By John R. Bolton
Thursday,
January 25, 2024
On June
18, 1935, the United Kingdom and Germany entered “a permanent and definite
agreement” that limited Germany’s total warship tonnage to 35 percent of the
British Commonwealth’s. This was a major concession from Great Britain, since
agreements at the Washington (1921–22) and London (1930) naval conferences had
already significantly reduced its own fleet. Hitler defined “permanent and
definite” to mean lasting less than four years: He abrogated the treaty on
April 28, 1939, four convenient months before the Molotov–Ribbentrop Pact
carved up Poland and started World War II. Arms control at work.
After
1945, America concluded a series of treaties that were, when signed or shortly
thereafter, almost uniformly disadvantageous to us. Considerable efforts to
eliminate these restraints have been made, but significant risk remains of
reverting to the old ways or not extracting ourselves from the remaining
harmful treaties. Whoever next wins the presidency should seek the effective
end of the usual arms-control theology before the tide turns again.
To
have any chance of bolstering U.S. national security, arms control must fit
into larger strategic frameworks, which it has not done well in the last
century. Even if they made sense in their day, many arms-control treaties have
not withstood changing circumstances. Preserving them is even less viable as we
enter a new phase of international affairs: the era after the post–Cold War
era. Russia’s invasion of Ukraine, Iran’s ongoing “ring of fire” strategy
against Israel, China’s aspirations for regional and then global hegemony, and
the Beijing–Moscow axis augur trying times. We need a post–post–Cold War
strategy avowedly skeptical of both the theoretical and the operational aspects
of the usual approaches to arms control.
Rethinking
arms-control doctrine down to its foundations began with Ronald Reagan’s 1983
Strategic Defense Initiative and resumed with George W. Bush. The partisan and
philosophical debates they launched have continued ever since, but the next
president will confront foreign- and defense-policy decisions that cannot be
postponed or ignored. Best to do some advance thinking now.
Bush’s
aspirations were more limited than what liberals derided as Reagan’s “Star
Wars.” Bush worried about American vulnerability to the prospect of “handfuls,
not hundreds,” of ballistic missiles launched against us by rogue states.
Providing even limited national missile defense, however, required withdrawing
from the 1972 Anti-Ballistic Missile (ABM) Treaty, as Bush did in December
2001. Arms control’s high priests and priestesses, and key senators such as Joe
Biden and John Kerry, were apoplectic. Missile defense was provocative, they
said. Leaving the ABM Treaty meant abandoning “the cornerstone of international
strategic stability” (a phrase commonly used by politicians, diplomats, and
arms controllers) and upsetting the premise of mutual assured destruction, they
said.
But
Bush persisted and withdrew. As the saying goes, the dogs barked and the
caravan moved on. In 2002, Bush turned to a new kind of strategic-arms
agreement with Vladimir Putin, the Treaty of Moscow, which set asymmetric
limits on deployed strategic nuclear warheads and was structured in ways very
different from earlier or later nuclear-weapons treaties. We abandoned the
complex, highly dubious counting and attribution metrics of prior
strategic-weapons deals, as well as verification procedures that Russia had
perfected means to evade. The Treaty of Moscow was sufficiently reviled by the
arms-control theocracy that Barack Obama replaced it in 2010 with the New START
(Strategic Arms Reduction Treaty), reverting to failed earlier approaches, more
on which below.
During
Bush’s first term, we also blocked efforts in the United Nations at
international gun control. We established the G-8 Global Partnership — to
increase funding for the destruction of Russia’s “excess” nuclear and chemical
weapons and delivery systems — and launched the Proliferation Security
Initiative to combat international trafficking in weapons and materials of mass
destruction. Neither effort required treaties or international bureaucracies.
We unsigned the Rome Statute, the treaty that had created the International
Criminal Court, to protect U.S. service members from the threat of criminal
action by unaccountable global prosecutors.
Finally,
the Bush administration scotched a proposed “verification” protocol to the
Biological Weapons Convention (BWC) that risked intellectual-property piracy
against U.S. pharmaceutical manufacturers but did not enhance the verification
of breaches. The BWC and the Chemical Weapons Convention express aspirations
not to use these weapons of mass destruction, but it is almost impossible to
verify compliance with them. Moreover, arms controllers forget that the BWC
sprang from Richard Nixon’s unilateral decision to eliminate
American biological munitions, which proved that we could abjure undesirable
weapons systems on our own.
The
Bush administration went a long way toward ending arms control, but the true
believers returned to power under Obama. Eager to ditch the heretical Treaty of
Moscow, his negotiators produced New START — the lineal descendant of two
earlier SALT (Strategic Arms Limitation Talks) and three START agreements —
which entered into force in February 2011 for ten years, extendable once for
five more. The Senate should never have ratified this execrable deal, as I
explained in these pages (“A Treaty for Utopia,” May 2010). Nonetheless, with a
Democratic majority it did so in a late-2010 lame-duck session, by 71 votes
(all 56 Democrats, two independents, and 13 Republicans) to 26. While the vote
seems lopsided, there were three nonvoters — retiring anti-treaty Republicans
who opposed ratification — and the Senate secured the constitutionally required
two-thirds ratification majority by only five votes. Today, given a possible
Republican majority ahead and the unlikelihood that so many Republicans would
defect again, ratifying a successor treaty is a dubious prospect at best.
The
Trump administration resumed untying Gulliver, exiting the Intermediate-Range
Nuclear Forces (INF) Treaty in 2019. While the INF Treaty may have made sense
in the 1980s, by the time of withdrawal only the United States was abiding by
its provisions. The likes of China and Iran, not treaty parties, were
accumulating substantial numbers of intermediate-range ballistic missiles, and
Russia was systematically violating INF Treaty limits. That left America as the
only country abiding by the treaty, an obviously self-inflicted handicap that
withdrawal corrected. Then, in 2020, the U.S. withdrew from the Open Skies
Treaty because Russia had abused its overflight privileges and because our
national technical assets made overflight to obtain information obsolete.
Russia subsequently withdrew from Open Skies.
But
the arms-control theology still has powerful adherents. On January 26, 2021,
newly inaugurated Joe Biden sent his first signal of weakness to Putin by
unconditionally extending New START for five years without seeking
modifications to it. This critical capitulation was utterly unwarranted by New
START’s merits or by developments since its ratification. The treaty was
fatally defective in that it did not address tactical nuclear weapons, in which
Russia had clear superiority. It remains true that no new deal would be
sensible for the United States unless it included tactical as well as strategic
warheads.
In
addition, technological threats that postdate New START (which deals with the
Cold War triad of land-based ballistic missiles, submarine-launched ballistic
missiles, and heavy bombers) need to be confronted, especially cruise missiles,
which can now reach hypersonic speeds.
Most
important, China has made substantial progress since 2010 toward becoming a
peer nuclear power. Beijing may not yet have the deliverable-weapons capacity
of Washington or Moscow, but the trajectory is clear.
A
tripolar U.S.–Russian–Chinese nuclear world (no other power has or will have
rates of warhead production comparable to China’s) would be almost
inexpressibly more dangerous than a bipolar U.S.–USSR world. The most critical
threat that China’s growing strategic-weapons arsenal poses is to the United
States. How will it manifest? Will we face periodic, independent risks of
nuclear conflict with either China or Russia? Or a combined threat
simultaneously? Or serial threats? Or all of the above? Answers to these
questions will dictate the nuclear-force levels necessary to deter first-strike
launches by either Beijing or Moscow or by both, and to defeat them no matter
how nuclear-conflict scenarios may unfold.
None
of this is pleasant to contemplate, but, as Herman Kahn advised, thinking about
the unthinkable is necessary in a nuclear world. These existential issues must
be addressed before we can safely enter trilateral nuclear-arms-control
negotiations. Beijing is refusing to negotiate until it achieves rough
numerical parity with Washington and Moscow. There is little room for diplomacy
anyway, since in February 2023 Russia suspended its participation in New START.
Further strategic-weapons agreements with Russia alone would be suicidal:
Bilateral nuclear treaties may be sensible in a bipolar nuclear world, but they
make no sense in a tripolar world. Russia and China surely grasp this. We can
only hope Joe Biden does as well. Next January, our president will have just
one year to decide how to handle New START’s impending expiration. We should
assess now which candidates understand the stakes and are likely to avoid being
encumbered by agreements not just outmoded but dangerous for America.
A
closely related challenge is the issue of U.S. nuclear testing. Unarguably, if
we do not soon resume underground testing, the safety and reliability of our
aging nuclear arsenal will be increasingly at risk, as America’s
Strategic Posture, a recent congressionally mandated report, shows. Since
1992, Washington has faced a self-imposed ban on underground nuclear testing
even though no international treaty in force prohibits it. The Limited Test Ban
Treaty of 1963 bars only atmospheric, space, and underwater testing, a gap that
the Comprehensive Nuclear-Test-Ban Treaty (CTBT), which would have banned all
testing, was intended to close. Because, however, not all five legitimate
nuclear powers under the Non-Proliferation Treaty (NPT) ratified the CTBT, it
never entered into force and likely never will. Though the U.S. signed the CTBT
in 1996, the Senate rejected its ratification by a vote of 51 to 48 in 1999.
Russia recently announced its withdrawal, thereby predictably dismaying Biden’s
advisers. The next U.S. president should extinguish the CTBT by unsigning it.
As was recently revealed, Beijing seems to be reactivating and upgrading its
Lop Nor nuclear-testing facility. We can predict confidently that neither China
nor Russia will hesitate to do what it thinks necessary to advance its
nuclear-weapons capabilities. We should not be caught short.
Additional
unfinished business involves the Conventional Armed Forces in Europe (CFE)
Treaty, another arms-control “cornerstone,” this one of European security.
Effective since 1990, as the Cold War ended, CFE became obsolete almost
immediately. The Warsaw Pact disbanded (its members largely joining NATO) and
the USSR fragmented. Russia suspended CFE Treaty compliance several times
before withdrawing formally in November 2023, having already invaded Ukraine,
another CFE Treaty party. In response, the United States and our NATO allies
suspended CFE Treaty performance. Like the CTBT, the CFE Treaty is a zombie
that the next president should promptly destroy.
The
list of arms-control-diplomacy failures goes on. The NPT, for example, has
never hindered truly determined proliferators such as North Korea (which now
has a second illicit nuclear reactor online) or Iran, much as arms-control
agreements have consistently failed to prevent grave violations by determined
aggressors.
This
long, sad history has given us adequate warning, and the next president should
learn from it. The array of threats the United States faces makes it imperative
that we initiate substantial, full-spectrum increases in our defense
capabilities, from traditional combat arms and cyberspace assets to nuclear
weapons. Instead of limiting our capabilities, we must ensure that we know what
we need and have it on hand. We are nowhere near that point.
No comments:
Post a Comment