CPH-NIP Formal Epistemology Workshop
Abstracts
Speaker: Olivier Roy (Groeningen)
Title: Agreement Theorems, Moral Cognitivism and Deliberative Democracy
Abstract: In this talk I investigate whether Agreement Theorems can provide a formal support for the type of moral cognitivism advocated by Jürgen Habermas in connection with his deliberative conception of democracy. I will argue that agreement theorems shed new lights on, and ultimately support moral cognitivism.
Speakers: Sonja Smets (Groeningen) and Alexandru Baltag (Oxford)
Title: Doxastic Attitudes and Norms for Interactive Belief Change
Abstract: The way an epistemic agent ("listener") should revise her beliefs after receiving information from a given source (e.g. via an announcement by a "speaker") depends on the listener's doxastic attitude towards the speaker: her opinion about the reliability of information coming from this particular source. Such doxastic attitudes are not "static" and purely "descriptive" as the attitudes that are usually investigated in doxastic-epistemic logics (e.g. knowledge, belief, strong belief, safe belief etc), but they are "dynamic" and "prescriptive": they embody "norms" for changing one's beliefs whenever given an input from a given source. We present a dynamic-epistemic logic that uses doxastic transformations to formalize this type of dynamic-normative attitudes of agents towards each other. In this setting, we analyze notions of "sincerity", "honesty" and "persuasiveness" of a communication act, distinguishing various types of lies (half-lies, honest lies, "sincere exagerations", etc). Next, we consider the long-term dynamics obtained by iterating such doxastic transformations: the way a group's beliefs evolve (converging or not to a fixed point) over a sequence of successive communication acts. Time-permitting, we apply this to two specific epistemological problems: (a) the issue of reaching inter-agent "doxastic agreement": aggregating agents' beliefs by communication and persuasion; and (b) "solving" the Surprise Examination paradox.
Speaker: Hannes Leitgeb (Bristol and LMU München)
Title: Reducing Belief Simpliciter to Degrees of Belief
Abstract: We prove that given quite reasonable assumptions, it is possible to give an explicit definition of belief simpliciter in terms of subjective probability, such that it is neither the case that belief is stripped of any of its usual logical properties, nor is it the case that believed propositions are bound to have probability 1. Belief simpliciter is not to be eliminated in favour of degrees of belief, rather, by reducing it to assignments of consistently high degrees of belief, both quantitative and qualitative belief turn out to be governed by one unified theory. Turning to possible applications and extensions of the theory, we suggest that this will allow us to see: how the Bayesian approach in general philosophy of science can be reconciled with the deductive or semantic conception of scientific theories and theory change; how the assertability of conditionals can become an all-or-nothing affair in the face of non-trivial subjective conditional probabilities; how knowledge entails a high degree of belief but not necessarly certainty; how primitive conditional probability functions (Popper functions) arise from conditionalizing absolute probability measures on maximally strong believed propositions with respect to different cautiousness thresholds; and how conditional chances may become the truthmakers of counterfactuals.
Speaker: Branden Fitelson (Rutgers)
Title: Knowledge from non-knowledge
Abstract: In this paper, I will first discuss some recent (alleged) examples of "knowledge from falsehood" (i.e., inferential knowledge that is epistemically grounded -- at least in part -- in a false premise), and I will explain how these examples can be strengthened in some interesting ways (if they are in fact bona fide cases of "knowledge from falsehood"). Then, I will discuss the broader question of whether one can have knowledge from non-knowledge (i.e., inferential knowledge that is epistemically grounded -- at least in part -- in a premise that is not known). I will show that some (standard) ways of resisting the existence of knowledge from non-knowledge are bound to fail. Finally, I will consider another (more promising) strategy for resisting the possibility of knowledge from non-knowledge (due to Vogel), as well as another recent argument FOR the existence of knowledge from non-knowledge (due to Luzzi).
Speaker: Tomoji Shogenji (Rhode Island)
Title: The Role of Coherence in the Transmission of Testimonial Justification
Abstract: It is sometimes thought that coherence of independently produced testimonies enhance their credibility, thereby indirectly enhancing the credibility of the hypothesis they support by the transmission of justification. This paper argues that coherence plays two different roles in the transmission of testimonial justification. First, it is shown that pairwise coherence of independently produced testimonies enhances their credibility by the transmission of justification between the pair. However, when it comes to the transmission of justification to the hypothesis they support, coherence of testimonies plays a negative role, viz. it is better for the hypothesis to be supported by less coherent testimonies (more "diverse" evidence), other things being equal. Distinguishing these two roles in the transmission of justification leads to a more balanced picture of how coherence affects the justification of beliefs.
Speaker: Erik Olsson (Lund)
Title: Setting the Threshold of Assertion: A Simulation Study in Social Epistemology
Abstract: Many philosophers have claimed that you are justified in asserting something only if you know that what you assert is true. This is the celebrated knowledge rule for assertion. As many philosophers have also pointed out, it is difficult for a potential asserter to separate knowledge from mere high subjective probability from her own subjective point of view. As a consequence, the knowledge rule reduces in practice to the high probability rule which states that you may asserting something only if you assign that what you assert a high enough probability, i.e., a probability that does not fall below some threshold value. As an extreme case, there is the rule requiring for assertion a subjective probability of 1. The high probability rule is a schema that gives rise to a whole spectrum of concrete assertion rules, depending on where, exactly, the threshold is taken to be located. This raises the question to what extent there is an optimal choice of a threshold. More precisely, is there a way of setting the threshold that maximizes veritistic value in the sense of Goldman (1999)? This question can be tackled by means of computer simulation, and in the talk some relevant results will presented that have been obtained using the Bayesian simulation framework Laputa developed at Lund University.
Speaker: Stephan Hartmann (Tilburg)
Title: No-Alternatives Arguments
Abstract: We construct a Bayesian model to show that the observation that no one has yet found an alternative to a proposed hypothesis supports the hypothesis in question. Our model has various applications in philosophy of science (such as the realism debate and IBE) which we also discuss. The talk is based on joint work with Richard Dawid (Vienna).
Speaker: Olivier Roy (Groeningen)
Title: Agreement Theorems, Moral Cognitivism and Deliberative Democracy
Abstract: In this talk I investigate whether Agreement Theorems can provide a formal support for the type of moral cognitivism advocated by Jürgen Habermas in connection with his deliberative conception of democracy. I will argue that agreement theorems shed new lights on, and ultimately support moral cognitivism.
Speakers: Sonja Smets (Groeningen) and Alexandru Baltag (Oxford)
Title: Doxastic Attitudes and Norms for Interactive Belief Change
Abstract: The way an epistemic agent ("listener") should revise her beliefs after receiving information from a given source (e.g. via an announcement by a "speaker") depends on the listener's doxastic attitude towards the speaker: her opinion about the reliability of information coming from this particular source. Such doxastic attitudes are not "static" and purely "descriptive" as the attitudes that are usually investigated in doxastic-epistemic logics (e.g. knowledge, belief, strong belief, safe belief etc), but they are "dynamic" and "prescriptive": they embody "norms" for changing one's beliefs whenever given an input from a given source. We present a dynamic-epistemic logic that uses doxastic transformations to formalize this type of dynamic-normative attitudes of agents towards each other. In this setting, we analyze notions of "sincerity", "honesty" and "persuasiveness" of a communication act, distinguishing various types of lies (half-lies, honest lies, "sincere exagerations", etc). Next, we consider the long-term dynamics obtained by iterating such doxastic transformations: the way a group's beliefs evolve (converging or not to a fixed point) over a sequence of successive communication acts. Time-permitting, we apply this to two specific epistemological problems: (a) the issue of reaching inter-agent "doxastic agreement": aggregating agents' beliefs by communication and persuasion; and (b) "solving" the Surprise Examination paradox.
Speaker: Hannes Leitgeb (Bristol and LMU München)
Title: Reducing Belief Simpliciter to Degrees of Belief
Abstract: We prove that given quite reasonable assumptions, it is possible to give an explicit definition of belief simpliciter in terms of subjective probability, such that it is neither the case that belief is stripped of any of its usual logical properties, nor is it the case that believed propositions are bound to have probability 1. Belief simpliciter is not to be eliminated in favour of degrees of belief, rather, by reducing it to assignments of consistently high degrees of belief, both quantitative and qualitative belief turn out to be governed by one unified theory. Turning to possible applications and extensions of the theory, we suggest that this will allow us to see: how the Bayesian approach in general philosophy of science can be reconciled with the deductive or semantic conception of scientific theories and theory change; how the assertability of conditionals can become an all-or-nothing affair in the face of non-trivial subjective conditional probabilities; how knowledge entails a high degree of belief but not necessarly certainty; how primitive conditional probability functions (Popper functions) arise from conditionalizing absolute probability measures on maximally strong believed propositions with respect to different cautiousness thresholds; and how conditional chances may become the truthmakers of counterfactuals.
Speaker: Branden Fitelson (Rutgers)
Title: Knowledge from non-knowledge
Abstract: In this paper, I will first discuss some recent (alleged) examples of "knowledge from falsehood" (i.e., inferential knowledge that is epistemically grounded -- at least in part -- in a false premise), and I will explain how these examples can be strengthened in some interesting ways (if they are in fact bona fide cases of "knowledge from falsehood"). Then, I will discuss the broader question of whether one can have knowledge from non-knowledge (i.e., inferential knowledge that is epistemically grounded -- at least in part -- in a premise that is not known). I will show that some (standard) ways of resisting the existence of knowledge from non-knowledge are bound to fail. Finally, I will consider another (more promising) strategy for resisting the possibility of knowledge from non-knowledge (due to Vogel), as well as another recent argument FOR the existence of knowledge from non-knowledge (due to Luzzi).
Speaker: Tomoji Shogenji (Rhode Island)
Title: The Role of Coherence in the Transmission of Testimonial Justification
Abstract: It is sometimes thought that coherence of independently produced testimonies enhance their credibility, thereby indirectly enhancing the credibility of the hypothesis they support by the transmission of justification. This paper argues that coherence plays two different roles in the transmission of testimonial justification. First, it is shown that pairwise coherence of independently produced testimonies enhances their credibility by the transmission of justification between the pair. However, when it comes to the transmission of justification to the hypothesis they support, coherence of testimonies plays a negative role, viz. it is better for the hypothesis to be supported by less coherent testimonies (more "diverse" evidence), other things being equal. Distinguishing these two roles in the transmission of justification leads to a more balanced picture of how coherence affects the justification of beliefs.
Speaker: Erik Olsson (Lund)
Title: Setting the Threshold of Assertion: A Simulation Study in Social Epistemology
Abstract: Many philosophers have claimed that you are justified in asserting something only if you know that what you assert is true. This is the celebrated knowledge rule for assertion. As many philosophers have also pointed out, it is difficult for a potential asserter to separate knowledge from mere high subjective probability from her own subjective point of view. As a consequence, the knowledge rule reduces in practice to the high probability rule which states that you may asserting something only if you assign that what you assert a high enough probability, i.e., a probability that does not fall below some threshold value. As an extreme case, there is the rule requiring for assertion a subjective probability of 1. The high probability rule is a schema that gives rise to a whole spectrum of concrete assertion rules, depending on where, exactly, the threshold is taken to be located. This raises the question to what extent there is an optimal choice of a threshold. More precisely, is there a way of setting the threshold that maximizes veritistic value in the sense of Goldman (1999)? This question can be tackled by means of computer simulation, and in the talk some relevant results will presented that have been obtained using the Bayesian simulation framework Laputa developed at Lund University.
Speaker: Stephan Hartmann (Tilburg)
Title: No-Alternatives Arguments
Abstract: We construct a Bayesian model to show that the observation that no one has yet found an alternative to a proposed hypothesis supports the hypothesis in question. Our model has various applications in philosophy of science (such as the realism debate and IBE) which we also discuss. The talk is based on joint work with Richard Dawid (Vienna).