Many observers decry an "imperial presidency" in war-making, equating unilateral force with an unconstrained executive. This research reframes the war powers debate by centering the powerful incentives lawmakers have to avoid blame for military action.
This article explains the pretrain–finetune paradigm for transformer language models and shows how applied political scientists can use it to classify political text with relatively little labeled data and modest computation. The author frames mechanisms behind pretraining and finetuning, addresses common misconceptions about costs and complexity, and demonstrates the approach by identifying toxic language in conversation threads that follow U.S. Senators’ tweets. An online tutorial accompanies the article to help nontechnical scholars adopt the workflow.
Are nonverbal reactions during parliamentary debate gendered? Do male and female members of parliament experience applause or jeering differently depending on the gendered nature of the speech?
Leaders' public displays of anger are a previously underappreciated source of coercive credibility. Anger expressions make threats appear more credible because targets infer that angry leaders are less sensitive to the costs of conflict, and therefore more likely to follow through on threats.
This paper asks whether US economic sanctions during the Trump presidency reduced support for Iran’s government or instead produced a "rally-around-the-flag" effect among Iranian influencers and their audiences. Using a large-scale supervised machine-learning analysis of nearly two million tweets from more than 1,000 Iranian influencers, the authors compare reactions to comprehensive versus targeted sanctions and examine variation by political faction and by which entity was targeted.
How do different types of MPs in mixed-member proportional systems signal territorial ties in public communication? This study examines whether directly elected (district) MPs and list MPs in Germany differ in the geographic focus of their social-media posts, introducing a two-dimensional measure that captures both how local referenced places are and how deprived those places are. The paper combines computational place extraction with deprivation linkage to compare the self-presentation of all members of the 19th German Bundestag on Twitter and Facebook.
This paper reviews and critiques prior embedding analyses of the democratic peace, proposes an improved methodology that leverages foundational embedding models, and reports replication results. The author evaluates Rathbun, Parker, and Pomeroy’s microstructural work—including surveys, computational studies, and meta-analysis—offers a systematic critique of embedding analysis practices, and introduces a new approach intended to address those issues while taking advantage of foundational embedding models. The replication presented in the abstract accords with the earlier finding that the democratic peace effect is racialized, and the paper frames its contribution as a guide for conducting embedding analysis in political science.
This article asks whether culture—understood as interconnected “cultural scripts” linking group identities to negative phenomena—helps explain rises in antisemitism. Using three linked studies, the authors isolate a script that historically links Judaism with disease, document an increase in antisemitic attitudes during the COVID-19 pandemic among Christian believers, and use association tasks and priming experiments to show that elements of the script are present and can be triggered in respondents’ minds.
This study asks which individual parliamentarians take an interest in foreign and defense policy and what predicts that attention, using parliamentary questions as a window into elected officials’ priorities. It introduces an original dataset of 10,180 questions on foreign and defense matters in India’s Lok Sabha and uses analysis of those questions to show which aspects of foreign security politicians emphasise and which politician-level characteristics are associated with greater attention.
This article asks how Members of the European Parliament (MEPs) represent Syrian and Ukrainian refugees and whether implicit hierarchies of race and gender shape those representations. Drawing on Critical Securitisation Theory, the authors compare discursive patterns to test a hypothesis that stigmatisation based on race and gender affects EU representations of refugees. The manuscript reports a computational comparison of EP speech acts to see when refugees are cast as urgent threats and when speech enables de‑threatening or protective framings.
This paper examines how Italian political parties positioned themselves on the European Union's response to the Russian invasion of Ukraine, focusing on four response dimensions: sanctions on Russia, military support for Ukraine, EU enlargement to include Ukraine, and reception of Ukrainian refugees. The study combines theory-driven expectations from cleavage politics, the inverted U curve and differentiated politicisation with an inductive reading of parties' social-media output to assess whether party family or Europeanist/Eurosceptic divides better explain variation in party stances.
This paper spotlights a largely overlooked actor in China’s external relations: the International Department of the Communist Party of China (ID‑CPC). It traces the ID‑CPC’s external engagement from the early 2000s onward, showing how party-to-party ties have been expanded and repurposed—especially since Xi Jinping took office—to advance Chinese interests and promote China’s vision for reforming the global order.
This article asks how a century of scholarship in the journal Public Administration has evolved in topics, methods and geography, and offers a systematic, data-driven portrait of the field using bibliometrics and natural language processing. The authors map thematic change across five eras of the journal, identify 63 linguistic clusters of research, and characterize shifts in disciplinary influence, geographical settings and methodological orientations without making causal claims beyond the descriptive evidence.
Political scientists increasingly use supervised machine learning to code multiple labels from the same texts. The current practice of training separate supervised models for each label ignores relationships among labels and is likely to under-perform as a result.
This article introduces a novel perspective on policy diffusion, shifting focus from adoption to the crucial yet understudied issue-definition stage.
There is an active debate about whether human rights standards have shifted over the past 30 years. Existing evidence rests on indicators produced by human coders who read the texts of human rights reports. This work reframes that debate as a supervised learning problem: if coders apply consistent standards over time, the mapping from textual features to human-coded scores should be time‑constant; if meanings have changed, identical textual features will receive different numerical scores at different times.
Existing approaches to measuring political disagreement from text perform well only on narrowly selected texts that discuss the same issues in the same style. This article presents the first viable approach for estimating legislator-specific position scores from the entire legislative speech corpus and for producing detailed, time-varying measures of speech polarization and politically loaded language.
The surge of reports from monitors like Amnesty International, Human Rights Watch, and the US State Department offers unprecedented detail on repression and rights protections—but most automatic approaches reduce texts to simple word counts that ignore syntax and word order, limiting what scholars and policy-makers can learn.