1. The Hidden Grammar of International Relations
International relations scholarship has produced sophisticated accounts of why states cooperate or conflict: power distributions, institutional design, interest congruence, domestic politics. What it has largely neglected is the layer of cognition beneath all these factors — the conceptual frameworks through which state actors perceive their interests, their counterparts, and the very nature of the issues they negotiate.
This is not a trivial omission. When negotiators sit across from one another at the table, they are not processing raw reality. They are processing it through cognitive frames — structured mental schemas that determine which facts are salient, which relationships are causally relevant, and which outcomes feel like victories or defeats. Two delegations with identical material interests but incompatible frames will frequently fail to reach agreement. Two delegations with divergent material interests but aligned frames will frequently find accommodation.
This study presents the first systematic, large-scale empirical investigation of conceptual framing in multilateral diplomacy. We analyse 6,400 documents spanning three decades and 180 countries, and we demonstrate that the cognitive architecture of diplomatic discourse is not epiphenomenal — it is constitutive of outcomes.
2. Conceptual Framing in Diplomatic Discourse: Theoretical Foundations
The theoretical lineage of this study runs from Goffman's frame analysis through Lakoff and Johnson's conceptual metaphor theory, Fillmore's frame semantics, and the policy framing literature initiated by Schön and Rein. What distinguishes our approach is the application of these frameworks at the scale of multilateral diplomacy and the operationalisation of frame alignment as a measurable, predictive variable.
We define a diplomatic frame as a structured set of conceptual elements — entities, relations, values, and entailments — that organises how a policy domain is understood by a state actor. Frames are not merely rhetorical positions; they determine what counts as a problem, what solutions are conceivable, and what trade-offs feel acceptable. The NATIONAL INTEREST frame, for example, constitutes the state as a bounded, competitive unit maximising relative gains; the GLOBAL COMMONS frame constitutes states as co-stewards of shared resources with common fate. These frames generate radically different inference patterns about the same set of geopolitical facts.
Frame alignment — the degree of conceptual overlap between the frames active in two or more negotiating parties — is distinct from preference alignment. Two states may have conflicting preferences within a shared frame (both want maximum share of a commons) or compatible preferences across incompatible frames (both want climate action but one frames it as ECONOMIC OPPORTUNITY while the other frames it as MORAL OBLIGATION). Our central empirical claim is that frame alignment, independently of preference alignment, predicts cooperative outcomes.
3. The Geopolitical Frame Stack: A New Analytical Construct
A central theoretical contribution of this study is the concept of the geopolitical frame stack — the layered set of compatible cognitive frames simultaneously active in a diplomatic negotiation. Unlike prior work treating frames as discrete and competing, our corpus analysis reveals that successful negotiations are characterised by the construction of multi-frame scaffolds in which several compatible frames reinforce each other across different dimensions of the negotiation.
Consider the negotiation of the 2015 Paris Agreement. Our analysis identifies at least four co-active frames in the final text and its preparatory discourse: CLIMATE AS COMMON THREAT (security frame), CLEAN ECONOMY AS OPPORTUNITY (economic frame), INTERGENERATIONAL STEWARDSHIP (moral frame), and SOVEREIGN PLEDGES (institutional frame). The agreement's resilience — and its broad ratification — can be attributed in part to this frame stack depth: parties with different primary frames could all find a compatible entry point into the agreement's conceptual structure.
By contrast, the failed Copenhagen Accord negotiations of 2009 were characterised by a shallow frame stack, dominated by competing NATIONAL BURDEN-SHARING frames that activated zero-sum inference patterns across major blocs. Our retrospective frame analysis of Copenhagen reveals that no significant frame convergence engineering was attempted by any facilitation team — a finding with direct implications for future multilateral process design.
4. Empirical Results: Frame Alignment as a Predictor of Treaty Outcomes
To test whether frame alignment predicted treaty outcomes independently of material variables, we constructed a dataset of 847 multilateral agreements (concluded and failed) from 1990–2024, coding each for: (1) frame alignment score between principal parties, (2) material interest alignment, (3) power asymmetry, (4) institutional support, and (5) negotiation duration. Frame alignment was scored using our dyadic resonance metric, which measures the proportion of conceptual entailments shared between the dominant frames of two parties.
Regression analysis revealed that frame alignment was the single strongest predictor of treaty completion (β = 0.67, 95% CI [0.61, 0.73]), accounting for 45% of variance in outcomes when entered alone. Material interest alignment (β = 0.29) and power symmetry (β = 0.22) showed significant but substantially smaller effects. Institutional support quality (β = 0.18) was the weakest predictor.
These findings hold across all five policy domains studied, though with variation in effect size. Frame alignment was most predictive in health policy negotiations (β = 0.74) — where the DISEASE AS COMMON ENEMY frame has achieved near-universal salience — and least predictive (though still dominant) in trade negotiations (β = 0.58), where material interests are more precisely quantifiable and frame diversity remains high.
5. Frame Convergence Engineering: The Practitioner Implications
The most immediately actionable finding of this study concerns frame convergence engineering — deliberate interventions designed to shift the cognitive framing of one or more parties toward greater alignment. Three facilitated negotiation pilots conducted in partnership with the UN Institute for Training and Research (UNITAR) provided the first experimental evidence that structured frame convergence interventions can measurably accelerate negotiations.
In each pilot, a frame diagnostic was conducted at the outset: a cognitive linguistic analysis of each party's position papers, public statements, and internal briefing documents to map their active frame systems. A frame convergence map was then constructed, identifying zones of existing alignment and potential bridging frames — frames that neither party was currently using but that were conceptually compatible with both parties' existing frame systems.
Facilitators then introduced bridging frames through carefully designed agenda-setting, process language, and proposal framing. Across the three pilots, mean negotiation time fell from 26 months (historical baseline for comparable negotiations) to 17 months — a 34% reduction. Critically, agreement durability was also higher: retrospective assessment at 3 years showed no significant erosion in any of the three agreements, compared to a 31% partial-collapse rate in matched historical comparators.
These results suggest that frame convergence engineering is not merely theoretically interesting but practically transformative for multilateral process design. We recommend its integration into standard diplomatic facilitation training programmes.
6. The Post-1995 Sovereignty-to-Stewardship Transition
One of the most significant longitudinal findings of this study concerns a macro-level conceptual transition detectable across the full 34-year corpus. From 1990 to approximately 1995, multilateral discourse was dominated by SOVEREIGNTY frames: the nation-state as inviolable, self-determining unit; international obligations as voluntary and strictly reciprocal; global problems as aggregations of national problems requiring national solutions.
From 1995 onward — accelerating sharply after 2005 — we detect a measurable shift toward STEWARDSHIP frames: the nation-state as trustee of global and intergenerational goods; international obligations as inherent in membership of a common civilisation; global problems as systemic phenomena requiring collective governance architectures. This shift is not merely rhetorical. Our entailment chain analysis shows that STEWARDSHIP framing generates qualitatively different inference patterns about appropriate policy responses, legitimacy conditions, and acceptable burden-sharing arrangements.
The transition is not complete, uniform, or uncontested. Counter-movements reasserting SOVEREIGNTY frames are detectable in roughly 30% of the corpus post-2016, concentrated in a subset of state actors. But the directionality of the trend is unmistakable, and it constitutes the most significant cognitive-architectural shift in multilateral discourse in the period studied — with profound implications for the future design of global governance institutions.
7. Conclusions: Toward a Cognitive Infrastructure for Global Governance
This study demonstrates that the conceptual frames through which state actors perceive and communicate about global challenges are not merely communicative packaging — they are the cognitive infrastructure of international cooperation itself. Frame alignment predicts treaty outcomes more strongly than material interests. Frame stack depth predicts agreement durability. Frame convergence engineering accelerates negotiation and improves long-term stability.
These findings have profound implications for how we design and conduct multilateral diplomacy. If the cognitive architecture of negotiation matters more than we have assumed, then investing in that architecture — through frame diagnostics, convergence facilitation, and the deliberate cultivation of shared conceptual frameworks — is among the highest-leverage interventions available to the international community.
We call for the establishment of a Diplomatic Cognitive Infrastructure programme within the UN system: a standing capacity for frame analysis and convergence facilitation, deployable across all major multilateral negotiation contexts. The tools exist. The evidence is now sufficient. The question is whether the international community has the conceptual sophistication to use them.