A preliminary study of science diplomacy networks in Central, Eastern and South-Eastern Europe

The paper addresses how science and diplomacy intersect and mutually shape each other in the context of Central, Eastern and South-Eastern Europe. It frames science diplomacy (SD) as a hybrid endeavor where diplomacy can enable scientific progress and science can facilitate policy and diplomatic breakthroughs. The study aims to contribute empirically by examining a relatively neglected European region and methodologically by applying social network analysis (SNA) to SD. Two research questions guide the analysis: (1) whether larger economies occupy more prominent positions in SD collaboration networks or smaller countries can be comparatively central; and (2) whether EU membership structures cooperation patterns, producing denser, distinct subgroups of EU versus non-EU countries within CEI member states.

The article situates SD within growing scholarship that examines national SD apparatuses, instruments (e.g., science attachés, S&T agreements), and case studies (e.g., Arctic, climate change). Prior work has focused largely on major powers and Global North countries (US, China, Germany, France, Switzerland, UK, Japan), with limited attention to Central, Eastern and South-Eastern Europe. It highlights the methodological value of SNA in international relations and public diplomacy research to capture how network structure and actor positions influence status, cooperation, and information flows. The paper notes a paucity of SNA applications to SD specifically, citing Paar-Jakli (2014) as a key exception, and proposes to extend a relational lens to MFAs in the CEI region.

Design: Online survey of Ministry of Foreign Affairs (MFA) officials responsible for SD in CEI Member States (MS). One designated respondent per MFA. Two waves: Sep–Nov 2019 (11/17 responses) and Dec 2020 (additional 4), totaling 15/17 MS (Albania, Belarus, Bulgaria, Croatia, Czech Republic, Hungary, Italy, Moldova, Montenegro, North Macedonia, Poland, Romania, Serbia, Slovakia, Slovenia). Survey scope: (1) National organization of SD (MFA roles, inter-ministerial coordination, goals); (2) Diplomatic tools for international S&T cooperation (science attachés, S&T protocols/agreements, joint initiatives); (3) SD capacity development (strategy integration, needed actions/partnerships, priority topics). Networks constructed: (a) Current cooperative relations among CEI MFAs in S&T, defined as the presence of any of three tools between dyads (attachés, agreements, joint initiatives). (b) Priority partners for future collaborations, as named by respondents among CEI MS. Operationalization: Binary directed networks. Prominence measured via indegree centrality and proximity prestige to capture direct and indirect nominations (Nooy et al., 2005). Country economic size proxied by GDP (World Bank 2019, current US$). Analysis tools: Pajek software for SNA. Community detection via the Louvain method to identify denser subgroups; modularity reported (current collaborations Q=0.053191; future priorities Q=0.100679), indicating weak but present community structure. Limitations acknowledged in design: two CEI MS (Bosnia and Herzegovina, Ukraine) did not respond; reliance on MFA self-reports; binary treatment of ties; small-N network.

- Respondents: 15 CEI MS participated. Respondents mostly trained in social sciences/humanities; seniority distributed across junior (7), mid-career (3), senior (5). - Current cooperation network (binary ties from attachés/agreements/joint initiatives): Average indegree 9.4; average proximity prestige 0.7356. Countries attracting most nominations on both metrics: Bulgaria, Czech Republic, Hungary, Italy, Montenegro, Poland, Serbia, Slovenia. Examples: Italy (indegree 12; proximity 0.8622), Serbia (12; 0.8750), Poland (11; 0.8047). Some larger economies (e.g., Romania) are less central; several smaller economies (e.g., Montenegro, Serbia, Slovenia, Bulgaria) are highly prominent. - Future cooperation priorities network: Average indegree 5.7; average proximity prestige 0.5977. Above-average on both metrics: Italy, Montenegro, Slovak Republic, Slovenia. Above-average indegree only: Moldova, Romania. Above-average proximity only: Hungary, Poland, Serbia. Italy (proximity 0.8235) and Montenegro (indegree 11) are notable hubs; smaller economies frequently appear as preferred partners. - Relationship to GDP: Prominence is only partially aligned with economic size. Italy and Poland are central as larger economies, but several smaller economies (Montenegro, Serbia, Slovenia, Bulgaria) also exhibit high prominence, while some larger economies (e.g., Romania) do not. - Community structure: Two subgroups identified in current collaborations: Subgroup 1—Albania, Croatia, Italy, Montenegro, North Macedonia, Romania, Serbia, Slovenia; Subgroup 2—Belarus, Bulgaria, Czech Republic, Hungary, Moldova, Poland, Slovakia. These are not cleanly split by EU status; instead they align broadly with subregional geography (South/South-Eastern vs Central/Eastern). In future-priority networks, the same partition holds except Bulgaria shifts from Subgroup 2 to Subgroup 1. - Overall, EU membership does not produce distinct, exclusive SD collaboration clusters within CEI; networks appear regionally structured rather than institutionally segmented by EU status.

Findings address RQ1 by showing that while material resources matter for some actors (e.g., Italy, Poland), network prominence in SD is not determined solely by economic size. Smaller states in the CEI region can achieve central positions and attract substantial collaboration ties, suggesting opportunities to 'punch above their weight' in SD networks. For RQ2, community detection indicates that collaboration patterns are not primarily divided by EU membership; instead, ties cluster along geographic subregions, and both EU and non-EU members appear within the same subgroups. This implies regional proximity and possibly historical or cultural linkages may be stronger drivers than EU status for MFA-led SD collaborations in this setting. Methodologically, the study illustrates how SNA reveals the interaction between actor attributes (GDP, membership) and relational structure (centrality, communities) in SD, highlighting cases of over- and under-performance relative to material capacity.

The study offers a preliminary SNA-based examination of MFA-led science diplomacy collaborations among CEI Member States. It contributes by: (1) demonstrating the applicability and value of SNA to SD; (2) showing that smaller economies can attain central roles alongside larger ones; and (3) finding that EU membership does not segment networks into distinct EU/non-EU clusters, with collaboration patterns aligning more with subregional geography. Future research should expand the empirical base beyond MFAs to include other stakeholders (universities, research councils, agencies), incorporate subnational and regional initiatives, and adopt longitudinal designs (e.g., pre/post EU accession) to better assess how attributes and institutional memberships shape network positions and evolution. Such evidence can inform CEI and national policies to strengthen ties, bridge gaps, and target priority collaborations.

- Sample and coverage: Only 15 of 17 CEI MS responded; Bosnia and Herzegovina and Ukraine are missing. Single respondent per MFA and reliance on self-reported data. - Network construction: Binary ties aggregating different instruments (attachés, agreements, joint initiatives) without weighting by intensity or recency. - Small-N network: Limited number of nodes constrains generalizability; close association of indegree and proximity prestige likely reflects small size and dense connectivity. - Temporal scope: Cross-sectional snapshot across two survey waves; no longitudinal assessment of dynamics. - Community structure: Low modularity scores indicate weak community structure; subgroup findings should be interpreted cautiously.