European Journal for Philosophy of Science最新文献

Research about the evolution of morality suffers from the lack of a clear, agreed-upon concept of morality. In response to this, recent accounts have become increasingly pluralist and pragmatic. In this paper, I argue that 1) both the concept of morality and the broader understanding of what makes us moral include ethical and metaethical assumptions; 2) there is no uncontroversial descriptive notion available, and therefore settling on a particular concept inevitably entails such assumptions; and 3) what is lacking is a reflection on the role that ethical and metaethical assumptions play, suggesting that the debate would benefit from making them explicit. Claims about “the true origin of morality” can fruitfully be analyzed as “mixed claims”: claims that combine a causal-historical hypothesis (e.g., about the evolution of a certain ability, such as empathy or joint intentionality) with ethical or metaethical assumptions about which abilities or norms make us moral. Making such assumptions explicit advances the epistemic aims of transparency and comparability, and thereby helps to avoid rash conclusions regarding, for instance, the nature of moral progress. Finally, it helps to unpack the normative knowledge shared by behavioral scientists and comparative psychologists and to give this knowledge its proper place in research.

The extraterrestrial hypothesis (ETH), the hypothesis that an extraterrestrial civilization (ETC) is active on Earth today, is taboo in academia, but the assumptions behind this taboo are faulty. Advances in biology have rendered the notion that complex life is rare in our Galaxy improbable. The objection that no ETC would come to Earth to hide from us does not consider all possible alien motives or means. For an advanced ETC, the convergent instrumental goals of all rational agents – self-preservation and the acquisition of resources – would support the objectives of removing existential threats and gathering strategic and non-strategic information. It could advance these objectives by proactively gathering information about and from inhabited planets, concealing itself while doing so, and terminating potential rivals before they become imminently dangerous. Other hypotheses of ETC behavior, including the zoo/interdict hypothesis and the dark forest hypothesis also undercut the claim that the ETH is highly improbable, and the ETH overturns none of our well-tested scientific knowledge. It follows that evidence offered in its support need not be extraordinary. The fact that most reports of unidentified anomalous phenomena (UAP) have natural or human explanations does not count against the ETH. Inference to the best explanation offers a way to find evidence for the hypothesis and some evidence exists, some of it taking the form of reliable witness reports. The most plausible alternative explanation for some UAP declines in probability over time. A hypothesis that does not contradict well-established facts or theories, is not highly improbable for other reasons, and explains otherwise unexplained evidence is a rational hypothesis. Since the ETH meets this test, it should be evaluated alongside other possibilities when the case-specific evidence warrants it.

Critics frequently target Ernest Nagel’s model of reduction for its purported inadequacy in addressing the issue of approximation. In response, proponents of Nagel’s model have integrated approximations into the more comprehensive Generalized Nagel-Schaffner model, or the GNS model. However, this article contends that the pertinent criticisms and responses are both misplaced: There are no barriers to Nagel’s model incorporating approximations, and it assimilates them in a manner distinctly dissimilar to the approach of the GNS model. Indeed, Nagel’s model is fundamentally static, providing invariant formal and informal conditions for reduction; and it is the dynamic history of science that fulfills relevant conditions for reduction, thereby achieving different degrees of reductive success. Consequently, approximations are essentially extraneous to Nagel’s model, since they pertain chiefly to temporally based scientific knowledge and concern merely the means (e.g., approximate or exact deductions) of fulfilling the conditions for reduction within specific problem contexts. This article also develops a Nagelian treatment of approximation, demonstrating how distinctive types of approximations aid in fulfilling various conditions for reduction and contribute to different degrees of reductive success. Two case studies are presented to illustrate the Nagelian treatment: The first examines the specific heat ratio anomaly within the traditional endeavor of reducing thermodynamics to statistical mechanics, and the second addresses the consumption puzzle in the context of reducing macroeconomics to microeconomics.

Popper’s (1983, 2002) philosophy of science has enjoyed something of a renaissance in the wake of the replication crisis, offering a philosophical basis for the ensuing science reform movement. However, adherence to Popper’s approach may also be at least partly responsible for the sense of “crisis” that has developed following multiple unexpected replication failures. In this article, I contrast Popper’s approach with that of Lakatos (1978) as well as with a related but problematic approach called naïve methodological falsificationism (NMF; Lakatos, 1978). The Popperian approach is powerful because it is based on logical refutations, but its theories are noncausal and, therefore, potentially lacking in scientific value. In contrast, the Lakatosian approach considers causal theories, but it concedes that these theories are not logically refutable. Finally, NMF represents a hybrid approach that subjects Lakatosian causal theories to Popperian logical refutations. However, its tactic of temporarily accepting a ceteris paribus clause during theory testing may be viewed as scientifically inappropriate, epistemically inconsistent, and “completely redundant” (Lakatos, 1978, p. 40). I conclude that the replication “crisis” makes the most sense in the context of the Popperian and NMF approaches because it is only in these two approaches that the failure to replicate a previously corroborated theory represents a logical refutation of that theory. In contrast, such replication failures are less problematic in the Lakatosian approach because they do not logically refute theories. Indeed, in the Lakatosian approach, replication failures can be temporarily ignored or used to motivate theory development.

This article mainly investigates whether common Bayesian confirmation measures are affected by stopping rules. The results indicate that difference measure d, log-ratio measure r, and log-likelihood measure l are not affected by non-informative stopping rules, but affected by informative stopping rules. In contrast, Carnap measure (tau ), normalized difference measure n, and Mortimer measure m are affected by (non-)informative stopping rules sometimes but sometimes aren’t. Besides, we use two examples to further illustrate that confirmation measures d, r, and l are better than (tau ,n), and m.

Despite its being one of Roger Penrose’s greatest contributions to spacetime physics, there is a dearth of philosophical literature on twistor theory. The one exception to this is Bain (2006)—but although excellent, there remains much to be said on the foundations and philosophy of twistor theory. In this article, we (a) present for philosophers an introduction to twistor theory, (b) consider how the spacetime–twistor correspondence interacts with the philosophical literature on theoretical equivalence, and (c) explore the bearing which twistor theory might have on philosophical issues such as the status of dynamics, the geometrisation of physics, spacetime ontology, the emergence of spacetime, and symmetry-to-reality inferences. We close with an elaboration of a variety of further opportunities for philosophical investigation into twistor theory.

A growing number of linguistic attempts to explain how languages change use cultural-evolutionary models involving selection or drift. Developmental constraints and biases, which take center stage in evolutionary developmental biology or evo-devo, seem to be absent within this framework, even though linguistics is home to numerous notions of constraint. In this paper, we show how these evo-devo concepts could be applied to linguistic change and why they should. This requires some conceptual groundwork, due to important differences between linguistic and biotic evolution. In biological evolution, development generates the organism's variable traits on which selection and drift act. In linguistic evolution by analogy, we say development generates the linguistic variants on which selection and drift can act. “Linguistic development” then picks out how individual speakers produce and comprehend language. It involves much more than just learning. Using this broad notion of development, we distinguish between different types of bias that could operate in the processes of linguistic innovation and transmission, which correspond to genetic mutation and biological reproduction, respectively. Having thus sharpened our conceptual toolbox, we then reanalyze two well-documented cases of linguistic change and show that, in both these cases, linguists have only considered Neo-Darwinian evolutionary explanations, falsely deploying an exclusive disjunction of selection and drift. We show that there is at least a third relevant alternative in these examples, namely developmental constraint or bias in the sense we explicate here.

There is a complex interplay between the models in dark matter detection experiments that have led to a difficulty in interpreting the results of the experiments and ascertain whether we have detected the particle or not. The aim of this paper is to categorise and explore the different models used in said experiments, by emphasizing the distinctions and dependencies among different types of models used in this field. With a background theory, models are categorised into four distinct types: background theory, theoretical, phenomenological, experimental and data. This taxonomy highlights how each model serves a unique purpose and operates under varying degrees of independence from their respective frameworks. A key focus is on the experimental model, which is shown to rely on constraints from both data and phenomenological ones. The article argues that while theoretical models provide a backdrop for understanding the nature of dark matter, the experimental models must stand independently, particularly in their methodological approaches. This is done via a discussion of the inherent challenges in dark matter detection, such as inconsistent results and difficulties in cross-comparison, stemming from the diverse modelling approaches.

Guiding principles are central to theory development in physics, especially when there is only limited empirical input available. Here I propose an approach to such principles looking at their heuristic role. I suggest a distinction between two modes of employing scientific principles. Principles of nature make descriptive claims about objects of inquiry, and principles of epistemic action give directives for further research. If a principle is employed as a guiding principle, then its use integrates both modes of employment: guiding principles imply descriptive claims, and they provide directives for further research. By discussing the correspondence principle and the naturalness principle as examples, I explore the consequences for understanding and evaluating current guiding principles in physics. Like principles of nature, guiding principles are evaluated regarding their descriptive implications about the research object. Like principles of epistemic action, guiding principles are evaluated regarding their ability to respond to context-specific needs of the epistemic agent.

In this paper, I explore the relation between actual scientific practice and conceptual interpretation of scientific theories by investigating the particle concept in non-relativistic quantum mechanics (NRQM). On the one hand, philosophers have raised various objections against the particle concept within the context of NRQM and proposed alternative ontologies such as wave function realism, Bohmian particles, mass density field, and flashes based on different realist solutions to the measurement problem. On the other hand, scientists continue to communicate, reason, and explain experimental phenomena using particle terms in the relevant regimes. It has been explicitly argued and, for most of the time, implicitly assumed in the philosophical literature that we do not need to take scientists’ particle talk seriously, and recovering position measurement of particles in our ontological accounts is sufficient to make contact with scientific practice. In this paper, I argue that although scientific discourse does not postulate a uniform and coherent ontology, it nevertheless postulates real properties. Our ontological accounts thus need to recover the various properties associated with the NRQM particle concept in scientific discourse. I show that recovering these particle properties is not trivially achievable by pointing out some particular challenges these revisionary ontologies face in the process.