preserve the realist/nondescriptionist insight that the extension of substances is determined by the world.

A response to Lindsay Craig's essay, The So-Called Extended Synthesis and Population Genetics.

Ich möchte philosophische Ideen von Hilary Putnam, Crispin Wright, Donald Davidson und George Eduard Moore zusammenbringen, um hieb- und stichfest zu beweisen, dass die beste skeptische Hypothese (gegen unser Wissen um die Beschaffenheit der Aussenwelt) nicht zutreffen kann. Putnams Externalismus, Wrights zusätzlicher Appell an Disquotationsprinzipien, Davidsons wahrheitskonditionale Semantik und Moores Verweis auf seine eigenen Hände lassen sich zu einem vierzeiligen Beweis verschmelzen, dessen Konklusion besagt, dass wir nicht von Anbeginn Gehirne im Tank sein können.

Analytische Sätze, die kraft Definition wahr sein sollen, schaden der Naturwissenschaft oder trivialisieren ihren Fortschritt: So lautet einer der Kritikpunkte, die Quine in seinem Feldzug gegen die Unterscheidung zwischen synthetischen und analytischen Sätzen vorgebracht hat. Sie schaden, so Quine, weil sie nicht revidiert werden dürfen und damit die Wahlfreiheit beim Theorienwandel über Gebühr einschränken. (Hätte sich z.B. Einstein vom analytischen Status der newtonischen Impulsdefinition beeindrucken lassen, so hätte er die Relativitätstheorie nicht formulieren können). Oder sie trivialisieren den Fortschritt, weil sich (...) durch Preisgabe analytischer Sätze bloß die Sprache ändert – und wie soll uns eine Änderung der Begriffe vorwärtsbringen? Um dieser Herausforderung zu begegnen, müssen wir Sprach- und Theorienwandel auseinanderdividieren. Im Falle einer wissenschaftlichen Revolution findet beides gleichzeitig statt. Trotzdem kann man die Beobachtungskonsequenzen der neuen Theorie (mithilfe des Ramsey-Satzes) mit denen der alten vergleichen; und mithilfe des Carnap-Satzes kann man den jeweiligen analytischen Anteil der beiden Theorien extrahieren. Es war Carnaps Fehler, diesen Schachzug zur Definition des Begriffs vom analytischen Satz heranzuziehen. Umgekehrt wird ein Schuh draus: Mithilfe einer holismustauglichen Definition von "analytisch" lässt sich zeigen, dass die Carnap-Sätze analytisch sind. Hier die Grundidee der fraglichen Definition: Ein Satz ist analytisch, wenn sich der empirische Gehalt keiner Gesamttheorie dadurch ändert, dass man ihr den Satz einpflanzt. (shrink)

Rollback arguments focus on long sequences of actions with identical initial conditions in order to explicate the luck problem that indeterminism poses for libertarian free will theories (i.e. the problem that indeterministic actions appear arbitrary in a free-will undermining way). In this paper, I propose a rollback argument for probability incompatibilism, i.e. for the thesis that free will is incompatible with all world-states being governed by objective probabilities. Other than the most prominently discussed rollback arguments, this argument explicitly focusses on (...) the ability to act otherwise. It argues that the negligible probability of the relative frequencies in overall rollback patterns being relevantly different indicates that even the ability to act otherwise with regard to individual actions is not free-will enabling. My proposed argument provides probability incompatibilists with a tool to argue against a classical event-causal response to the luck problem, while it can still motivate an agent-causal response to it. (shrink)

Wer humanitäre Kriege moralisch beurteilen will, muss sich in einem chaotischen Meer der Möglichkeiten auskennen; er muss (z.B. in der Rückschau) wissen, was geschehen wäre, hätten sich die Akteure anders entschieden. Solche Fragen betreffen keine Fakten, sondern Kontrafakten; mit kühlem Realitätssinn alleine ist diesen Fragen nicht beizukommen. Im Herzstück dieses Aufsatzes steht eine erkenntnistheoretische Analyse kontrafaktischer Sätze (VI-XIII). Wenn ich recht liege, müssen wir uns bei der Beurteilung solcher Sätze nicht nur an die harten Fakten halten; zusätzlich brauchen wir weichere (...) Beurteilungskriterien wie Einfühlungsvermögen, narrative Intelligenz, Phantasie – und Werthaltungen. Dass eine bestimmte Form von Pazifismus im Lichte dieser erkenntnistheoretischen These an Plausibilität gewinnt, versuche ich in den Schlussabschnitten anhand des Kosovo-Kriegs plausibel zu machen. Der Aufsatz beginnt mit ausserrationalen Betrachtungen zugunsten des Pazifismus und versucht insgesamt, allerlei angeblich vernünftigen Gegenargumenten den Wind aus den Segeln zu nehmen. (shrink)

There is, in some quarters, concern about high–level machine intelligence and superintelligent AI coming up in a few decades, bringing with it significant risks for humanity. In other quarters, these issues are ignored or considered science fiction. We wanted to clarify what the distribution of opinions actually is, what probability the best experts currently assign to high–level machine intelligence coming up within a particular time–frame, which risks they see with that development, and how fast they see these developing. We thus (...) designed a brief questionnaire and distributed it to four groups of experts in 2012/2013. The median estimate of respondents was for a one in two chance that high-level machine intelligence will be developed around 2040-2050, rising to a nine in ten chance by 2075. Experts expect that systems will move on to superintelligence in less than 30 years thereafter. They estimate the chance is about one in three that this development turns out to be ‘bad’ or ‘extremely bad’ for humanity. (shrink)