Affective positioning in China–US security communication in Asia: Evidence from Shangri-La Dialogue speeches (2014–2023)

This study examines affective positioning in Chinese and US Shangri-La Dialogue texts between 2014 and 2023 in order to explore how emotion functions in elite security communication. Drawing on a corpus of 16 country-year texts (2911 sentences) from the 2014–2019 and 2022–2023 meetings, the study combines sentence-level sentiment analysis using VADER with targeted NRC emotion classification and keyword-in-context (KWIC) validation. The results show that both Chinese and US texts are predominantly positive or neutral in overall sentiment orientation, reflecting the institutional and diplomatic constraints of the Shangri-La Dialogue as a multilateral security forum. At the same time, US texts display a significantly higher mean compound sentiment score and a lower proportion of negative sentences than Chinese texts, although the overall effect size is small. Temporal analysis further indicates that cross-national divergence is uneven rather than constant: the largest overall sentiment gap appears in 2014, whereas the strongest divergence in negative framing emerges in 2019, followed by 2022. Focused analysis of the 2019 negative subset shows that both sides rely primarily on fear- and anger-related language, but with different emotional configurations. Chinese negative sentences display relatively stronger anger and sadness and are embedded in frames of war, coercion, instability, and historical grievance, whereas US negative sentences retain relatively stronger trust and anticipation and are more closely tied to threat attribution, sovereignty, and the undermining of regional order. The study argues that China–US differences in elite security discourse are best understood not as stable national emotional styles, but as shifting patterns of affective positioning shaped by institutional context, temporal conditions, and divergent modes of negative framing.