Moving Past a Binary Framework: Openness and Security in International Research Collaboration

Abstract

The phrase “as open as possible and as closed as necessary” originated in 2016 as guidance for research data governance under the EU’s Horizon 2020 Open Research Data Pilot. Over the following decade, it migrated into national security guidelines, international policy frameworks, and institutional guidance as a governing principle for international research collaboration itself. This migration has had unintended consequences. Applied to collaboration rather than data, the phrase’s binary logic implies a zero-sum relationship between openness and security—one in which more of one requires less of the other. This framing alienates researchers who experience security measures as threats to their professional freedom, while simultaneously providing institutional cover for security services to overcorrect toward restriction. This issue brief argues that openness and security are not competing goals but complementary and mutually reinforcing ones, and offers four recommendations for revising policy language to reflect that reality and better support the international research ecosystem it is meant to protect.

Introduction

Over the past decade, the phrase “as open as possible and as closed as necessary” has become a common shorthand in discussions of research security. Often paired with calls to “balance openness and security,” it is now embedded in policy documents, institutional guidelines, and international frameworks[v] across the G7, the European Union, and beyond. Horizon Europe itself, which the European Commission describes as “designed to be as open as possible and as closed as necessary,” reflects this framing.

While well-intentioned, this framing has unintended consequences that go beyond mere linguistic precision. The language used to describe the relationship between openness and security shapes how researchers, institutions, and security services understand their respective roles and, crucially, whether they see themselves as partners or adversaries in the same project. When the dominant metaphor is one of finding a balance or a trade-off between two competing goals, the practical effect is often that security is understood as a constraint on collaboration rather than an enabler of it. Researchers interpret calls suggesting that it may be “necessary” to “close” their activities as pressure to limit their international partnerships. Security communities, applying the same binary framing, are inclined to err toward greater restriction when the framing suggests that restriction is the primary or only instrument of protection. Institutions caught between these two orientations struggle to develop coherent, workable policies that serve both while also arguing that the approach fails to allow for the necessity of academic freedom and open science.

A reconsideration of this language is overdue, not as an abstract intellectual exercise, but because the framing actively undermines the goals it is meant to serve. At a moment when international research collaboration is under pressure from multiple directions—geopolitical competition, foreign interference, and the securitization of scientific fields—the last thing the research community needs is policy language that inadvertently reinforces the perception that security and collaboration are in fundamental tension, if not entirely incompatible. Getting this language right is a precondition for getting the implementation right.