Facts and physics still matter: a response to Wilson’s historiography

Frank von Hippel chose to respond to what he rightly calls Benjamin Wilson’s takedown piece about Hans Bethe and Richard Garwin not by correcting his errors but with a reminder to Bulletin readers of the roles that each played in the development of nuclear weapons and in trying to control them. Their well-deserved legacies are important to issues of nuclear war and peace, to history, and to the Bulletin readership. But responding to Wilson appropriately also requires engagement on substantive intellectual grounds. His piece fails not because his historiography is provocative, but because his understanding of how scientists think is flawed.

Wilson says that “Garwin did not oppose missile defense per se.” This is true—Garwin has published at least one article on defenses for fixed ICBM silos that would protect them with higher effectiveness and lower cost than anything that the Defense Department had proposed. But Garwin was thoroughly opposed to expensive missile defenses that would not have a high likelihood of achieving their goals—a different point entirely that Wilson fails to mention in his discussion.

Wilson makes much of the editorial intervention in a Scientific American article, noting that editor Dennis Flanagan feared readers would think Bethe and Garwin “are generally in favor of missile defense” based on their original draft. But in the absence of draft text establishing that they endorsed area defense, the editorial intervention more likely reflects an editorial preference for rhetorical clarity over technical nuance in a magazine for the general public. Bethe and Garwin’s technical work differentiated systematically between anti-ballistic missile (ABM) defense types—ultimately opposing heavy and light area defense while supporting silo defense.

The key question is whether their Scientific American article represented their authentic view (as anyone would presume from their signoff on the changes) or public posturing for political credibility. The evidence supports the former: Their subsequent positions remained consistent with the article when understood as opposing area defense while supporting silo defense. Indeed, the Nixon administration’s shift from Sentinel (light area defense) to Safeguard (silo defense) actually moved policy toward the configuration they favored. The editorial process helped emphasize their opposition to Sentinel’s light area defense, but nothing in their subsequent advisory positions or writings, public or private, contradicts this opposition when ABM systems are properly differentiated.

Wilson says that if Bethe and Garwin had gotten their way in the late 1960s, the United States would have deployed a national ballistic missile defense system. He seems to base this claim on the fact that as “insiders,” they were providing advice to the US government on how to make a better (if thin) national ballistic missile defense system. What is the evidence that Bethe and Garwin preferred the US to deploy a thin national ABM system, rather than a constrained one resulting from a political environment insisting on some kind of ABM deployment? Wilson offers none.

Without such evidence, the story is entirely consistent with the idea that their consulting work and government service were their attempts to pursue what is known today as harm reduction, an age-old approach that accepts as inevitable an undesirable course of action but provides practical advice to minimize the negative consequences of pursuing that course. Harm reduction is an approach commonly used by those advising government on programs that are fundamentally flawed at their roots.

In the ABM context, harm reduction might well consist of providing technical advice to reduce financial cost or strategic instability and is entirely consistent with thinking ABM systems are on balance undesirable. Harm reduction would be especially important in an environment in which various political pressures would demand some form of ABM deployment. Furthermore, acknowledging that some arguments favor thin ABM defense does not constitute hypocrisy; an ABM opponent can acknowledge the validity of certain “pro” arguments, yet also believe that “anti” arguments are stronger.

Wilson correctly notes that Bethe and Garwin worked on ABM-related technologies in the early 1960s. But weapons scientists often work on systems they later oppose, once strategic or political implications become clear. Many Manhattan project scientists—such as Joseph Rotblat, Philip Morrison, and Victor Weisskopf—later became strong advocates for arms control and disarmament. Technical contribution doesn’t preclude subsequent opposition based on strategic assessment. Upholding a view that the opposition of Garwin and Bethe to ABM was political theater requires a showing that they knew their opposition was technically incorrect—which Wilson never makes.

Wilson asks: “How did Bethe and Garwin come to publish an article openly criticizing a defensive system resembling one they had supported for years?” His answer is that “they were drawn into a new kind of public performance by outsiders who desired displays of oppositional politics by scientists.” Wilson does not mention the possibility that these scientists learned new things and updated their assessments. The technical challenges of missile defense have always been more complicated and difficult to solve than initially thought. Both Bethe and Garwin had deepening engagement with the physics of re-entry vehicles and decoy discrimination throughout the mid-1960s (as Wilson acknowledges), and they could well have concluded that even light deployments were less feasible than their own earlier assessments suggested. In normal scientific practice, updating one’s beliefs in light of new evidence is praiseworthy behavior, not evidence of political opportunism.

Wilson correctly notes elsewhere that Bethe and Garwin would have argued that nuclear knowledge cannot be erased, granting the first rearmers under disarmament a decisive advantage that destabilizes peace. He calls this argument convenient for them and the research and development organizations for which they consulted, but never refutes its substance. Of course, he’d counter that engaging the argument at all legitimizes the stability artifice and its supporting establishments.

This points to the one substantive debate that Wilson’s various pieces do address, even if only implicitly: Is it better to work for change from the inside or the outside? In other forums, he contrasts incrementalists (my shorthand)—who prioritize “strategic stability”—to those seeking radical disarmament. The former view the latter as dangerously naive; the latter see the former as enablers of a militarized status quo.

In the real world, however, both are essential. I’m an incrementalist, and I do my work within the stability paradigm. But the disarmament advocates can generate public pressure that I cannot, and thus their role is vital, rather than peripheral. While such an acknowledgment will almost surely be seen as an example of incrementalist virtue signaling, I wish Wilson had recognized that both roles are necessary if we are to survive.

Wilson’s dismissal of technical realities also overlooks the emergence of a sophisticated, science-based research agenda that assists disarmament advocates, such as that offered by the Physicists’ Coalition for Nuclear Threat Reduction.  Rather than treating technical constraints as an excuse for the status quo as Wilson implies, these scientists seek solutions that could help to convert political impossibilities into manageable engineering challenges. By framing his view in historiographical terms, Wilson overlooks technical tools that might actually help make his desired outcome possible.

Moreover, I know from personal experience that some incrementalists do follow and learn important lessons from the aforementioned science-based research agenda.  For example, Garwin and I had conversations about zero-knowledge proofs as they apply to verification protocols, which are important to both the disarmament and incrementalist camps. Wilson is wrong in implying that the recognition of technical realities as such is a political choice and fails to understand that there are technical problems that any credible framework—incrementalist or disarmament-focused—must overcome.

Finally, Wilson is surely right in urging scrutiny for the commitments and interests of the insiders. But genuine intellectual honesty also requires engaging the scientific content of their arguments, even if their alleged convenient or self-serving nature might demand a more rigorous technical review. Facts and physics still matter, even in Wilson’s historiographical universe.