Whether To "Strangle

Whether to “Strangle William Burr and
Jeffrey T. Richelson
the Baby in the Cradle”
The United States and the Chinese
Nuclear Program, 1960–64
Less than a year after

President John F. Kennedy’s assassination, Arms Control and Disarmament
Agency (ACDA) Director William Foster told a historian that Kennedy had
been willing to “consider politically dangerous moves” to coerce the People’s
Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Republic of China (PRC) into complying with the 1963 Limited Test Ban Treaty.
Foster, whose comments remained classiªed until 1994, asserted that Kennedy
was even willing to sacriªce the proposed U.S.-European multilateral force
(MLF) to secure Soviet cooperation “in taking action, if necessary physically,
against China.” Foster told his interviewer that the president would “think out
loud,” saying, “You know, it wouldn’t be too hard if we could somehow get
kind of an anonymous airplane to go over there, take out the Chinese facili-
ties—they’ve only got a couple—and maybe we could do it, or maybe the So-
viet Union could do it, rather than face the threat of a China with nuclear
weapons.”1
Whether to “Strangle the Baby in the Cradle”
That Kennedy and his advisers considered using force against China’s nu-
clear facilities was ªrst documented publicly by historian Gordon Chang in
1988.2 In that same year, however, the man who had been Kennedy’s national
security assistant, McGeorge Bundy, downplayed Chang’s revelations, claim-
ing that White House discussions of preventive action against China had been
simply “talk, not serious planning or real intent.” Nonetheless, historians have
remained curious about the extent to which President Kennedy spurred the
The authors are senior analysts at the National Security Archive, George Washington University.
Earlier versions of this article appeared in the Bulletin of the Atomic Scientists as “The China Puz-
zle,” Vol. 53, No. 4 (July/August 1997), and were presented at the 1999 annual meeting of the Soci-
ety for the Historians of American Foreign Relations. The authors thank Barton Bernstein, William
Bundy, Gordon Chang, Lynn Eden, Raymond Garthoff, Robert Johnson, Robert Norris, David
Painter, John Prados, David Rosenberg, Walt Rostow, Gen. William Smith (ret'd), J. Samuel Walker,
and Paul Wolman for comments, and the W. Alton Jones Foundation for ªnancial support to the
National Security Archive, George Washington University.
1. Oral history interview with William C. Foster by Charles T. Morrissey, August 5, 1964,
declassiªed September 1994, copy at John F. Kennedy Library (hereinafter JFKL). The multilateral
force was a proposal to meet Western European, especially West German, concern about exclusive
U.S. control of nuclear weapons deployed in Europe by establishing a NATO-controlled sea-based
missile force.
2. Gordon Chang, “JFK, China, and the Bomb,” Journal of American History, Vol. 74, No. 4 (March
1988), pp. 1289–1310.
International Security, Vol. 25, No. 3 (Winter 2000/01), pp. 54–99

© 2000 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.
54
Whether to “Strangle the Baby in the Cradle” 55
national security bureaucracy to explore means of disrupting China’s nuclear

efforts.3
Now, newly declassiªed documents show that Kennedy and his advisers
did much more than talk. They reveal that the Kennedy administration initi-
ated a massive intelligence effort, including U-2 ºights and satellite reconnais-
sance programs, to break through the secrecy that surrounded the Chinese

nuclear program, which had begun in the mid-1950s. The documents show
that after Soviet Premier Nikita Khrushchev rebuffed U.S. overtures for joint
moves against the Chinese program, U.S. ofªcials explored military action
without the Soviets.
Newly declassiªed documents make it possible to trace, far more exten-
sively than before, the massive U.S. intelligence effort deployed against
China’s nuclear program and the interrelationship between that effort and the
analysis of China’s nuclear progress. Even more signiªcant, the new documen-
tation makes it possible to explore in greater detail than ever the policy de-
bates, discussion, and planning within the Kennedy and Johnson admin-
istrations on how to deal with Beijing’s nuclear effort.
By the onset of the Kennedy administration, if not earlier, senior U.S.
ofªcials saw a Chinese nuclear capability as a serious threat to national secu-
rity. Supporting this assessment were intelligence analysts, who emphasized
the consequences of a nuclear China for the prospect of nuclear “proliferation”
and for the stability of East Asia. U.S. intelligence sources also assumed that
the Chinese were pursuing the construction of a plutonium-based bomb,
which turned out not to be the case.
The new documentation corroborates Gordon Chang’s portrait of Kennedy’s
militancy and shows that Bundy was in fact the point man in countering the
Chinese nuclear effort, passing Kennedy’s instructions to the Central Intelli-
gence Agency (CIA), holding secret discussions with Nationalist Chinese
ofªcials, and seeking to enlist Soviet diplomats in joint efforts against China’s
nuclear program. Declassiªed material discloses that high-level U.S. ofªcials
initiated contingency plans for air attacks on Beijing’s nuclear facilities and
that the president offered support for “action” by the CIA. In particular, after
discussing Beijing’s nuclear program with Chinese Nationalist Gen. Chiang
Ching-kuo in 1963, CIA and Pentagon ofªcials began to shift their thinking to-
3. McGeorge Bundy, Danger and Survival: Choices about the Bomb in the First Fifty Years (New York:
Vintage, 1990), p. 532; and David Rosenberg, “The Origins of Overkill: Nuclear Weapons and
American Strategy, 1945–60,” in Steven E. Miller, ed., Strategy and Nuclear Deterrence: An Interna-
tional Security Reader (Princeton, N.J.: Princeton University Press, 1984), pp. 127, 143–144.
International Security 25:3 56
ward covert and paramilitary options, including the possibility of a raid by Re-
public of China (ROC) commandos.
The Kennedy administration explored the use of force in the absence of com-
plete information on the Chinese nuclear plan and without a thorough analysis
of the likely impact of China’s nuclear progress. By late 1963, however, a State
Department ofªcial, Robert Johnson, concluded that a Chinese nuclear capabil-

ity would not pose a major threat to U.S. interests, much less change the bal-
ance of power in East Asia. Johnson’s analysis undercut the alarmist thinking
that had motivated the administration’s search for coercive options and illus-
trated the heavy political costs the United States would suffer if Washington
acted without Moscow as a partner. Nonetheless, Bundy, now President
Lyndon B. Johnson’s national security assistant, remained attracted to the use
of force.
By late summer 1964, U.S. intelligence had concluded that the Chinese were
about to conduct an atomic test. By this time, however, with an election cam-
paign to conduct, the Johnson administration wanted to avoid a direct con-
frontation. Moreover, the Soviets remained unresponsive to proposals for a
joint effort. Thus the Chinese conducted their nuclear test unimpeded on Octo-
ber 16. Their use of highly enriched uranium rather than plutonium startled
U.S. intelligence and underlined its lack of solid intelligence on the Chinese ef-
fort. In the months that followed, Johnson administration ofªcials continued to
consider the use of force, but ofªcials at the top no longer saw such a strategy
as viable unless Beijing ªrst launched a major aggressive act against its neigh-
bors. Johnson and his advisers would monitor China’s nuclear progress closely
and search for ways to delay it through trade controls, but it was clear that the
president had tacitly decided to learn to live with the Chinese bomb.
Now, nearly forty years later, the Chinese nuclear weapons program and its
potential have been generating intense debate in the national security bureau-
cracy and the halls of Congress. As in the 1960s, some ofªcials are treating
China’s nuclear capabilities as a dire threat, while others are pointing to
Beijing’s emphasis on deterrence. Whether today’s policymakers will also
avoid overreactions remains to be seen.
We begin by reviewing Chairman Mao Zedong’s initial decisions during the
1950s to undertake a nuclear weapons program and the strict secrecy that en-
veloped it. We then turn to U.S. intelligence’s initially unsuccessful effort to
pierce Beijing’s nuclear secrecy, a problem that intelligence analysts com-
pounded by making faulty assumptions about the direction of Chinese nuclear
progress. In the next section, we look at the Kennedy administration’s concerns
about a Chinese nuclear capability and its early thinking about ways to offset
adverse international reactions to a nuclear China. We next explore the CIA’s
accelerated effort during 1962 and 1963 to glean intelligence on China, espe-
cially the role of U-2 photography in helping analysts delineate the physical
scope of China’s nuclear program. The estimate that China could test a device
within a few years raised alarm at the White House, and President Kennedy

encouraged the CIA and other agencies to explore possibilities for preventive
action against Chinese nuclear facilities.
After examining U.S. attempts to recruit Moscow for joint action against the
Chinese program, we turn to the Kennedy administration’s exploration with
the Nationalist Chinese government of possible means of preventing China’s
acquisition of nuclear weapons. We then focus on internal U.S. assessments of
the strategic implications of a nuclear China and the pros and cons of military
or covert action. A key special national intelligence estimate on a possible Chi-
nese nuclear test is the central element of the subsequent section. That leads to
our discussion of two interrelated topics—ªnal policy considerations in the
weeks prior to China’s October 16, 1964, and the reexamination by intelligence
analysts of some of their crucial judgments relating to the likely timing of the
Chinese test. We then examine U.S. and international reaction to the test as
well as the continued intelligence activities that allowed a better assessment of
the Chinese program. In our conclusion we appraise why the United States
chose not to act against China’s emerging nuclear program and consider the
signiªcance of the episode for today’s decisionmakers.
China’s Emerging Nuclear Program
It was during the 1954–55 confrontation between Beijing and Washington over
the offshore islands of Quemoy (Jinmen) and Matsu (Matzu) in the Taiwan
Strait that Mao made his initial decisions to develop at least a modest nuclear
capability. Recognizing the difªculty of neutralizing U.S. nuclear strength,
Mao nevertheless believed that even a few weapons would raise the interna-
tional prestige of the PRC and its leadership. In January 1955, in the midst of
the crisis, he authorized a full-scale effort to make China a nuclear power.4
4. John Wilson Lewis and Xue Litai, China Builds the Bomb (Stanford, Calif.: Stanford University
Press, 1988), pp. 11–46. See also Rosemary Foot, The Practice of Power: U.S. Relations with China since
1949 (New York: Oxford University Press, 1995), pp. 168–174; and Tian-Yu Cao, “Two Critical Mo-
ments in China’s Strategic Weapons Program,” presentation at the National Air and Space Mu-
seum, August 23, 2000.
Developing a nuclear arsenal requires scientiªc and technical expertise,

signiªcant industrial effort, and production or acquisition of ªssile material.
By the mid-1950s, no great mystery remained about the two major means by
which such resources could be employed to create nuclear weapons: One pro-
cess ended with warheads of enriched uranium (made up of 90 percent or
more of U-235), the other with plutonium warheads. Which process China

would pursue remained unclear to Western sources, however.
Following Mao’s decision, research facilities apparently aimed at atomic
weapons development began to spring up across China. A second Taiwan
Strait crisis in 1958 and U.S. nuclear weapons deployments on Taiwan rein-
forced, in Mao’s view, the wisdom of his earlier decision, and the Chinese gov-
ernment established the Beijing Nuclear Weapons Research Institute in 1958 (it
was replaced in 1962 by the Northwest Nuclear Weapons Research and Design
Academy—later known as the Ninth Academy—China’s version of the Los
Alamos Scientiªc Laboratory, in Qinghai). The government also established
the Tongxian Uranium Mining and Hydrometallurgy Institute in 1958.5
Also in 1958, construction began on the Baotou Nuclear Fuel Component
Plant and the Lanzhou Gaseous Diffusion Plant, which would produce, respec-
tively, uranium tetraºuoride (used in producing enriched uranium) and en-
riched uranium. In February 1960, ground was broken for construction of a
plutonium production reactor at the Jiuquan Atomic Energy Complex, which
also contained key facilities involved in the ªnal stages of producing China’s
atomic weapons. By 1960, construction work on a nuclear test site was also un-
der way in western China, at Lop Nur.6
At ªrst, the Soviet Union assisted the Chinese with plans to construct a plu-
tonium production complex. But the Sino-Soviet aid agreement began to fall
apart in June 1959, and at the time the Soviets ended their assistance, they had
delivered none of the key components for the facility, much less a promised
“sample” atomic bomb.7 The Chinese were now on their own. For their part,
5. Lewis and Xue, China Builds the Bomb, pp. 90, 140–141; and Robert S. Norris, Andrew S. Bur-
rows, and Richard W. Fieldhouse, Nuclear Weapons Databook, Vol. 5, British, French, and Chinese Nu-
clear Weapons (Boulder, Colo.: Westview, 1994), pp. 338, 340.
6. Norris, Burrows, and Fieldhouse, Nuclear Weapons Databook, pp. 338, 345; and Lewis and Xue,
China Builds the Bomb, pp. 177–178.
7. Central Intelligence Agency, “China: Plutonium Production Reactor Problems,” January 1988;
and Odd Arne Westad, Brothers in Arms: The Rise and Fall of the Sino-Soviet Alliance, 1945–63 (Stan-
ford, Calif.: Stanford University Press, 1998), pp. 157–159, 206–207. See also Victor Gobarev, “Soviet
Policy toward China: Developing Nuclear Weapons, 1949–1969,” Journal of Slavic Military History,
Vol. 12, No. 4 (December 1999), pp. 17–31.
U.S. intelligence sources knew that China was pursuing a nuclear develop-
ment strategy but had little speciªc knowledge of its extent and capability.
Piercing the Veil
By the early 1960s, Washington had become greatly concerned about China’s

pursuit of a nuclear weapons capability. The United States acquired aerial pho-
tography of Lanzhou in September 1959, although human as well as communi-
cations intelligence efforts appear to have produced only a small amount of
speciªc information.8
Thus when the Kennedy administration came to power, the United States’
knowledge about Beijing’s nuclear progress remained heavily conjectural, and
the strict secrecy that cloaked the Chinese program posed formidable obsta-
cles. The Chinese had made no public announcement of the establishment of
their nuclear research institutes, had conducted no ribbon-cutting ceremonies
at Baotou or Lanzhou, and issued no press releases of the search for a test site
in western China. Even within the Chinese program, facilities were generally
described by code names or by uninformative titles, such as Plant 404 for
Jiuquan.9
U.S. intelligence collection capabilities, particularly with respect to China,
were limited. Satellite imagery of China became available only in August 1960,
and, at ªrst, on an intermittent basis. Meanwhile, the location deep in China’s
interior of critical nuclear facilities such as Lop Nur and Lanzhou made it
difªcult to cover such targets with U-2 spy planes, given the bases from which
they were ºying.
A December 6, 1960, national intelligence estimate (NIE) reºected these limi-
tations. Its authors noted that “our evidence with respect to Communist
China’s nuclear program is fragmentary as is our information about the nature
and extent of Soviet aid.”10 What the United States did know was evident in an
NIE issued a week later. The NIE contained brief discussions of the organiza-
tion of the Chinese program, technical capabilities, uranium ore production,
8. Director of Central Intelligence (DCI), Special National Intelligence Estimate (SNIE) 13–2–63,
“Communist China’s Advanced Weapons Program,” July 24, 1963. Images of signiªcant primary
sources cited in this article, such as this document, may be found at the National Security Ar-
chive’s web site, http://www.nsarchive.org.
9. Norris, Burrows, and Fieldhouse, Nuclear Weapons Databook, pp. 338–341; and Lewis and Xue,
China Builds the Bomb, pp. 121, 318.
10. DCI, NIE 13–60, “Communist China,” December 6, 1960, p. 13, copy at National Security
Archive.
and ªssionable materials. The analysts observed that “Chinese development of

uranium resources and their probable construction of ore concentration and
uranium metal plants certainly would imply an intended use for the uranium
in plutonium production.” The NIE also noted that although uranium metal is
not required for U-235 production, the initial stages of the process could also
supply fuel for U-235 separation.11

CIA analysts thus guessed that a ªrst Chinese production reactor could go
critical in late 1961, with the ªrst plutonium possibly becoming available in
1962 (with the most probable date for a ªrst detonation being sometime in
1963).12 The NIE reºected the strong presumption that plutonium, not ura-
nium, would be the key ingredient in producing China’s ªrst bomb, as it had
been for the other nuclear powers. The process of enriching uranium to a level
suitable for use in weapons is laborious, and considerably less plutonium is
needed to make a ªrst-generation atomic weapon.13
Despite its limitations, the NIE had a signiªcant impact. One State Depart-
ment ofªcial saw it as “one of the most signiªcant among recent intelligence
products” because it conªrmed the existence of a Chinese nuclear program,
and it “advanced the earliest date by which Beijing might be able to explode a
nuclear device.”14
Assessing the Threat and Exploring the Options
The interest that the analysts had in Chinese nuclear developments was
matched by the Kennedy administration’s concerns about China. Although the
Chinese nuclear program was not at the top of the administration’s agenda—
Cuba, Laos, and Berlin all ranked higher—Kennedy was hostile to Mao’s re-
gime and found the prospect of a nuclear China disquieting. The president was
unaware of Mao’s limited objectives and, one adviser recalled, saw a Chinese
11. DCI, NIE 13–2–60, “The Chinese Communist Atomic Energy Program: Summary and Conclu-
sions,” December 13, 1960, U.S. Department of State, Foreign Relations of the United States, 1958–
1960, Vol. 19 (hereinafter FRUS, with appropriate year, volume, and page numbers) (Washington,
D.C.: U.S. Government Printing Ofªce [U.S. GPO], 1996), pp. 744–747.
12. Ibid.
13. Joel Ullom, “Enriched Uranium versus Plutonium: Proliferant Preferences in the Choice of Fis-
sile Material,” Nonproliferation Review, Vol. 2, No. 1 (Fall 1994), pp. 1–5.
14. John M. Steeves, Bureau of Far Eastern Affairs (State Department) to Director of State Depart-
ment Intelligence and Research (INR) Roger Hilsman, “National Intelligence Estimate on Implica-
tions of Chinese Communist Nuclear Capability,” April 12, 1961, Department of State Records
(hereinafter RG 59), Assistant Secretary for Far Eastern Affairs Subject, Personnel, and Country
File, 1960–1962, box 4, Communist China, January–June 1961.
nuclear test “as likely to be historically the most signiªcant and worst event of
the 1960s.”15
Kennedy worried enough about China’s nuclear ambitions that by June 1961
he wondered whether the increasingly rancorous Sino-Soviet split made it pos-
sible for Washington and Moscow to work together to restrain Beijing’s nu-
clear ambitions. As Soviet Premier Khrushchev showed at the June 1961

Vienna summit, he was far from reconciled to the idea of a permanent split and
showed no interest when Kennedy brought up the problem of a nuclear
China.16
During Kennedy’s ªrst two years in ofªce, no one in government had pre-
pared a comprehensive analysis of the problem of a nuclear China. Several
studies did appear, however. A Joint Chiefs of Staff (JCS) report in June 1961
concluded, without spelling out the political or military implications, that
Beijing’s “attainment of a nuclear capability will have a marked impact
on the security posture of the United States and the Free World, particu-
larly in Asia.” A few months later, a State Department report suggested that
a Chinese nuclear capability would pose more political and psychological
problems than military ones. According to the latter study, a nuclear
China could reap politically signiªcant “psychological dividends” by help-
ing to create feelings that “communism is the wave of the future.” Further,
for many Asians, a nuclear test would add to the credibility of the Chinese
model of economic development. They would raise their estimates of Chinese
“mililitary power relative to that of their own countries and the [United
States’] capabilities in the area.” The implication was that a heightened
sense of China’s power would create a bandwagon effect, with greater politi-
cal pressures on states in the region to accommodate Beijing and loosen
ties with Washington. Thus a nuclear China could only weaken Washington’s
inºuence in the region and its capabilities to intervene on behalf of allies
there.17
15. For Kennedy’s China policy, see James Fetzer, “Clinging to Containment: China Policy,” in
Thomas G. Paterson, ed., Kennedy’s Quest for Victory: American Foreign Policy, 1961–63 (New York:
Oxford University Press, 1989), pp. 178–197; Walt Rostow is quoted at p. 182.
16. Gordon Chang, Friends and Enemies: The United States, China, and the Soviet Union, 1948–72
(Stanford, Calif.: Stanford University Press, 1990), pp. 230–232. See also Chang, “JFK, China, and
the Bomb”; and Foot, The Practice of Power, pp. 179–180.
17. Memorandum from Joint Chiefs of Staff, “A Strategic Analysis of the Impact of the Acquisition
by Communist China of a Nuclear Capability,” June 26, 1961, FRUS, 1961–63, Vol. 22, pp. 84–85;
Policy Planning Council (PPC) Director George McGhee to Secretary of State Dean Rusk, “Antici-
patory Action Pending Chinese Demonstration of a Nuclear Capability,” September 13, 1961, Na-
tional Archives, RG 59, Records of Policy Planning Staff, 1957–61.
U.S. government ofªcials also saw developments in China as adding to the

problems of nuclear proliferation. In early 1961, CIA analysts mused that when
China or Israel went nuclear, “other nations might enter the ªeld if only to
counter the power and prestige which their rivals or their enemies might
gain.” Global instability could result because “even a small increase in the
number of nations possessing nuclear weapons will add to the dangers inher-

ent in critical situations when they arise.” In addition, an increase in the num-
ber of states with nuclear-threat capabilities would “increase the chances for
irrational and desperate action.”18
Their limited knowledge notwithstanding, U.S. government ofªcials were
anxious to consider ways to mitigate the effects of China’s impending nuclear
status. For example, in September 1961 the State Department’s Policy Planning
Council (PPC) director, George McGhee, proposed to Secretary of State Dean
Rusk that one way to reduce the psychological impact of a Chinese bomb was
to encourage, and perhaps even assist, India, whose own nuclear efforts were
already “sufªciently advanced” to test a device ªrst. Showing little interest in
the potential impact of a nuclear India for regional stability, McGhee wanted a
noncommunist Asian state to “beat Communist China to the punch.”19
McGhee’s scheme found uneven support at the State Department, and it was
diluted to a proposal for a quiet, exploratory effort by White House Science
Adviser Jerome Weisner during his upcoming trip to South Asia. Weisner
could meet with the chairman of India’s Atomic Energy Commission and ask
what effect a Chinese nuclear weapons capability might have on India’s nu-
clear program. The question might lead to an Indian request for assistance. The
proposal was approved by Undersecretary of State Chester Bowles but vetoed
by Secretary of State Rusk, who was unconvinced that “we should depart from
our stated policy that we are opposed to the further extension of nuclear weap-
ons capability.” If the United States abetted nuclear proliferation, Rusk argued,
it “would start us down a jungle path from which I see no exit.”20
18. ”Estimate of the World Situation,” January 17, 1961, FRUS, 1961–63, Vol. 8, p. 8. For back-
ground on U.S. nuclear nonproliferation policy, see Shane J. Maddock, “The Nth Country Conun-
drum: The American and Soviet Quest for Nuclear Nonproliferation, 1945–1970,” Ph.D.
dissertation, University of Connecticut, 1997.
19. McGhee to Dean Rusk, “Anticipatory Action Pending Chinese Demonstration of a Nuclear Ca-
pability.” For an air force proposal to arm U.S. allies with nuclear weapons to counter a Chinese ca-
pability, see memorandum to air force chief of staff, “Long-Range Threat of Communist China,”
February 8, 1961, Library of Congress, Thomas D. White Papers, box 44, Air Staff Actions, 1961.
20. Secretary of State Dean Rusk, memorandum for Executive Secretary Lucius Battle, October 7,
1961, Records of Policy Planning Staff, 1957–61, box 129, China.
Rusk did approve another McGhee proposal. In September 1962, he sanc-

tioned a coordinated overt-covert propaganda campaign to heighten Asian
awareness of “U.S. and Free World strength” and to neutralize “awe and un-
reasoned fear” of China. Besides emphasizing the United States’ strategic nu-
clear superiority, the campaign suggested that China’s nuclear program was
behind schedule, in hopes of producing a “What took you so long?” reaction to

any Chinese detonation.21
Concurrent with the efforts to shape world opinion, Robert Johnson, a PPC
East Asian specialist, began a series of major studies on the implications and
consequences of a Chinese nuclear test and a “regionally signiªcant” nuclear
capability. Johnson’s mandate was not only to determine the impact of a Chi-
nese nuclear capability but also to consider the policy changes that might be
needed to counter its political and diplomatic effects.22
New Intelligence
While Johnson worked on the ªrst of his studies, U.S. intelligence was improv-
ing its ability to monitor Chinese advanced weapons programs. One source of
high-resolution imagery were CIA U-2s ºown from Taiwan by Chinese Nation-
alist pilots. Although the missions were infrequent given the risks involved,
beginning in 1961 they covered a number of mainland targets.23
Further, the primary U.S. satellite reconnaissance program, code-named
CORONA, produced a quantum leap in the ability of the U.S. intelligence com-
munity to monitor activities in “denied areas” such as the Soviet Union and
China. There had been only two successful CORONA missions prior to the De-
cember 1960 NIE, whereas there were twenty-four such missions between Jan-
uary 1961 and June 1963. Moreover, improved resolution of the CORONA
cameras and larger ªlm supplies carried by successive camera systems meant
more and better photography. Thus a December CORONA 1961 mission pro-
vided the ªrst coverage of Lop Nur, although no one in Washington then rec-
ognized it as a prospective nuclear test site. In addition, the satellites could
21. PPC Director George McGhee to Secretary of State Dean Rusk, “Program to Inºuence World
Opinion with Respect to a Chinese Nuclear Detonation,” September 24, 1962, RG 59, Central Deci-
mal Files, 1960–63, 793.5611/9-2462.
22. “Nuclear Proliferation,” October 15, 1962, RG 59, Records of Policy Planning Council, 1962,
box 236, RH Johnson Chron File, 1962, Bulky Report.
23. For the U-2 and the Nationalists, see Chris Pocock, Dragon Lady: The History of the U-2 Spyplane
(Shrewsbury, England: Airlife, 1989), pp. 90–106, 143–163.
overºy targets that U-2s could not reach or reach only with great difªculty
from the available bases.24
An April 1962 NIE, whose increased length (twelve pages) reºected the ac-
cess to new information, also demonstrated continuing intelligence gaps.
Moreover, it reºected the continued assumption that plutonium would fuel
China’s ªrst bomb, even though there was no evidence of construction of a

plutonium production facility. The estimators argued that if one assumed “an
accelerated and highly successful program for the production of plutonium
(speciªcally the production of 30,000 tons of uranium metal a month) since
1960, the Chinese Communists could detonate an all-plutonium device in early
1963.” They considered it unlikely, however, that the Chinese would meet such
a schedule and predicted that the “ªrst Chinese test would probably be de-
layed beyond 1963, perhaps by as much as several years.” But whenever China
ªrst detonated a device, it appeared highly likely to the analysts that it would
be plutonium ªred.25
In early 1963, the CIA felt White House pressure for more information on
Beijing’s nuclear efforts. On January 10, Director of Central Intelligence (DCI)
John McCone met with National Security Assistant Bundy, who told him that
Cuba and the Chinese nuclear program were the “two issues foremost in the
minds of the highest authority and therefore should be treated accordingly by
CIA.” McCone had to acknowledge, however, that the agency knew very little
for certain about China’s progress—hence the need for an expanded effort.26
During the months that followed, Taiwanese U-2 ºights provided more in-
formation. A March 1963 ºight detected the nuclear complex at Baotou, al-
though U.S. intelligence mistakenly believed that it harbored a plutonium
reactor. State Department Bureau of Intelligence and Research (INR) ofªcial
George Denny told Rusk that this showed a capability to produce enough “ra-
dioactive material for a few nuclear weapons annually” and that Baotou’s dis-
covery conªrmed that the Chinese would be “capable of detonating a small
nuclear explosion by the end of 1963.”27
24. Robert A. McDonald, “CORONA: Success for Space Reconnaissance, A Look into the Cold War
and a Revolution for Intelligence,” Photogrammetric Engineering and Remote Sensing, Vol. 60, No. 6
(June 1995), pp. 689–720. For lack of knowledge about Lop Nur, see “U.S. and Soviet Knowledge
about CHICOM Advanced Weapons Programs,” July 9, 1963, RG 59, Executive Secretariat Country
Files, 1963–66, box 2, Communist China.
25. DCI, NIE 13–2–62, “Chinese Communist Advanced Weapons Capabilities,” April 25, 1962,
pp. 3, 11, FRUS, 1961–63, Vols. 7–9, microªche supplement, doc. 266.
26. Editorial note, FRUS, 1961–63, Vol. 22, p. 339.
27. ACDA, “Summary and Appraisal of Latest Evidence on Chinese Communist Advanced
Weapon Capabilities,” July 10, 1963, RG 59, Executive Secretariat Country Files, 1963–66, box 2,
The additional imagery was crucial in producing a July 1963 Special Na-
tional Intelligence Estimate (SNIE), “Communist China’s Advanced Weapons
Program.” The analysts reported that since the 1962 estimate, “we have re-
ceived a considerable amount of information, mainly from photography.” Yet
they also noted that “the gaps in our information remain substantial and we
are therefore not able to judge the present state or to project the future develop-

ment of the Chinese program as a whole with any very high degree of
conªdence.”28
The SNIE reºected the new information and the continued gaps. The ana-
lysts noted that recent photographic coverage of Baotou showed its “elaborate
security arrangements,” and that they believed the installation included a
small air-cooled plutonium production reactor, with associated facilities for
chemical separation and metal fabrication. That the installation resembled a
French plutonium reactor apparently led analysts to conclude that it too was a
plutonium plant. The Chinese were building a reactor for plutonium weapons
at Jiuquan, but it was years away from operation, although U-2 or satellite
photography discovered it sometime in 1964.29
March and June 1963 U-2 photography of the Lanzhou gaseous diffusion
plant showed progress being made on a nearby hydroelectric installation,
which intelligence experts believed was designed to supply the plant, as well
as transmission lines between the diffusion plant and a thermal electric plant.
According to ACDA ofªcials, the gaseous diffusion plant was large enough to
contain about 1,800 compressor stages, substantially less than the 4,000 stages
required to produce weapons-grade material. But their data were extraordi-
narily ºawed; the Lanzhou facility was only months away from producing
enough highly enriched uranium for a weapon. Interestingly, Chiang Kai-
shek’s intelligence services believed that the Lanzhou reactor was active dur-
ing 1963, but no one in Washington appears to have given any credence to that
report.30
Communist China; Pocock, Dragon Lady, pp. 92–95; author’s interview with Albert Wheelon,
Washington, D.C., April 9, 1997; and “The Secretary’s Staff Meeting,” May 1, 1963, RG 59, Execu-
tive Secretariat Brieªng Books, box 1, Secretary’s Large Staff Meetings, 1963.
28. DCI, SNIE 13–2–63, “Communist China’s Advanced Weapons Program,” July 24, 1963, p. 1,
copy at National Security Archive.
29. Ibid., pp. 2, 4, 5; and Ambassador-at-Large Llewellyn Thompson to the Secretary of State, “Plu-
tonium Plant in Communist China,” May 29, 1963, RG 59, Records of Ambassador-at-Large
Llewellyn Thompson (hereinafter cited as RAALLT), box 23, Secretary—Memos from Ambassador
T.; Lewis and Litai, China Builds the Bomb, p. 97; and DCI, NIE 13–3–65, “Communist China’s Mili-
tary Establishment,” March 10, 1965, copy at National Security Archive.
Weapon Capabilities”; Lewis and Xue, China Builds the Bomb, pp. 134–135; and memorandum of
Besides Baotou, U.S. intelligence identiªed several facilities of the Institute

of Atomic Energy, which was responsible for nuclear research and develop-
ment. One of them, located southwest of Beijing and secretly known to the
Chinese as Code 601, housed a Soviet-supplied research reactor as well as a cy-
clotron. Although the CIA did not know it at the time, Code 601 was playing
an important function: It produced ten tons of hexaºouride, an essential prod-

uct for uranium enrichment at the Lanzhou gaseous diffusion facility.31
Information gaps about China’s ªssile materials capability made it difªcult
to answer a fundamental question: When was China going to detonate its ªrst
atomic bomb? Based on the faulty identiªcation of Baotou as a plutonium pro-
duction facility and the assumption that plutonium would fuel China’s ªrst
atomic explosion, the analysts’ best guess was as soon as early 1964 but, if
“normal difªculties” unfolded, later in 1964 or in 1965. But they also noted that
the possible existence of another, undiscovered, plutonium reactor could mean
that “the Chinese could achieve a ªrst detonation at any time.”32
The estimate looked beyond questions of how and when China might ªrst
detonate a nuclear device and examined the policy impact of a Chinese bomb,
which is what ultimately concerned U.S. decisionmakers. On the one hand, the
estimators “did not believe that the explosion of a ªrst device, or even the ac-
quisition of a limited nuclear weapons capability, would produce major
changes in Chinese foreign policy in the sense that the Chinese would adopt a
general policy of military aggression or even be willing to take signiªcantly
greater military risks.” Chinese leaders, it was expected, would realize just
how limited their capabilities were. Yet the SNIE suggested that “the Chinese
would feel very much stronger and this mood would doubtless be reºected
in their approach to conºicts on their periphery. They would probably feel
that the U.S. would be more reluctant to intervene on the Asian mainland
and thus the tone of Chinese policy would probably become more assertive.”
In a footnote, INR’s acting director noted that the two conclusions appeared
contradictory.33
conversation (memcon), “Chinese Communist Atomic Capability,” October 28, 1963, W. Averell
Harriman Papers, box 588, Memcons “W.”
Weapon Capabilities”; and Lewis and Xue, China Builds the Bomb, pp. 99–100.
32. DCI, SNIE 13–2–63, “Communist China’s Advanced Weapons Program.” See Willis C.
Armstrong, William Leonhart, William J. McCaffrey, and Herbert C. Rothenberg, “The Hazards of
Single-Outcome Forecasting,” in H. Bradford Westerªeld, ed., Inside CIA’s Private World (New Ha-
ven, Conn.: Yale University Press, 1995) for a discussion of the estimate of when the ªrst detona-
tion would occur.
33. DCI, NIE 13–2–63, “Communist China’s Advanced Weapons Program,” p. 10.
Mission to Moscow
While intelligence ofªcers worked on improving their understanding of

China’s nuclear program, senior policy ofªcials, led by President Kennedy,
were considering how to rein in, even “take out,” China’s nuclear program.
With Sino-Soviet hostility increasing, Kennedy hoped that he could enlist

Khrushchev’s support. During his January 10, 1963, conversation with DCI
McCone, Bundy described the president’s fear that a nuclear China “would so
upset the world political scene [that] it would be intolerable.” Thus Kennedy
believed that “we should be prepared to take some form of action unless
they agreed to desist from further efforts in this ªeld.” If McCone initiated any
follow-up planning in response to Kennedy’s instructions, it remains classiªed.34
Kennedy soon expressed his thinking more directly and to a somewhat
broader ofªcial audience. At a meeting of the National Security Council (NSC)
on January 22, 1963, he explained that “our primary purpose [in the treaty ne-
gotiations with the Soviets] is to halt or delay the development of an atomic ca-
pability by the Chinese Communists.” Suggesting that a treaty without
Chinese involvement would “not be very meaningful,” Kennedy believed that
the Russians were as interested as he was in using an agreement as “pressure”
against China’s quest for a nuclear capability. For Kennedy, this was critically
important because the Chinese would be “our major antagonists of the late
60’s and beyond,” and a nuclear China would endanger the U.S. position in
Asia.35
Implicit in Kennedy’s thinking was the view, shared by area specialists at the
CIA and State Department, that the position of the Soviet leadership on peace-
ful coexistence and the dangers of nuclear escalation was substantially more
responsible and less dangerous than Beijing’s. Other Kennedy advisers agreed.
After the NSC meeting, Ambassador-at-Large W. Averell Harriman wrote to
Kennedy; noting Soviet aversion to the possibility of a nuclear-armed Ger-
many, a prospect Harriman also found daunting, he recommended a U.S.-
Soviet “understanding” on Germany and China. The components would be
nonproliferation and test ban agreements, which were then under discussion
34. Editorial note, FRUS, 1961–63, Vol. 22, p. 339.

35. “Remarks of President Kennedy to the National Security Council Meeting of January 22,
1963,” January 22, 1963, FRUS, 1961–63, Vol. 8, p. 462. See Kendrick Oliver, Kennedy, Macmillan, and
the Nuclear Test-Ban Debate, 1961–63 (New York: St. Martin’s, 1998), for an account of the relation-
ship between the Chinese nuclear program and the test ban discussions. A few weeks earlier, Sec-
retary of State Rusk had told Japan’s Foreign Minister Masayoshi Ohira that a Chinese nuclear test
would “destroy” a test ban treaty. Memcon, December 4, 1962, RG 59, 793.5611/12-462.
with the Soviets and the British. Alluding to his own conversations with Soviet
diplomats, Harriman seemed to think that they believed that a test ban agree-
ment could be turned against China. If Moscow and Washington reached an
agreement, “together we could compel China to stop nuclear development,
threatening to take out the facilities if necessary.”36
Even though the Soviets were vainly seeking to repair the schism with the

Chinese and were unlikely to agree to joint military or political action, Ken-
nedy’s and Harriman’s perceptions of mutual U.S.-Soviet concern over the
Chinese nuclear problem had a solid basis. They must have known from CIA
reporting that the Soviets had cut off assistance to the Chinese nuclear pro-
gram. More signiªcant, the implications of Moscow’s support for test ban and
nuclear nonproliferation agreements were not lost on Beijing. Beginning in Au-
gust 1962, the Soviets responded positively to U.S. proposals for an interna-
tional nuclear nonproliferation agreement. Secretary of State Rusk saw China
as a target of such an agreement, and its anti-Chinese aspect was evident to
Beijing, which bitterly protested Moscow’s participation in the talks.37
The possibility of a U.S.-Soviet coalition against the Chinese nuclear pro-
gram shaped a February 1963 request from Assistant Secretary of Defense for
International Security Affairs Paul Nitze to the JCS asking for a study of the
possibility of using “persuasion, pressure, or coercion” to induce China to sign
a test ban treaty. A few months later, on April 29, 1963, the chiefs responded
with a lengthy report listing the indirect measures, such as diplomatic and
propaganda campaigns, and the direct measures that might coerce Beijing.
Measures could be taken unilaterally or by allies, or with the active or tacit
support of the Soviet Union. The direct ones included inªltration, sabotage, or
invasion by Chinese Nationalists; maritime blockades; a South Korean inva-
sion of North Korea (to put pressure on China’s borders); conventional air at-
tacks on nuclear facilities; and the use of a tactical nuclear weapon on a
“selected CHICOM [Chinese Communist] target.”38
36. Deputy Assistant Director for National Estimates (CIA) Chester Cooper to Mose Harvey, PPC,
“Discussion of the Implications of the Sino-Soviet Dispute for U.S. Policy,” March 16, 1962, RG 59,
Ofªce of Soviet Union Affairs, Subject File 1957–63, box 6, 22.1 Soviet-Chicom Relations; and
Harriman to Kennedy, January 23, 1963, Library of Congress, W. Averell Harriman Papers, John
Kennedy General, 1963.
37. DCI, NIE 13–2–62, “Chinese Communist Advanced Weapons Capabilities,” p. 9; memcons,
August 8 and 23, 1962, FRUS, 1961–63, Vol. 7, pp. 541–547, 557; and INR, “Soviet Attitude toward
Chinese Communist Acquisition of a Nuclear-Weapons Capability,” September 11, 1963, RG 59,
Records of Policy Planning Council, 1963–64 (hereinafter cited as PPCR), box 250, China.
38. Gen. Curtis LeMay, acting chairman, JCS, to Secretary of Defense, April 29, 1963, “Study of
Chinese Communist Vulnerability,” with 31-page appendix attached, RG 59, Ofªce of the Country
If civilian policymakers hoped for unambiguous advice on the prospects

and possibilities of coercion, the JCS could not provide it. Acting Chairman
Curtis LeMay’s cover memo to the longer report concluded that it was “unreal-
istic to use overt military force” against China because even if Beijing was co-
erced into signing a test ban agreement, there was no guarantee that it would
adhere to one. Moreover, if the United States took unilateral action, whether

initiating a blockade or military action, policymakers had to consider the
strong prospects for retaliation and escalation, not to mention the difªculty of
justifying such an action to international opinion. If the United States acted
jointly with the Soviet Union, however, or at least had Moscow’s approval,
blockades or conventional military measures would be “more effective” in as-
suring Chinese compliance. Thus if the United States undertook air strikes
against PRC nuclear facilities, Soviet cooperation could “well be the difference
between escalation and quick acquiescence by the Chicoms.”39
The JCS were skeptical that Moscow would be interested in cooperating
with Washington against Beijing, but the possibility continued to entice civil-
ian ofªcials. Indeed, National Security Assistant Bundy may have been the
ªrst senior U.S. ofªcial to broach the issue directly with the Soviets. In mid-
May, during a meeting with Soviet Ambassador Anatoly Dobrynin, Bundy
suggested a “private and serious exchange of views” on the Chinese nuclear
problem. Dobrynin, however, was not interested. He took issue with U.S. plans
for a multilateral force (MLF), the proposed NATO-controlled medium-range
ballistic missile force designed to give West Germany and other nonnuclear al-
lies a role in decisions on nuclear weapons use. The MLF had been a sticking
point in U.S.-Soviet talks on nuclear nonproliferation and, Dobrynin argued,
“did not make it easier for the Soviet Government to deal with the question of
Chinese nuclear ambitions.” Although Bundy argued that the MLF was in
Moscow’s interest because it would bind West Germany closer to NATO, thus
checking nationalist tendencies and any “adventurous notions” in the Federal
Republic, Dobrynin refused to discuss China.40
Dobrynin’s rebuff did not halt discussion of a joint U.S.-Soviet approach,
however. Interest grew after President Kennedy publicly announced on June
Director for the Republic of China, 1954–65, box 4, Nuclear Capability. The JCS did not limit op-
tions to coercion, but also discussed possible incentives, such as increased trade, food aid, and
membership in international organizations.
39. Ibid., p. 26.
40. National Security Assistant McGeorge Bundy, “Memcon with Ambassador Dobrynin, at
lunch, May 17, 1963,” RG 59, RAALLT, box 10, White House, 1962.
10, 1963, that Premier Khrushchev had agreed to receive a high-level U.S. en-
voy to discuss a test ban treaty. Kennedy chose Ambassador-at-Large
Harriman as his representative, but the agenda for the talks with Khrushchev
was the subject of some controversy. In part because the JCS—if not Secretary
of Defense Robert McNamara—were opposed to a test ban, Arthur Barber, one
of Nitze’s assistants, argued that Harriman should also broach with Khrush-

chev the possibility of an agreement on nonproliferation of nuclear weapons.
Treating China as a central problem, Barber emphasized the possibility of joint
U.S.-Soviet cooperation and prepared a detailed paper, “Aborting the
CHICOM Nuclear Capability,” that posited U.S.-Soviet cooperation in a series
of moves, beginning with “political persuasion” (with Moscow taking the
lead) and concluding with “jointly conducted U.S.-Soviet air strikes” against a
“minimum complex of [Chinese] installations.” Recognizing that the MLF was
a stumbling block to Soviet interest in a nuclear nonproliferation agreement,
Barber believed that Moscow would never support joint measures against
China’s nuclear facilities unless Washington changed its nuclear policy toward
Europe by dropping or substantially modifying the MLF.41
Barber assumed that gaining Soviet cooperation on China by sacriªcing the
MLF was a strategy worth exploring, as did President Kennedy. Thus during a
cabinet meeting held a few weeks before Harriman’s mission to Moscow, when
ACDA Director Foster mentioned the possibility of U.S.-Soviet cooperation
against the Chinese nuclear program, Kennedy raised the possibility of “giving
up the MLF concept” as part of a nonproliferation agreement. National Secu-
rity Assistant Bundy saw the MLF as a “bargaining point,” but Secretary of
State Rusk promptly objected. The MLF “involved the Allies so deeply” that
giving it up was not “a possible position at this time.”42
Kennedy did not press the MLF point, but he remained interested in the pos-
sibility of joint U.S.-Soviet action against the Chinese nuclear program. When
he met with British Prime Minister Harold Macmillan in late June, both men
grappled with the problem of how to bind Beijing to a nuclear test ban agree-
ment, even discussing “methods” for preventing nuclear proliferation (the
minutes do not specify any details). Macmillan, however, raised the possibility
41. William Y. Smith, NSC staff, “Events Leading Up to the Harriman Moscow Mission,” August
21, 1963, FRUS, 1961–63, Vols. 7–9, microªche supplement; and “Brieªng Book on U.S.-Soviet Non-
Diffusion Agreement for Discussion at the Moscow Meeting,” c. June 12, 1963, JFKL, National Se-
curity Files, Carl Kaysen Papers, box 276, Nuclear Energy Matters. For internal Pentagon divisions
on a test ban, see FRUS, 1961–63, Vol. 7, pp. 719–722.
42. William C. Foster oral history; and editorial note, FRUS, 1961–63, Vol. 7, p. 735.
of a “joint note with the Russians to the Chinese,” and Kennedy agreed that
ways had to be found to restrain the Chinese to make a test ban effective. Once
the Chinese tested, Kennedy declared, the United States would have to resume
testing, suggesting that a test ban agreement could lapse under such circum-
stances. Discussions with the Russians were essential.43
Soon after arriving in Moscow on July 14, Harriman received a message

from Kennedy instructing him to emphasize to Khrushchev that a nuclear
China, even with small forces, “could be very dangerous to us all.” The presi-
dent wanted Harriman to explore Khrushchev’s thinking on “limiting or pre-
venting Chinese nuclear development and his willingness either to take Soviet
action or to accept U.S. action aimed in this direction.”44 Although Harriman
successfully ªnalized agreement on a Limited Test Ban Treaty, he failed to
catch Khrushchev’s interest in any discussion of political or other steps against
China. As long as France was going to hold out against the test ban, Khrush-
chev could not agree to isolate Beijing. Moreover, the premier played down the
Sino-Soviet split and rejected Harriman’s view that a nuclear China could
threaten Moscow. Khrushchev also played down the dangers of China’s nu-
clear ambitions, observing that a nuclear-armed Beijing would become “more
restrained.” “Whenever someone lacked [nuclear] means he was the one who
shouted the loudest.”45
Khrushchev had to be loath to discuss with U.S. envoys the possibility of ap-
plying pressure on China. A PRC delegation led by Deng Xiaoping had just
broken off a round of bitter ideological debates, possibly because Harriman
and his party were in Moscow negotiating the test ban. Some of the Soviet
leadership were dismayed by the polemics with Beijing, and Khrushchev
would have found it politically untenable to discuss action against a fraternal
socialist country, no matter how wayward.46
Continued Sino-Soviet tensions led U.S. policymakers to speculate about
a possible opening against the Chinese nuclear effort. Certainly Kennedy’s
apprehension did not lessen, and he may have instructed senior defense
ofªcials to look closely at the possibilities of military action. During a press
43. Memcons, June 29 and 30, 1963, in FRUS, 1961–63, Vol. 7, pp. 753–754, 758.
44. Telegram to Moscow embassy, July 15, 1963; and Moscow embassy telegram to State Depart-
ment, July 27, 1963, both in FRUS, 1961–63, Vol. 7, pp. 801, 860.
45. U.S. embassy, Moscow, telegram to State Department, July 27, 1963, in FRUS, 1961–63, Vol. 7,
p. 860; Oliver, Kennedy, Macmillan, and the Nuclear Test-Ban Debate, pp. 205–206; and Vladislav
Zubok, “Look What Chaos in the Beautiful Socialist Camp: Deng Xiaoping and the Russians, 1956–
1963,” Bulletin of the Cold War International History Project, Vol. 10 (Winter 1987/98), pp. 152–162.
46. Zubok “Look What Chaos in the Beautiful Socialist Camp.”
conference on August 1, he spoke of a “menacing situation”; acknowledging

that it would take some years before China could “become a full-ºedged nu-
clear power. . . . We would like to take some steps now which would lessen
that prospect.” Only the day before, Assistant Secretary of Defense for Interna-
tional Security Affairs William Bundy had secretly tasked the JCS to prepare a
contingency plan for a conventional attack designed to cause the “severest im-

pact and delay to the Chinese nuclear program.” Presumably this request had
the approval of Secretary of Defense McNamara, but the latter’s role in discus-
sions of China remains, so far, obscure.47
Exploring Options with the Chinese Nationalists
A visit to Washington in September 1963 by Chiang Kai-shek’s son, Gen.

Chiang Ching-kuo—Taiwan’s feared “security czar”—occasioned talk of possi-
ble “steps” against Beijing’s nuclear program, but this time with an old ally, the
Nationalist Chinese, instead of a Cold War adversary. DCI McCone had invited
Chiang for discussions, which centered on long-standing differences between
Washington and Taipei over military operations against the PRC. The dialogue
easily shifted to the possibility of action against the Chinese nuclear program.48
No doubt concerned that a nuclear-armed China would end any hopes for a
return to the mainland, Chiang Ching-kuo raised the issue of attacking China’s
nuclear facilities on several occasions. A day before his meeting with Kennedy,
Chiang visited CIA headquarters, where he took part in discussions on the
possibility of airborne raids against the facilities. Later, in the company of the
CIA’s deputy director of intelligence and former chief of station in Taiwan, Ray
Cline, and his successor, William Nelson, Chiang met National Security Assis-
tant Bundy, to discuss strikes against the mainland, including antinuclear op-
erations. Although Bundy favored measures to “weaken” the PRC, he doubted
that plans to seize territory would work and counseled against action that
could realign Beijing with Moscow or “trigger a major conºict.” Chiang
brought up possible action against Beijing’s nuclear installations, suggesting
that the United States provide “transportation and technical assistance” for a
commando operation. Bundy responded that the “United States is very inter-
47. “The President’s News Conference of August 1, 1963,” Public Papers of the Presidents of the
United States: John F. Kennedy, 1963 (Washington, D.C.: U.S. GPO, 1964), p. 616; and FRUS, 1964–68,
Vol. 30, p. 24, n. 7.
48. Memorandum for McGeorge Bundy, “Meeting of General Chiang Ching-kuo with the Presi-
dent,” September 10, 1963, JFKL, National Security Files, Countries, box 24, China General, Sep-
tember–October 1963. The public cover story was that the State Department had invited Chiang.
ested in whether something could be planned” that could have a “delaying

and preventive effect on the nuclear growth of China.” He believed, however,
that those measures needed “most careful study.”49
On September 11, an extended discussion between Kennedy and Chiang
prompted the president to question the proposal to send commandos against
Chinese nuclear installations. He asked “whether it would be possible to send

300 to 500 men by air to such distant . . . atomic installations as that at Baotou,
and whether it was not likely that the planes involved would be shot down.”
According to the record, Chiang replied that the commando raid proposal
“had been discussed by CIA ofªcials yesterday and that they had indicated
that such an operation was feasible.” Kennedy’s query suggested some doubts
about the proposal’s feasibility, and other comments emphasized the impor-
tance of “realistic” plans to “weaken the Chinese Communist regime.” To
avoid another Bay of Pigs operation “based more on hope than on realistic ap-
praisals,” Washington and Taipei needed better intelligence about conditions
on the mainland. In that way, Kennedy argued, “whatever action is under-
taken would ªt the actual situation.”50
A few days later, Chiang met with McCone to formalize the understandings
that the general had reached with Kennedy and his advisers. With respect to
action against PRC nuclear installations, McCone and Chiang agreed to estab-
lish a planning group to study the feasibility of attacks by Nationalist teams
against nuclear sites. Any operations would require joint approval by top au-
thorities. Unfortunately, details about the activities of the U.S.-ROC planning
group on Chinese nuclear targets are unavailable.51
Possibilities and Pitfalls of Direct Action
In the fall of 1963 and beyond, the issue of what, if anything, to do about the
Chinese nuclear program occupied not only meetings of high U.S. government
49. Deputy Assistant Secretary of State Edward Rice to Ambassador Ting-Fu Tsiang, October 14,
1963, enclosing record of Chiang-Kennedy meeting, RG 59, Subject-Numeric Files, 1963, POL
Chicom-AK; and CIA Far East Division Chief William E. Colby to McGeorge Bundy, “Visit of Gen-
eral Chiang Ching-kuo,” September 19, 1963, enclosing record of conversation, FRUS, 1961–63,
Vols. 12, 14, microªche supplement, doc. 67.
50. Deputy Assistant Secretary of State Rice to Ambassador Ting-Fu Tsiang, October 14, 1963.
51. Background Paper, “U.S.-G.R.C. Consultations Concerning Possible Action against the Main-
land,” and memorandum for the record, September 1963, on Understandings Reached during
Chiang Ching-kuo’s Visit,” n.d., both enclosed in State Department Executive Secretary Benjamin
Read to National Security Assistant McGeorge Bundy, “The President’s Meeting with Chinese
Minister of Defense Chiang Ching-kuo on September 23 at 11:30 A.M.,” September 18, 1965, RG 59.
Formerly Top Secret Foreign Policy Files, 1964–66, box 11, POL Chinat-U.S.
ofªcials but leading journalists as well—none more so than the well-connected

and inºuential columnist Stewart Alsop. In Saturday Evening Post columns
published in September and October, Alsop wrote about the “madness of Mao
Tse-tung” and the necessity and feasibility of military action against Beijing’s
nuclear program. He asserted that the “president and his inner circle . . . have
agreed in principle that China must be prevented, by whatever means, from

becoming a nuclear power.” “Nuclear sterilization” would require force, which
Alsop characterized as a “technically easy problem” that could be accom-
plished with a “few rather small bangs.”52
Behind locked doors, in the weeks after Chiang Ching-kuo’s visit, the Ken-
nedy administration continued to review options that would create a few small
bangs. Apparently, the possibility of air dropping Taiwanese sabotage teams
and other covert options got a close look at the CIA at least through the begin-
ning of 1964.53 Moreover, the Pentagon studied paramilitary options. On No-
vember 18, 1963, JCS Chairman Maxwell Taylor presented his colleagues with
a paper on “how we can prevent or delay the Chinese from succeeding in their
nuclear development program” slated for discussion at their next meeting. The
listing of the agenda item—”Unconventional Warfare Program BRAVO”—
indicates the paramilitary nature of the contemplated action. It may have been
this paper or some variant that Deputy Secretary of Defense Roswell Gilpatric
sent to McCone a month later, noting that the JCS had recommended an inter-
agency group to “consider ways and means for impeding the Chinese Com-
munist nuclear program.” That such action had been under serious
consideration is indicated by an unsuccessful U.S.-directed Chinese National-
ist attempt in the fall of 1963 to ºy a U-2 equipped with an infrared camera
52. Stewart Alsop, “The Real Meaning of the Test Ban,” and “The Madness of Mao Tse-tung,” Sat-
urday Evening Post, September 28 and October 26, 1963, respectively. Stewart was not as close to
Kennedy as was his brother, Joseph, but had access to the president and his advisers. See Robert W.
Merry, Taking on the World: Joseph and Stewart Alsop—Guardians of the American Century (New York:
Viking, 1996), pp. xviii, 374, 398.
53. Several documents mention that covert options, including a commando raid, had been
under tudy. See George Rathjens, ACDA, “Destruction of Chinese Nuclear Weapons Capabilities,”
December 14, 1964, copy available on the National Security Archive’s web page; Robert
H. Johnson, PPC, to PPC Director Walt Rostow, “Direct Action against Chicom Nuclear Facil-
ities,” February 12, 1964, PPCR, box 265, Chron File, R. Johnson, January–June 1964; and memo-
randum for the president, “The Implications of a Chinese Communist Nuclear Capability,”
April 17, 1964, RG 59, Subject-Numeric Files, 1964–66 (hereinafter SN 64–66), DEF 12–1 Chicom.
An excised version of the Rathjens paper, with commentary, was ªrst published by
Shane Maddock, “LBJ, China, and the Bomb,” SHAFR Newsletter, Vol. 27, No. 1 (March 1996),
pp. 1–5.
over the suspected plutonium reactor at Baotou, to determine whether it was

hot—and thus off limits to military attack.54
In addition, the JCS responded to Assistant Secretary of Defense William
Bundy’s earlier request for a contingency plan for a conventional attack to re-
tard Chinese nuclear development. In mid-December, they completed a plan
for a multiple-sortie attack designed to cause severe damage and delays. Nev-

ertheless, the large number of sorties required by the plan probably encour-
aged the JCS to propose looking into a possible nuclear attack on the same
facilities, a proposal that was undoubtedly disregarded.55
Sometime in the fall of 1963, additional pressure came from Chester Bowles,
who had become U.S. ambassador to India. No doubt, Bowles worried that un-
less Washington acted, Indian fear of a Chinese nuclear capability, especially in
light of the 1962 border war with Pakistan, could stimulate nuclear prolifera-
tion in South Asia. Bowles’s militant stance on China stood in interesting con-
trast to his opposition to the Bay of Pigs operation in 1961.56
Kennedy did not think exclusively in terms of military options; the president
and his advisers also worked to elicit Soviet cooperation on a nuclear
nonproliferation agreement partly aimed at China. Secretary of State Rusk dis-
cussed nonproliferation with Soviet Foreign Minister Andrei Gromyko at the
United Nations in the fall of 1963. When Gromyko raised the MLF problem,
Rusk needled him by observing that Moscow “had lost its virginity” when it
aided China’s nuclear program. The Soviets were nonetheless interested in a
nonproliferation agreement, and Gromyko discussed one with Kennedy on
October 10. Showing some willingness to exert indirect pressure on the PRC,
Gromyko acknowledged than an agreement would make China’s “political sit-
uation more difªcult and delicate,” presumably by increasing the PRC’s isola-
tion and raising pressures on it to follow nonproliferation standards.57
54. JCS Chairman Maxwell Taylor, memorandum for Gen. LeMay, Gen. Wheeler, Adm. McConald,
and Gen. Shoup, “Chinese Nuclear Development,” November 18, 1963, National Archives, RG
218, Taylor Papers, box 1; and Deputy Secretary of Defense Roswell Gilpatric to DCI McCone, De-
cember 13, 1963, summarized in FRUS, 1961–63, Vol. 22, p. 405, n. 3; and Wheelon interview.
55. FRUS, 1964–68, Vol. 30, p. 24, n. 7; and Johnson to Rostow, “Direct Action against Chicom Nu-
clear Facilities.”
56. Information from a retired CIA ofªcer who read Bowles’s message to the State Department.
57. Memcon, “World Reaction to Test Ban Treaty,” September 28, 1963, RG 59, Executive Secretar-
iat Conferences Files, 1949–63, box 327, CF 2318; memcon, “Non-Dissemination and the MLF,” Oc-
tober 10, 1963, JFKL, National Security Files, Countries, box 187, Gromyko—Talks. A few days
after the late-September Gromyko-Rusk meeting, the latter alluded to Soviet objections to the MLF
Kennedy asked when China would have the bomb, but Gromyko said he
did not know and added, rather defensively, that “the USSR did not give any-
thing to the Chinese.” Whether Kennedy took Gromyko aside to see if the So-
viets were interested in cooperative action against the Chinese nuclear
program is unknown.58
While Kennedy considered moves against China, State Department policy

planners had begun to step back and look skeptically at claims that Beijing’s
nuclear program would have an “intolerable” impact. For example, in early
July 1963, only a few weeks before Harriman tried to open discussion with
Khrushchev about the possibility of cooperation against the Chinese nuclear
program, PPC Director Walt Rostow informed him that the minimal nuclear
capability that Beijing could develop was unlikely to “convince . . . anyone”
that it could be “used as an umbrella for aggression.” Not only would “U.S.
overwhelming nuclear superiority” deter Beijing, its “desire to preserve its nu-
clear force as a credible deterrent might tend to make China even more cau-
tious than it is today in its encounters with American power.”59
Rostow’s argument about a Chinese nuclear capability was brief and im-
pressionistic because one of his staffers, Robert Johnson, had already com-
pleted the ªrst draft, more than 200 pages long, of a major study entitled “A
Chinese Communist Nuclear Detonation and Nuclear Capability.” By October
1963, Johnson had prepared a shorter (100-page) version for wider distribu-
tion. He had worked closely with a group of ofªcials from the Departments of
State and Defense, ACDA, the CIA, and the U.S. Information Agency, and the
shorter version had “[their] broad agreement.” Some of President Kennedy’s
advisers considered making the report available to him, but he probably never
saw it.60
Eschewing language about a “menacing situation,” Robert Johnson’s studies
were decidedly nonalarmist. Their conclusion about the implications of a Chi-
nese nuclear capability—that it would not require Washington to make any
when he told Gromyko that “the Soviet Union had slept with China and therefore could not pro-
test that we were holding the hand of Miss Germany.” Memcon, “Bombs in Orbit,” October 3,
1963, Executive Secretariat Conference Files, box 327, CF 2320.
58. Memcon, “Non-Dissemination and the MLF.
59. PPC Director Walt Rostow to Governor W. Averell Harriman, “Non-Proliferation,” July 2, 1963,
Harriman Papers.
60. “A Chinese Communist Nuclear Detonation and Nuclear Capability,” June 17, 1963, RG 59,
PPCR, box 276, S/P Papers June–July 1963; “Policy Planning Statement on a Chinese Communist
Nuclear Detonation and Nuclear Capability,” October 15, 1963, PPCR, box 275, S/P Papers; Henry
Owen, PPC, to Secretary of State, “A Chinese Communist Nuclear Detonation and Nuclear Capa-
bility,” October 7, 1963, PPCR, box 271, Chron File October–December 1963; and FRUS 1961–63,
Vol. 22, p. 405, n. 4.
signiªcant policy changes—ºowed from Johnson’s assumption that a Chinese

capability “will not, for the indeªnite future, alter the real relations of power
among the major states or the balance of military power in Asia.” For Johnson,
the “great asymmetry in Chinese Communist and U.S. capabilities and vulner-
abilities” minimized the Chinese nuclear threat. A nuclear China would be
within range of American striking power but not vice versa, forcing it to “take

account of the danger of a U.S. nuclear or non-nuclear counterforce attack as a
possible response to major . . . aggression.” This made it “exceedingly un-
likely” that the Chinese would use nuclear weapons ªrst, unless the mainland
was under “serious attack.” In Johnson’s view, the Chinese wanted a nuclear
force to deter an attack on their territory and were unlikely to change their es-
sentially prudent, risk-adverse military policy.61
Although he minimized the military risks, Robert Johnson was not sanguine
about the political implications of a Chinese nuclear test. He believed that Chi-
nese leaders were unlikely to make “gross public threats,” but they were apt to
believe that a nuclear capability could “weaken the will” of U.S.-allied neigh-
bors and induce them to accommodate China, thereby eroding the U.S. pres-
ence in East Asia. Although Beijing might be more willing to take risks in
military probing operations because of an overoptimistic assessment of its psy-
chological advantage, Johnson nonetheless argued that the Chinese would
avoid measures that could have “seriously adverse counter-effects.”
In emphasizing Chinese caution, Johnson showed the inºuence of the State
Department’s INR, which had prepared special studies designed to assist his
work. Its most comprehensive report held that a nuclear-armed China would
“eschew rash military actions” or even “nuclear blackmail”; instead it would
use its new capability as a “political weapon . . . to earn respect, to promote
neutralism, to encourage revolutionaries.”62
Robert Johnson argued that to neutralize any political beneªts that a small
nuclear capability would provide Beijing, the United States need not do much
more than it was already doing. On the political-military front, once Beijing
had tested a weapon, the United States had to be ready to provide “reassur-
ance” to friendly countries that it would help them in a confrontation with
China. Reassurances could counter “the effectiveness of Chinese . . . pressures
and the possibility of the development of independent nuclear capabilities by
61. Quotations in this paragraph and the next are from the October 15, 1963, study, pp. 6–12.
62. Author’s interview with Robert H. Johnson, March 3, 1999; and George C. Denney, INR, to the
Secretary, “Probable Consequences of a Chinese Communist Nuclear Detonation,” May 6, 1963,
RG 59, PPCR, box 250, China.
Asian countries.” To back up its assurances, Washington needed an “evident

ability to respond rapidly with adequate force to Communist military probes
without undue reliance upon nuclear weapons.” Any increase in Washington’s
reliance on nuclear weapons for deterrence or for responding to “non-nuclear
aggression,” Johnson argued, would make friendly countries more likely to
work with Beijing than with Washington. This meant that he expected that the

Kennedy administration’s ºexible response strategy, with its nonnuclear em-
phasis, could adequately deter Beijing.63
No version of Robert Johnson’s studies ever received high-level sanction,
but they had “ofªcial” standing as the basic State Department position and
were considered authoritative enough to be summarized for President Johnson
in April 1964. As Robert Johnson suggested many years later, he was interested
in exposing the bureaucracy to nonalarmist thinking about the implications of
a nuclear-armed China. “The point was to assure, as far as possible, that all
parts of the government were singing from the same sheet of music,” which
was not “going to happen automatically,” as was shown by the pressure from
President Kennedy and McGeorge Bundy.64
The PPC’s thinking had immediate impact. During a mid-October meeting
with senior State Department ofªcials, Rusk reviewed Robert Johnson’s study
and was favorably impressed. A few weeks later, Rostow’s interagency plan-
ning group discussed Johnson’s paper as well as the problem of “how to stran-
gle the baby in the cradle” before Beijing tested a weapon. At least one ofªcial
present, NSC staffer Robert Komer, believed that Johnson’s estimate had made
the issue of preventive action largely irrelevant. As he reported to McGeorge
Bundy, “If my reading of the thrust of the [Johnson] paper is correct,”—that
Beijing would remain “basically cautious”—”there would be less incentive for
us” to attack Chinese facilities.65
63. Quotations from the October 15, 1963, study, pp. 41–47.
64. Johnson interview; and letter from Robert H. Johnson, May 6, 1999. For the “ofªcial” status of
Johnson’s October 15, 1963, report, see Lindsay Grant, State Department Bureau of Far Eastern Af-
fairs, to Deputy Assistant Secretary of State Marshall Green, “Baguio Conference: Talking Points
on the Chinese Communist Nuclear Question,” May 25, 1964, RG 59, Bureau of Far Eastern Affairs,
Ofªce of Regional Affairs, SN 64–66, box 1, Politico-Military Conference (BAGUIO). One of the
highly abridged versions prepared for President Johnson unquestionably reached his desk; it was
attached to Secretary of State Rusk’s daily brieªng. See memorandum for the president, “Items for
Evening Reading,” May 1, 1964, RG 59, President’s Evening Reading Reports, box 1, President’s
Evening Reading Items, 1964.
65. “Highlights from the Secretary’s Policy Planning Meeting Held October 15, 1963,” October 24,
1963; and Robert W. Komer, NSC, to National Security Assistant McGeorge Bundy, November 5,
1963, both in FRUS, 1961–63, Vol. 22, pp. 399–402, 404–405.
Komer’s skepticism about preventive action was not enough to stop the on-
going planning at the CIA and Pentagon, but it may have been enough to give
some senior ofªcials pause for thought about the policy implications of the use
of force. For all of the talk about taking out Chinese nuclear facilities, no one on
the civilian side had subjected the idea to a detailed analysis. Sometime in the
fall of 1963, Rostow “committed” to Rusk that he would have the PPC prepare

a substantive study on forms of direct action against Chinese nuclear facilities.
Given his expertise, Robert Johnson was selected to lead this project, which he
coordinated with ofªcials at the CIA and the Defense Department.66
While Robert Johnson was working on his study, President Kennedy was as-
sassinated, leading to Lyndon Johnson’s succession. The new president recog-
nized the need for a more ºexible China policy, but he was nevertheless
ambivalent. In a phone conversation with Senator Richard Russell, only weeks
after he came to power, Johnson said that “there’s [no] question” that sooner or
later the United States would have to recognize the PRC, although as Russell
observed, such action was then “politically poison.” The strength of the Tai-
wan lobby undoubtedly inºuenced the president’s calculations, but he was
also troubled by China’s inºuence in East Asia, seeing it as a signiªcant threat
to American power and credibility in the Paciªc.67
It would be months before Lyndon Johnson would take a position on mili-
tary action against Chinese nuclear facilities, but NSC ofªcials such as Robert
Komer still saw no need for overreaction. Nevertheless, Komer’s superior, Na-
tional Security Assistant Bundy, wanted to see more work on preventive ac-
tion, noting that he was “for it” (whether he meant action or more analysis is
not clear). And the president had his own concerns. Showing some anxiety
about the impending presidential campaign and a nuclear-capable China,
Johnson told Time magazine correspondent Hugh Sidey that “we can’t let
[Barry] Goldwater and Red China both get the bomb at the same time. Then
the shit would really hit the fan.” Nevertheless, unlike his predecessor, John-
son made no public comments about China’s nuclear program, much less re-
marks about taking “steps” against it.68
66. Johnson to Rostow, “Direct Action against ChiCom Nuclear Facilities.”

67. Editorial note, FRUS 1964–68, Vol. 30, pp. 3, 23–24; Walter LaFeber, The American Age: United
States Foreign Policy at Home and Abroad since 1750 (New York: W.W. Norton, 1989), p. 579; and
Lloyd Gardner, Pay Any Price: Lyndon Johnson and the Wars for Vietnam (Chicago: Ivan R. Dee, 1995),
pp. 115–117.
68. Robert W. Komer, NSC, to National Security Assistant McGeorge Bundy, February 6, 1964,
FRUS, 1964–68, Vol. 30, pp. 23–24; for Bundy’s “I’m for it,” ibid., p. 24, n. 8; Johnson quotation from
Gardner, Pay Any Price, p. 119. For the Komer-LBJ relationship, see John Prados, Keepers of the Keys:
In mid-April 1964, Robert Johnson completed his study “An Exploration of

the Possible Bases of Action against the Chinese Communist Nuclear Facil-
ities.” Owing to its highly sensitive subject matter, the study was classiªed
“top secret.” It remains classiªed, although its excised conclusions appear in
the State Department’s Foreign Relations series. Fortunately, several declassiªed
reports, one prepared by Johnson himself, summarize major issues and argu-

ments in the study.69
Robert Johnson identiªed four methods of attack: an overt nonnuclear air at-
tack by the United States, an air attack by the ROC, covert ground attacks em-
ploying agents in China, and an air drop of ROC sabotage teams. A U.S. air
attack, Johnson suggested, was problematic, because it would take too many
sorties to destroy targets completely. A ground attack involving Chinese agents
was impractical because such assets were unavailable. The Nationalists lacked
the capabilities to stage their own air raid, and the United States could not con-
vincingly disguise any assistance that it provided. If, however, the Nationalists
dropped a 100-man team, they could cause great damage but not destroy a fa-
cility completely.70
Preventive military action (whether overt or covert) had, Johnson wrote,
signiªcant advantages: Destroying China’s nuclear facilities would remove
“the immediate incentive and justiªcation for Indian development of nuclear
weapons and the possible movement of Japan in the same direction.” Further,
China would lose the “political-psychological” and “defense” advantages pro-
vided by a nuclear capability.71
Robert Johnson also identiªed great disadvantages to any violent option.
First, with all the gaps in intelligence about the PRC’s nuclear program, Wash-
ington may not have identiªed all of the relevant targets. Second, an attack
would only “buy some time,” perhaps four or ªve years. Given Beijing’s deter-
mination to acquire a nuclear deterrent, it was likely to try to restart the nu-
clear program and would probably build underground facilities and
A History of the National Security Council from Truman to Bush (New York: William T. Morrow, 1991),
p. 120.
69. “An Exploration of the Possible Bases for Action against the Chinese Communist Nuclear Fa-
cilities,” April 14, 1964, FRUS, 1964–68, Vol. 30, pp. 39–40; Robert Johnson to members of PPC,
“The Chinese Communist Nuclear Capability and Some ‘Unorthodox’ Approaches to the Problem
of Nuclear Proliferation,” June 1, 1964, PPCR, box 264, Johnson, Chron File, January 1964, Bulky
Reports; and Rathjens, “Destruction of Chinese Nuclear Weapons Capabilities.”
70. Rathjens, “Destruction of Chinese Nuclear Weapons Capabilities.”
71. Quotations in this and the next paragraph are from Johnson, “The Chinese Communist Nu-
clear Capability and Some ‘Unorthodox’ Approaches to the Problem of Nuclear Proliferation,”
June 1, 1964.
strengthen air defenses to prevent another attack. Third, Chinese retaliation,

perhaps against Taiwan or U.S. bases in East Asia, could not be dismissed.
Finally, an unprovoked attack could entail heavy foreign policy costs. Overseas
and domestic critics would argue that the attack exempliªed the United States’
“unwillingness to accept the existence of Communist China as a major world
actor.” Others would argue that the action contradicted U.S. efforts to down-

play the signiªcance of a Chinese nuclear capability. Still others would charge
the United States with racism by singling out China or point out the “grave
risks of precipitating war.”
A case for action that could win international acceptance would be one
based on worldwide arms control agreements, such as treaties on nonprolif-
eration and inspected suspension of nuclear materials production. Once such
agreements had been negotiated, international opinion might accept action
against China if it was overtly ºouting them. By the time that such treaties had
been negotiated, however, they would have little force against China because it
would probably have already tested a nuclear weapon; like France, it would be
an “existing” nuclear power. In any event, the French were unlikely to support
such agreements. Unless they did, the Soviets would not support action sin-
gling out China. Further, Johnson saw Soviet acquiescence in U.S. action
against China as “highly unlikely.”72
Johnson believed that world opinion would be more likely to accept military
action in response to aggression, for example, if Chinese forces openly inter-
vened on the communist side in limited wars in Laos or Vietnam. The problem
was that complete destruction of the known facilities would take sizable
bomber attacks against sites deep within China’s. Beijing or Moscow could
readily misinterpret what was supposed to be a “limited war action” as the be-
ginnings of a major attack on the mainland. The possibility of “retaliatory ac-
tion . . . could not be ruled out.”73
72. The last quotation is from “An Exploration of the Possible Bases for Action against the Chinese
Communist Nuclear Facilities,” p. 40. Some State Department experts on the Soviet Union argued
that, from a pure balance-of-power standpoint, the United States should not work with Moscow
against Beijing because a small Chinese nuclear capability “could interpose something of a deter-
rent to Soviet military pressure but . . . not threaten the U.S.” A balance-of-power approach reso-
nated with the thinking of inºuential China hands such as Deputy Assistant Secretary of State
Marshall Green, but whether it carried any weight elsewhere remains to be seen. See Helmut
Sonnenfeldt, INR, to INR Director Thomas Hughes, April 14, 1964, enclosing “The U.S. Interest in
Communist China,” October 14, 1963, RG 59, Records of the Ofªce of the Counselor, 1955–77, HS
Chron File, July–December 1963.
73. Johnson, “The Chinese Communist Nuclear Capability and Some ‘Unorthodox’ Approaches to
the Problem of Nuclear Proliferation,” June 1, 1964.
An earlier conclusion of Robert Johnson’s study—that the Chinese nuclear

threat was not serious enough to “justify . . . actions which would involve great
political costs or high military risks”—ºowed from his earlier analyses of Chi-
nese prudence. Given the possible advantages of direct action, such as prevent-
ing a “chain reaction” of nuclear proliferation in Asia, Johnson allowed that a
covert approach was the “most politically feasible form of action” as long as

the United States could disassociate itself from it. Nevertheless, although a
Nationalist sabotage team could conceivably destroy the Baotou facility, if it
was determined that action against the others (e.g., at Lanzhou) was also nec-
essary, then the problem was more complex. “Near-simultaneous attacks
would be necessary and perhaps difªcult to manage.” Further, a Chinese retal-
iatory attack on Taiwan would put Washington “in a very difªcult interna-
tional position.” In general, Johnson’s preference was to reserve the covert
option in the event of open Chinese aggression because its implementation
would not involve the high risks associated with a U.S. bomber attack.74
Robert Johnson’s analysis was implicitly contrary to President Kennedy’s
thinking in 1963. That is, Johnson argued that preventive action against
China’s nuclear program was dangerous and likely to fail and that it could
hurt the United States’ image and weaken its prestige, the intangible assets of
world power.
Robert Johnson later wrote that his study reºected a “broad interdepartmen-
tal consensus,” which suggested that its conclusions would carry no small
weight when senior ofªcials considered Chinese nuclear potential. But its im-
mediate reception—even whether President Johnson saw it or received a
brieªng on it—remains obscure.75 On April 20, 1964, only a few days after the
report reached the White House, however, National Security Assistant Bundy
and NSC staffer Michael Forrestal told colleagues that they were dissatisªed
with another report from Rostow that summarized Robert Johnson’s thinking
on the implications of a nuclear China. According to JCS Chairman Taylor’s as-
sistant, Col. William Smith, Bundy and Forrestal believed that the paper “de-
fused the issue too much” because a Chinese nuclear capability would “have
74. FRUS, 1964–68, Vol. 30, pp. 39–40; and Johnson “The Chinese Communist Nuclear Capability
and Some ‘Unorthodox’ Approaches to the Problem of Nuclear Proliferation.”
75. Robert Johnson to Assistant Secretary of State William Bundy, “The Secretary’s Speech on the
Far East and the Chicom Nuclear Problem,” September 4, 1964, SN 64–66, DEF 12–1 Chicom. The
copy of Johnson’s April 14 report at the Johnson Library has no markings on it indicating that the
president saw it or that Bundy read it. Telephone conversation with Senior Archivist Regina
Greenwell, Lyndon Johnson Library, May 5, 1999.
far greater political consequences” than Rostow believed. Given Bundy’s incli-
nation toward preventive action against China, he may also have felt that Rob-
ert Johnson’s analysis of military attacks was not positive enough.76
Perhaps to counter Bundy’s apprehensions, and certainly to ensure that the
president saw the PPC’s analysis, at the end of April Rusk sent the president a
highly condensed summary of Robert Johnson’s thinking on the Chinese nu-

clear problem. Besides mentioning the continuing study of the covert option,
the summary included one of the major conclusions of the report on direct ac-
tion: that preventive military action was “undesirable except possibly as part
of general action . . . in response to major ChiCom aggression.” Given Rusk’s
friendly reception of Johnson’s earlier work on China, this action served as a
tacit and high-level endorsement for PPC thinking generally and that conclu-
sion in particular.77
While U.S. discussion continued, the ªrst eight months of 1964 saw a steady
march toward production of China’s ªrst atomic bomb.78 Even though faulty
intelligence data led Robert Johnson to conclude that the Lanzhou facility was
“incomplete and possibly incompleteable,” by mid-January it had produced its
ªrst satisfactory highly enriched uranium and then began initial operations.
Moreover, in April, the Jiuquan complex produced the ªrst nuclear compo-
nents for the bomb. In June, the Ninth Academy conducted a successful full-
scale simulation blast test. On August 19, workers at Jiuquan assembled
China’s ªrst nuclear test explosive, including the nuclear core, to verify its
design.79
A New Estimate on Chinese Progress
To get a ªx on China’s nuclear progress, the U.S. intelligence establishment

worked overtime to penetrate the ring of secrecy surrounding the effort. Some
clues came from clandestine sources. Early in 1964, Robert Johnson and other
76. Memorandum by William Y. Smith, “Daily White House Staff Meeting,” April 20, 1964, in
FRUS, 1964–68, Vol. 30, pp. 56–57.
77. Memorandum for the president, “Items for Evening Reading,” May 1, 1964, RG 59, President’s
Evening Reading Reports, box 1, President’s Evening Reading Items, 1964.
78. Enough interest in covert or military action against China’s nuclear facilities continued for
Johnson to recapitulate for a PPC discussion the arguments in his top secret study. See Johnson,
“The Chinese Communist Nuclear Capability and Some ‘Unorthodox’ Approaches to the Problem
of Nuclear Proliferation.”
79. Lewis and Xue, China Builds the Bomb, pp. 135–136, 167–169; Norris, Burrows, and Fieldhouse,
Nuclear Weapons Databook, p. 333; and Johnson, “The Chinese Communist Nuclear Capability and
Some ‘Unorthodox’ Approaches to the Problem of Nuclear Proliferation.”
ofªcials at the State Department read CIA reports stating that Chinese ofªcials
had said that the ªrst test would “deªnitely” occur in 1964. Former INR China
specialist Allen Whiting recalls reading agent reports on Premier Zhou Enlai’s
visit to Mali. According to one report, Zhou told Premier Mobido Keita, who
was very close to Beijing, that China would conduct an atomic test in October.
Nevertheless, no one yet regarded such reports as decisive; thus Robert John-

son wrote that “we really don’t know when the ªrst detonation would
occur.”80
Overhead reconnaissance efforts were particularly important. Lop Nur had
been added to the list of U-2 targets in the spring of 1964, when the planes ºew
two or three missions out of Charbatia in eastern India.81 Nine successful or
partially successful CORONA missions took place between the publication of
the July 1963 estimate and August 25, 1964. In addition, ten GAMBIT satellites
carrying high-resolution cameras, which could distinguish objects about
eighteen inches apart, had been launched beginning in July 1963.82
The intelligence establishment collected data, but as DCI McCone told Presi-
dent Johnson on July 24, he could not “foretell when the Chinese would ex-
plode a device.” Within a few days, however, CORONA photography of Lop
Nur would make more accurate predictions possible. A SNIE published on
August 26, 1964, reported that early-August CORONA photography of Lop
Nur showed a tower and instrumentation sites. Thus the analysts concluded
that “the previously suspect facility at Lop Nur is a nuclear test site which
could be ready for use in two months.”83
CIA analysts believed, however, “that [the detonation] will not occur until
sometime after the end of 1964.” That conclusion was driven by another: that
China “will not have sufªcient ªssionable material for a test of a nuclear de-
vice in the next few months.” The conviction that there was insufªcient mate-
rial resulted from the continued belief that China’s ªrst bomb would be fueled
by plutonium, not uranium (the Lanzhou plant that had already produced the
80. Robert Johnson to Walt Rostow, “Possible APAG Discussion of the ChiCom Nuclear Problem,”
March 9, 1964, PPCR, box 265, Chron File, R. Johnson, January–June 1964; and interview with Al-
len S. Whiting, Crystal City, Virginia, December 13, 1996.
81. Pocock, Dragon Lady, p. 98; and Wheelon interview.
82. McDonald, “CORONA: Success for Space Reconnaissance,” p. 716; and Jeffrey T. Richelson,
America’s Secret Eyes in Space: The U.S. Keyhole Spy Satellite Program (New York: Harper and Row,
1990), p. 358.
83. McCone, memorandum for the record, July 24, 1964; and DCI, SNIE 13–4–64, “The Chances
of an Imminent Communist Chinese Nuclear Explosion,” August 26, 1964, in Kevin Ruffner, ed.,
CORONA: America’s First Satellite Program (Washington, D.C.: CIA, 1995), pp. 237–245, at p. 239.
required U-235 was described as “behind schedule”), and that only one pluto-
nium reactor—the one believed to be at Baotou—could not produce enough
plutonium for a bomb until at least 1965.84
The intelligence analysts believed that even if there were no major obstacles,
it would take at least eighteen, and more likely twenty-four, months after the
startup of the Baotou reactor before a nuclear device would be ready for test-

ing. The earliest date that the Chinese could test, given such assumptions,
would be mid-1965.85
The estimators went on to raise the possibility that China might have an-
other source of ªssionable material. One possible source would be from a facil-
ity started with Soviet help, prior to the withdrawal of Soviet assistance, at
about the same time as work on the Lanzhou gaseous diffusion facility began.
So far, overhead photography had not identiªed it.86
Intelligence analysts also raised the possibility that China might have ac-
quired ªssionable material from a non-Soviet foreign source.87 The version of
the estimate released in 1995 contains a sentence noting the possibility, with
the remaining portion of the paragraph (more than seven lines) blacked out.
Logic and other documents suggest, however, that the only possible foreign
source was France. An August 15, 1963, State Department cable referred to in-
dications of “French-Soviet and French-Chinese cooperation in the atomic en-
ergy ªeld prior to the withdrawal of Soviet technicians from Communist
China.” It also noted a continuing personal relationship between the high com-
missioner of the French Atomic Energy Agency and several members of the
PRC’s Institute of Atomic Energy.88
The analysts were also unsure what the test site activity signiªed. On the one
hand, they noted that it was incongruous to bring the site to a state of readi-
ness without having a device nearly ready for testing—that it would be techni-
cally undesirable to install much of the instrumentation more than a few weeks
before the actual test. On the other hand, analysts also noted that it would not
84. DCI, SNIE 13–4–64, “The Chances of an Imminent Communist Chinese Nuclear Explosion.”
85. Ibid., p. 241.
86. Ibid., pp. 241–242.
87. Ibid., pp. 242–243.
88. Department of State airgram CA-1908, “French-Soviet and French-Chinese Cooperation in the
Atomic Energy Field,” August 15, 1963, RG 59, Subject-Numeric Files, 1963, AE 11–1 FR-USSR. The
issue was of enough concern four years later that the chef du cabinet in the Ministry of Science,
Space, and Atomic Affairs found it necessary to assure U.S. diplomats that there was “zero cooper-
ation between the French and Chicoms on nuclear matters, military or nonmilitary.” U.S. embassy,
Paris, airgram A-1641, “Comment of French Ofªcial and Rumors of Franco-Chinese Nuclear Coop-
eration,” April 18, 1967, RG 59, Subject-Numeric Files, 1967–69, Def 12-1 Chicom.
be surprising if there was uneven progress in various phases of the Chinese

program. In addition, given Lop Nur’s remote location and the poor transpor-
tation available, China might take a long lead time in preparing the installa-
tion. On balance then, the estimators believed that the detonation would not
occur until at least early 1965.89
Such conclusions were disputed both within and outside the CIA. Two

prominent nuclear advisers, Albert and Richard Latter, told the CIA deputy di-
rector for science and technology, Dr. Albert Wheelon, that the CIA’s Ofªce of
Scientiªc Intelligence, which had responsibility for studying foreign nuclear
programs, was “screwing up,” in assuming that a ªrst bomb would rely on
plutonium. Wheelon took the Latters to see DCI McCone. Moreover, Allen
Whiting argued that a test was imminent. He doubted that the Chinese would
put up the tower at Lop Nur that showed up in CORONA imagery unless they
were planning a test. The agent reports of Zhou’s statements about a nuclear
test in October further convinced Whiting that the CIA estimates were too
cautious.90
Final Policy Choices, Second Intelligence Thoughts
With their estimate under scrutiny, CIA analysts began to restudy the data. In
the meantime, some U.S. ofªcials were thinking about military options or at
least threatening to use force. On September 4, 1964, Assistant Secretary of
State Bundy suggested to his staff the possibility that a speech by Rusk could
include a suggestion that Washington might take preventive action against
Chinese nuclear facilities. Bundy’s proposal quickly produced opposition from
Robert Johnson, because any advance warning could help the Chinese foil an
attack, and because it would have a negative political impact internationally,
by stirring fears of war while providing Beijing with justiªcation for its nuclear
weapons program.91
How William Bundy responded to Johnson’s advice is unknown, but the
seemingly imminent Chinese test made the question of preventive action ripe
for a presidential decision. The Chinese “nuclear danger” had been an agenda
89. DCI, SNIE 13–4–64, “The Chances of an Imminent Communist Chinese Nuclear Explosion,”
pp. 243–244.
90. Wheelon interview; and Whiting interview.
91. Robert Johnson to Assistant Secretary of State William Bundy, “The Secretary’s Speech on the
Far East and the Chicom Nuclear Problem,” September 4, 1964, DEF 12-1 Chicom.
item at several of President Johnson’s Tuesday lunches, a gathering of his top

national security ofªcials, during the summer, but critically important deci-
sions were made on September 15 when top advisers—McCone, McNamara,
Rusk, and McGeorge Bundy—met at the State Department. They decided that
it would be better to let the Chinese test occur than to take “unprovoked uni-
lateral U.S. military action.” Attacks on Chinese nuclear facilities would be

possible only in the event of “military hostilities.” Although cautious on uni-
lateral action, the advisers had enough concern about Beijing’s nuclear prog-
ress to consider the possibility of joint steps with the Soviets, such as a
“warning . . . against tests” or “even a possible agreement to cooperate in pre-
ventive military action.” Whether anyone at the table expected the Soviets to
be any more receptive is unknown. In any event, Rusk was to make early con-
tact with Ambassador Dobrynin. Further, to acquire more information, the ad-
visers proposed another U-2 ºight out of Taiwan over the test site. Later that
day, President Johnson approved these recommendations.92
Only Soviet archives can conªrm if Rusk met with Dobrynin to discuss a
joint approach; if they did, no U.S. records of the talk have survived. But on
September 25, McGeorge Bundy attempted to sound out the ambassador. A
statement that Khrushchev had made on September 15, the same day that
Johnson and his advisers discussed the Chinese nuclear problem, may have
encouraged Bundy to believe that Moscow might be in the mood to consider
joint action. Responding to Mao’s hostile comments on Soviet border rights in
the Far East, Khrushchev warned that the Soviets would use all “means at their
disposal” to protect the borders, including “up-to-date weapons of annihila-
tion.” The Soviets had made their ªrst nuclear threat against China.93
92. Agendas for Tuesday lunch, July 28, 1964, and August 19, 1964, Lyndon B. Johnson Library,
McGeorge Bundy papers, box 18/19, ªle Luncheon with President, Vol. 1 (1); memorandum for the
record, September 15, 1964, FRUS, 1964–68, Vol. 30, pp. 94–95; a copy of same with fewer excisions
appears on the National Security Archive’s web site. See also the discussion by Chang, Friends and
Enemies, p. 250; and John Lewis Gaddis, Strategies of Containment: A Critical Appraisal of Postwar
American National Security Policy (New York: Oxford University Press, 1982), pp. 210–211.
93. In Deborah Shapley, Promise and Power: The Life and Times of Robert McNamara (Boston: Little,
Brown, 1993), p. 388, Shapley claims that Rusk did not meet with Dobrynin. However, Rusk’s ap-
pointment calendar shows two meetings with the ambassador scheduled during the two weeks af-
ter the White House meeting on September 15; telephone conversation with Senior Archivist
Regina Greenwell, Lyndon Johnson Library, April 17, 1997. For Khrushchev’s threat, see Harry
Gelman, The Soviet Far East Buildup and Soviet Risk-Taking against China (Santa Monica, Calif.: Rand
Corporation, 1982), p. 17.
However Khrushchev felt, Dobrynin was not interested in talking with

Bundy about any anti-Chinese initiatives. Just as in May 1963, Bundy proposed
a “private and serious talk about what to do about this problem.” In response,
Dobrynin admitted the “depth and strength” of the Sino-Soviet split, which he
blamed on Mao’s “personal megalomania,” but he implicitly took a Chinese
nuclear capability “for granted.” He argued that Chinese nuclear weapons had

“no importance against the Soviet Union or against the U.S.” A Chinese test
would have a “psychological impact” in Asia, but that was of “no importance
for his government.”94
The Soviet Union’s negative response effectively settled the argument over
direct action. Whether or not President Johnson was acquainted with any of
the details of previous discussions of direct action, his disapproval of “unpro-
voked” unilateral action suggested that he had come around to a substantially
less alarmist view than the one taken by President Kennedy. Whether White
House ofªcials still found the covert option attractive is unknown: When CIA
Deputy Director for Plans Richard Helms raised the question of action against
Beijing’s nuclear facilities, White House staff told him “to keep his mouth
shut.”95
The election, only weeks away, undoubtedly had some impact on Johnson’s
thinking about the Chinese nuclear problem. In the heat of the campaign, with
Johnson running on a “peace platform” against the hawkish Goldwater, the
last thing he wanted to contemplate was any military action against China,
with all of the risks that involved. Whether election concerns were a bottom-
line consideration, however, is an imponderable. After all, in spite of the forth-
coming election, he authorized feelers to Dobrynin; of course, Johnson may
well have suspected that the Soviets would not be responsive.
Johnson’s determination to avoid confrontation with China, made evident
by his Vietnam policy, very likely shaped his stance on preemption. Although
he worried that inaction on Vietnam would beneªt China, Johnson wanted to
avoid military measures that could provoke a wider war. That determination
did not mean that senior advisers would not consider action against Chinese
nuclear facilities as part of a Vietnam strategy. A few months later, McNamara
told the JCS that China’s nuclear capability was “a greater threat over a long
time period” than North Vietnam; that made it worth targeting PRC nuclear
94. Memcon, September 25, 1964, FRUS, 1964–68, Vol. 30, pp. 104–105.
95. For “keep your mouth shut,” see memo from Robert W. Komer, NSC, to Bundy, September 18,
1964, FRUS, 1964–68, Vol. 30, p. 99.
facilities if the United States began heavy bombing of North Vietnam. Johnson,
however, would hold the line; when he escalated the war in 1965, he purpose-
fully avoided action that could trigger conºict with China.96
Around the time of the Bundy-Dobrynin meeting on September 25, the U.S.
intelligence establishment, probably drawing on new satellite photography ob-
tained in late August and mid-September, had decided that the preparations at

the Lop Nur site were basically complete. Also suggestive was an agent report
from a member of a Malian government delegation that had recently visited
China; it stated that the Chinese had scheduled a test for October 1, China’s na-
tional day. Moreover, on September 11 Dobrynin had told Ambassador-at-
Large Llewellyn Thompson that Beijing could test “any time now.”97
By this time, State Department ofªcials, prompted by ACDA Deputy Direc-
tor Adrian Fisher, were giving thought to an anticipatory statement about an
impending Chinese test. As former INR ofªcial Allen Whiting later recalled,
such publicity would allow Washington to upstage the Chinese, lessen the
test’s political impact, and “reassure neighboring countries that the U.S. was
watching and aware.” As part of that process, DCI McCone, one step ahead of
his analysts, told NATO’s North Atlantic Council in mid-September that the
Chinese would test within sixty days.98
By late September, the White House and the State Department were ready to
make an announcement. After Whiting indirectly leaked word of a Chinese
test to CBS News, on September 29 reporters queried State Department
spokesman Robert McCloskey about the accuracy of television reports of an
96. For McNamara’s statement, see memorandum for the record, “Summary of Meeting of Joint
Chiefs with Secretary of Defense Following Regular Secretary of Defense Meeting 9:30 a.m., No-
vember 2, 1964, re. Situation in South Vietnam,” Wallace Green Papers, Marine Corps Historical
Archives. The McNamara quote is also cited in David E. Kaiser, American Tragedy: Kennedy, Johnson,
and the Origins of the Vietnam War (Cambridge, Mass.: Harvard University Press, 2000), p. 254. For
Johnson’s emphasis on avoiding conºict with China over Vietnam, see Larry Berman, Planning a
Tragedy: The Americanization of the War in Vietnam (New York: W.W. Norton, 1982), pp. 125, 142–143.
97. Memorandum for the Record, October 16, 1964, FRUS, 1964–68, Vol. 30, p. 109; memcon, “Mis-
cellaneous Matters,” September 11, 1964, RG 59, RAALLT, box 21, Chron File, July 1964. For the
Malian delegation and the reports of an October 1 test, see “The Secretary’s Staff Meeting,” Octo-
ber 28 and November 27, 1964, RG 59, Executive Secretariat Brieªng Books, box 1, Secretary’s Staff
Meeting, August–December 1964. A Malian delegation visited China on June 19–July 4 and met
with Mao and Zhou, among other senior ofªcials. See U.S. Foreign Broadcast Information Service,
Daily Report–Far East, Nos. 120 (June 19, 1964) through 132 (July 8, 1964).
98. Memorandum for the record, October 16, 1964, FRUS, 1964–68, Vol. 30, p. 109; Journals of Glenn
Seaborg, Vols. 7–9 (Berkeley, Calif.: Lawrence Berkeley Laboratory, 1989), p. 254; and Robert John-
son to Assistant Secretary of State William Bundy, “The Secretary’s Speech on the Far East and the
ChiCom Nuclear Problem,” September 4, 1964, SN 64–66, Def 12–1 Chicom; and Whiting inter-
view.
impending test. With President Johnson’s consent, Secretary of State Rusk had
already approved a statement that McCloskey read to the press. Noting that a
Chinese nuclear explosion “might occur in the near future,” McCloskey
stated—for background only and not for attribution—that “from a variety of
sources, we know that it is quite possible that [an] explosion could occur at any
time.” Downplaying the event’s immediate signiªcance, he observed that the

Chinese were a “long way” from having nuclear delivery systems.99
Two weeks later, on October 15, Donald Chamberlain, the CIA’s assistant di-
rector for scientiªc intelligence, informed Deputy Director of Central Intelli-
gence Marshall Carter that the most recent information had conªrmed that
Lop Nur was probably ready to host an atomic test. Beyond describing speciªc
items that turned up in overhead photography—including a 340-foot tower
surrounded by a double fence, arrays for instrument emplacement, two small
towers, and various bunkers and platforms—Chamberlain observed that the
“high priority given to the completion of site construction suggests that a test
is scheduled in the fairly near future.” He also noted that the high level of
ºight activity to and from the area halted in September 1963, when the site was
essentially complete, but had resumed in late September 1964, possibly
reºecting ªnal preparations.100
According to CIA scientists, a restudy of the Baotou reactor site indicated
that adequate primary and backup electric power circuits for reactor operation
had been installed by March 1963, resulting in a reduction in the conªdence
about the August 1964 judgment that the reactor did not begin operation until
early 1964. In addition, the scientists wrote that “we no longer believe that evi-
dence on plutonium availability justiªes the on-balance judgment reached in
August 1964. We believe the Lop Nur evidence indicates that a test could occur
99. Whiting interview; PPC Director Walt Rostow to the Secretary, “The Handling of a Possible
Chinese Communist Nuclear Test,” September 26, 1964, and Henry Owen, PPC, to the Secretary,
“The Handling of a Possible Chinese Communist Nuclear Test,” September 28, 1964, both in PPCR,
box 265, RJ Chron File, July–December 1964; President Lyndon Johnson telephone conversation
no. 5,890, with National Security Assistant McGeorge Bundy, September 26, 1964, tape WH6409.15,
LBJ Library. For Johnson’s role approving the announcement, see memcon, October 27, 1964, RG
59, Executive Secretariat, Conference Files, 1949–72, box 359, Visit of Patrick Gordon Walker 10/
26–27/64, Administration, Sub Misc, and Memcons; and “Transcript of Daily Press Brieªng, Tues-
day, September 29, 1964,” RG 59, Records of Special Assistant to Undersecretary for Political Af-
fairs, 1963–65, box 2, Psychological Preparations of Chinese Test, October 16, 1964.
100. Assistant Director for Scientiªc Intelligence Donald Chamberlain to Deputy DCI Marshall
Carter, “Estimated Imminence of a Chinese Nuclear Test,” October 15, 1964, FRUS, 1964–68,
Vol. 30, pp. 107–108.
at any time.” But they hedged their bets by concluding that “we believe a test
will occur sometime within the next six to eight months.”101
Detonation and Aftermath
The next six to eight months included, of course, the very next day. On October

16, 1964, Beijing announced the detonation of its ªrst atomic device; U.S. nu-
clear intelligence platforms had already picked up its acoustic and electromag-
netic signals. Eleven of thirteen U.S. stations with the capability to detect the
detonation’s electromagnetic pulse did so. Of particular importance were the
TOE DANCER ºights, which detected and collected nuclear debris from the
test. The ºights involved a variety of Air Weather Service aircraft loaded with
special collection equipment. Debris was picked up on more than thirty indi-
vidual sorties by aircraft ºying from Yokota Air Base in Japan.102
Once the Atomic Energy Commission (AEC) began to examine debris from
the radioactive cloud, however, analysts had to abandon their assumption that
the device employed plutonium. On October 20, the same day that a CORONA
satellite snapped a picture of ground zero that showed clear signs of the deto-
nation, AEC Chairman Glenn Seaborg told a cabinet meeting that, to the com-
mission’s surprise, radiological analysis had shown that the bomb contained
U-235. Seaborg told McCone the next day that the Chinese bomb was “a ura-
nium-235 device”—that is, an implosion design using uranium instead of plu-
tonium. U.S. ofªcials quickly saw this as “quite an accomplishment” that, as
William Bundy put it, “probably advanced the date that Chicoms would be [a]
full nuclear power.”103
Where the highly enriched uranium came from, neither McCone nor
Seaborg knew. They recognized that Lanzhou was one source, but statements
that Khrushchev and Deputy Premier Anastas Mikoyan had made, some days
before Khrushchev was deposed, added to the confusion. In conversations
101. Ibid.
102. The History of the Air Force Technical Application Center, 1 July–31 December 1964, n.d., pp. 11, 25,
National Security Archive.
103. Journals of Glenn Seaborg, Vol. 9, pp. 254, 261; V. Gupta and D. Rich, “Locating the Detonation
Point of China’s First Nuclear Explosive Test on 16 October 1964,” International Journal of Remote
Sensing, Vol. 17, No. 10 (1996), pp. 1969–1974; and circular cable 727, October 23, 1964, SN 1964 64–
66, DEF 12-1 Chicom. See also Glenn T. Seaborg with Benjamin S. Loeb, Stemming the Tide: Arms
Control in the Johnson Years (Lexington, Mass.: Lexington Books, 1986), p. 116. Allen Whiting later
recalled that he was “so happy that [he] danced a jig in the ofªce.” Whiting interview.
with foreign visitors, Khrushchev and Mikoyan had observed that the Chinese
were behind schedule even though the Soviets had provided the information
and materials necessary for a test. These statements motivated Rusk to protest
to Dobrynin that this information was inconsistent with Soviet claims that as-
sistance to China had been strictly for “peaceful uses.” But later reports sug-
gested that Soviet assistance had not been enough and that the Chinese may

have found a new way to separate uranium. In December 1964, a U-2 ºight
ended some of the uncertainty; infrared detection systems conªrmed that
Lanzhou was indeed active.104
On the day of the test, the PRC launched a propaganda campaign, stressing
three points: China’s purpose in developing nuclear weapons was “to break
the superpower monopoly,” China would never be the ªrst to use nuclear
weapons, and all nuclear weapons should be eliminated. In effect, the Chinese
were conªrming the State Department’s analysis that they would be a cautious
nuclear power.105
In an attempt to neutralize any political fallout, hours after the detonation
President Johnson issued a reassuring statement, based on a draft prepared
long in advance. Johnson emphasized “Free World nuclear strength,” reaf-
ªrmed “defense commitments to Asia,” and downplayed the test’s signiª-
cance or any imminent threat. Perhaps because of the U.S. government’s ear-
lier efforts to minimize the test’s importance, the world reaction was like John-
son’s, “relatively restrained.” “There was no panic”; rather in some Asian
countries with signiªcant Chinese populations, there was “pride in an ‘Asian
explosion.’” Beijing’s scientiªc accomplishment had raised China’s prestige.106
104. Journals of Glenn Seaborg, Vol. 9, p. 261; memcon, “Soviet Leadership Change: Chinese Com-
munist Nuclear Explosion,” October 23, 1964, RG 59, Secretary’s Memoranda of Conversation,
1953–64, box 30, SecMemcon, October 1964; memcon with Soviet Ambassador Anatoly Dobrynin,
November 19, 1964, ibid., SecMemcon, November 1964; “Secretary’s Staff Meeting Minutes,” De-
cember 21, 1964, RG 59, Executive Secretariat Brieªng Books, box 1, Secretary’s Staff Meetings,
August–December 1964; Westerªeld, Inside CIA’s Private World, p. 246; R.E. Lawrence and Harry
W. Woo, “Infrared Imagery in Overhead Reconnaissance,” Studies in Intelligence, Vol. 11, No. 2
(Summer 1967), pp. 17–40, at p. 23; and Pocock, Dragon Lady, pp. 100–115.
105. Defense Intelligence Agency, China: Views on Nuclear Arms Control, January 1987, p. 7.
106. Public Papers of the President, Lyndon B. Johnson, 1963–64, Book 2 (Washington, D.C.: U.S.
GPO, 1965), p. 1357; State Department Executive Secretary Benjamin Read to National Security As-
sistant McGeorge Bundy, “Standby Statement for Chinese Communist Nuclear Test,” September
30, 1964, RG 59, RAALLT, box 26, China Nuclear Capacity; Background Paper, “Effect of Chicom
Nuclear Explosion,” December 8, 1964, RG 59, Records of Special Assistant to Undersecretary for
Political Affairs, 1963–65, box 2, Comments on Chinese Test; and Lindsay Grant, Bureau of Far
Eastern Affairs, to Assistant Secretary of State McGeorge Bundy, “Policy Implications of Far East-
ern Reactions to CCNE,” October 23, 1964, RG 59, SN 1964–66, DEF 12–1 Chicom.
The reaction to the test was far from calm in Taiwan, however. There “pride”
could also be found, but the test shocked Chiang Kai-shek; his reaction was
“convulsive.” Chiang had played down U.S. as well as his own intelligence
service’s estimates of an early test, and the explosion represented a “severe
blow” to his hopes for “Mainland recovery.” Seeing his regime as Beijing’s
chief target, Chiang demanded military action against mainland nuclear facili-

ties before China had a capability to deliver weapons. Rusk would try to as-
sure the Nationalists that Beijing would “pursue a cautious policy” to avoid
destruction of its nuclear facilities, but that did not discourage Chiang or his
spouse from repeating demands for action in the months that followed. John-
son, Rusk, and McNamara, however, were much more interested in ensuring
that the escalating Vietnam War did not involve direct conºict with China.
They were content to make the usual assurances that U.S. nuclear forces were
sufªcient to deter Beijing.107
Even if reaction from much of the world was “restrained,” the Chinese nu-
clear test quickly raised concerns about a possible “chain reaction” of nuclear
proliferation in Asia. Intelligence reports from the capitals of two key coun-
tries, Japan and India, suggested that heads of state there were considering the
possibility of developing national nuclear weapons programs, even though
antinuclear public opinion made early action unlikely. Even before such re-
ports were available, President Johnson decided to appoint a panel of “wise
men”—former senior government ofªcials—to make recommendations on
“means to prevent the spread of nuclear weapons.”108
The panel, headed by former Deputy Secretary of Defense Roswell Gilpatric,
scrutinized a variety of strategies to curb proliferation, one of which included
the possible use of force against China’s nuclear facilities. Action against China
would have been part of a radical nonproliferation policy that would have in-
cluded a no-ªrst-use policy, substantial withdrawal of tactical nuclear weap-
107. Grant to Bundy, “Policy Implications of Far Eastern Reactions to CCNE”; U.S. embassy, Tai-
pei, cables 324, 334, 339, and 363, October 17, 20, 23, and 29, respectively, RG 59, SN 1964–66, DEF
12–1 Chicom; and memcon, “Chinese Communist Nuclear Capability,” December 3, 1964, SN
1964–66, UN 6 Chicom. For the conversation with Madame Chiang Kai-shek, see memcon,
“China,” FRUS 1964–68, Vol. 30, pp. 207–208. For Chiang’s disregard of warnings from his intelli-
gence service, see U.S. embassy, Taipei, airgram 328, “President Chiang on CCNE’s Impact on GRC
Recovery Plans,” December 15, 1965, SN 1964–66, Pol 1 Chinat.
108. Robert Rochlin, ACDA, to Raymond Garthoff, Committee on Nuclear Proliferation, “Com-
ments on Non-Proliferation Paper of December 12, 1964,” December 31, 1964, copy at National Se-
curity Archive; editorial notes, FRUS, 1964–68, Vol. 11, pp. 120–121, 126.
ons from Europe, a blockade of France’s Paciªc nuclear test sites, and a nuclear
strategy of minimum deterrence.109
As part of the review of policy options, George Rathjens, an ACDA ofªcial,
critiqued Robert Johnson’s report on the basis of a worst-case analysis. He ar-
gued that Johnson had underestimated the effects of Chinese nuclear capabili-
ties, asserting, for example, that the United States would be far more

devastated than China by the destruction of two or three of its top cities. He
further claimed that Johnson had given “inadequate weight to the near term
anti-proliferation effects of destroying Chinese nuclear capabilities.”110
Rejecting a radical antiproliferation policy, the Gilpatric Committee tacitly
followed Johnson’s approach by eschewing proposals for attacks on China’s
nuclear facilities. Nevertheless, the committee saw nuclear proliferation as a
“grave threat” and called for a vigorous nonproliferation policy, including the
formulation and passage of an international treaty. Implicitly accepting a nu-
clear China, it suggested a reexamination of U.S. policy toward China and ad-
vised encouraging China to participate in arms control agreements.111
In part because the Gilpatric Committee raised troubling questions about the
MLF, which still enjoyed signiªcant State Department support, President John-
son did not see its report as an opportunity to push for major initiatives on nu-
clear proliferation policy. Moreover, despite his work on a nonproliferation
agreement, Secretary of State Rusk was far from sold on nonproliferation pol-
icy. He even encouraged top secret State-Defense studies on the possibility of a
“nuclear weapons bank” to which India and possibly other countries would
have access in the event of a confrontation with China, giving them an alterna-
tive to their own national nuclear weapons programs. Although the Chinese
nuclear test had encouraged the ªrst high-level policy review of nuclear prolif-
eration, it would take more political pressure before Johnson was willing to
make a signiªcant political investment in a nonproliferation treaty.112
109. For alternative policies considered by the committee, see papers entitled “Problem Areas,”
National Security Archive; FRUS, 1964–68, Vol. 11, p. 165; and Raymond L. Garthoff, “A Comment
on ‘LBJ, China and the Bomb,’” SHAFR Newsletter, Vol. 28, No. 3 (September 1997), pp. 28–29.
110. Rathjens, “Destruction of Chinese Nuclear Weapons Capabilities.”
111. Committee on Nuclear Proliferation, “A Report to the President by the Committee on Nuclear
Proliferation,” January 21, 1965, FRUS, 1964–68, Vol. 11, pp. 173–182.
112. For Johnson’s ambivalence about the Gilpatric report, see Seaborg, Stemming the Tide, pp. 136–
149. For Rusk’s interest in a “weapons bank,” see Ambassador Llewellyn Thompson to Secretary
of State Dean Rusk, “Review of Non-Proliferation Policy,” December 4, 1964, RG 59, Records Re-
lating to Arms Control and Disarmament, 1961–66, box 9, Defense Affairs 18-10 Non-Proliferation,
November–December 1964. For U.S. policy toward the Indian nuclear problem, see George
Perkovich’s pathbreaking study, India’s Nuclear Bomb: The Impact of Global Proliferation (Berkeley:
University of California Press, 1999).
Wariness about Beijing’s nuclear activities guaranteed continued U.S. efforts

to monitor them. Even before the October 16 test, the National Reconnaissance
Ofªce (NRO) accelerated its schedule of satellite launches so that it could pho-
tograph and monitor the progress of the identiªed Chinese nuclear and missile
facilities while trying to discover new sites. Given China’s expanse, this was
a huge task, and a higher priority—the completion of satellite photographic

coverage of the Soviet Union—limited the resources that could be devoted
to China. Indeed, after the CIA published a NIE on the Chinese military, an
analytical postmortem pointed to the problem of “insufªcient information,”
the “insufªcient priority assigned to the analytic effort,” and “some lack of
resources available for that effort.” During the 1960s the CIA and NRO would
develop a variety of systems—including satellites, remotely piloted vehicles,
and emplaced sensors—designed to monitor Chinese nuclear and missile
developments.113
Not long after the Gilpatric report was issued, rumblings of interest in a pre-
emptive approach would be heard among U.S. ofªcials. In the spring of 1965,
U.S. naval intelligence predicted that the Chinese were only a few years away
from a rudimentary submarine-launched ballistic missile (SLBM) capability,
thus giving Beijing the ability to strike U.S. territory—unquestionably Hawaii,
possibly the U.S. West Coast—from a submarine in Paciªc waters. The navy
even suggested that the Chinese might consider initiating a “catalytic war”—
attacking the United States with submarine-launched missiles in such a way as
to make the United States believe the attacker was the Soviet Union and re-
spond accordingly. During a brieªng to State Department ofªcials, Paul Nitze,
now secretary of the navy, and other navy ofªcials proposed sinking a Chinese
missile-launching submarine on its maiden voyage. The navy’s fears were pre-
mature, however; the Chinese were trying to develop an SLBM, but they
would not even test-launch one from a submarine until the early 1980s.114
113. Committee on Overhead Reconnaissance, “Photographic Coverage Contemplated during the

Next Six Months of Signiªcant Chicom Targets,” October 2, 1964, copy at National Security Ar-
chive; and Department of State airgram CA-13358 to U.S. embassy, Bangkok et al., “Post-Mortem
on NIE 13-3-65, ‘Communist China’s Military Establishment,’” June 11, 1965, SN 1964–66, INT 2-2.
See Jeffrey T. Richelson, The Wizards of Langley: The CIA’s Directorate of Science and Technology (Boul-
der, Colo.: Westview, forthcoming).
114. Garthoff, “A Comment on the Discussion of ‘LBJ, China, and the Bomb,” p. 30; PPC ofªcials
Richard G. Golbert and W.E. Gathright to PPC Director Walt Rostow, “The ChiCom ‘G’ Class (Mis-
sile-Launching) Submarine,” May 4, 1965, RG 59, Records of Policy Planning Staff, 1965–69, Sub-
ject and Country Files, 1965–69, box 328, miscellaneous folder. Using a confrontation with China
over Vietnam to bomb nuclear facilities would also continue to interest senior ofªcials. See Ambas-
China’s nuclear test did not generate rapid changes in U.S. China policy, but it
created pressures for change. Indeed, shortly after the test, U.S. diplomats real-
ized that international support was “eroding” for their policy against admit-
ting the PRC into the UN. But Washington would continue to hold the line. For
Rusk, admitting China could only encourage Beijing to “continue the Commu-
nist push into Southeast Asia.” To contain Chinese inºuence, the administra-

tion waged war in Vietnam even though Vietnamese nationalists were deter-
mined to preserve their independence from the giant to their north.115
Conclusion
While carefully monitoring the situation, President Johnson and his advisers
avoided military confrontation with China during 1964–68. Instead, they fo-
cused on U.S. and allied nation trade controls to ensure that high-tech products
did not reach Beijing’s nuclear program. Moreover, ofªcials in the Pentagon
and the State Department would justify antiballistic missile programs by
pointing to a Chinese intercontinental ballistic missile threat anticipated to ma-
terialize during the 1970s. Although that is another story, defense against a
Chinese missile threat would be a signiªcant thread in U.S. ABM
policymaking through the early Nixon administration.
The reliance on export controls, military containment, and continued intelli-
gence monitoring was a long way from Kennedy’s fancies about “anonymous”
planes striking at Chinese nuclear facilities. Whether Kennedy would have
taken a cautious approach or pushed for military action, with all of its dangers,
will never be known. He had made his pronouncements about a “menace” and
an “intolerable” Chinese nuclear threat in the absence of systematic analytical
work on the implications of the Chinese nuclear capability and the pros and
cons of an attack. Kennedy never saw Robert Johnson’s 1963 study of a nuclear
China and was dead by the time that Johnson had completed his cost-beneªt
analysis of military action. Lyndon Johnson, who was rhetorically more cau-
tious about China and Johnson’s analyses (which were available to the White
sador Llewellyn Thompson to Secretary of State Dean Rusk, “China Study,” July 15, 1965, RG 59,
Records of Policy Planning Staff, 1965–69, box 325, miscellaneous folder.
115. U.S. embassy, Paris, POLTO 654 to Secretary of State Dean Rusk, October 28, 1964, SN 64–66,
NATO 8–2; Foot, The Practice of Power, p. 42; and memcon, “Communist China and the UN: South-
east Asia,” November 13, 1964, Secretary of State’s Memoranda of Conversations, 1953–64, box 30,
November 1964. For Rusk, Johnson, and China, see Warren Cohen, Dean Rusk (Totowa, N.J.: Coo-
per Square, 1980), pp. 280–289; and Garson, “Lyndon B. Johnson and the China Enigma,” Journal of
Contemporary History, Vol. 32, No. 1 (January 1997), pp. 63–80.
House) made a strong case against proposals for military action unless Wash-
ington had world opinion on its side or Beijing was menacing its neighbors.
Enlisting world opinion required tacit if not explicit Soviet support. As a
number of historians have pointed out, Kennedy wanted to enroll Moscow in a
political or even military campaign to halt Beijing’s nuclear effort. But the Sovi-
ets rebuffed U.S. overtures in 1963 and 1964. Whatever fears of a nuclear China

Khrushchev may have had, he was perhaps more interested in mending fences
with Beijing and certainly in avoiding a credibility crisis in the world commu-
nist movement if even a word of the U.S.-Soviet talks on China leaked.
President Johnson was troubled enough by the implications of a nuclear
China to approve an overture to Moscow, but his rejection of unilateral action
suggested that he saw no “intolerable” threats if Beijing successfully tested a
weapon. If the president had read or remembered the synopses of Robert John-
son’s studies, he could have concluded that Beijing’s cautious military policy
and its weak strategic position made a nuclear-capable China a tolerable incon-
venience. Most important, President Johnson was wholly averse to confronta-
tion with China. To prevent a U.S.-China war and the risks of global conºict,
he abstained from unilateral attacks on Baotou and Lanzhou just as he would
limit escalation against North Vietnam or fend off proposals from Chiang for
operations on the mainland. No doubt Johnson also assumed that the forth-
coming presidential election made inaction on the China nuclear problem even
more desirable. Thus he preserved his freedom of action; Johnson could pur-
sue attacks on nuclear sites if the Chinese military ever acted in ways that
threatened American power in Asia and the Paciªc.
The Chinese nuclear test did not bring the foreign policy or military debacle
that President Kennedy feared, but it may have accelerated Sino-American
rapprochement. Signiªcant shifts in the U.S. stance toward China occurred late
in the Johnson administration, and the next president, Richard Nixon, believed
that China’s nuclear potential made a fresh approach, not preventive action,
mandatory. Nixon, though, was hardly blasé about Mao’s China, as evidenced
by his emphasis on a possible Chinese ICBM threat. Signiªcantly, Nixon would
learn that the Soviets were far more worried than he was about a nuclear
China. In August 1969, with Sino-Soviet forces engaged in border ªghting,
Moscow threatened action against China’s nuclear capability, although
whether with actual intent remains to be seen.116
116. Raymond Garthoff, Detente and Confrontation: American-Soviet Relations from Nixon to Reagan
(Washington, D.C.: Brookings, 1994), pp. 236–237; and Gobarev, “Soviet Policy toward China,”
The history of the Kennedy and Johnson administrations’ decisions and ac-
tions with respect to China’s emerging nuclear capability concerns a key point
in U.S.-China relations, the history of the Cold War, and the effort to curb nu-
clear proliferation. The U.S. intelligence community’s efforts to monitor and
forecast China’s nuclear progress and its possible impact on world affairs rep-
resents an important part of the community’s history. Yet policymakers at

other agencies established the parameters for intelligence work; thus presi-
dents demanded aggressive intelligence collection on China’s nuclear efforts,
and one even backed preventive covert action, while State Department ana-
lysts may have helped discourage senior ofªcials from approving a unilateral
attack. Thus when the intelligence and policymaking stories are intertwined,
one obtains a more comprehensive understanding of the contribution that each
made to U.S. policy on a momentous issue.
The account also represents a case where, despite a leader’s (speciªcally
Mao Zedong’s) cavalier rhetoric about nuclear war, acquisition of a nuclear ca-
pability did not produce an onslaught against his neighbors. Here the estimate,
from CIA and State Department ofªcials, that a nuclear China would not be-
have aggressively proved far more accurate than the worst-case fears of Presi-
dent Kennedy and National Security Assistant Bundy. There would probably
be little dispute that restraint (on both sides) was wise or that an attack on
China’s nuclear facilities would undoubtedly have done far more harm than
good.
Today’s world is different in many ways from that of the early 1960s. U.S. in-
telligence capabilities—both in terms of collection and analytic capabilities—
are much greater. Preventive attacks do not seem to be a particularly viable op-
tion because of the undoubted adverse international reaction in all but the
most dire circumstances, and possibly even then. As a result, missile defense is
today proposed to deal with situations that in the past would have produced
proposals for preventive action. In addition, there is not one but a number of
rogue regimes seeking weapons of mass destruction (WMD)—with Saddam
Hussein’s Iraq at the top of the list.
Yet the history of how the United States dealt with China’s emerging nuclear
capability is not irrelevant today. The Johnson administration’s restrained ap-
proach toward the Chinese nuclear program is signiªcant not because it sug-
p. 47. For the argument that Nixon seriously considered tacit support for Soviet action against Chi-
nese nuclear facilities, see Patrick Tyler, A Great Wall: Six Presidents and China, An Investigative His-
tory (New York: PublicAffairs, 1999), pp. 62–63.
gests that restraint is always the best choice, but because it highlights the
different variables that must be considered in choosing among diplomatic, eco-
nomic, and military (including preventive action and missile defense) options.
Those variables include the limits of intelligence collection and uncertainties in
intelligence analysis concerning the nature and status of a WMD program, the
likely impact a speciªc WMD capability will have on the conduct of a nation

and its leadership, probable overseas reactions to various U.S. decisions, the
possibility of international action, and the deterrent effect of U.S. nuclear capa-
bilities on decisions to develop or employ WMD. The necessity for astute as-
sessment of those variables when making decisions on how to deal with WMD
proliferation will be an enduring challenge for U.S. and international
policymakers.

Whether To "Strangle

Uploaded by

Copyright:

Available Formats

Whether To "Strangle

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Whether To "Strangle

Uploaded by

Copyright:

Available Formats

Whether to “Strangle William Burr and

Less than a year after

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

International Security, Vol. 25, No. 3 (Winter 2000/01), pp. 54–99

national security bureaucracy to explore means of disrupting China’s nuclear

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

China’s Emerging Nuclear Program

Developing a nuclear arsenal requires scientiªc and technical expertise,

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Piercing the Veil

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

and ªssionable materials. The analysts observed that “Chinese development of

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Assessing the Threat and Exploring the Options

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

U.S. government ofªcials also saw developments in China as adding to the

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Rusk did approve another McGhee proposal. In September 1962, he sanc-

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Besides Baotou, U.S. intelligence identiªed several facilities of the Institute

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

While intelligence ofªcers worked on improving their understanding of

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

34. Editorial note, FRUS, 1961–63, Vol. 22, p. 339.

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

If civilian policymakers hoped for unambiguous advice on the prospects

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

conference on August 1, he spoke of a “menacing situation”; acknowledging

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Exploring Options with the Chinese Nationalists

A visit to Washington in September 1963 by Chiang Kai-shek’s son, Gen.

ested in whether something could be planned” that could have a “delaying

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Possibilities and Pitfalls of Direct Action

ofªcials but leading journalists as well—none more so than the well-connected

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

over the suspected plutonium reactor at Baotou, to determine whether it was

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

signiªcant policy changes—ºowed from Johnson’s assumption that a Chinese

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Asian countries.” To back up its assurances, Washington needed an “evident

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

66. Johnson to Rostow, “Direct Action against ChiCom Nuclear Facilities.”

In mid-April 1964, Robert Johnson completed his study “An Exploration of

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

strengthen air defenses to prevent another attack. Third, Chinese retaliation,

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

An earlier conclusion of Robert Johnson’s study—that the Chinese nuclear

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

Downloaded from http://direct.mit.edu/isec/article-pdf/25/3/54/692462/016228800560525.pdf by guest on 20 April 2021

A New Estimate on Chinese Progress