Entropy, Information, and the Updating of Probabilities

This paper is a review of a particular approach to the method of maximum entropy as a general framework for inference. The discussion emphasizes pragmatic elements in the derivation. An epistemic notion of information is defined in terms of its relation to the Bayesian beliefs of ideally rational agents. The method of updating from a prior to posterior probability distribution is designed through an eliminative induction process. The logarithmic relative entropy is singled out as a unique tool for updating (a) that is of universal applicability, (b) that recognizes the value of prior information, and (c) that recognizes the privileged role played by the notion of independence in science. The resulting framework—the ME method—can handle arbitrary priors and arbitrary constraints. It includes the MaxEnt and Bayes’ rules as special cases and, therefore, unifies entropic and Bayesian methods into a single general inference scheme. The ME method goes beyond the mere selection of a single posterior, and also addresses the question of how much less probable other distributions might be, which provides a direct bridge to the theories of fluctuations and large deviations.

Probabilistic truthlikeness, content elements, and meta-inductive probability optimization

The paper starts with the distinction between conjunction-of-parts accounts and disjunction-of-possibilities accounts to truthlikeness (Sects. 1, 2). In Sect. 3, three distinctions between kinds of truthlikeness measures (t-measures) are introduced: (i) comparative versus numeric t-measures, (ii) t-measures for qualitative versus quantitative theories, and (iii) t-measures for deterministic versus probabilistic truth. These three kinds of truthlikeness are explicated and developed within a version of conjunctive part accounts based on content elements (Sects. 4, 5). The focus lies on measures of probabilistic truthlikeness, that are divided into t-measures for statistical probabilities and single case probabilities (Sect. 4). The logical notion of probabilistic truthlikeness (evaluated relative to true probabilistic laws) can be treated as a subcase of deterministic truthlikeness for quantitative theories (Sects. 4–6). In contrast, the epistemic notion of probabilistic truthlikeness (evaluated relative to given empirical evidence) creates genuinely new problems, especially for hypotheses about single case probabilities that are evaluated not by comparison to observed frequencies (as statistical probabilities), but by comparison to the truth values of single event statements (Sect. 6). By the method of meta-induction, competing theories about single case probabilities can be aggregated into a combined theory with optimal predictive success and epistemic truthlikeness (Sect. 7).

Demonstration and the Indemonstrability of the Stoic Indemonstrables

Since Mates’ seminal Stoic Logic there has been uncertainty and debate about how to treat the term anapodeiktos when used of Stoic syllogisms. This paper argues that the customary translation of anapodeiktos by ‘indemonstrable’ is accurate, and it explains why this is so. At the heart of the explanation is an argument that, contrary to what is commonly assumed, indemonstrability is rooted in the generic account of the Stoic epistemic notion of demonstration (apodeixis). Some minor insights into Stoic logic ensue.

The Verisimilitudinarian approach to ‘the Truth’

The Verisimilitudinarian approach to scientific progress (VS, for short) is traditionally considered a realist-correspondist model to explain the proximity of our best scientific theories to the way things really are in the world out there (ʻthe Truthʻ, with the capital ʻtʻ). However, VS is based on notions, such as ʻestimated verisimilitudeʻ or ʻapproximate truthʻ, that dilute the model in a functionalist-like theory. My thesis, then, is that VS tries to incorporate notions, such as ʻprogressʻ, in a pre-constituted metaphysical conception of the world, but fails in providing a fitting framework. The main argument that I will develop to support this claim is that the notions that they use to explain scientific progress (ʻestimated verisimilitudeʻ or ʻapproximate truthʻ) have nothing to do with ʻthe Truthʻ. After presenting Cevolani and Tamboloʻs answer (2013) to Birdʻs arguments (2007), I will claim that VS sacrifices the realist-correspondist truth in favor of an epistemic notion of truth, which can obviously be compatible with certain kinds of realism but not with the one the authors have in mind (the correspondence between our theories and the way things really are).

Evidentiality

This chapter surveys the different ways in which evidentiality is conceived of and referred to. It first outlines the history of the term evidentiality and other terms that have been associated with evidentiality. It then discusses different definitions and ways of understanding evidentiality. Most often evidentiality is defined in terms of the notion of information source or related notions. This notion has been understood as an epistemic notion, as a deictic or grounding notion and/or as having subjective and intersubjective aspects. Subsequently, the chapter discusses three distinctions that are significant to understanding information source and evidentiality, viz. distinctions pertaining to coding (coded versus non-coded), way of encoding (grammatical versus lexical encoding), and discourse prominence (discursively secondary versus discursively primary status).

The Long Goodbye from Relativism

Chapter 4 rethinks Protagoras’s Measure Doctrine from a non-relativist, realist perspective. It coins the term Measure Realism, according to which human beings are the measure of what is good for them. This proposal is developed via a series of arguments against contemporary Truth Relativism and against a purely epistemic notion of standards of assessment. Human beings, it is argued, are the measure in a metaphysical way: qua the beings we are, we are the standard that is needed in ethical theory. It also makes sense to ask “what is the good for animals?” and “what is the good for the universe?” And it makes sense to imagine better cognizers than us, aiming to supersede what appear to be our limitations. Still, ethical theorizing examines how we are to live as human beings, which includes concern with the good of other beings and how we should aim to be better cognizers and agents.

Speaking like “us”: self- and other-categorization as Norwegian speakers in student interactions

This article investigates the manner in which students in Norwegian upper secondary schools negotiate “legitimate speaker” category membership in spoken interactions. It is concerned with adolescents’ linguistic identity as Norwegian speakers, and the language ideologies that inform categorization practices. Three collaborative features are distinguished to identify membership categorization practices: the pragmatic notion of scales of time and place; pronouns, in particular the personal pronouns “we”, “us”, and “they”; and the socio-epistemic notion of object-side assessments. The article asks whether the students’ categories correspond to those of education policy and research, and finds that students operate with more complex membership categorizations than the binary Norwegian/non-Norwegian division found in applied linguistic research and policy documents. It also finds that status as “expert” speaker of Norwegian is negotiable in student interactions, where social factors such as interpersonal relations, perceived social differences and performance of category-bound activities are as important as linguistic factors. The article shows how both non-native and native speakers of Norwegian can discursively construct themselves and others as more or less expert speakers of the language. As such, it contributes with insights on how institutionally defined linguistic markets, linguistic capital and legitimate speakers can be contested at the local level of student interactions.