Research Forum
All welcome!
Master's students get credit for attendance. Please make sure you indicate your name on the sheet that will be passed around during the talk.
If you have difficulty accessing the eLearning site (https://elearning.uni-bayreuth.de/course/view.php?id=32952), please let us know (Patricia Rich / Paolo Galeazzi).
Winter 2024/25
November 26th I Joint Talk I Conrad Heilmann and Stefan Wintein (Erasmus University Rotterdam) I Objective Fairness Theories in Economics and Philosophy
Abstract: Fairness is an important normative concept, central to theories in both philosophy and economics, as well as to moral theory and practice. We advance a methodological thesis about normative fairness research in philosophy and economics: these theories have established conceptual insight about fairness objectively – at least in some important ways and to some degree. We demonstrate that there is more to objectivity of fairness than meets the eye, in three ways: (i) there are important non-subjective theories of fairness in both economics and philosophy, (ii) the content of subjective fairness theories in both economics and philosophy is established more objectively than one might think, (iii) there is much intersubjective agreement between fairness researchers on both the subjective and non-subjective fairness research in both disciplines. More generally, our methodological thesis about normative fairness research in philosophy and economics reveals several important, and hitherto unacknowledged, parallels between the two disciplines. It is also contrary to the widely held opinion that fairness is a quintessentially subjective concept.January 14th I Marc Cheong (University of Melbourne) I "Do you come from a Land Down Under?" AI & digital ethics perspectives from the Antipodes
Abstract: tbaJanuary 22th I Joint Talk I Vincent Merlin (Université de Caen) I Measuring power in French inter-municipal councils [Note that this talk will take place in room S 62 (RW)]
Abstract: tbaJanuary 28th I Friederike Wall (Alpen-Adria-Universität Klagenfurt) I Hidden-Action Models from a Critical Rationalist Perspective
Abstract: tba
Preceding Talks
November 5th I Roman Heil (Goethe-Universität Frankfurt/M) I Epistemic Justification and Third Parties
Abstract: Whether a belief is epistemically justified is widely considered to be an affair between the believer and the way they formed their belief. In this paper, I explore the view that the justificatory status of a belief is linked to what is appropriate for third parties to believe. In doing so, I will draw inspiration from debates in legal theory, where the view that legal justification is linked to the appropriateness of third-party conduct has been systematically worked out and widely codified. I will show that these legal debates are strikingly analog to recent discussions about the function of epistemic assessment, where it has been argued that there is a link between a belief’s being epistemically justified and the appropriateness of third-party deference to said belief (Dogramaci, 2012, 2015; Greco and Hedden, 2016). In this paper, I will spell out this link and argue that it provides us with a powerful explanation of our division of epistemic labour. However, I will also show that this explanatory proposal is threatened by cases of (what I call) deference conflicts. I will argue that the most promising response to these cases is to take a “civil law–style” approach and favour the view that epistemic justification is factive.
October 22nd I Florian Boge (TU Dortmund) I Re-Assessing Machine Cognition in the Age of Deep Learning
Abstract: How seriously should we take the “I” in AI? Do ChatGPT and co literally understand our prompts? This question has long puzzled philosophers and scientists alike, with verdicts ranging from outright enthusiasm to profound pessimism. In this lecture, I will re-address the issue from two vantage points. First, I will suggest that Searle’s classic “Chinese Room Argument” can be revived in the age of Deep Learning, but in ways quite different from those Searle himself envisioned. Combining a more careful approach to Deep Learning theory with a slight alteration of the original scenario I call “The Chinese Library”, I will show that, insofar as Searle’s arguments were applicable in the 1980s, they are still applicable today. Second, I will suggest a close connection between understanding and the possession of concepts and, based on evidence from the technical literature, suggest that we should not assume that present-day DNNs have concepts – and hence, that they do understand anything.July 9th I Alex Gregory (University of Southampton) I Take In Your Hen: How to Evaluate Hedonic Adaptation
Abstract: Humans have a strong tendency to hedonically adapt to their circumstances, so that something that once brought joy eventually brings only indifference. Does this tendency guarantee a kind of failure on our part? Happiness, like other emotions, seems subject to evaluation in terms of its fittingness. But it’s not clear how hedonic adaptation could possibly maintain fittingness: it involves changing one’s level of happiness even if one’s circumstance stay the same. This paper mounts a tentative defence of hedonic adaptation against this concern. It does so by articulating a key difference between the scale of happiness and the scale of goodness, and shows how that difference guarantees an inability to perfectly track goods with our levels of happiness. Given this background constraint, hedonic adaptation may be the most appropriate way for our happiness to change over time, even if we thereby fall short of some more perfect ideal.July 2th I Anette Stimmer (University of St. Andrews) I Norm Contestation and Change in International Politics
Abstract: My book project studies contestation over established international legal norms such as territorial integrity, due process rights and the prohibition of torture in the post-Cold War era. Legal ambiguities and tensions give rise to debate when international law is applied to concrete situations. What outcomes can such norm contestation have and what are their implications for norm strength? What influences the duration of norm contestation? I use insights from rhetoric to show that states can agree or disagree on both the norm frame (justification) and/or claim (action) when applying international law. Thus, norm contestation can have four different outcomes: norm clarification (frame and claim agreement), norm recognition (frame agreement/claim disagreement), norm neglect (frame disagreement/claim agreement) and norm impasse (frame and claim disagreement). These alternate endings have different effects on the clarity and strength of the contested norms, as well as on subsequent debate over them. States act strategically when interpreting norms, but social dynamics intervene in the process, and influence the outcome of contestation. Norms are collectively held standards of appropriate behaviour, and therefore the verdict of the international community over norm interpretations decides over whether they take hold and whose reputation will suffer. I find that the three elements of debate influence the duration and outcome of norm contestation: 1) audience reactions, that is, what kind of social support the contesting parties receive; 2) argumentation, or the ability to legitimate interpretations; and, 3) the characteristics of speakers – in particular, when delegation takes place, the ability of agents to legitimate their work with their output and identity. The book illustrates this theory with eight case studies. In the talk, I will focus on two of these case studies: the contestation over the Ombudsperson to the ISIL and Al-Qaida Sanctions Committee of the UN Security Council and the maritime boundary dispute between China and the Philippines in the South China Sea.+++ this talk had to be cancelled +++ June 25th I Roman Heil (Goethe-Universität Frankfurt) I Epistemic Justification and Third Parties +++ this talk had to be cancelled +++
Abstract: Whether a belief is epistemically justified is widely considered to be an affair between the believer and the way they formed their belief. In my talk, I will explore the view that being epistemically justified in believing concerns others as well. Although unorthodox to consider in epistemology, legal theorists have systematically developed the view that the legal justification of conduct guides third parties in how they should engage with said conduct. In my talk, I will argue that epistemic justification guides third parties in what they should believe. More specifically, I will defend the view that epistemic justification is tightly linked to the normativity of deference and show that it provides a promising story about how we efficiently organize the division of epistemic labor. However, depending on how we think about what I call deference conflicts, the resulting view predicts that there are externalist defeaters to epistemic justification.+++ this talk had to be cancelled +++ June 26th I Erik Voeten (Georgetown University) I Joint Talk: The Energy Transition and Political Polarization Along Occupational Lines: Evidence from Germany I (will be taking place on Wednesday in S 61 (RW I)) +++ this talk had to be cancelled +++
- June 11 I Ophelia Deroy (LMU München) I The public-private distinction in the mind: How we draw it and why it matters
Abstract: If you are on a bus engrossed in your phone, you recognize that the visual content on your screen remains private to you. Any sudden sound emanating from your device however becomes public to those nearby. But how, and where is drawn the difference between a public and a private observable or observed object? Are those two even the same? While answering these questions, I want also to explain why absence of publicity and illusions of publicity matter to our understanding of some phenomena like polarisation and other social dynamics. - +++ this talk had to be cancelled +++ May 28th I Florian Boge (TU Dortmund) I Put it to the Test: Getting Serious about Explanations in Explainable Artificial Intelligence +++ this talk had to be cancelled +++
Abstract: Artificial Intelligence (AI) now pervades both science and society, but many present AI systems are known to be notorious black boxes. This can become a pressing issue, especially in high-stakes contexts such as decision-making in medical practice. Here, explanations of the outputs of an AI system seem desirable for the sake of calibrating trust, or may even constitute the basis of claims to moral and legal accountability. However, in contrast to standard scientific practice, current practice in the field of eXplainable AI (XAI) falls short in an important respect: While scientific explanations are usually required to be accompanied by testable predictions (Douglas, 2009), explanations in XAI are usually only validated on existing and well-known data. In our paper, we integrate insights from the philosophy of testability with recent developments in XAI to suggest a way out of the loop. This will be done by building on the framework for ‘Falsifiable eXplanations of Artificial Intelligence’ (FXAI), recently proposed by Schumacher et al., as a case study, and on its applications in cancer-research. - May 14th I Marta Halina (University of Cambridge) I Intuitive Physics in Nonhuman Animals
Abstract: Comparative psychologists have spent the last few decades examining whether nonhuman animals understand the physical world in a way that is similar to humans. Broadly, human intuitive physics is thought to include capacities such as knowing that solid objects continue to exist even when no longer perceived, that objects tend to fall unless prevented from doing so, and that objects do not pass through each other. In this talk, I introduce the empirical research program dedicated to investigating intuitive physics in nonhuman animals. I then show how current research in this area encounters problems of holistic and contrastive underdetermination. Finally, I propose a route forward: computational modelling combined with signature testing. January 30th I Ella Whiteley (University of Sheffield) I Attentional Objectification
Abstract: This talk brings precision to a pervasive but under-theorised way in which objectification can occur: through attentional patterns alone. Further, it introduces particularly subtle forms of attention-based objectification, where the attentional pattern’s problems are revealed in its comparative nature. For instance, a person might listen to a woman’s conversational contributions, and so not ignore something meaningful about her, and yet find her figure comparatively more noticeable. Alternatively, a person might not fixate on the bodies of black men, and yet find their bodies comparatively more salient than the bodies of white men. Recognising these particularly elusive forms of objectification requires acknowledging that, in contrast with influential interpretations of objectification, one needn’t be reduced to a body, or to have one’s autonomy denied, to count as being objectified. Moreover, the subtlety of these forms of objectification grants them an insidious immunity from criticism, which results in distinctive harms for the victim.January 23rd 2024 I Viktoria Knoll (TU Dresden) I The Normativity of Gender Revisited <- Due to train strikes, this session must be moved online. We will meet in the following zoom room at the usual time: https://uni-bayreuth.zoom.us/j/4145888814?pwd=NnpzUXAyQllrMGc5MXExaWw0YmZLZz09
January 16th 2024 I Gil Hersch (Virginia Tech - Kellog Center for Philosophy, Politics, and Economics) I Joint Talk: Weighting Waiting
Abstract: Imagine a case in which there is some good that many people want, for example a refreshment at a kiosk. People know to line up and queue, wait their turn to place their order, based on the order in which they arrived. Now imagine that someone rashes in, yelling that their partner just fainted outside and is in dire need of some water. I assume most people would find it absurd if those already in the queue would insist that the person get in the queue just like everyone else. Whole we generally treat line cutting as reprehensible, we also recognize that there are times in which people's claim for a good override our entitlement to be served before them just because we were ahead of them in the queue. What is much less commonplace is the recognition that there exists a continuum between everyone receiving the good in the order in which thy join the queue, and some people having a sufficiently strong claim to justify their jumping to the front of the queue. Between these extremes of completely equal treatment and lexicographic priority to very strong claim, I propose a weighting system for queuing based on different claim strengths.December 12th 2023 I Laura Jahn (Copenhagen) I Curbing Amplification Online — Towards Improving the Quality of Information Spread on Social Media Using Agent-Based Models and Twitter Data
Abstract: This talk presents a research project that studies ways to curb the amplification of low-quality content, such as misinformation, on social media using agent-based models and data from the social media platform Twitter (now X). The work focuses explicitly on the amplification through one-click user reactions such as likes and shares. Liking and sharing are central ways by which information spreads in a social network while informing platforms’ content-sorting algorithms, further increasing reach. Amplification through likes and shares may be driven by coordinated and/or inauthentic actors such as social bots. Yet, also authentic human users may spread low-quality content. In light of social influence and cognitive biases, authentic users may engage with high-engagement posts allocating little to no attention to assess accuracy or quality. Both inauthentic and authentic dynamics amplify misinformation online and undermine the wisdom of crowds: High engagement does not reliably point to high quality. While the inflation of engagement metrics is a readily available manipulation strategy undermining the wisdom-of-crowds effect, research has yet to extensively study the amplification of low-quality content through likes and shares. A major reason is that data on one-click user reactions is non-trivial to collect. From different angles, the research project addresses threats to the wisdom of crowds and aims to improve the (epistemic) quality of the information that gets amplified on social media. We present computational methods to detect inauthentic, coordinated metric inflation and suspicious correlations in reactions data. This part of the project is based on computer-simulated data from an agent-based model and novel empirical data live-collected through Twitter with a scripted algorithm written with the purpose of overcoming the data shortage on one-click user reactions. Another part of the project studies behavioral interventions based on friction to prevent the amplification of low-quality content analyzed with an agent-based model.
November 28th 2023 I Katharine Browne (Oslo) I What is wrong with how attention is commodified?
Abstract: Our attention is commodified: it is bought and sold in market transactions when individuals lend out the ability to control their attentional capacities in exchange (for example) for technological services. What is wrong with that? Attention markets, we argue, resemble labor markets. By drawing on the ethics of commodification and core features of attention, we show that attention markets, while not always morally wrong, carry special moral risks: because of how attention shapes beliefs and desires, subjective experience and action, they are prone to be disrespectful, alienating, and provide fertile grounds for domination. Our analysis calls for regulatory interventions.
November 21th 2023 I André Bachtiger (Stuttgart) I Designing Democracy on Mars and Earth
Abstract: The talk presents the DDME (Designing Democracy on Mars and Earth) project which sets up a bottom-up design to obtain a deeper understanding of citizens’ democratic preferences (principles and designs). Based on input from democratic theorists, DDME explores how citizens imagine “ideal” democracy (on “Mars”) and mend “real” democracy (i.e., how they would reform the political systems they live in (on “Earth”)) when they had the chance to reflect or deliberate on the pros and cons of the various conceptions and schemes of democracy. DDME is the first large-scale project to delegate democratic designing to citizens adopting a systematic and global approach where citizens together with democratic theorists reflect on advanced theoretical inputs (e.g., problem-based thinking on democracy) and think creatively about optimal democratic designs. In the talk, I will present first results from the DDME project.- April 18th 2023 I Maren Behrensen (Twente) I Dragging Ontology back into Politics: Transphobia, Moral Panics, and Populism
- May 16th 2023 I Alessandro Galeazzi (Brescia) I The Evolution of Polarization in Twitter Debates: The Cases of U.S. Presidential Election and Climate Change
Abstract: The public sphere is a complex environment where diverse ideas are discussed and contribute to shaping public opinion. With the rise of social media, a portion of these debates has shifted online, where communication is faster and disintermediated. Often, online debates around controversial issues become polarized, with two opposing factions clashing over ideological differences. This contrast may lead to an excessive level of polarization that can harm society in various ways, ranging from the risk of political deadlocks to individual and collective violent actions.
Here, we present two studies on polarization in Twitter debates surrounding the U.S. presidential election and climate change. We first introduce the techniques used to study polarization and then we analyze the evolution of the debate over time. Finally, we present some open questions and future research directions. Overall, our findings contribute to the understanding of the role of polarization in online debates and its evolution over time. May 23th 2023 I Emanuele Ratti (Linz) I Automated Science, Machine Learning, and Values
Abstract: Dreams of automated science have accompanied the development of modern natural sciences. However, it is difficult to pinpoint a specific conception of automated science. Recent developments in artificial intelligence (AI), and machine learning in particular (ML), have been associated to the idea of automated science, suggesting that ML-driven contemporary automated science is just another step towards what science has always promised to fulfil. In this talk, I investigate the relation between ML and automated science. First, I reconstruct and identify two views of automated science. The first, called traditional automated science (TAS), sees science as mechanical, fostering intersubjective agreement, and suppressing scientists' subjectivity. The second view is Paul Humphreys' conception, and it differs from the first because it does minimize the suppression of subjectivity, and it attributes to automated science a non-human epistemic horizon. I compare these two views to ML to see which kind of automated science contemporary AI can possibly foster.June 13th 2023 I Matthias Brinkmann (LMU) I Joint Talk: What, if anything, is morally wrong with inflation?
Abstract: Laypeople often discuss inflation in moralised terms, such as greed, political corruption, dishonesty, and so forth. However, contemporary philosophers have not had much to say about inflation at all. In this paper, I ask whether there is any philosophically respectable argument against inflation. After introducing the basics, I first discuss and re-construct several consequentialist arguments against inflation one can find implicit in the economic literature. Second, I explore various non-consequentialist arguments against inflation: for example, that it constitutes a hidden tax, arbitrary re-distribution, “counterfeiting”, or a violation of property rights. Third, I develop a novel argument against inflation based on the notion of legitimate expectations. Many socially valuable practices, such as saving or entering debt contracts, rest on money expectations—expectations that money keep its value, or at least only change its value in predictable ways. I argue that central banks as the controlling institutions of money have special moral duties to respect money expectations, and thus, not to unexpectedly change the value of money. In closing, I discuss several objections to, and implications of, the view.June 14th 2023 I Dean Lueck (Indiana) I Joint Talk: The Institutional Legacy of the Mexican Rancho System in California
July 04th 2023 I Zoey Lavallee (CRÉ Montreal/McGill) I Affordances and the shape of addiction
Abstract: Recently, a number of researchers have posited what might be called an ‘affordance model’ of addiction, according to which addiction shapes what possibilities for action addicted agents experience the world as offering. Combating addiction, then, involves altering the addicted person’s world by removing addiction-related affordances and creating new affordances. In this paper, we develop a more detailed affordance model of addiction. We argue that what is significant about the addicted person’s world is not only what affordances are experienced as salient, but also the way in which the addicted person’s world is shaped by a dominant concern, resulting in a comparably inflexible way of experiencing and interacting with the environment. Our analysis of addiction foregrounds the dynamic relationship between an individual and their environment, and suggests that strategies for recovery need to involve mechanisms that not only alter the environment, but that also counteract the dominant concern in addiction.July 11th 2023 I Chiara Lisciandra (LMU) I Explanatory Norms and Interdisciplinary Science
Abstract: This paper provides resources from the philosophy of science to identify differences between explanatory norms across disciplines and how such differences affect interdisciplinary work. This research field has its roots in Humphreys' seminal work (2002) and its refinement in a series of publications (2004, 2019). The body of literature on explanatory norms is rapidly growing. However, there is still no consensus on a theoretical framework that allows us to identify norms across disciplines in a systematic manner. The aim of this paper is to provide such a framework and use it to i) identify explanatory norms across domains; and ii) predict patterns of interdisciplinary work accordingly. By pursuing these goals, this work promises to be theoretically significant and practically relevant. It contributes to the work on domain-specific explanatory norms; and provides recommendations for science-policy analysis of interdisciplinary science.