Washington D.C. June 1, 2005 - The failure of the recently
concluded review conference on the Nuclear Non-Proliferation Treaty,
at a time when the future of the non-proliferation system is in
question (Note 1), makes it an opportune time
to look at how the U.S. intelligence establishment analyzed the
proliferation issue during the years before the Treaty was negotiated.
National Intelligence Estimates (NIEs) from the 1960s and earlier
shed light on how U.S. intelligence thought about the problem
of nuclear proliferation, especially which countries had the will
and the capability to produce nuclear weapons.
Beginning in the late 1950s, the intelligence community began
to assess systematically the developing problem of the spread
of nuclear weapons. (Note 2) Responding to policymakers
demands for information and analysis, intelligence analysts began
looking the considerations that shaped decisions to go nuclear
and the range of countries, from Israel to China, which could
initiate independent nuclear weapons programs. Moreover, estimators
looked at the broader implications of a situation where more countries
had nuclear weapons. What kind of a world would it be?
This briefing book shows how analysts addressed those problems
by publishing for the first time the first ten years of NIEs on
nuclear proliferation. Among the findings in the documents:
- the spread of "nuclear know-how" around the world
was creating widespread capabilities for national nuclear weapons
- an NIE from 1963 estimates that a basic capability to build
two nuclear weapons a year was relatively cheap; no more than
$180 million in 1963 dollars (a bit under $1 billion in today's
dollars), although building advanced weapons delivery systems
would greatly add to the expense
- countries sought nuclear capabilities for reasons of "prestige"
and "military effectiveness"; countries like China
sought nuclear weapons as a "deterrent to the use of US
nuclear weapons in the Far East"
- during the late 1950s, analysts saw France, China, West Germany,
Japan, Sweden, and Israel as among the countries with the greatest
potential to develop nuclear weapons, although many of them
faced domestic and international constraints to producing them
- estimates on Israeli potential found, in 1961, "considerable
evidence" that Israel was "developing capabilities";
by 1966, analysts believed that Israel had "imported and
stockpiled sufficient unsafeguarded uranium for a few weapons."
- in the aftermath of the 1964 Chinese test, India appeared
to be a "serious" candidate for nuclear status; if
India went in that direction, analysts saw Pakistan likely to
follow suit, probably with help from China.
- by the late 1950s and early 1960s, analysts worried that the
proliferation of nuclear weapons could "materially increase"
the chance of world war and raise the risk of an "unintentional
or unauthorized detonation of [nuclear] weapons." Also
of concern were the acquisition of nuclear weapons by "irresponsible"
governments and the "risk" of stationing nuclear weapons
- multilateral nonproliferation agreements and nuclear test
bans could restrain national nuclear programs but could not
stop a determined government from initiating a nuclear weapons
- a report from the mid-1970s shows concern about nuclear terrorism
as well as the possibility that "threshold states"
such as Iran, Libya, Taiwan, and South Africa could acquire
Even before the first atomic test at Los Alamos, the U.S. government
worried that other powers might get the bomb. (Note
3) Whether Nazi Germany would produce nuclear weapons before
the United States did was a real, if short-lived, apprehension.
After World War II, when U.S.-Soviet Union tensions became a significant
element of world politics, the prospects and possibility of a
Soviet bomb was unnerving to many, and the U.S. government developed
intelligence systems to track the Soviet project. Through a huge
effort, including espionage, the Soviets acquired the bomb in
1949; the United Kingdom's test in 1952 meant that there were
three nuclear weapons states. Developments during the 1950s would
increase potential for nuclear proliferation: Chinese and French
decisions to begin nuclear weapons programs and President Eisenhower's
"Atoms for Peace" program spreading nuclear power production
capabilities. By the late 1950s, with the prospect for new nuclear
states beginning to crystallize -- the so-called "Fourth"
country problem -- U.S. intelligence faced pressure to look systematically
at the problem and its broader implication. National Intelligence
Estimate 100-6-57, "Nuclear Weapons Production in Fourth
Countries - Likelihood and Consequences," June 18, 1957,
was the first of many estimates, coordinated and published by
the Central Intelligence Agency on a more-or-less annual basis,
to gauge the changing problem of nuclear proliferation.
While the CIA has produced many secret estimates of nuclear proliferation,
it has been reluctant to declassify them. During the 1990s, it
released a few, in whole or in part, for use in the State Department's
Foreign Relations series, but when it came to FOIA or
mandatory review requests, the answer was "no." When
CIA did not deny NIEs altogether it would release them only in
bits and pieces, denying the analysis of specific countries. Recently,
however, the Agency's reviewers took steps toward greater openness
when they responded to FOIA requests and mandatory review appeals
by releasing NIEs from 1958, 1960, and 1967 on nuclear weapons
and missile proliferation. Instead of typically withholding the
names of the countries deemed most capable of producing nuclear
weapons and ballistic missiles, the CIA released them. What influenced
this change in approach is a matter of conjecture, although it
may reflect Agency concerns about recent criticisms of its record
on nuclear proliferation. Whatever motivates these declassification
decisions, the Agency should be applauded for beginning to take
a more forthcoming approach to the NIEs.
Besides publishing the declassified NIEs on nuclear proliferation,
this briefing book includes documents, acquired through archival
research and FOIA requests, that shed light on the context for
the production of estimates of nuclear proliferation. The documents
show how the NIEs were requested, how they were followed up, the
role of embassies in the estimating process, and the contributions
of the State Department's Bureau of Intelligence and Research
(INR) to the NIEs.
Note: The following documents are in PDF format.
You will need to download and install the free Adobe
Acrobat Reader to view.
1: Memorandum from W. Park Armstrong, Special Assistant to
the Secretary of State for Intelligence to Fisher Howe et al.,
"National Intelligence Estimate on the `Fourth Country Problem'",
May 3, 1957, enclosing letter to Director of Central Intelligence
Allen W. Dulles, April 25, 1957
Source: National Archives, Record Group 59, Records of
the Department of State (hereinafter RG 59), Central Decimal Files,
1955-1959, 101.21 NIS/5-357
Growing concern among State Department officials about the impact
of nuclear proliferation on the U.S. position in world affairs
produced a demand for an NIE that would estimate the extent of
the problem as well as the "probable effects of the development
of unconventional capabilities by fourth countries." As the
most senior State Department official with intelligence responsibilities,
it was plainly matter of course for W. Park Armstrong to write
to CIA chief Allen Dulles, ask for an NIE, and to schedule completion
within two months.
2: National Intelligence Estimate 100-6-57, "Nuclear
Weapons Production in Fourth Countries - Likelihood and Consequences,"
June 18, 1957
Source: CIA FOIA release
After denying this NIE when the National Security Archive initially
requested it in the mid-1990s, in response to a second request
the CIA declassified it in full in the fall of 2001. Looking at
capabilities and intentions as well as the consequences of proliferation,
the estimators saw up to ten countries that had the potential
to produce nuclear weapons on a limited basis. One of the main
conclusions, and a pessimistic one at that, was the near certainty
that "over the next decade an increasing number of countries
will obtain possession of nuclear weapons and that effective international
control will be increasingly difficult to achieve." From
current perspectives, some of the analysis seems wrongheaded,
such as the alleged "even" chance for a Japanese bomb,
the likelihood of a Swedish capability, and the discussion of
Canada's potential. Nonetheless, it should be kept in mind that
countries like Japan and Sweden would continue to figure in discussions
of proliferation risk during the l950s and 1960s. Indeed, for
some years the Swedish military was strongly interested in nuclear
weapons and ran a covert weapons research program during the 1950s
and 1960s, despite growing public anti-nuclear sentimen. (Note
4) The estimate also mentioned Australia in passing noting
that it had been "receiving important assistance [from the
United Kingdom] in the development of nuclear energy programs."
The estimators may have been unaware that the Australians had
wanted nuclear reactors so they would have a nuclear weapons option.
The analysis of other countries, such as Israel, France and China,
is more credible in retrospect but also shows the limited knowledge
of national decisionmaking, for example, the estimators did not
know that France had already made the basic decisions on a nuclear
capability. The discussion of West Germany and possible Western
European cooperative arrangements reflects the concerns of those
days but also the conviction that it would be possible to head
off an independent West German nuclear program by giving it access
to the U.S. nuclear stockpile ("nuclear sharing") if
world war broke out. The multilateral force (MLF) proposal of
the early 1960s was another attempt at forestalling a nuclear
West Germany. (Note 6) While the estimators were
relatively sanguine about the implications of wider nuclear capabilities
for the possibility of general nuclear war, they acknowledged
the imponderables, e.g., the acquisition of nuclear weapons by
an "irresponsible" government or "radical"
political changes in the countries that had nuclear potential.
3a and 3b: NIE 100-2-58
Document 3a. National
Intelligence Estimate 100-2-58, "Development of Nuclear Capabilities
by Fourth Countries: Likelihood and Consequences," July 1,
Source: FOIA request by National Security Archive
3b. Annex to National Intelligence Estimate No. 100-2-58,
"Development of Nuclear Capabilities by Fourth Countries:
Likelihood and Consequences," July 1, 1958, Secret, Excised
Source: Dwight D. Eisenhower Library, mandatory review
NIE 100-6-57 may have created a demand for even more analysis
and information because the next estimate of the "fourth
country problem" was twice as long. NIE 100-2-58, recently
released in its entirety by the CIA (which had initially denied
Archive requests for this document), provides more detailed analysis
of capabilities of potential nuclear states, "probable courses
of action" including motives for proliferation, possible
deterrents to proliferation, and the implications of proliferation
for international relations. The blanket certainty of the 1957
estimate about an "increasing number of countries" obtaining
nuclear weapons was replaced by a more nuanced analysis, but without
any generalizations about the most probable course of events.
Nevertheless, the analysts were even more certain that France
was on the road to testing nuclear weapons; also, as before, they
saw Sweden as a strong candidate under certain circumstances.
China was also most likely to have begun a development program
and Israel would follow suit if it had access to the necessary
materials. The analysts paid even more attention to the possibility
of West European cooperation to produce the bomb but were unsure
whether the French-Italy-German (FIG) group would go beyond missile
research into a cooperative program to build the bomb. (Note
The estimators had a far more ominous view of the consequences
of proliferation than they did a year earlier. While the 1957
estimate opined that "fourth power production of nuclear
weapons over the next 10 years is not likely to … materially
increase the likelihood of general war" the 1958 estimate
asserted that nuclear proliferation is "certain to produce
difficulties and in most cases would tend to increase the chances
of general war by an expansion of local conflicts." While
the interested public would have not seen these estimates, the
scholarly community was already raising searching questions about
the risks of nuclear proliferation. Moreover, only weeks after
this NIE was published Irish Foreign Minister Frank Aiken introduced
the first of a series of resolutions-the "Irish Resolutions"--at
the United Nations calling for a nuclear non-proliferation policy.
4a and 4b: Reactions to NIE 100-2-58
Document 4a. Untitled
memorandum from U.S. Embassy, Sweden, 2 September 1958
Source: FOIA request by National Security Archive
4b. State Department instruction, CA-3456, to various embassies,
"Development of Nuclear Weapons
on the European Continent: Request for Information," October
Source: RG 59, Records of Component Offices of the Bureau
of Intelligence and Research, 1947-1963, Lot 61D1597, box 4, OSI
4c. State Department instruction, CA-5044, to various embassies,
"Development of Nuclear Weapons Capabilities by `Fourth Powers,':
Political Aspects," December 3, 1958, enclosing U.S. Intelligence
Board "Post-Mortem on NIE 100-2-58", October 30, 1958
RG 59, Central Decimal Files, 1955-1959, 100.21-NIS/12-958
Some interesting reactions to NIE 100-2-58 have turned up through
archival research and FOIA requests. The economics counselor and
the Army attaché at the embassy in Sweden assessed the
estimate of Swedish nuclear potential and produced highly divergent
reactions, with the counselor arguing that the estimate exaggerated
Swedish potential and the attaché generally agreeing with
the NIE. Apparently the mission chief, Ambassador Francis White,
leaned toward the thinking of the counselor, because the cover
memo included a criticism of the "flat statement" about
the likelihood of a Swedish bomb in the absence of progress on
Plainly the NIE led to demands for more information. In October
the Department of State asked Western European embassies to raise
their antennae for information on the French nuclear program as
well as on European cooperative nuclear weapons projects. Within
that same month, in another effort to encourage information gathering
by the embassies, the State Department sent them a rarely released
"post-mortem" of an NIE by the U.S. Intelligence Board,
the body that vetted the NIEs in their final stages. It shows
that the intelligence establishment felt that it was not getting
the information that it needed for full analysis of the proliferation
problem. For example, it found the sources on French plans to
be not good enough and wanted the CIA-led Guided Missiles Intelligence
Committee to look more closely at the ballistic-missile potential
of those countries with "fairly imminent" nuclear capability.
Moreover, as the State Department underlined, the USIB wanted
more reporting from the "field," from embassy officers
but also from collectors in the military and the CIA.
5: NIE 100-4-60, "Likelihood and Consequences of the
Development of Nuclear Capabilities by Additional Countries",
9 September 1960, Secret, excised copy
Source: Mandatory review request
With the French nuclear test in February 1950, the "Fourth
Country" problem turned into the "Nth" country
problem; no one could foresee how many new nuclear states would
materialize. This NIE, recently released in slightly excised form
after previous denials, takes into account the impact of the French
test on the situation. Still there was much uncertainty about
future prospects. This NIE treated France as the only formerly
non-nuclear state that "is known to have programs underway."
Unlike in 1958, when China's nuclear ambitions were discussed
more speculatively, the analysts believed that "Communist
China almost certainly has started a weapons program." Yet,
they were not absolutely sure. They also assumed that Soviet nuclear
weapons assistance to China was still a possibility when in fact
Moscow had already called off its nuclear assistance program.
In a dissent on page 10, naval intelligence rightly made much
of the withdrawal of Soviet technical assistance in light of "current
dissensions" between Beijing and Moscow. (Note
9) The CIA excised all references to Israel from this NIE;
in contrast to its decisions on NIE 100-2-58, the Agency followed
its more routine procedure on the Israeli nuclear question.
As in 1958, the analysts discussed Sweden, West Germany, Japan,
and a European cooperative effort, seeing all of them as unlikely
prospects for nuclear weapons programs barring changes in the
world situation and domestic policy and politics. Also as in previous
analyses, the estimators discussed the potentially dangerous impact
of the spread of nuclear weapons; this NIE mentioned the peril
of nuclear sharing schemes: "the stationing of nuclear weapons
on foreign soil and training indigenous forces in their use is
not without risk." This was becoming a concern beyond the
intelligence agencies. For example, not long after this NIE appeared,
the Joint Committee on Atomic Energy pointed out the danger that
that "revolutionary coup d'etats by the Communists or rightists
in certain NATO countries" could lead to seizure of U.S.
nuclear weapons and "actual, attempted, or accidental use."
6a and b: NIE 4-3-61
Document 6a. Guided
Missiles and Astronautics Intelligence Committee, "GMAIC
Contribution to NIE 4-3-61," July 20, 1961, secret, excised
Source: CIA Records Search Tool (CREST), National Archives
library, College Park, MD
6b. National Intelligence Estimate, NIE 4-3-61, "Nuclear
Weapons and Delivery Capabilities of Free World Countries Other
than the US and UK," September 21, 1961, Secret, Excised
Source: microfiche supplement to U.S. Department of State,
Foreign Relations of the United States, 1961-1963, Volume
VII, Arms Control and Disarmament (Washington, D.C., U.S.
Government Printing Office, 1997)
The next NIE, published in mid-1961, shows a greatly broadened
scope of inquiry, not only weapons program but also the proliferation
of delivery system capabilities. Unlike the earlier estimates,
this one did not look at the broader international implications
of nuclear spread, limiting itself to the national considerations
that prompted a decision to build the bomb. It did, however, hone
the analysis of the problem. Leaving the problem of China to separate
estimates, and perhaps in response to earlier critiques, the estimators
created categories of "likely" and "unlikely candidates."
Countries like Canada were in the latter category, but Sweden
was still in the "likely group," along with Israel,
France, West Germany, India, and Japan, as well as "Western
European groupings." But the estimators also divided countries
that had made a positive decision, such as France and Israel,
from a somewhat larger number of countries that only had the potential
to develop nuclear weapons. For example, with respect to India,
a Chinese nuclear test "would greatly strengthen the view
… that there is a pressing need for an Indian nuclear capability."
Japan could also go nuclear in the wake of a Chinese test or if
it lost confidence in U.S. security guarantees.
Document 6a, a background studies prepared for NIE 4-3-61 by
the GMAIC, is one of the thousands and thousands of documents
released in scanned images through the CIA Records Search Tool
(CREST), available on computers only at the National Archives
library (College Park, Md) and the Lyndon B. Johnson Library (Austin,
Tx.). The CREST reviewers deleted the list on page 2 naming countries
with the capability to produce ballistic missiles, although more
or less the same information had been released when the NIE was
published four years earlier.
7: Special National Intelligence Estimate, 30-2-63, "The
Advanced Weapons Programs of the UAR and Israel," May 8,
1963, secret, excised copy
Source: CIA FOIA release
The sensitivity of the Israeli nuclear issue and that nation's
commitment to a policy of opacity on its nuclear weapons status
(Note 11) has generally encouraged the U.S.
government to take a highly restrictive approach to declassifying
intelligence analysis on the Israeli program. While the information
in document 6b is an example of uncharacteristic openness on this
issue, this document exemplifies the restrictive policy. The CIA
withheld almost all of it. (Note 12)
8: National Intelligence Estimate Number 4-63, Likelihood
and Consequences of a Proliferation of Nuclear Weapons Systems,
June 28, 1963, secret, excised copy
CIA FOIA appeal release
This NIE has gone through several rounds of declassification
review in recent years (Note 13); this is the
most complete release so far. In an earlier release CIA withheld
the discussion of the major candidates for nuclear proliferation;
now that has been declassified except for the review of Israel's
nuclear potential. The list of eight countries that were under
special scrutiny is essentially the same as in NIE 4-3-61, although
Italy has been added. Also partly declassified for the first time
is the "Special Case," on pages 13-14, on a possible
transfer to third countries, such as Egypt, of nuclear weapons
technology or even a complete weapon.
Showing that NIEs were committee products, paragraph 25 opined
that a nuclear-armed China would not have a significant impact
on Chinese policy while the next paragraph held that the "tone
of Chinese policy would probably become more assertive."
Pointing out the contradiction, the State Department Bureau of
Intelligence and Research dissented from the more foreboding analysis
held elsewhere in the bureaucracy.
Unlike the 1957 analysis, the estimators were more confident
that the nuclear proliferation problem would be limited: "there
will not be widespread proliferation of nuclear weapons over the
next 10 years" even though a growing numbers of countries
would have the capability to join the nuclear club. Nevertheless,
limited or not, the estimators saw a more unstable world. If new
nuclear states became involved in conflicts with non-nuclear or
nuclear-neighbors, the "situation would be potentially more
dangerous … because of the added uncertainties introduce
by [the] presence" of nuclear weapons. In addition, the estimators
believed that as the number of nuclear weapons states increased
"the risk of unintentional or unauthorized detonation of
such weapons will also rise," partly because of the absence
of safety measures.
An annex discusses the costs of starting a small nuclear weapons
program. A "minimal" capability to build two plutonium
implosion weapons a year was relatively cheap; ranging from $140
million (about $670 million in today's dollars) to 180 million
in 1963 dollars (a bit under $1 billion in today's dollars), with
annual operating costs running between $20 and 30 million (between
$100 to 150 million in today's dollars). If, however, a country
wanted modern weapons delivery systems, e.g., ballistic missiles
or bombers, "costs rise steeply." (Note
9: NIE 4-2-64, "Prospects for a Proliferation of Nuclear
Weapons over the Next Decade," October 21, 1964
Source: FOIA appeal release
This NIE, prepared only days after China's first nuclear test,
is heavily redacted with almost half of it withheld in its entirety.
In light of the Chinese test, the estimators were more confident
that India would follow suit: "we believe the chances are
better than even that India will decide to develop nuclear weapons
within the next few years." Instead the Indians would take almost 10 years to make a decision to "develop," or produce and test a nuclear device ("peaceful nuclear explosive") and quite a few more years before they weaponized their nuclear program. (Note 15) The estimators
also saw a greater possibility of an Israeli and a Swedish decision
to begin a nuclear weapons program. The analysis of the implications
of nuclear proliferation was substantially the same as NIE 4-63.
10: Central Intelligence Agency, "Nuclear Weapons Programs
Around the World," 3 December 1964, top secret, excised copy
Source: Mandatory review request by National Security
Archive to Lyndon B. Johnson Library, National Security File,
Committee File, Committee on Nuclear Proliferation, box 6, Prospects
for Proliferation of Nuclear Weapons
Shortly after the first Chinese test, President Lyndon Johnson
commission former Deputy Secretary of Defense Roswell Gilpatric
to head a panel to investigate the nuclear proliferation problem.
The Gilpatric report turned out to be a milestone in U.S. nuclear
proliferation policy, although its import remains a matter of
some controversy. (Note 16)
The Gilpatric panel asked military and intelligence agencies
to prepare background reports on the problem and this is one of
the CIA responses. Significantly, it provides details on the state
of nuclear weapons programs that CIA withheld from the NIE of
six weeks earlier. What explains this inconsistency is unknown;
perhaps, declassification guidelines for the NIEs are more stringent
because of their inter-agency status. Like NIE 4-3-61, this estimate
analyzed both progress on nuclear weapons and delivery systems,
although the analysis was slightly different than NIE 4-64. While
the latter saw a greater possibility for Israeli and Swedish nuclear
programs, the present study tended to rule them out: "Israel
has probably decided to not build nuclear weapons,' while "the
chances are better than even that Sweden will not initiate a weapons
program during this decade."
11a and b: INR Contributions to NIEs
Document 11a. U.S.
Department of State, Bureau of Intelligence and Research, "
Contribution to NIE 4-1-65," 4 October 1965.
11b. U. S. Department of State, "Addendum INR Contribution
to NIE 4-65 Likelihood of Further Nuclear Proliferation,"
4 November 1965
Source: State Department FOIA release
In the fall of 1965, the intelligence community began work on
another NIE on the proliferation issue; while the NIE was provisionally
called NIE 4-65 it was not ready until 1966 and the number was
changed to 4-66 (see document 12). Some of the proposed State
Department Bureau of Intelligence Research input into the NIE
has been released and it gives some insights into it. Suggesting
that the INR had a broader scope than the previous two, the first
document reviews the likelihood of nuclear weapons decisions by
Japan, Australia, and Indonesia; the last two had not previously
been on the list of candidates. Another document estimated the
possibility of a nuclear weapons decision by Pakistan. Not only
did it see such a decision likely if an Indian nuclear weapons
program emerged. If Pakistan so decided, the estimate astutely
suggested that Pakistan was likely to seek help from China, another
adversary of India. Pakistan would either seek Chinese security
guarantee or more "tangible assistance" in the form
of completed weapons, fissionable material, or technical aid.
12: NIE 4-66, "The Likelihood of Further Nuclear Proliferation,"
January 20, 1966, Secret, excised copy
Lyndon B. Johnson Library, mandatory review release; currently
The Archive has an earlier release of this document under appeal
at the Interagency Security Classification Appeal Panel (ISCAP)
but the CIA recently declassified more information from this NIE,
including the country analysis, in a separate release. Interestingly,
in contrast to the 1963 and 1964 estimates, CIA released some
details on Israel: it had "imported and stockpiled sufficient
unsafeguarded uranium for a few weapons." The excised analysis
probably indicates how U.S. intelligence assessed Israel's production
capabilities from the Dimona reactor as well as the source(s)
of imports. In any event, the analysts concluded that Israel had
not yet produced any weapons but that it would be hard for the
international community to check such a decision if Israel believed
that "the threat from the Arab states could no longer be
contained by conventional means." According to Avner Cohen's
major study, by May 1967, Israel had an operational nuclear weapon.
Besides Israel, the analysts saw India as the only other "serious
[contender] for nuclear status," with the latter "likely
to undertake a nuclear weapons program in the next several years."
In keeping with the 1965 INR contributions, Pakistan was on the
list of potential nuclear states; so were the Australia, Switzerland,
Taiwan, Indonesia, and the UAR. The CIA partly declassified an
interesting analysis of the "Snowball Effect" suggesting
the larger impact of decisions on a nuclear capability by one
or two countries.
13: U.S. Embassy Bonn Airgram A-1512 to Department of State,
"NIE 23-66: West German Capabilities and Intentions to Produce
Nuclear Weapons and Delivery System," 12 April 1966
Source: RG 59, Subject-Numeric Files 1964-1966, INT 2-2
Showing the variety of inputs for NIEs, the U.S. embassy in West
Germany produced this airgram to contribute to an estimate on
West Germany's nuclear weapons potential. Who sought an NIE on
this topic is presently unknown. Over the course of 1966, however,
the United Nation's Eighteen-Nation Disarmament Committee (ENDC)
was holding intensive discussions on a nuclear non-proliferation
treaty of which the West German government was highly critical.
(Note 18) Perhaps someone in the White House
or elsewhere in the federal bureaucracy wanted U.S. intelligence
to look more closely at the West German nuclear problem to see
if there was any risk. Plainly the embassy saw little danger;
it reaffirmed the "judgment expressed in NIE 4-66 - that
West Germany will almost certainly not embark on a program to
develop or acquire her own nuclear weapons." Risks would
materialize, however, if West Germany began to fear for its security.
That would happen only in the event of a "radical change
in the basic structure of the political relationships in the postwar
world," such as a reversal in U.S. security policy toward
Western Europe. An NIE on West German capabilities and intentions
was soon produced although it remains classified.
14a and b: Ballistic Missile Proliferation
Document 14a. NIE
4-67, "Proliferation of Missile Delivery Systems for Nuclear
Weapons," 26 January 1967, secret, excised copy, as released
14b. NIE 4-67, "Proliferation of Missile Delivery Systems
for Nuclear Weapons," 26 January 1967, secret, excised copy,
as released under appeal, March 2005
Source: Lyndon B. Johnson Library, National Security
Files, National Intelligence Estimates, box 1, 4 Arms and Disarmament
During 1967, instead of its usual estimate on the likelihood
of nuclear proliferation, the intelligence community looked at
a related problem--the extent to which new nuclear states would
have the capability to produce missile delivery systems. A year
ago, the National Security Archive submitted a mandatory review
request to the Johnson Library for a copy of this NIE. The initial
request produced a heavily excised version, withholding most of
the analysis (see document 12a). An appeal from the National Security
Archive recently produced a lightly excised version (see document
12b), with most of the discussion of specific countries intact.
Noting that one of the most "difficult and expensive"
aspects of a nuclear weapons project is the development and deployment
of a missile delivery system, the estimate looked closely at six
non-nuclear states that were deemed "serious candidates for
acquiring such systems": Israel, India, the UAR, Sweden,
Japan, and West Germany. Of the six, however, a number of them
faced significant domestic inhibitions and opposition, international
prohibitions, or resource constraints that made them less likely
candidates in the short-run. The discussion of Israel, one of
most motivated countries, has great excisions although a pending
appeal to ISCAP may produce more details.
15: Eight Years Later: New "Threshold States"
Central Intelligence Agency, Directorate of Intelligence, Office
of Political Research, Research Study, "Managing Nuclear
Proliferation: The Politics of Limited Choice," December
1975, Secret, excised copy
Source: CREST, National Archives Library, College Park,
A report prepared by CIA several years later suggests that the
problem of widespread nuclear knowledge raised by the early NIEs
had only grown worse: "the requisite materials and technology
are already too widely available for technical safeguards and
international regulation to be effective." While the author
believed that there was "no hope of preventing nuclear proliferation,"
the study argued that it might be possible to influence the choices
made by would-be nuclear states, e.g., to not go too far down
the road to weaponization. The NIEs from the 1950s and 1960s did
not anticipate the possibility of nuclear terrorism, but that
prospect was under discussion by the mid-1970s: "the same
increasing availability of nuclear materials and technology which
has made nuclear explosives accessible to developing states can
also be expected sooner or later to bring them within the reach
of terrorist groups."
* The author thanks Jeffrey Richelson for helpful comments and
for contributing documents.
1. For assessments of the issues, see Fred Kaplan, "The
Real Nuclear Option," at http://www.slate.com/id/2117940/;
and "The Proliferation Crisis," Washington Post,
4 May 2005, at http://www.washingtonpost.com/wp-dyn/content/article/2005/05/03/AR2005050301388.html;
Alexander Barnes Dyer, “Nuclear (Non-Proliferation Treaty)
Meltdown,” at http://www.slate.com/id/2119831/;
and Lawrence Wittner, “Has the Bush Administration Made
Nuclear Proliferation Inevitable?,” History News Network,
30 May 2005, at http://hnn.us/articles/12185.html.
2. For a history of U.S. nuclear intelligence, see Jeffrey Richelson,
Spying on the Bomb: American Nuclear Intelligence from the
Nazis to Iran and North Korea (New York: W.W. Norton, forthcoming).
3. See Richelson, Spying on the Bomb and Robert S.
Norris, Racing for the Bomb: General Leslie R. Groves, the
Manhattan Project's Indispensable Man (South Royalton, Vt.:
Steerforth Press, 2002).
4. For discussions of the Swedish program, see Thomas B. Johansson,
"Sweden's Abortive Nuclear Weapons Project," The
Bulletin of the Atomic Scientists 41 (March 1986): 31-34,
and Tor Larsson, "The Swedish Nuclear and Non-Nuclear Postures,"
Storia Delle Relazioni Internationali 13 (1998): 101-119.
5. Wayne Reynolds, "Rethinking the Joint Project: Australia's
Bid for Nuclear Weapons, 1945-1960," Historical Journal
41 (1998): 853-873.
6. For the German nuclear question during this period, see Marc
Trachtenberg, A Constructed Peace: The Making of the European
Settlement, 1945-1963 (Princeton, Princeton University Press,
7. For the France-Italy-West German consortium, see Leopoldo
Nuti, "The F-I-G Story Revisited," Storia Delle
Relazioni Internationali 13 (1998): 69-101.
8. See George Bunn, Arms Control by Committee: Managing
Negotiations with the Russians (Stanford: Stanford University
Press, 1992), 64-66.
9. For a fascinating account of Chinese-Soviet nuclear cooperation
during the 1950s, see Evgeny A. Negin and Yuri N. Smirnov, "Did
the USSR Share Atomic Secrets with China?" published by the
Parallel History Project on NATO and the Warsaw Pact, at http://www.isn.ethz.ch/php/documents/collection_11/texts/Negin_Smirnov_engl.htm.
10. Joint Committee on Atomic Energy, [Report of Ad Hoc Subcommittee
on U.S. Policies Regarding the Assignment of Nuclear Weapons to
NATO], 11 February 1961, copy at National Security Archive.
11. For the Israeli nuclear program and the policy of opacity,
see Avner Cohen, Israel and the Bomb (New York: Columbia
University Press, 1998).
12. The same excisions show up in the Foreign Relations
volume on the region; see U.S. Department of State, Foreign
Relations, 1961-1963, Vol. XVIII, Near East, 1961-1963 (Washington,
D.C., Government Printing Office, 1995), at http://www.state.gov/r/pa/ho/frus/kennedyjf/xviii/26209.htm,
13. For a version released in 2001, see http://nsarchive.gwu.edu/NSAEBB/NSAEBB90/dubious-secrets-9B.pdf.
14. Thanks to Steven Koziak, Center for Strategic and Budgetary
Assessments, for showing how to use the Office of Management and
Budget GDP deflator to convert 1963 dollars into today's dollars.
15. George Perkovich, India's Nuclear Bomb: The Impact on
Global Proliferation (Berkeley: University of California
Press, 1999), 84-85.
16. For a recent upgrading of the significance of the Gilpatric
report, see Francis Gavin, "Blasts from the Past: Proliferation
Lessons from the 1960s," International Security 29
(Winter 2004), 29-130. Interestingly, an account of the NPT's
origins and negotiations by one of the negotiators does not mention
the report. See Bunn, Arms Control by Committee, 59-82.
17. Cohen, Israel and the Bomb, 273-275.
18. For West Germany and the NPT, see Susanna Schrafstetter
and Stephen Twigge, Avoiding Armageddon: Europe, the United
States, and the Struggle for Nuclear Proliferation, 1945-1970
(Westport, CT: Praeger, 2004), 182-194.