Why Trump Let Nuclear Arms Control Die
The lapse of the New START treaty has done more than remove numerical ceilings on the nuclear weapons arsenals of Russia and the United States. The treaty’s demise has dismantled a system of shared knowledge – inspections, data exchanges, and notification regimes – that made credible commitments possible.
This outcome reflects more than a discrete policy failure. It reflects Trump’s worldview – call it a dealmaker’s epistemology. In this view, durable institutional knowledge is not an asset but a constraint. Negotiations are not cumulative processes through which states learn about one another over time, but discrete transactions in which leverage matters more than memory.
Hence Trump’s confidence in dispatching amateur envoys like Steve Witkoff and Jared Kushner – unburdened by knowledge of prior negotiations – to resolve immensely complex conflicts. Inexperience is not a handicap in this view; it is a qualification.
Trump’s business career – littered with hundreds of payment disputes with contractors and suppliers – offers a revealing microcosm. Former associates describe a routine practice: negotiate aggressively, then renegotiate again after delivery, effectively forcing counterparties to accept less or resort to costly litigation. Contract law scholars describe this way of operating as “selling out goodwill” – extracting value by consuming reputational capital rather than preserving it.
The same logic is apparent in Trump’s approach to alliances. His repeated claim that European allies must “pay up” for US protection – coupled with suggestions that Russia should feel free to “do whatever the hell they want” to non-compliant NATO members – resembles something closer to extortion than to traditional burden-sharing diplomacy.
But Trump is no mafia don. He cannot keep his word. Even as European allies are told to increase defense spending to secure American protection, they are made to doubt America’s commitment to NATO’s mutual-defense guarantee. The result resembles a protection racket in which the fee is collected but the protection is withheld.
Such double-crossing can be profitable under very specific conditions. It works best in one-shot interactions, when counterparties do not expect repeated dealings and information about past behavior cannot easily circulate. Trump’s geographically dispersed business model – a hotel in one market, a casino in another, a licensing deal elsewhere – fits these conditions unusually well.
International diplomacy does not. Relations among states are necessarily iterative. They observe one another, share intelligence, and update expectations collectively. Reputation is not local; it is global and cumulative. A government that abandons one agreement is judged not only by the immediate counterparty but by every state evaluating future commitments.
This is why the loss of New START matters far beyond US-Russian relations. The treaty sustained a shared epistemic infrastructure: on-site inspections, telemetry exchanges, and movement notifications. These mechanisms reduced uncertainty. And in nuclear deterrence, reducing uncertainty is often more stabilizing than reducing arsenals. Verification regimes are not bureaucratic accessories to arms control; they are the mechanisms through which states make credible commitments about tomorrow.
When verification disappears, suspicion fills the void. And suspicion, in nuclear strategy, is self-reinforcing. Each side must assume the worst about the other’s capabilities and intentions. The rational response becomes hedging – deploying more warheads, raising alert levels, and accelerating modernization programs. Arms racing, in this sense, is not always the product of aggressive intent. It is often the product of informational decay.
When the principal architect of the post-1945 security order signals indifference to institutional continuity, other states rationally shorten their time horizons. The international system becomes less future-oriented not because states suddenly become reckless, but because the informational foundations of long-term restraint erode.
Underlying these choices is a deeper conflict about political time. The crucial distinction is not between leaders who care about the future and those who do not. It is between two conceptions of how the future is governed.
Institutional time is cumulative. It depends on expertise, verification systems, alliances, and memory. Trust is constructed slowly and maintained across administrations and even generations.
Episodic time is transactional. Negotiations are dramatic encounters between principals, each largely independent of the past and weakly constrained by the future. Success is measured by the outcome of the moment rather than by the durability of the agreement it produces. When Trump promises to end wars in one day, it isn’t just grandiosity; it is the abolition of the future as something that unfolds, that requires patience, cultivation, and investment.
The dealmaker’s epistemology is inherently episodic. Deep knowledge of a counterpart can seem like a liability rather than an asset. Understanding the constraints that shaped previous agreements can look like weakness rather than realism. Institutional memory becomes a form of capture by the past rather than a resource for navigating the future.
But nuclear stability belongs to institutional time. Deterrence depends not only on capabilities but on predictable expectations about how those capabilities will evolve. When verification systems erode, states lose not only constraints but shared ways of interpreting one another’s behavior. Worst-case assumptions become rational default positions.
The dealmaker’s epistemology assumes that if a negotiation fails, one can walk away and try again with someone else. In nuclear strategy, there is no next market, no new counterparty, and no second chance to correct a catastrophic miscalculation.
In New York real estate, the dealmaker’s epistemology eventually met its natural limit. Contractors, banks, and suppliers burned by Trump’s serial double-dealing simply refused to do business with him again. Reputation caught up with practice.
But not for the American electorate, which has now twice entrusted the presidency to a man whose defining commercial skill was betraying the “reliance interest” of others. Whether this proves gravely damaging or irreversibly catastrophic remains to be seen. But the lapse of New START should make one thing clear: Trump’s willingness to break commitments is not an American domestic problem, nor even an Atlantic one. It is a problem for the planet. For the first time in over half a century, roughly 8,000 warheads sit in two arsenals with no binding limits and no verification. It is horrifying to realize that nuclear stability, built over decades, can be unmade in months by a leader who thinks only “losers” make promises they have to keep.
*Stephen Holmes, is the Professor of Law at New York University School of Law and Berlin Prize Fellow at the American Academy in Berlin and co-author (with Ivan Krastev) of The Light that Failed: A Reckoning (Penguin Books, 2019).