{"id":"e7183431-3fec-47e5-b1c5-65260c8ea2a8","arxiv_id":"2505.22476","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":3.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A coalition of physicists and historians argues that civilian-only civil clauses at European research institutions must be preserved, and that current EU and German dual-use research policies are eroding them.","lead":"This paper is a position statement from the Science4Peace Forum, a panel of mostly physicists and historians, arguing that European universities and research bodies should keep their commitments to purely civilian and peaceful research. It reviews the history of these civil clauses, surveys how different countries handle military research at universities, and warns that recent EU and German policy pushes toward dual-use research threaten open international science.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central policy claim rests on an untested empirical premise: that admitting military or dual-use research into civilian institutions will actually restrict openness, deter international partners, and endanger staff; the paper argues this by plausibility and anecdotes, not evidence.","rationale":"The reader's weakest assumption identifies exactly the same load-bearing concern: the policy conclusions rest on an unproven causal premise about the effects of civilian-only versus dual-use research. My stress test adds specificity by pointing to the sections where the premise is asserted without evidence (4.3, 5.1) and by noting that Section 5.2's fence example is a partial counterexample to the claim that military links inevitably contaminate civilian research. Because the paper is a policy position essay rather than an empirical study, and because it presents no data that could test the mechanism, the appropriate verdict remains UNVERDICTED. My concern does not move the verdict; it sharpens the reason. A natural experiment exists and could settle the causal question, so the concern is concrete and testable. I do not see an internal inconsistency or a fraudulent element; the issue is that a central empirical claim is unsupported. The secondary factual assertion about sanctions being unprecedented without a UN Security Council mandate is also contestable and unreferenced, but it is less central to the paper's main recommendation than the causal premise, so I focus on the latter.","tokens_in":20311,"tokens_out":2706,"duration_ms":33802,"concrete_test":"Identify institutions that relaxed civil clauses or accepted military/dual-use funding (e.g., Flemish universities post-2018, Bavarian universities post-2024, Japanese universities receiving ATLA funds post-2015) and compare their open-access publication share, international co-authorship, foreign researcher visits, and partner-institution collaboration intent before versus after the policy change, using difference-in-differences against matched civilian-only institutions. If these openness indicators do not decline, the paper's central causal premise fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing concern is the causal premise connecting civilian-only research to beneficial outcomes and military involvement to harms. Section 5.1 ('From Open Science to Military Secrecy') claims that military research would necessarily restrict open access, make future international contributions unlikely, increase cyber threats, and make DESY a military target, but offers no comparative evidence; it argues from plausibility. Section 4.3 asserts 'Those who talk with each other are less likely to shoot at each other' with no empirical support. Section 7's conclusion ('a major threat to universal, peaceful and international research') depends on these mechanisms. The paper itself weakens the claim in Section 5.2, noting CERN collaborates with institutes that separate military researchers by a fence, implying contamination is not inevitable. Without evidence that dual-use research measurably reduces openness or collaboration, the policy recommendation to preserve civil clauses loses its empirical foundation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper, authored by members of the Science4Peace Forum, argues for preserving Civil Clauses—commitments by universities and research institutions to pursue only peaceful, civilian research—in the face of recent European and German policy initiatives that promote dual-use and military research. It reviews the historical origins of Civil Clauses in postwar Germany and Japan, surveys the current status of civilian-military research separation across several countries, discusses the risks of opening institutions such as DESY to military research, and criticizes sanctions against Russian and Belarusian scientists. The paper concludes that relaxing Civil Clauses is a major threat to universal, peaceful, and international research.","tokens_in":20381,"tokens_out":4150,"duration_ms":52260,"significance":"As a documented position statement, the paper usefully assembles primary policy documents (EU White Paper, ReArm Europe Plan, German coalition agreements, institutional charters) and historical material on Civil Clauses, and it explicitly acknowledges nuance, including acceptable cooperation for arms control and the existence of fenced separations at CERN partner institutes. Its main weakness is that the central empirical claims—that military involvement necessarily restricts openness, deters partners, and endangers staff, and that purely civilian international science builds peace—are asserted from plausibility rather than demonstrated with comparative evidence. The paper provides no machine-checked proofs or new quantitative data; its value is as a policy-oriented synthesis and normative argument. If the causal premises are accepted, the policy recommendations are coherent; if not, the recommendations lack evidentiary support.","major_comments":[{"comment":"The load-bearing causal premise is unsupported. Section 4.3 asserts 'Those who talk with each other are less likely to shoot at each other,' and Section 5.1 claims that introducing military research 'would necessarily' restrict open access, deter international partners, increase cyber threats, and make DESY a military target. No comparative evidence is provided for these mechanisms. The authors should either supply empirical evidence—for example, by examining institutions that relaxed similar restrictions, such as Flanders after 2018, Japan's ATLA-funded university research, or Sweden's Campus Total Defense collaboration—or explicitly reframe the argument as a value-based position that does not depend on contested causal claims. As written, Section 7's conclusion that this is 'a major threat to universal, peaceful and international research' rests on assertion rather than evidence.","section":"Sections 4.3 and 5.1"},{"comment":"The factual claim that 'Now is the first occasion that sanctions are imposed against scientists in national and international institutions without the mandate of the UN Security Council' is stated without citation and is contestable. The paper later notes the 1993 CERN case under a UN mandate, but the unqualified 'first occasion' claim needs a precise definition of scope (which institutions, which sanctions, which historical period) and supporting evidence, or it should be softened to 'to our knowledge.' Because this claim anchors the sanctions critique, its accuracy is important.","section":"Section 3"},{"comment":"There is an internal tension between the claim that military research necessarily harms civilian institutions and the acknowledgment in Section 5.2 that CERN collaborates with institutes that fence off military research. If a fence is sufficient to preserve the civilian character of collaboration, then the harms described in Section 5.1 (restricted access, security risks, partner deterrence) are not inevitable consequences of co-location but depend on specific organizational arrangements. The authors should reconcile these positions and specify the conditions under which military co-location would actually produce the predicted harms.","section":"Sections 5.1 and 5.2"},{"comment":"The paper cites its own earlier position papers (refs. [16] and [17]) as support for central arguments, and Section 5.4 reproduces a FONAS statement co-authored by two of the present authors. This self-referential material should be explicitly identified as position statements rather than independent evidence, and the authors should supplement it with external sources where available. The circularity is not fatal because the argument also relies on external documents, but it should be disclosed for the reader.","section":"Section 5.4 and references [16], [17]"}],"minor_comments":[{"comment":"There are numerous typographical errors, including 'Civilan' in the Section 5.3 title, 'Commision' in references [1] and [4], 'Nucelar' in reference [18], 'CERN?s' in Section 5.2, and an unclosed parenthesis in Section 5.1 after '[76]'. A careful proofreading pass is needed.","section":"General"},{"comment":"The opening speech by Major-General Nares is dated 1949 in Section 2 but appears to be associated with the 1946 reopening in Section 4.3. Please verify the date and make the narrative consistent.","section":"Sections 2 and 4.3"},{"comment":"The first bullet in the call to action has a subject-verb agreement error: 'using scientific results obtained in public universities and research institutions are used' should be 'that scientific results obtained in public universities and research institutions are used.'","section":"Section 6.2"},{"comment":"The country survey is necessarily selective, but the criteria for inclusion are not stated. For example, Palestine is represented by a single mention of 'research topic priorities' without a reference, while other countries receive detailed discussion. Please indicate the basis of selection or mark the section as illustrative rather than systematic.","section":"Section 4.6"},{"comment":"Many references lack access dates, document numbers, or stable identifiers (e.g., refs. [5], [79], [81], [82]). For a paper intended to document policy developments, full bibliographic details would strengthen verifiability.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper is best read as a multi-author advocacy statement rather than as a conventional empirical study. If the journal's scope includes well-documented position papers in physics and society, it could be publishable after the empirical claims are either supported or explicitly reframed as normative. The self-citation pattern is understandable for a forum's own position paper but should be disclosed more prominently."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a policy position paper, not a research preprint. What is actually new is the way it assembles a current snapshot of European policy shifts—EU White Paper, ReArm Europe, the German coalition agreement, the DESY debate—and pairs that with a useful country-by-country survey of how civilian and military research are separated (or not). The historical material on TU Berlin, the Science Council of Japan, and the CERN Convention is well sourced and gives the argument real roots. It is also honest in places: it acknowledges that the civilian–military line is sometimes hard to draw, and John Ellis's piece candidly notes that CERN already collaborates with institutes that use a literal fence to separate military and civilian researchers. That sort of nuance earns credibility.\n\nThe soft spots are concentrated in the causal argument. The paper repeatedly asserts—without comparative evidence—that letting dual-use or military research into civilian labs will restrict openness, scare off international partners, raise cyber risks, and make staff targets. Those outcomes may well occur, but the paper argues from plausibility and anecdote, not from cases where such opening actually happened (the country survey is the natural place for that, but it doesn't systematically mine it). The claim in Section 3 that current sanctions are the first science sanctions imposed without a UN Security Council mandate is contestable and unreferenced. And the use of its own earlier position papers as support is noticeable, though not disqualifying. These flaws matter because the paper's concluding recommendation—preserve civil clauses—is only as strong as the premise that abandoning them brings measurable harms. That premise is sensible but unproven here.\n\nWho gets value from this? Anyone tracking European science policy or the militarization debate, especially around CERN/DESY and the German Zivilklausel movement. It is a useful briefing document, not a contribution to empirical knowledge. I would send it to peer review because the topic is consequential and the documentary foundation is solid, but I would expect reviewers to push hard on the unsupported empirical claims and on the sanctions history.\n\nRecommendation: engage with it as a serious statement, whose recommendations are worth debating. It is not a research paper and should not be evaluated as one.","headline":"A coherent and well-documented position paper arguing to preserve civil clauses; the policy claim rests on plausible but unproven causal premises, and the historical/documentary material is its real strength.","tokens_in":21122,"tokens_out":1368,"would_cite":false,"duration_ms":19074,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper argues that Civil Clauses—institutional commitments to purely civilian research—are under coordinated political pressure, and that defending them is essential to keeping science open, universal, and peaceful.","keywords":["civil clauses","dual-use research","military research","science for peace","open science","science sanctions","research policy","international scientific cooperation"],"falsifier":"Compare institutions that have dropped or weakened Civil Clauses, such as Flanders after 2018, Bavaria after 2024, and Japan's Ministry of Defence funding scheme, with matched civilian-only institutions over five to ten years; if openness, foreign collaborations, and researcher retention do not measurably decline in the dual-use group, the paper's central causal claim would be undercut.","tokens_in":20022,"feed_emoji":"🕊️","tokens_out":6616,"duration_ms":65874,"temperature":0.7,"pith_summary":"After World War II, many universities and research centres adopted Civil Clauses: formal commitments to pursue only civilian and peaceful research. This paper argues that those commitments are now under coordinated pressure from EU and national policies that promote dual-use and military research, notably the EU White Paper and ReArm Europe plan, and from German policy documents and coalition agreements calling for civil-military cooperation. The authors contend that opening civilian institutions to military work would erode open access, deter international partners, turn laboratories into military targets, and ultimately damage science's ability to address climate change, poverty, and war. They also argue that science sanctions against Russian and Belarusian researchers are unprecedented and counterproductive, and that expelled scientists will be slow to return. The paper's positive claim is that preserving the civilian character of public research is a practical condition for universal, open, and peaceful science, not merely a symbolic gesture.","feed_headline":"Keep military research out of civilian labs, paper urges","feed_subtitle":"A 30-author review says EU and German dual-use policies would erode openness, trust, and international collaboration.","key_machinery":"The central object is the Civil Clause, a formal, often self-imposed commitment by a university or research institution to limit research and teaching to civilian and peaceful purposes. The paper treats the 1946 founding of the Technical University of Berlin, Japan's 1950 Science Council pledge, and the 1954 CERN Convention ('no concern with work for military requirements') as the templates. The mechanism doing the work is institutional: a Civil Clause creates a transparent, generalizable criterion for international collaboration, keeps results open, and gives scientists and students a collective basis to resist political pressure to take military funding. The paper argues that removing or weakening these clauses removes a barrier that has protected openness and trust for seventy years.","core_discovery":"The central claim is that the strict separation between civilian and military research, embodied in Civil Clauses, is a load-bearing institution for science and peace. The paper documents the spread of these clauses since 1946, their current legal and political erosion in Germany, Belgium, Japan, Spain, and at the European Union level, and argues that the shift toward dual-use research would irreversibly change the character of laboratories such as DESY and CERN. It holds that international, open, civilian research builds trust between nations, that sanctions against scientists from Russia and Belarus break that trust, and that allowing military research into civilian facilities would restrict access, endanger staff and infrastructure, and drive away the researchers who chose fundamental science precisely because it is non-military. The authors do not claim that the civil-military boundary is always clean, but they maintain that it is worth defending and that the burden of proof lies on those who would blur it.","pith_inferences":["A testable extension would track institutions that removed or weakened Civil Clauses, such as Flanders after 2018, Bavaria after 2024, and Japan with Ministry of Defence funding, against matched civilian-only institutions to measure changes in open-access output, foreign collaboration, and researcher retention.","The paper's trust-building claim implies an empirical prediction: if scientific contact between researchers from rival states increases, cooperation and reduced hostility should follow, a pattern that could be studied with bibliometric data from Cold War and post-Cold War collaborations.","Even if the immediate threat is limited to a few laboratories, the cited policy documents create a norm shift that may be hard to reverse, so the relevant comparison is long-run rather than today's funding flows.","The paper's framing suggests an alternative policy path: instead of opening civilian laboratories, governments could fund dedicated defense research institutes with clear security arrangements, preserving the civilian character of universities."],"forward_implications":["If Civil Clauses are abandoned or weakened, public research institutions will face pressure to accept military and dual-use funding, shifting resources away from civilian research on climate, health, and poverty.","Open access and international collaboration will shrink, because military and dual-use projects require secrecy, export controls, and restricted participation.","Laboratories that host military research become attractive cyber and physical targets in conflicts, endangering staff and nearby civilian populations.","Science sanctions against Russian and Belarusian researchers will deepen mistrust and make it harder to rebuild cooperation after the war, and may push affected scientists toward military work.","The universality and credibility of science will be damaged if science is seen as a geopolitical tool rather than a shared human enterprise."],"supporting_citations":[{"why":"EU White Paper on dual-use potential, the policy the paper opposes.","marker":"[1]"},{"why":"ReArm Europe Plan/Readiness 2030, the EU push for dual-use investment.","marker":"[4]"},{"why":"Nares speech at the founding of TU Berlin, the template for the first Civil Clause.","marker":"[5]"},{"why":"Science Council of Japan statement committing to non-military research.","marker":"[6]"},{"why":"CERN Convention's 'no concern with work for military requirements', the core institutional principle.","marker":"[7]"},{"why":"DESY directorate measures imposing sanctions, evidence of the science-sanctions regime.","marker":"[15]"},{"why":"Science4Peace arguments against excluding scientists, supporting the sanctions critique.","marker":"[16]"},{"why":"Leopoldina-DFG report illustrating the geopoliticization of science.","marker":"[22]"},{"why":"Overview of existing Civil Clauses in Germany, documenting the commitments at stake.","marker":"[24]"}],"fun_headline_variants":["Civil clauses protect science and peace, review says","Defend civil clauses in research, paper urges","Preserve civilian-only research, reviewers argue","Don't blur civil-military line, science review warns"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that purely civilian, open, international scientific cooperation itself builds trust between nations, and that admitting military research into a civilian laboratory would actually restrict openness, deter partners, and endanger staff; the paper argues this by plausibility and historical analogy rather than direct evidence.","fun_headline_variants_meta":{"raw":{"variants":["Civil clauses protect science and peace, review says","Defend civil clauses in research, paper urges","Preserve civilian-only research, reviewers argue","Don't blur civil-military line, science review warns"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000248,"raw_usage":{"total_tokens":1512,"prompt_tokens":877,"completion_tokens":635,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":493,"completion_tokens_details":{"reasoning_tokens":575}},"tokens_in":493,"tokens_out":635,"duration_ms":6674,"temperature":1.0,"reasoning_tokens":575,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T13:07:43.936860+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compare institutions that have dropped or weakened Civil Clauses, such as Flanders after 2018, Bavaria after 2024, and Japan's Ministry of Defence funding scheme, with matched civilian-only institutions over five to ten years; if openness, foreign collaborations, and researcher retention do not measurably decline in the dual-use group, the paper's central causal claim would be undercut.","supporting_citations":[{"cited_title":"Beyond a Year of Sanctions in Science","cited_arxiv_id":"2311.02141","evidence_quote":"Science4Peace arguments against excluding scientists, supporting the sanctions critique."},{"cited_title":"Hochschulen für den Frieden – Ja zur Zivilklausel","cited_arxiv_id":null,"evidence_quote":"Overview of existing Civil Clauses in Germany, documenting the commitments at stake."}],"review_version":1}