Conciliationism holds that it is rational to modify one’s beliefs in the face of disagreement. But extant conciliatory norms yield incorrect results in cases involving excessive confidence—cases in which one’s interlocutors are sure, or almost sure, that their opinions are correct. After explaining the problem of excessive confidence, I show that a Bayesian approach to Conciliationism handles the problem elegantly and effectively. Further, it has desirable—indeed essential—features, including the ability to (i) contend with multiple interlocutors, (ii) deal with gradations in competence, (iii) incorporate ubiquitous interdependence, and (iv) account for critical contextual features of cases of disagreement.

Recent governments, both in the UK and internationally, have increasingly used their power to attempt to alter the behaviour of people in receipt of social security benefits. This can be seen in the case of the UK’s benefit cap, a policy introduced with the specific goal of changing behaviour by capping social security support at the household level. Alongside promoting transitions into employment, there was also a focus on encouraging households to move to cheaper accommodation, something which was portrayed as achievable by those defending the policy. Drawing on case studies from qualitative longitudinal research with parents affected by the benefit cap, this article demonstrates that individuals are, in fact, relatively powerless to change their housing situations, which are routinely already overcrowded and of poor quality, even where rents are very high. Instead, affected households experience state-imposed hardship. We problematise both the cap itself and the governmental narrative that knowingly ascribes social security recipients with a power they do not have.

Recent years have seen record numbers of applications to UK psychiatry training, yet consultant vacancies remain high and substantial workforce gaps persist. This contradiction reflects a growing recruitment–retention paradox: increasing pressure at the point of entry has not translated into sustainable workforce capacity. This feature introduces the pressurised leaky funnel, a systems-based conceptual model that reframes the psychiatry workforce as a pathway shaped by upstream recruitment pressures and downstream attrition across five stages: exposure and intent, application, selection, training environment and career outcomes. Drawing on established workforce models and educational psychology theory, the model explains how application volume can expand while misalignment, motivational erosion and identity strain drive cumulative workforce loss across the pipeline. We argue that recruitment, selection and retention should not be treated as separate policy domains but understood as interacting components of a single system. By linking where doctors enter psychiatry with how commitment is sustained or eroded, the model offers a framework for moving beyond short-term recruitment metrics towards progression, retention and long-term workforce sustainability, while highlighting new opportunities for selection reform, training environment redesign and retention-focused workforce planning.

Generative artificial intelligence (GenAI) is a powerful technology that has vast potential to support various aspects of second language acquisition (SLA). Given GenAI’s capabilities, it is particularly relevant for the teaching and learning of second language (L2) writing. Despite its potential, there are also clear hazards and a range of potentially negative side effects, many of which have yet to be explored. Building on existing research, in this piece, I propose a series of six future research tasks that may prove useful for further understanding the affordances and limitations of GenAI for L2 writing. These six research tasks are organized into three interrelated themes, which cover 1) learning processes and outcomes, 2) student use and interactions, and 3) teaching. For each theme, two research tasks are proposed. Each task includes a discussion of what research is needed, why it is needed, along with how scholars might investigate that research task by adopting quantitative, qualitative, or mixed methods. The goal of this piece is to provide potential research ideas for graduate students and faculty and, ultimately, to foster research–pedagogy connections involving GenAI, L2 writing, and SLA.

Corruption persists because feedback between individual behaviour, social norms, and institutional rules creates self-reinforcing dynamics. Although laboratory experiments provide growing evidence on anti-corruption interventions, this literature remains fragmented, failing to explain why enforcement succeeds in some contexts and fails in others. To address this gap, this paper develops a Dynamic Corruption Equilibrium (DCE) Framework. Drawing on a Bibliometric-Systematic Review of 132 experimental studies, it identifies six intervention classes across institutional, social, and individual levels, with behavioural dispositions acting as cross-cutting moderators. While existing studies examine these interventions in isolation, overlooking cross-level interactions and behavioural heterogeneity, the DCE Framework integrates insights from complex adaptive systems theory and institutional economics to conceptualise corruption as a dynamic, multi-level system. By specifying three mechanisms: cross-level feedback loops, conditional pathways, and system bistability, the framework explains how corruption equilibria become self-reinforcing or shift, offering a diagnostic lens for analysing intervention effectiveness within complex institutional environments.

Social scientists have paid significant attention to the study of ethnic and religious minorities in Europe, and yet one group that evaded such scrutiny is the Tatars residing in modern-day Lithuania, Belarus, and Poland, who are unique in being Europe’s only Muslim community that survived under Catholic rule since the late medieval period. While Muslims in medieval and early modern France, Italy, Portugal, and Spain were eradicated through a mix of mass expulsions, forced conversions, and massacres, Lithuanian-Polish Tatars survived over six centuries. This article examines this unique case to understand the comparative political dynamics of persecution and toleration across medieval and early modern Europe. The article argues that the interstate and societal configuration of power explains the Tatars’ exceptional survival. The interstate and domestic dynamics are linked in that Lithuanian rulers successfully resisted forced conversion and eventually adopted Christianity on their own terms, which allowed for the preservation and perpetuation of religious sectarian diversity backed up by multiple political stakeholders. In the domestic struggle between monarchs, Papal allies, the Catholic nobility, and non-Catholics, none of the religious sectarian factions could achieve a hegemonic majority, let alone monopolistic control of political and military power, necessary for a coercive religious sectarian homogenization.

In expectations-driven liquidity traps (LTs), a higher inflation target is associated with lower inflation and consumption. As a result, introducing the possibility of expectations-driven LTs to an otherwise standard model lowers the optimal inflation target. Using a calibrated New Keynesian model with an effective lower bound (ELB) constraint on nominal interest rates, we find that even a very small probability of falling into an expectations-driven LT lowers the optimal inflation target nontrivially. Our analysis provides a novel reason to be cautious about the argument that central banks should raise their inflation targets in light of a higher likelihood of hitting the ELB.

As the Association for European Paediatric and Congenital Cardiology, the largest paediatric cardiology organisation globally, we are dedicated to safeguard the health, safety, and wellbeing of children, adolescents, and adults with congenital and acquired heart disease, as well as to support the professionals who care for them across the globe. As both a scientific and ethical community, we are profoundly troubled by the devastating humanitarian consequences of armed conflicts. Across the world, we are witnessing an immense toll on civilians, particularly children, who are always the most vulnerable in times of war. The destruction of healthcare infrastructure, obstruction of access to medical care, and the severe psychological and physical trauma endured by children are deeply alarming. While the Association for European Paediatric and Congenital Cardiology does not engage in political debate, we cannot remain silent when children’s lives and futures are endangered. We stand firmly by our values of compassion, equity, and peace and voice our deep concern for all (children) affected by violence and conflict, irrespective of their country, ethnicity, nationality, or faith. We invite our members and partner societies to join in upholding the principle that every child, everywhere, deserves to grow up in peace, with access to healthcare, and to live free from violence.

This report presents the provenance and technological analysis of two large obsidian bifaces recovered from a Middle Horizon (ca. AD 600–1000) ritual abandonment context at the coastal site of Cerro de Oro in the Cañete Valley, Peru. Portable X-ray fluorescence (pXRF) analysis identifies Quispisisa as the obsidian source, thereby documenting its circulation beyond highland Wari contexts. Technological analysis indicates contrasting trajectories: one biface shows no evidence of use, whereas the other exhibits macro use wear consistent with woodworking. Its location in a residential complex suggests community participation in the acquisition and ritual re-signification of prestigious nonlocal materials.

This article considers the interaction of the expansion of international law and the rising politicization of domestic institutions. As international legal processes frequently incorporate domestic institutions, how citizens react to the development of international law may become influenced by their perceptions of those institutions. We argue that involving politicized domestic institutions in the international legal decision-making process affects support for rulings through individuals’ perceptions of the domestic judiciary vis-a-vis an international court. Contra our preregistered expectations, a survey experiment fielded in Hungary shows that opposition partisans withdraw support for a European Union law decision when it is issued by a Hungarian court rather than directly by the European Court of Justice, while the involvement of a national court does not increase support for European Union law among government supporters. Further exploratory analyses of government supporters suggest that court decisions are broadly unlikely to move these individuals’ attitudes toward international law.

This study investigated cross-linguistic influence (CLI) in spoken word recognition in a typologically distinct language pair. Japanese–English bilinguals performed auditory lexical decision tasks in both L1 (Japanese) and L2 (English), presented in counterbalanced order, responding to cognates varying in cross-linguistic overlap and to matched nonwords. This design allowed direct comparison of L1 and L2 processing within the same individuals. Response times measured from stimulus onset and offset were compared to capture changes in effects across the time course of processing. In both languages, phonological and semantic similarities significantly facilitated responses, though phonological similarity effects varied slightly over time. Cognate frequency inhibited responses later in time, varying by language, and L2 proficiency further modulated performance. Importantly, these effects emerged spontaneously, without priming, demonstrating bidirectional cognate facilitation even across linguistically distant languages. The results support the applicability of the BIA+ model to auditory processing, even for different-script bilinguals.

Banco do Brasil played a central role in Brazil’s macroeconomic management during the Military Regime (1964–1985), a period marked by ambitious development strategies and heightened monetary instability. While previous research has emphasized its involvement with the Treasury and balance-of-payments financing, the broader scope of its functions has remained underexplored. Drawing on newly digitized balance-sheet data, this article reconstructs and classifies Banco do Brasil’s operations across its main areas of activity. It shows that Banco do Brasil acted as a multipurpose policymaking institution, mediating between competing objectives of growth, stabilization, and external adjustment. By tracing changes over time, the article highlights how Banco do Brasil’s functions evolved with shifting policy priorities across military governments, revealing a more complex and adaptive role in shaping Brazil’s macroeconomic outputs.

Śrīvidyā is a Tantric tradition which includes a form of religious initiation into an esoteric practice. Descriptions of initiation allude to and often describe the initiator sharing an experience with the initiand. In this article, we will suggest that such experiences can be understood in terms of a ‘shared token experience’. To make a case for taking such an experience into account, we use an argument for the epistemic value of religious experience. Our defence of the sharing of token experience across subjects also opens up questions as to the limits of consciousness or subjecthood. The phenomenon of Śrīvidyā initiatory experience also undermines two assumptions regarding what it means to be a subject. The first assumption is that each conscious subject is a unified whole insofar as its different experiential parts are mereologically connected to one subject. The second assumption is that each subject’s experiential field is bounded to one subject such that no two subjects can share the same experiential field. We argue that these assumptions do not necessarily preclude the possibility of a shared token experience.

Black constituents often receive lower-quality responsiveness from elected officials, particularly white representatives. This issue is exacerbated by partisan gerrymandering, which packs Black voters into districts with high majority populations. This trend has led to heightened concerns about increased vote dilution for minority and Democratic voters. This poses an important question: do legislators respond to shifts in their district’s racial demographics? Using data from before and after the 2010 redistricting cycle, we examine whether MCs engaged in policy responsiveness to Black constituents. Consistent with prior research, MCs did not sponsor or cosponsor more civil rights bills in response to increases in their Black constituency. However, Democratic MCs are more likely to mention civil rights in floor speeches, and Republican MCs are more likely to vote on bills in a manner consistent with LCCR priorities. These findings provide new insight into the representation Black Americans receive in Congress.

Natural language processing (NLP) has moved from a specialized research field into the everyday infrastructure of writing, search, translation, education, journalism, public administration, and scientific work. This transition changes what counts as progress. Accuracy, fluency, and benchmark performance remain important, but they are no longer sufficient when language technologies shape knowledge, decisions, identities, and public trust. This column introduces Responsible NLP as a research orientation that integrates fairness, transparency, privacy, safety, cultural diversity, environmental awareness, and human agency across the full life cycle of language technologies. It argues that responsibility is not an external constraint on innovation, but a condition for meaningful and trustworthy innovation. Future research must therefore ask not only whether an NLP system works but also for whom it works, under which assumptions, with what risks, and with what forms of accountability.

Across 13 surveys of 590 residents in seven ventilator-capable SNFs, residents in ventilator-designated beds had markedly higher ESBL (48.1% vs 28.2%; aOR = 1.64) and C. auris (38.6% vs 15.2%, aOR = 2.89), but lower MRSA colonization (35.2% vs 45.5%; aOR = 0.47), supporting the need for MDRO prevention beyond current Enhanced Barrier Precautions.