One common approach to solve multi-objective reinforcement learning (MORL) problems is to extend conventional Q-learning by using vector Q-values in combination with a utility function. However issues can arise with this approach in the context of stochastic environments, particularly when optimising for the scalarised expected reward (SER) criterion. This paper extends prior research, providing a detailed examination of the factors influencing the frequency with which value-based MORL Q-learning algorithms learn the SER-optimal policy for an environment with stochastic state transitions. We empirically examine several variations of the core multi-objective Q-learning algorithm as well as reward engineering approaches and demonstrate the limitations of these methods. In particular, we highlight the critical impact of the noisy Q-value estimates issue on the stability and convergence of these algorithms.

In this article, we examine the determinants of citizens’ democratic preferences in federal states with politically significant national or linguistic diversity. Using original survey data from Belgium, Canada and Switzerland, we test whether members of national or linguistic minorities prefer different (electoral, direct or deliberative) forms of decision-making than majority members – since some give advantage to them more than others. While we find effects of citizens’ objective and subjective minority-majority position on their democratic preferences, individual-level predictors such as satisfaction with the current functioning of democracy, economic well-being and political ideology remain at least as strong predictors. These findings enrich the literatures on democratic fatigue, reform and innovation by showing that even in states with significant national-linguistic diversity, democratic preferences seem to transcend communities, indicating room for cross-group consensus. Yet, since group-level factors have some relevance, democratic reforms need to pay attention to them to be inclusive of all societal segments.

Over the past two decades, political science has produced varied examples of ethnographic approaches. These approaches have not only tackled epistemological dilemmas but also exposed a second methodological dimension of ethnographic practice that is not so systematically explored: the relationship of researchers to their research participants. In this article, we focus on this second dimension, using emblematic texts in political science, especially in comparative politics, to develop a fourfold typology of political ethnographies that takes into account the emotional dynamics between researchers and participants. We use this typology to analyze various gradations through which these emotional dynamics develop in fieldwork. Focusing on the navigation between distance and proximity that these dynamics entail, we propose the concept of “emotional proximity” to capture relations between the researcher and the participant. We investigate the validity of this novel typology by applying it to ethnographic studies of far-right actors. The political distance separating researchers and participants in these studies allows us to investigate the methodologies of disconnecting political from emotional dynamics across this fourfold schema of ethnographic varieties. We argue that the “infidelity” of emotional distance (instead of proximity) is not an objectivist epistemological necessity but a methodological tool that is indispensable to the practice of participant observation.

In natural speech, phonetic cues that distinguish lexical items can be hyperarticulated when there is a minimal pair competitor, a process known as contrastive hyperarticulation. Building upon prior work, this article examines the cue-specific nature of contrastive hyperarticulation in Japanese, focussing on stop voice onset time (VOT) using a speech corpus. We confirmed that the existence of a voicing minimal pair competitor in the lexicon triggers hyperarticulation of VOT duration on the target segment (shorter for voiced stops and longer for voiceless stops), while other contrasts (singleton vs. geminate) did not. The results also suggest that contrastive hyperarticulation (a) is more compatible with casual speech than slow/clear speech, (b) is sensitive to position in a word (greater in word-initial position than in non-initial position) and (c) applies to a greater degree in Japanese than in English due to properties of stops. This provides further evidence that the phonetic specificity of contrastive hyperarticulation is cross-linguistically relevant.

Tell settlements often provide a unique window into prehistoric lifeways due to remarkable preservation and safeguarding from modern disturbances. Vésztő-Mágor in Hungary is one such tell with stratigraphy, features and finds that reflect thousands of years of prehistoric settlement. In 2021, the Vésztő-Mágor Conservation and Exhibition Program began the work of stabilizing, documenting and preserving prehistoric deposits, features and artefacts exposed in an in situ exhibition trench at Vésztő-Mágor. In the process, an exceptionally well-preserved carbonized item was discovered embedded in a series of Middle Bronze Age house floors. We describe the object and context of discovery, and interpret it as matting inside a wattle-and-daub house. We expand our discussion to similar contexts known from Vésztő-Mágor, in the Carpathian Basin, and beyond, to highlight the technologies involving organic materials used at prehistoric tell sites and their significance for understanding lifeways at these settlements.

In this paper, we generalise to the family of Fermat quartics $X^4 + Y^4 = 2^m, m \in \mathbb {Z}$, a result of Aigner [‘Über die Möglichkeit von $x^4 + y^4 = z^4$ in quadratischen Körpern’, Jahresber. Deutsch. Math.-Ver. 43 (1934), 226–228], which proves that there is only one quadratic field, namely $\mathbb {Q}(\sqrt {-7})$, that contains solutions to the Fermat quartic $X^4 + Y^4 = 1$. The $m \equiv 0 \pmod 4$ case is due to Aigner. The $m \equiv 2 \pmod 4$ case follows from a result of Emory [‘The Diophantine equation $X^4 + Y^4 = D^2Z^4$ in quadratic fields’, Integers 12 (2012), Article no. A65, 8 pages]. This paper focuses on the two cases $m \equiv 1, 3 \pmod 4$, classifying for $m \equiv 1 \pmod 4$ the infinitely many quadratic number fields that contain solutions, and proving for $m \equiv 3 \pmod 4$ that $\mathbb {Q}(\sqrt {2})$ and $\mathbb {Q}(\sqrt {-2})$ are the only quadratic number fields that contain solutions.

Existing measures of Actively Open-Minded Thinking (AOT) primarily assess the acceptance of rational thinking norms and standards, rather than actual thinking and resulting behavior. These scales can be susceptible to impression management, often yield inflated scores, and may not accurately capture how individuals think in real-life contexts. To address these limitations, we developed and validated a novel Situational Judgment Test for Actively Open-Minded Thinking (AOT-SJT), designed to assess behavioral tendencies related to AOT in realistic scenarios. AOT is conceptualized as the disposition to consider alternative viewpoints, seek disconfirming evidence, and revise beliefs in light of new information. Across 4 studies, we constructed and refined the AOT-SJT using scenarios that simulate everyday decision-making. In Study 1, we tested initial items among Croatian participants, resulting in a 13-item measure with solid psychometric properties. Study 2 confirmed the test’s convergent validity with cognitive and personality constructs and its predictive power for different forms of rational thinking. In Study 3, new items were introduced to enhance construct coverage, particularly around evidence search direction. Study 4 extended validation to an English-speaking sample, supporting cross-linguistic applicability, although effect sizes related to convergent validity were somewhat lower than before. Findings across studies show that the AOT-SJT aligns with theoretical expectations, demonstrates solid convergent validity with existing AOT scales, and effectively distinguishes levels of open-mindedness. By measuring behavioral intentions rather than standards acceptance, the AOT-SJT offers an externally valid assessment of AOT.

According to the traditional conception of epistemic responsibilist virtue, virtuous traits are evaluatively and conceptually primary. Virtuous acts are derived from, or proceed from, virtue. If an agent performs an act based on good epistemic motives but does not possess the virtue, it is merely an act that a virtuous agent would perform, but is not itself virtuous. In contrast, according to the view I propose and call Act First Responsibilism (AFR), virtuous acts are evaluatively and conceptually primary. Acts based on good motives are virtuous regardless of whether they issue from a virtuous trait. The possession of a virtuous trait is defined derivatively in terms of the performance of virtuous acts. With the new dispositional model of virtues I propose, AFR provides the metaphysical foundation for taking virtuous acts as evaluatively and conceptually primary. I argue that AFR is not normatively slight, meets the demands of an exemplarist virtue theory, and fares better than the traditional view in explaining, predicting, understanding, and evaluating an agent, as well as considering someone worthy of imitation.

The breakdown of liberal hegemony, the rise of the New Right, and the violent realignment of world order have been accompanied by a retreat from traditional humanist concerns in critical international theory, including emancipation, political subjectivity, social totality, universal history, and the anticipatory-utopian dimension of critique. Scholars have identified numerous shortcomings in first-generation and contemporary critical International Relations (IR), and our discipline still questions its purpose and object of study. This article proposes a more radical and realistic approach to critical international theory based on a reappraisal of Andrew Linklater’s oeuvre. It frames the critical project in IR as a Lakatosian research programme and calls for a progressive problem shift that foregrounds what Linklater, drawing from Kant and Marx, calls the necessarily tripartite structure of critical theory. I argue that by tracing an alternative path through classical sources of the tradition, pivoting from Hegel and the deep social relationalism that follows, while integrating a tripartite commitment with a more rigorous reflexive methodology, we can revitalise the emancipatory approach to IR and provide renewed purpose and direction to the discipline. Grounded in a left-Hegelian tradition of thought, the argument aspires to resonate with other critical theoretical traditions both within and beyond IR.

The Avicennan text De congelatione et conglutinatione lapidum had a great influence on the alchemical thought of the thirteenth century. This Latin text disputed both the veracity of alchemy and the possibility of alchemical transmutation by arguing that art is inferior to nature and that the alchemists cannot manipulate a metal because its true characteristics are hidden from our senses; thus an alchemist cannot change something which is unknown to him. Newman’s pioneer studies examined the diffusion and impact of the first Avicennan argument on medieval alchemy and he shed light on the art-versus-nature debate. This paper has a twofold aim: on the one hand it aims to further Newman’s study by focusing on the second Avicennan argument, which is closely related to the problem of substantial form, and on the other hand it aims to show how the aforesaid problem paved the way for the emergence of corpuscularianism, which flourished during the early modern period. In this regard, it will become clear that the historiographical case of alchemy and its problem of substantial form can serve as an exegetical tool for ‘bridging’ the Middle Ages and the early modern period with respect to the relation between Aristotelianism and corpuscularianism.

When we learn that someone holds irrational beliefs, we often respond by reducing our epistemic trust in them. In this paper, I will propose a novel account of such trust reductions. The recently popular relationship-modification account (RMA) of epistemic blame will serve as a foil for this project. RMA says that epistemically blaming others for their epistemic failings involves modifying our epistemic relationships with them, paradigmatically via a reduction of epistemic trust. RMA has recently faced two challenges of extensional inadequacy, which I will show result from a mistaken view about what type of response trust reductions are. I will draw on resources from legal theory to show that trust reductions bear all the hallmarks of so-called non-punitive measures, which serve preventative collective purposes. I will argue for an account of trust reductions as non-punitive epistemic measures that serve the purpose of preventing unreliable informants from negatively affecting the epistemic commons. My account explains when and why it is appropriate to reduce epistemic trust and shows where RMA goes wrong.

This article examines the financial strategies employed by Genoese businessmen in Atlantic Castile during the late fifteenth and early sixteenth centuries, focusing on the relative value of private enterprises. It reveals how the interplay between Castilian market characteristics and the unique financial culture of Genoese merchants predominantly supported short-term and transient economic ventures. These methods significantly reduced the need for centralised corporate structures and redundant accounting procedures, supported by a robust notarial system. Consequently, trading in securities and credit extensions became prominent features of this financial market. The research demonstrates that the widespread use of account money, credit and bills of exchange allowed Ligurians to engage in trade and access markets with flexibility and agility through quantitative and functional economic analyses. It also highlights the importance of various credit arrangements, such as loans, bills of trade, or IOUs, which extend beyond merely deferring payments.

Following the announcement of lunar programs by the US and China in recent years, a new round of lunar competition has commenced. In the absence of effective international regimes, there is a pressing need to develop legal measures to preserve lunar security and safety. Rules on demilitarization and deconfliction are particularly crucial. The 1967 Outer Space Treaty establishes the principles of ‘peaceful purposes’ and ‘exclusively for peaceful purposes’, but their interpretation remains contested. The rise of cislunar space strategies further complicates the application of these principles. Meanwhile, for managing daily operations on the lunar surface, the US-led Artemis Accords propose safety zones to prevent harmful interference. However, non-signatories like China and Russia are not bound by these provisions. As a significant player in the lunar competition, China should contribute to the establishment and improvement of legal measures to ensure lunar security and safety. For lunar demilitarization, China should refine the PPWT draft to address the specific security demands of cislunar space and the Moon. Concurrently, China should actively participate in formulating norms, rules, and principles of responsible behavior to mitigate threats in cislunar space and on the Moon until a universally accepted legally binding treaty is achieved. For lunar deconfliction, considering that the creation of safety zones may become the earliest practice to fulfill the obligation to avoid harmful interference, China needs to coordinate with the US to negotiate detailed rules on the size, scope, nature, behavior, and dispute settlement measures related to safety zones.