Authors: Akanksha Singh, Medha Bhardwaj, Aneeka Dhanuka, Akshay Srivastava, Ananya Singh 

Abstract 

The shift in the international community from bipolarity to multipolarity has determined the new patterns of cooperation and competition between states by defining the specificities of emerging geopolitical alliances, which constitute the subject of this research. This article aims to contribute to the understanding of voting behaviour in the United Nations General Assembly (UNGA) as an indicator of geopolitical alliances by applying Social Network Analysis to the UNGA roll-call voting data. Although there is an extensive body of literature documenting the trends in voting behaviour, alliance-building processes, and network analysis methods in international relations, the application of SNA to the recent UNGA voting data to identify emerging geopolitical alliances has not been explored in detail. 

This research undertakes a mixed-methodology approach by design, as it relies on both qualitative and quantitative sources of information. On the one hand, it uses the literature review as a research method which allows the formulation of the research questions based on the critical analysis of the existing scholarly works. On the other hand, the study implements a quantitative method of network analysis in order to operationalize the qualitative hypotheses derived from the review of literature. Specifically, the Python coding of Voeten’s dataset was used to identify voting communities using similarity scores, Louvain’s clustering algorithm, and centrality measures. 

The findings demonstrate that the network of voting alliances has fragmented into a weakly-aligned, non-aligned European-pivot bloc, a cohesive and structurally dominant Global South bloc, and a fully isolated two-country component comprising the United States and Israel, alongside a small number of singleton states with no significant voting ties to the rest of the network. Overall, the analysis shows that UNGA voting behaviour reflects the shift from military pacts to more fluid and issue-based partnerships, thus indicating the move towards a more decentralized international system. 

The study’s qualitative component consisted of a critical review of the existing literature on voting alliances in the UNGA voting data. The key themes addressed by the review include voting blocs, geopolitical alliances, and the network analysis methods used in the study of international relations. This component of the research enabled them to formulate the key research questions and contextualize the findings of the quantitative component. Aggregately, the findings demonstrate that Social Network Analysis (SNA) is an effective method for studying the evolving geopolitical alliances and identifying influential actors in Global Politics. Particularly, the study illustrates how voting blocs in the UNGA transform over time and the types of partnerships that emerge within the international system. 

Introduction 

The United Nations General Assembly (UNGA) provides a comprehensive forum for collective decision making by assessing the country’s voting patterns that capture how the states position themselves in the international context. Voting behaviour in the UN General Assembly is used to study countries’ foreign policy preferences, alliances, and changes in international politics, revealing a state’s position on global issues (Voeten, 2013). Critically, countries belong to different blocs based on different issues resulting in the evolution of voting alignments. (Russett, 1966). However, the UN voting patterns tend to get influenced by power dynamics beyond the UN (Russett, 1966) revealing their true geopolitical positions, forge coalitions and a contested challenge to negotiate global governance. Alignment strategies pursued by these rising powers on issues like security, human rights, economic development, and sovereignty can heavily determine the geopolitical landscape in the upcoming years (Voeten, 2013). To examine how emerging alliances actually position themselves across multiple issues requires moving beyond formal diplomatic channels. 

Patterns of alignment and opposition in the United Nations General Assembly (UNGA) voting determine the positioning of the international community, further raising critical questions on the functioning of state sovereignty and structural autonomy (de Oliveira et al., 2025). The ongoing shift towards the emergence of multipolarity reveals significant challenges as major powers such as China, Russia, Brazil, and India compete to assert greater influence over the Western dominance of the post-Cold War consensus (Muhammad, 2023). Strategic interests and ideological affinities are required to understand the scope for emerging geopolitical alliances. Yet scholars have struggled to map these shifting alliances by relying on tools and frameworks developed for a bipolar or unipolar world. This paper primarily attempts to bridge this gap by systematically analysing UNGA voting patterns through comparative analysis of pre-existing research. Additionally, in order to identify and characterize the geopolitical alliances shaping the 21st-century international order, a computational analysis utilising advanced network analysis methods has been conducted. 

Historical Evolution 

The United Nations General Assembly was established on October 24, 1945, with the primary purpose being “to maintain international peace and security”, and “to develop friendly relations among nations based on respect for the principle of equal rights and self-determination of peoples” (UN Charter, Article 1). The General Assembly serves as “the parliament of mankind,” where all Member States have equal representation with one vote each (UN Charter, Article 10), making UNGA voting a critical arena for expressing geopolitical positions and the emergence of alliances. Article 18 of the UN Charter states the voting process of the UN General Assembly, stipulating that “each Member of the General Assembly shall have one vote” and that, “decisions of the General Assembly on important questions shall be made by a two-thirds majority of the members present and voting.”  

During the Cold War (1946-1989), there was a stable confrontation between the two superpowers, the US and the USSR, suggesting the dominance of Western-led institutions (Fotheringham, 1963). The creation of a bipolar world masked crucial complexity, parallels of which were formed around the notion of anti-colonialism, the quest for racial equality, and development of the Global South. During the period of anti-colonialism, voting was not unified in nature and revealed a cross-cutting pattern. Between 1960 and 1970, African countries increasingly voted together following the establishment of the Organization of African Unity (OAU) in 1963 and strengthening the broader Afro-Asian and Global South coalition (Abate & Brunn, 1977). The presence of a stable and interpretable second dimension is reflected in the mid-1960s to mid-1980s with the North-South conflict, tied to the rise of the Non-Aligned Movement and the G-77 (Bailey & Voeten, 2018).  

The post-Cold War transition (1990-2000) revealed that bipolarity had masked an emerging North-South cleavage. Many former Soviet bloc countries in Eastern Europe began voting more similarly to Western European countries (Kim & Russett, 2006), reflecting their political and economic transition after communism’s collapse and integration with Western institutions. Today, the international system displays multipolar fragmentation with systematic divisions between Global North and Global South (de Oliveira et al., 2025), and the expansion of new alliances such as QUAD, AUKUS, BRICS, SCO, and NATO is reshaping global security. (Dahal, 2024). 

Research Question 

The changing nature of the global order and the increasing levels of complexity in international relations have spurred a number of academic inquiries into the ways that states may interact within the framework of the United Nations General Assembly in the future. While previous research has highlighted the importance of voting behaviour, alliances, as well as the use of related network-based approaches in International Relations, it remains crucial to understand what contemporary voting patterns can suggest about the likely nature of alliances. This research aims to address the following question: 

How can Social Network Analysis of United Nations General Assembly voting patterns be applied to identify and analyse geopolitical alliances in the modern geopolitical landscape? 

Research Objectives 

The present study is designed to achieve the following four objectives: 

• To examine voting patterns in the United Nations General Assembly (UNGA) as assessing geopolitical alignment among member states. 

• To apply Social Network Analysis (SNA) for the identification of emerging geopolitical alliances, i.e. reflected in UNGA voting behaviour. 

• To analyse the restructuring of geopolitical voting alignments in the context of an increasingly multipolar international system. 

• To evaluate the impact of network-based approaches in understanding patterns of global cooperation and alliance formation. 

Research Scope 

This study focuses on identifying emerging geopolitical alliances by analyzing voting patterns in the United Nations General Assembly (UNGA) through the application of Social Network Analysis (SNA). It examines how voting similarity patterns among participating member states reflects the evolution in diplomatic alignments and political cooperation within the twenty-first-century international system. The study is primarily centered on understanding alliance formation as observed through UNGA voting behavior in the context of a global order marked by increasing multipolarity. While acknowledging that geopolitical alignments are also influenced by such other factors like, historical, economic, regional, and domestic political factors. Thus, the scope of this research is limited to the analysis of voting linkages within the UNGA. Consequently, the study does not seek to explain the full range of political, economic, or security determinants of state behavior but instead focuses on how network-based analysis of UNGA voting patterns can identify and interpret emerging geopolitical alliances. 

Considering a progressively complex multipolar global order, the study in the field of geopolitical alliances is a challenging task that requires new approaches in identification patterns of interaction between states. The analysis of the United Nations General Assembly voting data allows one to track the trends in diplomatic interactions and relations of countries to one another beyond geopolitical alliances. This paper explores this issue additionally with the help of a social network analysis(SNA) by detecting the possible patterns and contributions to the expanding work of literature aiming at global alliances. Through the study of this topic, the research aspires to shed light on the contemporary dynamics of international relations and demonstrate the effectiveness of applying network analysis methods in the study. 

Literature Review 

Statistical Trends and Patterns 

Some common statistical patterns seen across time to study geopolitical alliances by analyzing voting patterns were S-score or S-index developed by Curtis Signorino and Jeffrey M. Ritter (1999), Kendalls’ Tau-b popularized by Bruce Bueno de Mesquita (1975), Q-factor analysis which was systematized by Bruce Rusett in 1966, and Diffusion geometry and thematic clustering proposed by Minh-Tam Le, Mathew Lawlor, Bruce Russett, and Steven W. Zucker in 2013. Although more recently, models like the Bailey, Strezhnev, and Voeten (2017) approach, have replaced the traditional models for being flawed as UN voting patterns reflect but do not perfectly represent world politics. Voting patterns get influenced by international relations and power dynamics beyond the United Nations. Even though statistical patterns keep replacing each other for accuracy, they laid the groundwork for recognizing major BLOC formations and similarities in foreign policy choices amongst nations. 

Post Cold-War Geopolitical Alliances 

Establishment of the UN created a global platform where state voting behaviour could reflect international alignments. During the Cold war, UNGA voting was largely structured around the ideological divide between the United States and its Western allies and the Soviet Union and its Eastern allies. In a study by Yohannis Abate and Stanley D. Brunn (1977), they observed that between 1960 and 1970, African countries increasingly coordinated their voting rather than acting individually after the establishment of OAU (1963). Their affinity towards Asia and other developing countries gave rise to the Afro-Asian coalition forming an identifiable sub-group that shifted the focus from East-West ideological rivalry towards North-South concerns. The shift caused a wave of economic development, decolonization and racial equality which formidably increased the representation of newly independent African states within the UNGA. Following the end of the cold war in 1991, voting alignments increasingly reflected political, economic, and institutional relationships. Records consistently revealed geopolitical blocs based on shared foreign policy preferences rather than solely on formal military alliances. (Kim & Russett, 2006).  

-Shifting alliances with shifting preferences 

Modern issue-based divisions emerged around human rights, globalization and sovereignty. A coordinated voting pattern amongst regional organizations and coalitions like the EU and developing country groups was noted. The shift in global power from the Atlantic region towards Asia with countries like China, India and Russia taking the center stage as an influence in international politics became growingly evident in the increasingly transitioning international system that pointed towards a multipolar world order (Dahal,2024). Geopolitical competition has expanded beyond traditional military platforms to include geoeconomics, cybersecurity, technology, artificial intelligence, and space. The influence of multilateral institutions has weakened while nationalism, strategic competition, and geopolitical rivalry have increased. New geopolitical partnerships and security arrangements-including BRICS, the Shanghai Cooperation Organization (SCO), the Quadrilateral Security Dialogue (Quad), AUKUS, China’s Belt and Road Initiative (BRI), and NATO expansion- are reshaping regional and global power dynamics. 

UNGA Voting Behavior and Contemporary Dynamics 

UNGA voting is an effective tool for analyzing policy references and diplomatic behavior as assessed by Khan, M. Z. I. (2021). Major geopolitical changes are consistently reflected in long-term voting behavior. A comprehensive analysis of the UNGA voting pattern may reveal states’ foreign policy preferences and diplomatic alignments. They also help in identifying political proximity and geopolitical blocs beyond formal military alliances. Geopolitical interests remain the primary driver of alliance formation (Dahal 2024). His study shows that security concerns influence voting decisions although in another study Kim & Russett (2006) suggested that ideological orientations influence voting behavior. Dahal (2024) also observes that emerging issues such as climate change, pandemics, technology, cybersecurity, artificial intelligence and space reforms are increasingly shaping geopolitical dynamics and alliance formation.  

-Voting Behaviour 

Voting behavior reflects positions on issues such as security, sovereignty, human rights, and economic development. Economic interdependence and trade relationships affect diplomatic alignment (A.G. Muhammad, 2023) but regional cooperation and historical relationships contribute to coordinated voting (Kim & Russett, 2006). Expanded datasets improve empirical reliability and strengthen research findings on international alignments. More comprehensive voting datasets, including amendments, failed resolutions, procedural votes, and consensus decisions, provide a more complete picture of state behavior than traditional datasets. Bangladesh’s voting behavior demonstrates how UNGA records can be used to study foreign policy orientation while recognizing that geopolitical and bilateral considerations also influence decisions. Major geopolitical events, including France’s NATO policy shift and the collapse of the Soviet Union, are reflected in changes in voting behavior. Even though UN voting records reliably capture long-term geopolitical alignments and changes in international politics, voting records have limitations because governments may compromise, abstain strategically, or conduct diplomacy outside formal voting. UNGA voting should therefore be interpreted alongside broader geopolitical, diplomatic, and bilateral factors.  

Network Analysis and International Cooperation.  

Hafner-Burton et al. (2009) established network analysis for understanding international relations by arguing that states’ positions within international networks influence their power, cooperation, and policy outcomes. Based on this perspective, Lee and Stek (2016) applied social network analysis to the co-sponsorship of United Nations General Assembly (UNGA) resolutions and the study revealed that states formed flexible partnerships that often extended beyond formal military alliances. Similarly, Pauls and Cranmer (2017) analyzed UNGA voting networks and identified affinity communities of states with similar voting behaviour, demonstrating that countries within the same community were more likely to cooperate and less likely to engage in any armed conflict. Collectively, these studies highlight the use of network analysis for better understanding of alliance formation, political cooperation, and patterns of state interaction in international relations. While these studies largely rely on binary agreement measures or resolution co-sponsorship to identify affinity communities, the present study builds on this foundation by applying a thresholder cosine-similarity network together with Louvain community detection to more recent UNGA voting records, offering a finer-grained account of how these voting communities are structured today. 

Measuring Geopolitical Alignment and Fragmentation 

A study by Bailey and Voeten (2017) reveals that geopolitical alignments are understood better through a two-dimensional ideal point model that captures both broad geopolitical preferences and issue-specific policy positions reflected in UNGA voting. Airaudo et al. (2025) reevaluated UNGA voting as a measure of geopolitical distance and the study revealed that estimate of geopolitical fragmentation depends on the voting data and methodology, with economic votes providing a more reliable measure of alignment. Further, Kirsamer and Riha (2025) developed a bilateral index of geopolitical fragmentation that combines UNGA voting records, governance indicators and diplomatic representation, identifying three major geopolitical blocs through machine learning techniques. Later studies by Voeten (2026) concluded that different indicators capture different dimensions of geopolitical alignment. These studies emphasize on the growing importance of developing robust and appropriate measures of geopolitical alignment and fragmentation. 

Determinants of UNGA Voting Behaviour 

Multiple studies have revealed that UNGA voting behaviour is influenced by both international and domestic political factors. The study by Carter and Stone (2015) reveals shared democratic values and strategic foreign aid incentives shape the support for U.S. positions in the UNGA.  Further, Lees (2024) emphasizes that states foreign policy preference is influenced by both domestic class coalitions as well as the social forces that further influences the governments with different social bases to acquire different voting positions in the UNGA. Together, these studies reflect a combination of strategic incentives and domestic political dynamics in the voting behaviour in the General Assembly. 

Geopolitical Realignment in a Multipolar World 

Recent studies have examined how changes in the international system had affected the voting patterns of the UNGA. The study by Binder and Payton (2020) found that BRICS countries had adopted multiple pragmatic, issue-specific voting strategies, with the aim of balancing cooperation with both Western and non-Western powers rather than consistently supporting or opposing the existing international order. Similarly, Mosler and Potrafke (2024) observed that although U.S. allies largely maintained close voting alignment with the United States during the Trump presidency, many other countries became less aligned, reflecting changes in U.S. foreign policy. Focusing on the present Russia–Ukraine conflict, de Oliveira et al. (2025) conducted a study and the findings revealed that liberal democracies and Western allies are more likely to support UNGA resolutions condemning Russia, while many Global South countries adopted different voting positions, demonstrating an increasingly multipolar world order. Collectively, these studies indicate that contemporary UNGA voting reflects evolving geopolitical alignments and the growing complexity of international relations in a multipolar era. 

Evolving Alliances Through Evolving Analysis 

The study and understanding of Geopolitical alliances have increased and have gone through significant changes as the overall international order has transformed significantly after the post-Cold War era. The alliances are formed based on treaties, which can be through political, military and ideological commitments, or collective security interests. In the modern world, the patterns of cooperation and relationship between states have become more issue-specific, and the meaning of formal alliances has become insufficient for explaining modern patterns of cooperation. In recent times, the UNGA, United Nations General Assembly, voting record has become a record to examine geopolitical alignment as it provides its preferences for foreign policy measures and how countries vote in the United Nations to discover which countries are forming new political partnerships or alliances.  

A major contribution to this field is Erik Voeten’s work on conceptualising geopolitical alignment. Voeten’s “Conceptualizing and Measuring Geopolitical Alignments” establishes that geopolitical alignment is a multidimensional concept, in which diplomatic, ideological, economic, and security relationships constitute different forms of cooperation rather than a single, universally measurable phenomenon. Voeten believes that UNGA voting records remain one of the strongest datasets in analysing foreign policy preferences because they provide consistent, long-term, and globally comparable information on state behaviour. The results of his findings significantly influenced succeeding research and encouraged scholars to combine voting analysis with broader political contexts.  

Further studying these conceptual findings, in the paper “Community structure in the United Nations General Assembly”, Maco, Mucha and Porter’ methodological contribution of introducing network analysis to the study of UNGA voting resulted in the analysis of voting patterns as interconnected networks and used community detection to identify clusters of states with shared diplomatic behaviour instead of comparing countries through simple voting similarity. It revealed that these voting groups change over time in response to major international developments, highlighting the dynamic nature of geopolitical alignments. 

By exhibiting that network analysis can expose patterns of political cooperation beyond traditional statistical methods, the authors established a valuable framework for examining emerging alliances in international politics. 

Since diplomatic behaviour is also influenced by historical, political, economic, and regional factors, many scholars argue that network analysis should be complemented with qualitative approaches to achieve a more comprehensive understanding of international alignments. 

The study by Binder, M., & Payton, A. L- “With Frenemies Like These: Rising Power Voting Behaviour in the UN General Assembly”- uncovers the increasing popularity of emerging middle powers in shaping the current multipolar international system. The authors observed that many countries now adopt flexible foreign policies that cooperate on specific issues rather than being fixated on maintaining set geopolitical alignments. Homogeneous patterns are visible in studies of UNGA voting which was observed during the Trump administration when the Russia-Ukraine conflict took place, where countries often debated with making diplomatic decisions only after considering political, economic and regional interests. Together, these studies highlight that contemporary geopolitical alliances are becoming more flexible, overlapping, and influenced by changing international conditions.  

While existing research has significantly advanced the understanding of geopolitical alignments and UNGA voting behaviour, several gaps remain in understanding contemporary geopolitical alignments. Most studies examine specific organisations, regional initiatives, or individual geopolitical events, offering insights within limited contexts. As a result, broader changes in alliance formation across multiple international issues receive comparatively less attention. In addition, relatively few studies apply network analysis to UNGA voting over diverse issues to identify emerging patterns of cooperation. This study addresses these limitations by using network analysis to examine evolving voting relationships in the UN General Assembly and explore how they reflect the formation of new geopolitical alliances in an increasingly multipolar world. 

Methodology 

This paper is a mix of quantitative and qualitative research designs, with the core of the quantitative and qualitative approaches being computational analysis and comparative analysis, respectively. A computational analysis of Voeten’s UNGA voting dataset has been conducted using the ‘Louvain Community Detection’ and other ‘Network Analysis Libraries’ with centrality metrics to identify fragmentation in international order. The extracted voting data records each member state’s vote in textual form, such as Y (for “yes”), N (for “no”), and A (for abstention) through vectorization. The data is organised using computational libraries in Python, and the Louvain Community Detection method is then used to provide a non-biased, mathematical grouping of nations. This method generally identifies patterns in the voting behaviour of the states with respect to the UNGA resolutions. 

To make this pipeline reproducible, the parameters used are stated explicitly here. The analysis uses the 2021-03-23 UN Votes dataset derived from Erik Voeten’s UN General Assembly voting data, merged with the roll-call metadata. The contemporary analysis is restricted to the date window 27 December 2014 to 27 December 2019. After excluding the explicitly listed dissolved/historical state labels and retaining countries with at least 10 recorded votes, the network contains 192 country nodes and 508 roll-call resolutions. Votes are encoded as 1 for ‘Yes’, -1 for ‘No’, and 0 for ‘Abstain’. The voting matrix is constructed with missing country–resolution observations filled with 0 before cosine similarity is calculated. Thus, in the current implementation, missing/non-voting observations are not distinguished from abstentions; this is a limitation of the present pipeline and should be stated explicitly. Pairwise voting similarity is calculated as cosine similarity between the resulting country vote vectors. An undirected weighted edge is retained when cosine similarity is greater than 0.60. Louvain community detection is applied with random state = 42 and the package’s default resolution parameter (γ = 1.0). Degree centrality and weighted betweenness centrality are calculated on the final graph. Eigenvector centrality is calculated separately within each connected component because the final network is disconnected. 

Comparative analysis requires researchers to compare two or more papers or cases to identify similarities and differences among the studies. Several research papers have been analysed to understand the voting behaviour of multiple states, their shared preferences, strategic partnerships, and ideological affinities through which states are forming new geopolitical alliances. As such the literature review also forms the major basis of this study. By studying multiple research papers, the researchers have tried to analyse the voting pattern of the states and alliances. These papers discussed different states and their biases in the UNGA resolutions and how they reflect the emergence of blocs and alliances in global politics.  

Additionally, maps, figures, flowcharts, tables and graphs have been used to analyse the UNGA data. Several papers have used various methods to understand voting behaviour. Techniques like Hierarchical Thematic Clustering groups countries with similar UN voting behaviour into blocs. Diffusion Geometry examines how one country’s voting behaviour may influence others over time. Centrality Metric Selection finds which country is the most influential within a UN voting alliance. Ideal Point Estimation Places countries on a spectrum (e.g., pro-West to anti-West) according to UN votes. These methods are used by researchers when studying international relations, data science and machine learning. Since the following research paper also examines the UNGA voting behaviour, these techniques have been carried forward through the review of literature here as well. 

Data Analysis 

Quantitative Data Findings 

To add to the existing literature, this study bypasses the standard binary agreement framework in favour of a relational network design via vectorization. Similarity methods were used to turn UN votes into numeric vectors, and cosine similarity was performed after. This method is more accurate because it accounts for states’ preferences in degrees rather than putting countries into binary boxes of agreed/disagreed. This formulation closely follows how Bailey, Strezhnev, and Voeten’s papers measure preference/magnitude. 

Additionally, the generated similarity matrix was converted into a unipartite network graph. Before settling on a final edge-weight threshold, a sensitivity sweep was run across cutoffs ranging from 0.45 to 0.75, tracking how modularity, community count, and network connectivity responded at each step. Modularity rose steadily across this range, from 0.149 at the loosest threshold up to 0.233 at the tightest, and the network held at two communities up to 0.55, consistently settling into three from 0.60 onward, with the number of nodes left disconnected also climbing from 0 to 9 across the same range. This confirmed 0.60 as the threshold used to isolate meaningful structural alignments going forward to the point at which modularity begins improving meaningfully without fragmenting the network as aggressively as thresholds beyond 0.70 do. 

Threshold sensitivity table (1)

As depicted in the table above, the lowest level for a value of threshold at which the active network forms three meaningful clusters with only two isolated nodes is 0.60. Further increase in the threshold value to 0.65 and 0.70 results in slight improvement in modularity, while there is an increase in the number of isolates to 3 and 5, respectively; however, at a threshold value of 0.75, there is an increase in modularity, while the size of the largest cluster decreases sharply from 185 to 131. 

To group states without imposing subjective geopolitical categories, we then applied the Louvain community detection algorithm at this threshold. This optimization strategy partitions the global network based purely on mathematical patterns of voting consistency, effectively mitigating the agenda-voting bias that historically skews simple agreement indices. 

Eigenvector centrality found nodes with the largest/core influencers tied to other countries with power, computed separately within each connected component of the network, since a node’s standing in a small, disconnected cluster is not comparable to its standing in a large one, and collapsing the two onto a single scale would understate the influence of countries in smaller components. Betweenness centrality found states that could be considered “bridge states” and hedgers in this new multipolar system. 

Results 

The macro UNGA network, drawn from the most recent five-year window available in the dataset (27th December 2014 to 27th December 2019), exhibits a baseline network density of 0.5548, meaning it has a moderate global consensus. It is to note that high density can mean the majority of states agree on trivial issues, which can obscure severe underlying polarization on critical security or human rights issues. 

This study therefore examines voting alignments during 2014–2019 and interprets them in the broader context of contemporary geopolitical research discussed above. 

Applying the Louvain community detection, we can see the 3-bloc system emerged cleanly : Bloc 0 (Non-aligned/European Pivot) had San Marino, Malta, and Cyprus as head parties and did not have very clean voting consistency; Bloc 1 (Cohesive Global South) was the largest and strongest connected bloc, containing Mozambique, Singapore, and Thailand; and a third, structurally distinct pairing consisting of only the US and Israel emerged as its own two-node component, entirely disconnected from the rest of the network. Average voting consistency was highest in Bloc 1. Eigenvector centrality showed that Bloc 1 had by far the most countries with large power cores or influencers, at ~0.095. 

The US-Israel pairing is worth treating separately from the three-bloc structure above, because it isn’t merely a loosely-connected fourth bloc but a fully isolated component with no ties above the similarity threshold to any other country in the network. Within that two-node component, both the US and Israel score an eigenvector centrality of 0.7071 each, which is the mathematical maximum possible for a component of that size: each is the other’s entire structural universe, while being completely severed from everyone else. Two further countries, Nauru and South Sudan, appeared as complete singleton isolates with no ties above threshold to any other state at all, plausibly reflecting a genuinely idiosyncratic voting record in Nauru’s case, and a short, sparse voting history for South Sudan, which only joined the UN in 2011. 

Betweenness centrality created the “bridge” and “swing” states. Betweenness calculates countries that may be considered geopolitical swing states necessary for network connectivity: Cyprus (Bloc 0) serves as the largest structural bridge at 0.0968, acting as a connective point between the non-aligned pivot and the wider network. 

Selected within-component eigenvector centrality values reported by the analysis. 

For polarity questions, the analysis looked at voting on resolutions 5394 and 5549. The highest possible polarizing score was 2.00 (Agree with Non-aligned state – Disagree with West-aligned bloc). The mean polarizing vote was for the West that agreed strongly on ‘NO’ (-1.00), and Global South which agreed strongly on ‘YES’ (1.00). 

Resolution Comparison Table(2)

Link for the data analysis (to be opened in Google Colab)- https://drive.google.com/file/d/1KKtfRo4QzDKmMTH6rwoScjYvFdDAKtsx/view?usp=sharing  

Interpretation of the results 

These results of our analysis provide empirical weight to several key theories of global politics. Our data shows the United States and Israel forming an isolated two-country component, each with an eigenvector centrality of 0.7071, the theoretical ceiling for a component of that size. Rather than reading as “low-influence” nodes, the two are maximally tied to each other while being cut off from the rest of the network entirely. This confirms the findings of the paper (Benati and Capurri, 2022), in which they noted that “the USA was statistically isolated on its own, especially in 2004 when it supported Israel on its own.” These results suggest that this isolation is not an anomaly but a structural feature of the modern UNGA voting trend, and the threshold sensitivity analysis shows that this feature strengthens more isolated nodes and higher modularity, as the similarity threshold rises, indicating it is a genuine property of the voting data rather than an artefact of one arbitrarily chosen cutoff. 

The high eigenvector centrality of Bloc 1 (Mozambique and Singapore at ~0.095) supports the observations of Macon, Mucha and Porter (2012), who argued in his paper that “modularity based on G77 membership has steadily increased, especially in the post-cold war period.” This bloc now defines the “global consensus,” making the US-led position an “outlier” in the General Assembly. 

The analysis also surfaced two singleton isolates, Nauru and South Sudan, that sit outside all three blocs entirely. Unlike the US-Israel dyad, these are not cases of maximal alignment with one partner but genuine outliers with no similarity-network ties above threshold to anyone, a distinction worth keeping separate from the “Bloc of Two” framing, since it points to a different underlying mechanism (idiosyncratic or under-sampled voting behaviour) rather than deliberate bilateral alignment. 

While the US is isolated, the polarization data shows that the Global South (often aligned with China) maintains a consistent “YES” position on resolutions that divide the house. This corroborates the finding in Adarkwah et al. (2026) in which he states that “China exhibits broader global voting alignment than the US,” and is thus effectively positioning itself as the leader of the “global consensus” while the US retreats into a “Bloc of Two.” 

Qualitative Data Findings Across the Reviewed Literature 

The corpus of studies scrutinized in the review encompasses a period of eight decades, the findings of which can be divided into four analytical topics: the formation and dissolution of blocs, measurement of alignment, reliability of voting data, and decline in multilateral agreement. 

-Evolution of Voting Blocs 

Concerning bloc formation, Fotheringham’s (1946–1959) pioneering analysis determined that the politics of the Cold War were characterized by almost completely bipolar conflict, as the United States’ score of 74.6% far exceeded the Soviet Union’s 38.8% score, and the parties only voted together in 20.4% of cases. Importantly, Fotheringham revealed a second, cross-cutting axis — anti-colonialism — along which they inverted their positions: Arab and Asian countries voted together with the Soviet bloc not because of ideological kinship but rather because of their solidarity in terms of anti-colonialism, providing what is called “cross-cutting pattern” by the present thesis.  

The existence of this dual-axis framework is further supported in later results in the corpus: the two-dimensional construct created by Bailey and Voeten (2018) presents a statistically significant North-South axis from the mid-1960s to the mid-1980s, overlapping with the emergence of the Non-Aligned Movement and G77. The network-communities approach used by Macon, Mucha, and Porter (2012) verifies the findings confirming that the pattern of countries interconnected via implementing modularity due to G77 connection formed in 1964 has substantially changed from 1964 onwards, particularly in the post-Cold War period. The alignment index developed by Benati and Capurri (2026) supports these findings, extending them to 2022, and detecting the very stable Western core of 50-58 countries through the Eastern bloc’s disintegration after the fall of USSR (1990-1993) with some involvement of Russia in the Western cluster. 

-Evolution of Measurement Methods 

A different category of results deals with the instruments that have been used to identify such structures. The literature has chronicled a gradual evolution of the methods used: Russett’s (1966) Q-technique was replaced by Signorino and Ritter’s (1999) S-score, which was in turn surpassed by the ideal-point model of Bailey, Strezhnev, and Voeten (2017). Both transitions were brought about by the empirical deficiencies of its predecessor:  

Bailey et al. show that the S-scores only recognized the beginning of Gorbachev’s rapprochement with the West in 1991, while the ideal-point estimates revealed the occurrence of the event in the mid-1980s. The discrepancy is attributed to the mixing of the genuine preference changes with the annual fluctuations in the voting agenda of the UNGA by the S-score, which is responsible for 36% of S-score variability as opposed to only 3% for the ideal points. The next attempt to solve this issue was made by Benati and Capurri, who introduced the A-index. 

-Reliability and Validity of UNGA Voting 

A third category pertains to the external validity of UNGA voting as a means of gauging state preference. According to Adarkwah, Sabel, and Zilja (2024), UNGA agreement indices are essentially uncorrelated with global economic sanction data (correlation near zero), and that for U.S. dyads, the correlation is only −0.35. This result indicates that voting behaviour and coercive foreign policy are two distinct empirical constructs. Moreover, the same study has found UNGA data to be comparatively “sticky,” failing to account for any change occurring in the US–Russia relations that has been registered by more sensitive measures. Carter and Stone (2015) have shown that vote-buying through foreign aid payments further complicates the inferential application of voting data in that the voting behaviour of weak democracies reflects not so much their preferences as the strategic costs of aid conditions. 

Table (3)

Illustrating the Divergence of Blocs Through Selected UNGA/UNSC Resolutions. 

Note: This information was compiled from the findings of Fotheringham (1946–1959), Mosler and Potrafke (2020), and Brangwin (2025) as discussed in the previous section. 

-Institutional Fragmentation 

The fourth category emphasizes the increase of institutional strain, documented by Mosler and Potrafke (2020). The researchers found a 7.2 percentage- point reduction in the agreement of the Western allies when voting along with the US during Donald Trump’s presidency, with that decline taking place in relation to resolutions regarding the Middle East and with NATO states suffering the hardship most of all. Brangwin (2025) accounts for the decline of the number of resolutions adopted by the UN Security Council (UNSC). As of 2021, there were 57 resolutions and only 1 veto. By 2025, the number of resolutions dropped to 24 against 2 vetoes, whereas the number of votes in favour of pro-Ukrainian resolutions similarly dropped from 141 to 93 votes within two years. In addition to that, it was for the first time ever that the USA voted along with Russia against Ukraine both in the UNGA and UNSC. 

Interpretation and Analysis 

These collective findings support three interconnected theoretical claims about the nature and significance of studying UNGA voting. 

The first claim focuses on the existence of “blocs.” The presence of a consistent East-West/ North-South divide spanning nearly eighty years of data suggests that this dual structure is a lasting feature of the post-war international system rather than a result of specific measurement methods. However, the shifts observed within this structure, such as the dissolution of the Soviet bloc and changes in alignment, indicate that blocs should be seen as evolving formations that depend on resolutions and renegotiations, rather than fixed alliances. There are methodological implications on this reframing, which emphasises the need for careful consideration of historical context, while comparing bloc memberships over time. 

The second claim addresses the measurement methods used in this research. The differences between S-scores and ideal-point estimations of data in capturing the Gorbachev transition, highlight that these indices are representative of very distinct theoretical concepts. Evolving measurement methods in the field should be viewed as a series of interpretations of “alignment,” each based on different assumptions about how preference changes are expressed. Consequently: findings from different measurement traditions are not advertently compatible; prevalent and amalgamated disquisitions urge an acknowledgement towards this methodological diversity. 

The third claim posits a specificity on recent data’s broader implications. Changes in the adoption of the UNSC resolution, shifts regarding pro-Ukraine voting support, and U.S.-Russia voting convergence- suggest a departure from previous bloc realignments, particularly owing to the involvement of a key permanent member. These observations, in convergence with the slow response of UNGA data to rapid geopolitical shifts, indicate an underlying potentiality to expound upon underestimation of contemporary institutional fragmentation. This allocates furthermore avenues for prospective research, such as synthesising UNGA data with more dynamic indicators to determine if the current period signifies a temporary disruption in the stable Western core; or the plausible start of a fundamental restructuring of the postwar multilateral order. 

Conclusion 

This study set out to move beyond bipolar frameworks and ask whether a multipolar structure of alignment is now visible in UNGA voting. Network density was calculated as 0.5548, meaning there is somewhat of a global consensus, though high density can mean the majority of states agree on trivial issues creating an illusion of no polarity. The community detection showed the same structure reported in the Results section: Bloc 0 (Non-aligned/European Pivot), led by San Marino, Malta, and Cyprus, with weak voting consistency; a fully isolated two-node component containing only the US and Israel; and Bloc 1 (Cohesive Global South), the largest and strongest connected bloc, containing Mozambique, Thailand, and Singapore, with the highest average voting consistency and eigenvector centrality of ~0.095. These results confirm the findings of Benati and Capurri (2022), who noted that the USA was statistically isolated on its own, especially in 2004 when it supported Israel on its own, suggesting this isolation is not an anomaly but a structural feature of modern UNGA voting. Bloc 1’s centrality also supports Macon, Mucha, and Porter’s observation that modularity based on G77 membership has steadily increased, especially in the post-Cold War period, positioning the Global South rather than the West as the Assembly’s new centre of gravity. The qualitative review situates these results within a much longer arc. Fotheringham’s Cold War data showed the U.S. and the U.S.S.R. disagreed 57.6% of the time, achieving common ground in only 20.4% of the total votes, with a success index of 74.6% for the U.S. against 38.8% for the U.S.S.R. The cross-cutting anti-colonial axis he identified, later formalized as a North-South dimension by Bailey and Voeten (2018) as a statistically significant North-South axis from the mid-1960s to the mid-1980s tied to the rise of the Non-Aligned Movement and G-77, anticipates precisely the bloc realignment detected here. 

At the same time, Adarkwah, Sabel, and Zilja found that UNGA agreement indices are essentially uncorrelated with global economic sanction data, and Carter and Stone showed that vote-buying through foreign aid payments further complicates the inferential application of voting data cautions that apply equally to this paper’s own method. Taken together, three conclusions follow: First, bloc structure persists but is no longer purely East-West; it now centres on a US-Israel isolate against a consolidated Global South.  Second, measurement differences mean cross-study comparisons must be made cautiously. Third, the institutional strain documented by Mosler and Potrafke, and Brangwin’s finding that it was for the first time ever that the USA voted along with Russia against Ukraine both in the UNGA and UNSC, suggests this may be the start of a deeper restructuring of the postwar multilateral order rather than a temporary disruption.  

Acknowledgement 

We express our gratitude towards Mehar Kaur, Namburi Gayatri, Anushruti Rathore, Akshat Kumar Loomba and Harshul Bhadoria for their contributed to this research paper. 

​​References​ 

-Abate, Y., & Brunn, S. D. (1977). The Emergence of African Voting Blocs and Alliances in the United Nations, 1961–1970. The Professional Geographer, 29(4), 338–346. https://doi.org/10.1111/j.0033-0124.1977.00338.x 

-Adarkwah, G. K., Sabel, C. A., & Zilja, F. (2026). Measuring Geopolitics: The Promise and Limits of UNGA Voting Data for IB Research. Journal of International Business Policy, 9(1), 102–123. https://doi.org/10.1057/s42214-025-00237-3 

-Airaudo, F. S., De Soyres, F., Richards, K., & Santacreu, A. M. (2025, March 21). Fragmentation? Revisiting the Ideal Point Distance Measure of Geopolitical Distance. https://www.federalreserve.gov/econres/notes/feds-notes/fragmentation-revisiting-the-ideal-point-distance-measure-of-geopolitical-distance-20250321.html?hl=en-US#:~:text=Second%2C%20economic%20votes%20may%20better%20reflect%20true,U.S.%2C%20albeit%20from%20a%20strong%2DChina%20starting%20point. 

-Bailey, M. A., Strezhnev, A., & Voeten, E. (2015). Estimating Dynamic State Preferences from United Nations Voting Data. Journal of Conflict Resolution, 61(2), 430–456. https://doi.org/10.1177/0022002715595700 

-Bailey, M.A. and Voeten, E. A Two-Dimensional Analysis of Seventy Years of United Nations Voting. Public Choice 176, 33–55 (2018). https://doi.org/10.1007/s11127-018-0550-4 

-Benati, S., & Capurri, A. (2026). The Alignment Index and its Application to Voting at the United Nations General Assembly. Quality & Quantity. https://doi.org/10.1007/s11135-026-02814-x 

-Binder, M., and Payton, A. L. (2022). With Frenemies Like These: Rising Power Voting Behavior in the UN General Assembly. British Journal of Political Science, 52(1), 381–398.  https://doi.org/10.1017/S0007123420000538?utm_source=chatgpt.com

-Brangwin, N. (2026, April 28). UN Voting Patterns Reflect the Changing Geopolitical Landscape. https://www.aph.gov.au/About_Parliament/Parliamentary_departments/Parliamentary_Library/Research/FlagPost/2025/September/UN_voting_patterns_reflect_the_changing_geopolitical_landscape 

-Carter, D. B., & Stone, R. W. (2014). Democracy and Multilateralism: The Case of Vote Buying in the UN General Assembly. International Organization, 69(1), 1–33. https://doi.org/10.1017/s0020818314000186 

-Dahal, D. R. (2024). The Shifting Geopolitics. Journal of Political Science, 24(1), 185-201. https://doi.org/10.3126/jps.v24i1.62863 

-De Oliveira, A. N. C., Geiger, L. M., & Ugar, E. T. (2025). The United Nations General Assembly in a Multipolar World Order: An Analysis of Roll Call Votes on UNGA Resolutions Addressing Russia’s Military Actions in Ukraine. Chinese Political Science Review. https://doi.org/10.1007/s41111-025-00292-4 

-Destradi, S., & Vüllers, J. (2024). Populism and the Liberal International Order: An Analysis of UN Voting Patterns. The Review of International Organizations, 20(4), 1053–1080. https://doi.org/10.1007/s11558-024-09569-w 

-Draguljic, ́ G. and Anspach, N. M. (2025) Southern Solidarity? A Mixed-Methods Analysis of The G77 in the United Nations General Assembly. Global Studies Quarterly, https://doi.org/10.1093/isagsq/ksaf027 

-Fjelstul, J., Hug, S., & Kilby, C. (2021, May 31). What can we learn by collecting more complete UNGA voting records? University of Geneva. https://www.unige.ch/ses/spo/static/simonhug/unga.html  

-Fotheringham, P. (1961). Voting Patterns in the General Assembly of the United Nations [MA Thesis, University of Manitoba]. https://mspace.lib.umanitoba.ca/bitstream/1993/29332/1/Fotheringham_Voting_patterns.pdf 

-Gallop, M. and Minhas, S. (2021). A Network Approach to Measuring State Preferences. Network Science, 9(2), 135–152. doi:10.1017/nws.2020.44 

-Hafner-Burton, E. M., Kahler, M., & Montgomery, A. H. (2009). Network Analysis for International Relations. International Organization, 63(3), 559–592. https://doi.org/10.1017/s0020818309090195 

-Khan, M. Z. I. (2020). Is Voting Patterns at the United Nations General Assembly a Useful Way to Understand a Country’s Policy Inclinations: Bangladesh’s Voting Records at the United Nations General Assembly. SAGE Open, 10(4). https://doi.org/10.1177/2158244020961117 

-Kim, B., Niu, X., Hunter, D., & CaO, X. (2023). A Dynamic Additive and Multiplicative Effects Network Model with Application to the United Nations Voting Behaviors. The annals of applied statistics, 17(4), 3283–3299. https://doi.org/10.1214/23-aoas1762 

-Kim, S. Y., & Russett, B. (1996). The new politics of voting alignments in the United Nations General Assembly. International Organization, 50(4), 629–652. https://doi.org/10.1017/s0020818300033531 

-Kirsamer, N., & Riha, K. (2025). Mapping Geopolitical Fragmentation: A Bilateral Index for Economic Analysis. In OeNB Bulletin Q4/25 (p. 33). https://www.oenb.at/dam/jcr:4fd70043-bd6b-49ad-aa6a-b8bac2973e93/oenb-bulletin-q4-25-mapping-geopolitical-fragmentation.pdf?hl=en-US 

-Kusari, C., Roy, S., & Sengupta, S. (2026, June). Understanding geopolitical alignments through covariate augmented spectral clustering of heterogeneous UNGA voting data. https://arxiv.org/html/2606.25490v1 

-Le, M.-T., Lawlor, M., Russett, B. M., Sweeney, J., & Zucker, S. W. (2013). Analysis of UN voting patterns via diffusion geometry and thematic clustering. In 2013 IEEE 13th International Conference on Data Mining Workshops (ICDMW) (pp. 544–551). https://doi.org/10.1109/ICDMW.2013.81  

-Lee, E., & Stek, P. E. (2016). Shifting Alliances in International Organizations: A Social Networks Analysis of Co-Sponsorship of UNGA Resolutions, 1976-2012. Journal of Contemporary Eastern Asia, 15(2), 191–210. https://doi.org/10.17477/jcea.2016.15.2.191 

-Lees, N. (2023). The Endurance of the G77 in international relations: South–South ideology and voting at the United Nations 1970–2015. Japanese Journal of Political Science, 24(3), 310–330. doi:10.1017/S1468109923000105 

-Lees, N. (2024). Is World Politics Class Politics? States, Social Forces and Voting in the United Nations General Assembly 1946–2020. Cooperation and Conflict, 61(2), 346–372. https://doi.org/10.1177/00108367241269623 

-Macon, K. T., Mucha, P. J., & Porter, M. A. (2011). Community Structure in the United Nations General Assembly. Physica a Statistical Mechanics and Its Applications, 391(1–2), 343–361. https://doi.org/10.1016/j.physa.2011.06.030 

-Matsuoka, T., Tsuchiya, T., Tohkairin, T., Otsuka, K., & Sakamoto, T. (2026). From Speeches to Resolutions: Veto Politics and Discursive Dominance in the UN Security Council. Journal of Economic Policy Reform, 1–20. https://doi.org/10.1080/17487870.2026.2689940 

-Mosler, M., & Potrafke, N. (2020). International political alignment during the Trump presidency: voting at the UN general assembly. International Interactions, 46(3), 481–497. https://doi.org/10.1080/03050629.2020.1719405 

-Muhammad, A. (2023). The Geopolitical Implications of Shifting Alliances in a Multipolar World. Zenodo (CERN European Organization for Nuclear Research). https://doi.org/10.5281/zenodo.8382207 

-Pauls, S. D., & Cranmer, S. J. (2017). Affinity Communities in United Nations Voting: Implications for Democracy, Cooperation, and Conflict. Physica a Statistical Mechanics and Its Applications, 484, 428–439. https://doi.org/10.1016/j.physa.2017.04.177 

-Russett, B. M. (1966). Discovering voting groups in the United Nations. American Political Science Review, 60(2), 327–339. https://doi.org/10.2307/1953359 

-Signorino, C. S., & Ritter, J. M. (1999). Tau-b or Not Tau-b: Measuring the Similarity of Foreign Policy Positions. International Studies Quarterly, 43(1), 115–144. https://doi.org/10.1111/0020-8833.00113 

-Steinert, C. V., & Weyrauch, D. (2024). Belt and Road Initiative Membership and Voting Patterns in the United Nations General Assembly. Research & Politics, 11(1). https://doi.org/10.1177/20531680241233784 

-United Nations. (n.d.). United Nations Charter (full text) | United Nations. https://www.un.org/en/about-us/un-charter/full-text 

-Voeten, E. (2013). Data and Analyses of Voting in the United Nations. In Routledge Handbook of International Organization. Routledge. 

-Voeten, E. (2026). Conceptualizing and Measuring Geopolitical Alignments. Annual Review of Political Science 29:273-288. https://doi.org/10.1146/annurev-polisci-041924-012436