When scientists seek further confirmation of their results, they often attempt to duplicate the results using diverse means. To the extent that they are successful in doing so, their results are said to be robust. This paper investigates the logic of such "robustness analysis" [RA]. The most important and challenging question an account of RA can answer is what sense of evidential diversity is involved in RAs. I argue that prevailing formal explications of such diversity are unsatisfactory. I propose a (...) unified, explanatory account of diversity in RAs. The resulting account is, I argue, truer to actual cases of RA in science; moreover, this account affords us a helpful, new foothold on the logic undergirding RAs. (shrink)

How should we reason in science? Jan Sprenger and Stephan Hartmann offer a refreshing take on classical topics in philosophy of science, using a single key concept to explain and to elucidate manifold aspects of scientific reasoning. They present good arguments and good inferences as being characterized by their effect on our rational degrees of belief. Refuting the view that there is no place for subjective attitudes in 'objective science', Sprenger and Hartmann explain the value of convincing evidence in terms (...) of a cycle of variations on the theme of representing rational degrees of belief by means of subjective probabilities (and changing them by Bayesian conditionalization). In doing so, they integrate Bayesian inference—the leading theory of rationality in social science—with the practice of 21st century science. Bayesian Philosophy of Science thereby shows how modeling such attitudes improves our understanding of causes, explanations, confirming evidence, and scientific models in general. It combines a scientifically minded and mathematically sophisticated approach with conceptual analysis and attention to methodological problems of modern science, especially in statistical inference, and is therefore a valuable resource for philosophers and scientific practitioners. (shrink)

This paper develops axiomatic foundations for a probabilistic-interventionist theory of causal strength. Transferring methods from Bayesian confirmation theory, I proceed in three steps: I develop a framework for defining and comparing measures of causal strength; I argue that no single measure can satisfy all natural constraints; I prove two representation theorems for popular measures of causal strength: Pearl's causal effect measure and Eells' difference measure. In other words, I demonstrate these two measures can be derived from a set of plausible (...) adequacy conditions. The paper concludes by sketching future research avenues. (shrink)

Carl Hempel argued that probabilistic hypotheses are limited in what they can explain. He contended that a hypothesis cannot explain why E is true if the hypothesis says that E has a probability less than 0.5. Wesley Salmon and Richard Jeffrey argued to the contrary, contending that P can explain why E is true even when P says that E’s probability is very low. This debate concerned noncontrastive explananda. Here, a view of contrastive causal explanation is described and defended. It (...) provides a new limit on what probabilistic hypotheses can explain; the limitation is that P cannot explain why E is true rather than A if P assign E a probability that is less than or equal to the probability that P assigns to A. The view entails that a true deterministic theory and a true probabilistic theory that apply to the same explanandum partition are such that the deterministic theory explains all the true contrastive propositions constructable from that partition, whereas the probabilistic theory often fails to do so. (shrink)

This paper proposes an analysis of surprise formulated in terms of proximity to the truth, to replace the probabilistic account of surprise. It is common to link surprise to the low probability of the outcome. The idea seems sensible because an outcome with a low probability is unexpected, and an unexpected outcome often surprises us. However, the link between surprise and low probability is known to break down in some cases. There have been some attempts to modify the probabilistic account (...) to deal with these cases, but as we shall see, they are still faced with problems. The new analysis of surprise I propose turns to accuracy and identifies an unexpected degree of inaccuracy as reason for surprise. The shift from probability to proximity allows us to solve puzzles that strain the probabilistic account of surprise. (shrink)

ABSTRACT When scientists seek further confirmation of their results, they often attempt to duplicate the results using diverse means. To the extent that they are successful in doing so, their results are said to be ‘robust’. This article investigates the logic of such ‘robustness analysis’. The most important and challenging question an account of RA can answer is what sense of evidential diversity is involved in RAs. I argue that prevailing formal explications of such diversity are unsatisfactory. I propose a (...) unified, explanatory account of diversity in RAs. The resulting account is, I argue, truer to actual cases of RA in science; moreover, this account affords us a helpful new foothold on the logic undergirding RAs. 1Robustness Analysis in Science 2RA Diversity and Independence 2.1Unconditional probabilistic independence 2.2Reliability independence 2.3Confirmational and conditional independence 2.4Partial independence? 3Robustness Analysis as Explanatory Reasoning 3.1Explanatorily discriminating means of detection 3.2The logic of robustness analysis 3.2.1Mutually exclusive competitors 3.2.2Consistent epistemic competitors 4Conclusion Appendix. (shrink)

The proposition that Tweety is a bird coheres better with the proposition that Tweety has wings than with the proposition that Tweety cannot fly. This relationship of contrastive coherence is the focus of the present paper. Based on recent work in formal epistemology we consider various possibilities to model this relationship by means of probability theory. In a second step we consider different applications of these models. Among others, we offer a coherentist interpretation of the conjunction fallacy.

Recently there have been several attempts in formal epistemology to develop an adequate probabilistic measure of coherence. There is much to recommend probabilistic measures of coherence. They are quantitative and render formally precise a notion—coherence—notorious for its elusiveness. Further, some of them do very well, intuitively, on a variety of test cases. Siebel, however, argues that there can be no adequate probabilistic measure of coherence. Take some set of propositions A, some probabilistic measure of coherence, and a probability distribution such (...) that all the probabilities on which A’s degree of coherence depends (according to the measure in question) are defined. Then, the argument goes, the degree to which A is coherent depends solely on the details of the distribution in question and not at all on the explanatory relations, if any, standing between the propositions in A. This is problematic, the argument continues, because, first, explanation matters for coherence, and, second, explanation cannot be adequately captured solely in terms of probability. We argue that Siebel’s argument falls short. (shrink)

This paper discusses two senses in which a hypothesis may be said to unify evidence. One is the ability of the hypothesis to increase the mutual information of a set of evidence statements; the other is the ability of the hypothesis to explain commonalities in observed phenomena by positing a common origin for them. On Bayesian updating, it is only mutual information unification that contributes to the incremental support of a hypothesis by the evidence unified. This poses a challenge for (...) the view that explanation is a confirmatory virtue that contributes to such incremental support; its advocates must ground it in some relevant difference between humans and a Bayesian agent. (shrink)

Several philosophers propose probabilistic measures of how well a potential scientific explanation would explain the given evidence. These measures could elaborate “best” in “inference to the best explanation”. This paper argues that none of these measures succeeds. The paper considers the various rival explanations that scientists proposed for the parallelogram of forces. Scientists regarded various features of these proposals as making them more or less “lovely”. None of these probabilistic measures of loveliness can reflect these features. The paper concludes by (...) considering the kinds of probabilities that could reflect these features. (shrink)

The Problem of Old Evidence is a perennial issue for Bayesian confirmation theory. Garber famously argues that the problem can be solved by conditionalizing on the proposition that a hypothesis deductively implies the existence of the old evidence. In recent work, Hartmann and Fitelson :712–717, 2015) and Sprenger :383–401, 2015) aim for similar, but more general, solutions to the Problem of Old Evidence. These solutions are more general because they allow the explanatory relationship between a new hypothesis and old evidence (...) to be inductive, rather than deductive. In this paper, I argue that these solutions are either unsound or under-motivated, depending on the case of inductive explanation that we have in mind. This lends support to the broader claim that Garber–Style Bayesian confirmation cannot capture the sense in which new hypotheses that do not deductively imply old evidence nevertheless seem to be confirmed via old evidence. (shrink)

The Problem of Old Evidence is a perennial issue for Bayesian confirmation theory. Garber (Test Sci Theor 10:99–131, 1983) famously argues that the problem can be solved by conditionalizing on the proposition that a hypothesis deductively implies the existence of the old evidence. In recent work, Hartmann and Fitelson (Philos Sci 82(4):712–717, 2015) and Sprenger (Philos Sci 82(3):383–401, 2015) aim for similar, but more general, solutions to the Problem of Old Evidence. These solutions are more general because they allow the (...) explanatory relationship between a new hypothesis and old evidence to be inductive, rather than deductive. In this paper, I argue that these solutions are either unsound or under-motivated, depending on the case of inductive explanation that we have in mind. This lends support to the broader claim that Garber–Style Bayesian confirmation cannot capture the sense in which new hypotheses that do not deductively imply old evidence nevertheless seem to be confirmed via old evidence. (shrink)

Recent literature in philosophy of science has addressed purported notions of explanatory virtues—‘explanatory power’, ‘unification’, and ‘coherence’. In each case, a probabilistic relation between a theory and data is said to measure the power of an explanation, or degree of unification, or degree of coherence. This essay argues that the measures do not capture cases that are paradigms of scientific explanation, that the available psychological evidence indicates that the measures do not capture judgements of explanatory power, and, finally, that the (...) measures do not provide useful methods for selecting hypotheses. 1. Introduction2. Some Proposed Measures of Explanatory Virtues3. Descriptive Inadequacy3.1 Excellent but false explanations3.2 Causal explanation4. Psychological Inadequacy5. Finding the Truth6. Conclusion. (shrink)

How good is an explanation and when is one explanation better than another? In this paper, I address these questions by exploring probabilistic measures of explanatory power in order to defend a particular Bayesian account of explanatory goodness. Critical to this discussion is a distinction between weak and strong measures of explanatory power due to Good (Br J Philos Sci 19:123–143, 1968). In particular, I argue that if one is interested in the overall goodness of an explanation, an appropriate balance (...) needs to be struck between the weak explanatory power and the complexity of a hypothesis. In light of this, I provide a new defence of a strong measure proposed by Good by providing new derivations of it, comparing it with other measures and exploring its connection with information, confirmation and explanatory virtues. Furthermore, Good really presented a family of strong measures, whereas I draw on a complexity criterion that favours a specific measure and hence provides a more precise way to quantify explanatory goodness. (shrink)

I propose a qualitative Bayesian account of explanatory goodness that is analogous to the Bayesian account of incremental confirmation. This is achieved by means of a complexity criterion according to which an explanation h is good if the reduction in the complexity of the explanandum e brought about by h (the explanatory gain) is greater than the additional complexity introduced by h in the context of e (the explanatory cost). To illustrate the account, I apply it in the context of (...) ad hoc hypotheses. (shrink)

Schupbach and Sprenger introduce a novel probabilistic approach to measuring the explanatory power that a given explanans exerts over a corresponding explanandum. Though we are sympathetic to their general approach, we argue that it does not adequately capture the way in which the causal explanatory power that c exerts on e varies with background knowledge. We then amend their approach so that it does capture this variance. Though our account of explanatory power is less ambitious than Schupbach and Sprenger’s in (...) the sense that it is limited to causal explanatory power, it is also more ambitious because we do not limit its domain to cases where c genuinely explains e. Instead, we claim that c causally explains e if and only if our account says that c explains e with some positive amount of causal explanatory power. (shrink)

Schupbach and Sprenger introduce a novel probabilistic approach to measuring the explanatory power that a given explanans exerts over a corresponding explanandum. Though we are sympathetic to their general approach, we argue that it does not adequately capture the way in which the causal explanatory power that c exerts on e varies with background knowledge. We then amend their approach so that it does capture this variance. Though our account of explanatory power is less ambitious than Schupbach and Sprenger’s in (...) the sense that it is limited to causal explanatory power, it is also more ambitious because we do not limit its domain to cases where c genuinely explains e. Instead, we claim that c causally explains e if and only if our account says that c explains e with some positive amount of causal explanatory power. 1Introduction 2The Logic of Explanatory Power 3Subjective and Nomic Distributions 3.1Actual degrees of belief 3.2The causal distribution 4Background Knowledge 4.1Conditionalization and colliders 4.2A helpful intervention 5Causal Explanatory Power 5.1The applicability of explanatory power 5.2Statistical relevance ≠ causal explanatory power 5.3Interventionist explanatory power 5.4E illustrated 6Conclusion. (shrink)

In this paper, we examine possible formalisations of the riskiness condition for hypothesis testing. First, we informally introduce derivability and riskiness as testing conditions together with the corresponding arguments for refutation and confirmation. Then, we distinguish two different senses of confirmation and focus our discussion on one of them with the aid of a historical example. In the remaining sections, we offer a brief overview of the main references to the risky condition in the literature and scrutinise different options for (...) formally capturing riskiness; we show why none of them works. We conclude with some remarks about the relation between derivability and riskiness and claim that riskiness essentially involves a contextual, pragmatic component that eludes a complete formal reconstruction. (shrink)

Probability ratio and likelihood ratio measures of inductive support and related notions have appeared as theoretical tools for probabilistic approaches in the philosophy of science, the psychology of reasoning, and artificial intelligence. In an effort of conceptual clarification, several authors have pursued axiomatic foundations for these two families of measures. Such results have been criticized, however, as relying on unduly demanding or poorly motivated mathematical assumptions. We provide two novel theorems showing that probability ratio and likelihood ratio measures can be (...) axiomatized in a way that overcomes these difficulties. (shrink)

The current state of inductive logic is puzzling. Survey presentations are recurrently offered and a very rich and extensive handbook was entirely dedicated to the topic just a few years ago [23]. Among the contributions to this very volume, however, one finds forceful arguments to the effect that inductive logic is not needed and that the belief in its existence is itself a misguided illusion , while other distinguished observers have eventually come to see at least the label as “slightly (...) antiquated” .What seems not to have lost any of its currency is the problem which inductive logic is meant to address. Inference from limited ascertained information to uncertain hypotheses is ubiquitous in learning, prediction and discovery. The logical insight that such kind of inference is fallible m .. (shrink)

This paper brings together results from the philosophy and the psychology of explanation to argue that there are multiple concepts of explanation in human psychology. Specifically, it is shown that pluralism about explanation coheres with the multiplicity of models of explanation available in the philosophy of science, and it is supported by evidence from the psychology of explanatory judgment. Focusing on the case of a norm of explanatory power, the paper concludes by responding to the worry that if there is (...) a plurality of concepts of explanation, one will not be able to normatively evaluate what counts as good explanation. (shrink)

Explanation is a central concept in human psychology. Drawing upon philosophical theories of explanation, psychologists have recently begun to examine the relationship between explanation, probability and causality. Our study advances this growing literature in the intersection of psychology and philosophy of science by systematically investigating how judgments of explanatory power are affected by the prior credibility of a potential explanation, the causal framing used to describe the explanation, the generalizability of the explanation, and its statistical relevance for the evidence. Collectively, (...) the results of our five experiments support the hypothesis that the prior credibility of a causal explanation plays a central role in explanatory reasoning: first, because of the presence of strong main effects on judgments of explanatory power, and second, because of the gate-keeping role it has for other factors. Highly credible explanations were not susceptible to causal framing effects. Instead, highly credible hypotheses were sensitive to the effects of factors which are usually considered relevant from a normative point of view: the generalizability of an explanation, and its statistical relevance for the evidence. These results advance current literature in the philosophy and psychology of explanation in three ways. First, they yield a more nuanced understanding of the determinants of judgments of explanatory power, and the interaction between these factors. Second, they illuminate the close relationship between prior beliefs and explanatory power. Third, they clarify the relationship between abductive and probabilistic reasoning. (shrink)

Jonah N. Schupbach and Jan Sprenger and Vincenzo Crupi and Katya Tentori have recently proposed measures of explanatory power and have shown that they are characterized by certain arguably desirable conditions or axioms. I further examine the properties of these two measures, and a third measure considered by I. J. Good and Timothy McGrew . This third measure also has an axiomatic representation. I consider a simple coin-tossing example in which only the Crupi–Tentori measure does not perform well. The Schupbach–Sprenger (...) and Good–McGrew measures are based on different notions of explanatory power, but both are tenable. 1 Introduction2 Measures of Explanatory Power3 The Schupbach–Sprenger Measure4 The Crupi–Tentori Measure5 The Good–McGrew Measure6 Affirmative versus Comprehensive Explanatory Power7 Coin-Flipping Examples8 Decomposition9 Conclusion. (shrink)

Jonah N. Schupbach and Jan Sprenger have proposed conditions of adequacy for measures of explanatory power. They derive and defend a measure of explanatory power satisfying their conditions of adequacy. This article furthers the development of their measure. The requirement that the measure be multidimensional analytic is avoided. Several proofs are simplified, and gaps in proofs are filled.

Recent years have witnessed an effort to explicate the concept of explanatory power in a Bayesian framework by constructing explanatory measures. It has been argued that those measures should not violate the principle of explanatory justice, which states that explanatory power cannot be extended “for free.” I argue, by formal means, that one recent measure claiming to be immune from explanatory injustice fails to be so. I end by concluding that the explanatory justice criticism can be dissolved, given a natural (...) interpretation of the concept of negative explanatory power. (shrink)

Defenders of Inference to the Best Explanation claim that explanatory factors should play an important role in empirical inference. They disagree, however, about how exactly to formulate this role. In particular, they disagree about whether to formulate IBE as an inference rule for full beliefs or for degrees of belief, as well as how a rule for degrees of belief should relate to Bayesianism. In this essay I advance a new argument against non-Bayesian versions of IBE. My argument focuses on (...) cases in which we are concerned with multiple levels of explanation of some phenomenon. I show that in many such cases, following IBE as an inference rule for full beliefs leads to deductively inconsistent beliefs, and following IBE as a non-Bayesian updating rule for degrees of belief leads to probabilistically incoherent degrees of belief. (shrink)

The paper investigates measures of explanatory power and how to define the inference schema “Inference to the Best Explanation”. It argues that these measures can also be used to quantify the systematic power of a hypothesis and the inference schema “Inference to the Best Systematization” is defined. It demonstrates that systematic power is a fruitful criterion for theory choice and IBS is truth-conducive. It also shows that even radical Bayesians must admit that systemic power is an integral component of Bayesian (...) reasoning. Finally, the paper puts the achieved results in perspective with van Fraassen’s famous criticism of IBE. (shrink)

The present paper investigates the first step of rational belief acquisition. It, thus, focuses on justificatory relations between perceptual experiences and perceptual beliefs, and between their contents, respectively. In particular, the paper aims at outlining how it is possible to reason from the content of perceptual experiences to the content of perceptual beliefs. The paper thereby approaches this aim by combining a formal epistemology perspective with an eye towards recent advances in philosophy of cognition. Furthermore the paper restricts its focus, (...) it concentrates on the case of color perception and perceptual beliefs about color. (shrink)

Probabilistic dependence and independence are among the key concepts of Bayesian epistemology. This paper focuses on the study of one specific quantitative notion of probabilistic dependence. More specifically, section 1 introduces Keynes’s coefficient of dependence and shows how it is related to pivotal aspects of scientific reasoning such as confirmation, coherence, the explanatory and unificatory power of theories, and the diversity of evidence. The intimate connection between Keynes’s coefficient of dependence and scientific reasoning raises the question of how Keynes’s coefficient (...) of dependence is related to truth, and how it can be made fruitful for epistemological considerations. This question is answered in section 2 of the paper. Section 3 outlines the consequences the results have for epistemology and the philosophy of science from a Bayesian point of view. (shrink)

This dissertation aims to determine the optimal level of granularity for the variables used in probabilistic causal models. These causal models are useful for generating explanations in a number of scientific contexts. In Chapter 1, I argue that there is rarely a unique level of granularity at which a given phenomenon can be causally explained, thereby rejecting various causal exclusion arguments. In Chapter 2, I consider several recent proposals for measuring the explanatory power of causal explanations, and show that these (...) measures fail to track the comparative depth of explanations given at different levels of granularity. In Chapter 3, I offer a pragmatic account of how to partition the measure space of a causal variable so as to optimally explain its effect. My account uses the decision-theoretic notion of value of information, and indexes the relative depth of an explanation to a particular agent faced with a particular decision problem. Chapter 4 applies this same decisiontheoretic framework to answer the epistemic question of how to discover constitutive relationships in nature. In Chapter 5, I describe the formal details of the relationship between random variables that are meant to be coarse-grained and fine-grained representations of the same type of phenomenon. I use this formal framework to rebut a popular argument for the view that special science probabilities can be objective chances. Chapter 6 discusses challenges related to the causal interpretation of Bayes nets that use imprecise rather than precise probabilities. (shrink)

The aim of this thesis is to improve our understanding of how to assess and communicate uncertainty in areas of research deeply afflicted by it, the assessment and communication of which are made more fraught still by the studies’ immediate policy implications. The IPCC is my case study throughout the thesis, which consists of three parts. In Part 1, I offer a thorough diagnosis of conceptual problems faced by the IPCC uncertainty framework. The main problem I discuss is the persistent (...) ambiguity surrounding the concepts of ‘confidence’ and ‘likelihood’; I argue that the lack of a conceptually valid interpretation of these concepts compatible with the IPCC uncertainty guide’s recommendations has worrying implications for both the IPCC authors’ treatment of uncertainties and the interpretability of the information provided in the AR5. Finally, I show that an understanding of the reasons behind the IPCC’s decision to include two uncertainty scales can offer insights into the nature of this problem. In Part 2, I review what philosophers have said about model-based robustness analysis. I assess several arguments that have been offered for its epistemic import and relate this discussion to the context of climate model ensembles. I also discuss various measures of independence in the climate literature, and assess the extent to which these measures can help evaluate the epistemic import of model robustness. In Part 3, I explore the notion of the ‘weight of evidence’ typically associated with Keynes. I argue that the Bayesian is bound to struggle with this notion, and draw some lessons from this fact. Finally, I critically assess some recent proposals for a new IPCC uncertainty framework that significantly depart from the current one. (shrink)

Hempel’s Converse Consequence Condition (CCC), Entailment Condition (EC), and Special Consequence Condition (SCC) have some prima facie plausibility when taken individually. Hempel, though, shows that they have no plausibility when taken together, for together they entail that E confirms H for any propositions E and H. This is “Hempel’s paradox”. It turns out that Hempel’s argument would fail if one or more of CCC, EC, and SCC were modified in terms of explanation. This opens up the possibility that Hempel’s paradox (...) can be solved by modifying one or more of CCC, EC, and SCC in terms of explanation. I explore this possibility by modifying CCC and SCC in terms of explanation and considering whether CCC and SCC so modified are correct. I also relate that possibility to Inference to the Best Explanation. (shrink)

One of the main goals of scientific research is to provide a description of the empirical data which is as accurate and comprehensive as possible, while relying on as few and simple assumptions as possible. In this paper, I propose a definition of the notion of few and simple assumptions that is not affected by known problems. This leads to the introduction of a simple model of scientific progress that is based only on empirical accuracy and conciseness. An essential point (...) in this task is the understanding of the role played by measurability in the formulation of a scientific theory. This is the key to prevent artificially concise formulations. The model is confronted here with many possible objections and with challenging cases of real progress. Although I cannot exclude that the model might have some limitations, it includes all the cases of genuine progress examined here, and no spurious one. In this model, I stress the role of the state of the art, which is the collection of all the theories that are the only legitimate source of scientific predictions. Progress is a global upgrade of the state of the art. (shrink)