International Scientific Collaboration As A Unity Practice

Science wasn't always international. The Royal Society of London, founded in 1660, was an English institution. The French Academy of Sciences was French. For most of the 17th and 18th centuries, national scientific cultures competed more than they cooperated. Isaac Newton and Gottfried Wilhelm Leibniz fought a famous bitter priority dispute over calculus that was as much about English vs. Continental prestige as it was about mathematics.

What changed? Three forces.

First, the cost of instruments exceeded the budget of any single nation. By the 19th century, observatories required expensive equipment that could only be justified by sharing data. By the 20th century, particle accelerators were beyond any individual nation's means. Cooperation became a technical necessity.

Second, the phenomena studied were planetary. You can't understand global weather from one country. You can't track migratory species from one country. You can't observe an eclipse from one longitude. International Geophysical Years, starting in 1882, were the first sustained response.

Third, the scientific method itself demands multiple perspectives. A result you cannot show to a hostile reviewer is not a robust result. Internationalism in science is not a value imported from elsewhere. It is what the epistemic method requires.

The most extraordinary story of scientific collaboration is its survival through the Cold War.

The International Geophysical Year (1957–58) brought 67 countries together, including the US and USSR at the height of their nuclear confrontation. It launched Sputnik (the USSR's contribution to the IGY), produced the first evidence of the Van Allen radiation belts, and laid the groundwork for the Antarctic Treaty of 1959.

The Pugwash Conferences on Science and World Affairs, founded in 1957, brought American and Soviet scientists together to discuss nuclear weapons. Joseph Rotblat, one of the founders, later won the Nobel Peace Prize for this work. The conferences continue today.

The Apollo-Soyuz Test Project (1975) docked an American and Soviet spacecraft in orbit. The astronauts and cosmonauts shook hands on live television. Kissinger and Brezhnev used it as symbolic cover for a political thaw, but the engineers who made it work had been collaborating quietly for years.

The Human Frontier Science Program (1987), initially a Japanese initiative, funds cross-border collaboration on fundamental biology. It was explicitly designed to keep scientific exchange open regardless of political tensions.

None of this was easy. Scientists were surveilled. Papers were censored. Some researchers lost careers or worse for their contacts across the Iron Curtain. But the channel stayed open when nothing else did. The collaboration survived because the knowledge-making process required it.

CERN (European Organization for Nuclear Research), 1954–present. Founded to reverse European brain drain after WWII, now a consortium of 23 member states plus participating nations. The Large Hadron Collider is the largest machine ever built — a 27km ring under the French-Swiss border. The 2012 Higgs boson discovery involved the ATLAS and CMS collaborations, each with over 3,000 authors on the discovery papers. Author lists on modern physics papers sometimes stretch past 5,000 names. Nationality is invisible in the author list. The collaboration forms the nationality.

The International Space Station, 1998–present. NASA, Roscosmos, ESA, JAXA, CSA. Continuously inhabited since 2000. The largest single structure ever assembled in space. The ISS survived even the collapse of US-Russia relations after 2014 and the war in Ukraine — not because Western governments wanted to keep the partnership going but because the engineering and safety dependencies were so deeply intertwined that breaking the collaboration was practically impossible. This is the infrastructure-as-politics point made concrete. Once you build something that must be operated together, the partnership calcifies even when it should politically dissolve.

The Hubble Space Telescope, 1990–present, and JWST, 2021–present. Joint NASA/ESA (and CSA for JWST). Every astronomer on Earth can apply for observing time through a common proposal process. The data goes into the Mikulski Archive for Space Telescopes, where anyone can download it after a one-year proprietary period. The proprietary period itself is short by commercial standards — the default is openness, with a small exclusive window to reward the proposer.

ITER, 1985–ongoing. The International Thermonuclear Experimental Reactor, being built in Cadarache, France. 35 nations participating. Aims to demonstrate sustained fusion power. The project has been late and expensive — first plasma was expected in 2025, now pushed further. But the fact that China, Russia, the EU, India, Japan, Korea, and the US are still building it together, through every geopolitical shock of the last forty years, is astonishing.

LIGO (Laser Interferometer Gravitational-Wave Observatory), 1992–present. First detection of gravitational waves in 2015, confirming a prediction of Einstein's general relativity. LIGO has partner observatories in Italy (Virgo), Japan (KAGRA), and soon India. The worldwide network allows triangulation of sources. First observations of neutron star mergers (2017) triggered follow-up by dozens of telescopes within hours. This is a global rapid-response scientific infrastructure.

The Event Horizon Telescope, 2017. The first image of a black hole (M87*, released 2019) was produced by synchronizing eight radio telescopes on four continents into a single Earth-sized virtual telescope. It required atomic clocks, hard drives flown between continents, and supercomputing on data sets that couldn't be transferred over networks. A planetary-scale instrument assembled from parts.

The Human Genome Project, 1990–2003. Led by the US NIH and UK Wellcome Trust, with major contributions from France, Germany, Japan, and China. The most consequential decision was the 1996 Bermuda Principles: all sequence data released within 24 hours of generation, placed in the public domain, no patents on raw sequence. This was a deliberate pre-emption of Celera Genomics' attempt to commercialize the genome. The public consortium won the race — or rather, refused to make it a race, and released the reference genome freely.

Robert K. Merton's 1942 essay The Normative Structure of Science identified four norms that he argued define science as an institution. They are often summarized as CUDOS:

- Communalism: scientific knowledge is shared, not privately owned. - Universalism: claims are evaluated by impersonal criteria, not by the identity of the claimant. - Disinterestedness: scientists act for the advancement of knowledge, not personal gain. - Organized Skepticism: no claim is exempt from critical scrutiny.

Merton's norms are aspirational. Real science violates them regularly. But they remain the explicit ideal, and the ideal shapes behavior. When a scientist commits fraud, the community reacts as if a trust has been violated — not just because fraud is bad, but because the norms specifically forbid it.

John Ziman's Real Science (2000) offered a counter-view: what Merton described is academic science, funded by governments, pursued in universities, rewarded through publication and tenure. Industrial science operates on different norms: PLACE — Proprietary, Local, Authoritarian, Commissioned, Expert. When science moves from the academy to the corporation, communalism becomes proprietary, universalism becomes local, disinterestedness becomes commissioned. This is not a criticism of industrial science per se. It's a description of a different institution that happens to use the same techniques.

The tension between Mertonian science and Zimanian science is where most of the problems in contemporary science live.

Commercial capture. Pharmaceutical research is the paradigm case. A drug that could save lives becomes an asset that must be monetized. The WHO reports that billions of people lack access to essential medicines, many of which are patented. The TRIPS Agreement under the WTO extended patent protection globally, making this a structural feature of the world trade system. COVID-19 vaccines were partly developed with public money, but their intellectual property stayed private. Moderna's mRNA platform was built on years of public research. When the vaccine was released, the company earned billions. The platform was not shared globally. Low- and middle-income countries waited years for doses.

Patent warfare. The CRISPR patent dispute between the Broad Institute (Feng Zhang) and UC Berkeley (Jennifer Doudna and Emmanuelle Charpentier) consumed a decade of litigation. Doudna and Charpentier won the Nobel Prize in 2020. The patent fight continues. The field advances slower and more cautiously than it otherwise would.

Access barriers. Scientists from African, Middle Eastern, and South Asian countries are regularly denied conference visas in Europe and North America. Publication fees for open-access journals can exceed a year's budget for researchers in low-income countries. Scientific English is a cost. Subscription-based databases are a cost. The norm of universalism is contradicted by the infrastructure of access.

Dual-use research. Some science is weapons research by another name. The 2011 controversy over gain-of-function research on H5N1 influenza — where researchers created more transmissible versions of a deadly virus to study them — split the scientific community. The openness norm conflicts with biosecurity. There is no good resolution.

Fraud and replication. The Schön scandal (physics, 2002), the Hwang scandal (stem cells, 2005), the replication crisis in psychology (2010s) and biomedicine (ongoing) — each revealed fault lines in the norm system. Peer review failed. Prestige journals failed. The system corrected, but slowly, and at a cost to public trust.

Geopolitics is reasserting itself. The China Initiative in the US (2018–2022) investigated researchers with Chinese ties, chilled collaboration, and harmed careers. Export controls on AI chips and models are now shaping what can be published. Russia was expelled from CERN experiments in 2022. The bridge that survived the Cold War is under more strain now than it has been in a long time.

The Intergovernmental Panel on Climate Change is worth studying separately because it operationalizes scientific collaboration as direct input to policy.

Founded in 1988 by the WMO and UNEP. Three working groups: I (physical science), II (impacts and adaptation), III (mitigation). Assessment reports every 6–8 years. The Sixth Assessment, completed in 2023, involved over 700 authors from 90+ countries. Thousands of expert reviewers. Government representatives line-by-line approve the Summary for Policymakers.

The tension is built in. Scientists produce the chapters. Governments approve the summaries. Governments sometimes demand language changes that soften findings. Scientists have walked out. The system persists because the alternative — purely political negotiation without a scientific baseline — would be worse.

The IPCC is a template. IPBES (biodiversity) followed. A future iteration for oceans, for AI governance, for planetary commons management, can be built. The model is imperfect and the model works.

Scientific collaboration demonstrates several things about large-scale human cooperation:

1. Trust is an output, not a starting condition. You don't start by trusting your Russian colleague. You start by checking their data. Trust accrues from the accumulated experience of checking and finding it sound.

2. Procedures can substitute for shared values. Scientists from every political and religious background cooperate on the LHC because the procedures — instrumentation, statistics, peer review — are narrow and specified. The procedures don't require agreement on values. They require agreement on method.

3. Infrastructure is politics at a slower clock speed. Building shared infrastructure creates cooperation that persists through political shocks. ISS. CERN. Hubble. These are physical facts on the ground that override short-term political tensions.

4. Openness is a strategy, not just a value. The Human Genome Project's Bermuda Principles were a deliberate move to pre-empt enclosure. The moral claim and the strategic claim coincided. Open beat closed.

5. Sanctions for norm violation come from peers, not police. Fraud in science is punished by loss of reputation, retraction, exclusion. The sanctions are real but decentralized. No central authority exists. The community is the authority.

6. Failure modes are instructive. Where science fails unity — commercial capture, patent warfare, geopolitical interference — shows us where the system needs to be strengthened. The breakdowns are signals, not indictments.

The Republic of Science (Michael Polanyi, 1962). Polanyi argued science operates like a market with invisible coordination. No central authority tells scientists what to work on, but the collective result is coherent progress. Competitive cooperation.

Research Infrastructures as Meta-Institutions. Big Science facilities (CERN, ITER, LIGO) are not just tools. They are institutions that generate their own norms, communities, and governance practices. A framework worth expanding to other planetary challenges.

The Bermuda Model. Data-sharing norms negotiated by scientific communities before commercial or governmental interests can capture the field. Applicable to genomics, neuroscience, AI, astronomy, climate.

Post-Normal Science (Funtowicz and Ravetz, 1993). For problems where facts are uncertain, values are disputed, stakes are high, and decisions urgent, the classical scientific method is insufficient. New practices of extended peer review that include affected stakeholders are needed. Relevant to climate, pandemics, AI.

1. Read one open-access paper. Go to arXiv, bioRxiv, or PubMed Central. Pick a paper from a country you've never been to. Read the abstract, scan the methods, read the conclusion. You just participated in the most successful cooperative institution humans have built.

2. Audit your science consumption. Where does your information about science come from? Journalists? Social media? Press releases? When was the last time you read a primary source? The cooperation is producing the documents. Are you reading them?

3. Find a citizen science project. Zooniverse hosts hundreds. Contribute to one. You are now part of a distributed research collaboration.

4. Map a collaboration. Pick a recent major discovery — the Higgs boson, an exoplanet, a new cancer drug, a gravitational wave. Trace its author list. How many countries? How many institutions? How many disciplines? You will find it impossible to tell the "national" story.

- Robert K. Merton, "The Normative Structure of Science" (1942) - John Ziman, Real Science: What It Is and What It Means (2000) - Michael Polanyi, "The Republic of Science: Its Political and Economic Theory" (1962) - Silvio Funtowicz and Jerome Ravetz, "Science for the Post-Normal Age," Futures (1993) - James Gleick, Genius and The Information (various) - Daniel Kevles, The Physicists (1978) - Alex Soojung-Kim Pang, Empire and the Sun (history of international scientific expeditions) - Francis Collins, The Language of God (for inside accounts of the Human Genome Project) - Lisa Randall, Knocking on Heaven's Door (for inside account of the LHC) - IPCC Sixth Assessment Synthesis Report (2023) - Nicholas Mulder, The Economic Weapon (on the history of sanctions and their effect on scientific exchange)

Science is not a neutral activity. It is a civilizational commitment, embodied in specific institutions, maintained by specific norms, vulnerable to specific failures. It is also proof — the most rigorous proof we have — that the statement "we are human" is not just an ethical claim. It is an operational reality that produces knowledge no nation could produce alone.

The telescope sees for all of us. The accelerator accelerates for all of us. The genome is ours. Whatever we do with what science teaches — that is the next question. The fact that science can teach it across every line we have drawn is the answer to the question of whether we can cooperate at planetary scale.

We already are. The question is whether we scale the practice to match the moment.