Nuclear Weapon Initiatives: Low-Yield R&D, Advanced Concepts, Earth Penetrators, Test Readiness
CRS Report for Congress
Nuclear Weapon Initiatives:
Low-Yield R&D, Advanced Concepts,
Earth Penetrators, Test Readiness
Updated March 8, 2004
Specialist in National Defense
Foreign Affairs, Defense, and Trade Division
Congressional Research Service ˜ The Library of Congress
Nuclear Weapon Initiatives:
Low-Yield R&D, Advanced Concepts,
Earth Penetrators, Test Readiness
The Bush Administration completed its congressionally-mandated Nuclear
Posture Review in December 2001. The review led to major changes in U.S. nuclear
policy. It found that the Cold War relationship with Russia was “very inappropriate”
and that this nation must be able to deal with new threats. It planned to retain Cold
War-era nuclear weapons, which would suffice for many contingencies, though at
reduced numbers. To complement these weapons so as to improve U.S. ability to
deal with new, more dispersed threats in various countries, the Administration sought
to explore additional nuclear capabilities.
Accordingly, the FY2004 request included four nuclear weapon initiatives: (1)
rescinding the ban that Congress imposed in 1993 on R&D on low-yield nuclear
weapons; (2) $6 million for the Advanced Concepts Initiative (ACI) to begin certain
studies of weapon-related science and technology; (3) $15 million to continue a study
of the Robust Nuclear Earth Penetrator (RNEP), in which an existing bomb would
be converted into a weapon able to penetrate into the ground before detonating to
improve its ability to destroy buried targets; and (4) $25 million to enable the United
States to conduct a nuclear test within 18 months of a presidential order to test, and
for related purposes, as compared with the current 24-36 month time that was set
shortly after the end of the Cold War. Congress acted on these requests in the
FY2004 National Defense Authorization Act (P.L. 108-136), and acted on the latter
three in the FY2004 Energy and Water Development Appropriations Act (P.L. 108-
137). For FY2005, the Administration requests $9.0 million for ACI, $27.6 million
for RNEP, and $30.0 million for improving nuclear test readiness.
These initiatives are controversial. Supporters claim that the first three
initiatives would enhance deterrence, thereby reducing the risk of war, and that some
weapons that might result from the initiatives could enable the United States to
destroy key targets in nations that may pose a threat. Critics are concerned that these
initiatives would lead to nuclear testing, increase the risk of nuclear proliferation, and
make U.S. use of nuclear weapons more likely. Regarding enhanced test readiness,
the Administration argues that nuclear testing might be needed, for example, to check
fixes to weapon types with defects, and that 24 to 36 months is too long to wait;
critics are concerned that shortening the time to test could signal a U.S. intent to test,
and that renewed testing could lead to a renewed interest in testing by other nations.
This report provides the policy context for the four initiatives. For each, it then
presents a description, history, FY2004 legislative actions, the FY2005 request (for
all but low-yield R&D), and issues for Congress. It is designed for those who want
a detailed introduction to the debate, those seeking arguments and counterarguments,
and those looking for answers to specific questions. It will track congressional and
executive actions on these initiatives through updates as developments warrant.
The Broader Context for the Four Initiatives.............................2
R&D on Low-Yield Nuclear Weapons.................................6
Military Utility of 5-kt Weapons..............................7
History Through the FY2003 Budget Cycle.........................8
Legislative Actions in the FY2004 Budget Cycle....................15
Conference Issues and Outcomes.................................18
Developments in the FY2005 Budget Cycle........................20
Issues for Congress...........................................20
Would lifting the ban lead to acquisition and use of
Would low-yield weapons make nuclear proliferation more likely?..21
Would low-yield weapons offer military value?.................22
Advanced Concepts Initiative.......................................24
History Through the FY2003 Budget Cycle........................25
Legislative Actions in the FY2004 Budget Cycle....................26
Conference Issues and Outcomes.................................28
The FY2005 Budget Cycle......................................30
Issues for Congress...........................................30
Does deterrence require new types of nuclear weapons?...........31
Do ACI programs offer significant military value?...............33
Is it technically feasible to use nuclear weapons to destroy
stored bioweapon stockpiles? If so, is such use advisable?....34
Can ACI help in the fight against nuclear proliferation?...........40
Does ACI offer unique value for the nuclear weapons enterprise?...42
Robust Nuclear Earth Penetrator.....................................44
History Through the FY2003 Budget Cycle........................48
Legislative Actions in the FY2004 Budget Cycle....................50
Conference Issues and Outcomes.................................51
The FY2005 Budget Cycle......................................52
Issues for Congress...........................................53
Would RNEP promote deterrence?...........................53
Would RNEP provide added military value?....................54
Would fallout from RNEP bar use of the weapon?...............55
Would RNEP have an adverse effect on nuclear nonproliferation?..56
Will Targets Be Available for RNEP?.........................56
Nuclear Test Readiness............................................58
History Through the FY2003 Budget Cycle........................63
Legislative Actions in the FY2004 Budget Cycle....................66
Conference Issues and Outcomes.................................67
The FY2005 Budget Cycle......................................68
Issues for Congress...........................................68
Is an 18-month test readiness posture desirable?.................68
Is an 18-month posture feasible for stockpile stewardship or
weapons development tests?............................71
Four small issues, or one large one?..........................71
Nuclear preemption and use: ambiguities and uncertainties........72
Will the new weapons deter?................................72
Will research lead to testing, acquisition, and use?...............73
Any new weapon has limits to its military value.................74
List of Figures
Figure 1. Earth penetration reduces yield needed to destroy buried targets....45
Figure 2. Lower yield reduces fallout; shallow depth of burst increases it.....46
Nuclear Weapon Initiatives:
Low-Yield R&D, Advanced Concepts,
Earth Penetrators, Test Readiness
For FY2004, the Administration proposed four nuclear weapon initiatives that
were small in terms of dollars and immediate impact, but that would have large
policy ramifications and could lead to major outlays if implemented over time.
Congress considered all of them in the FY2004 National Defense Authorization Act
(NDAA) (P.L. 108-136), and considered all but the ban on R&D on low-yield
weapons, which does not involve funding, in the FY2004 Energy and Water
Development Appropriations Act (EWDAA) (P.L. 108-137) as well.1 For FY2005,
the low-yield R&D ban was not at issue, as Congress lifted it in the FY2004 NDAA;
the Administration requested funds for the other three items below.
Rescission of a ban on R&D on low-yield nuclear weapons. FY1994 legislation
barred R&D that could lead to production of low-yield nuclear weapons. The
Administration proposed rescinding this provision.
Advanced Concepts Initiative (ACI): For FY2004, the Administration requested
$15.0 million for the Robust Nuclear Earth Penetrator and $6.0 million for other
ACI programs. The latter include computer modeling, remotely monitoring
warheads and controlling their firing systems, and studying how to design
warheads with specific radiation outputs and other effects. The FY2005 request
is $27.6 million for the earth penetrator and $9.0 million for ACI.
Robust Nuclear Earth Penetrator (RNEP): The request would continue a study
of modifying an existing weapon to penetrate completely into the ground before
detonating, increasing its ability to destroy buried targets.
Nuclear Test Readiness: A “test readiness posture” is the maximum time
between a presidential order to conduct a nuclear test and the test itself. Since
1996, the U.S. posture has been 24 to 36 months. For FY2004, the
Administration requested $24.9 million to maintain this posture (for which at
least $15.0 million is needed) and to begin changing it to 18 months. The
FY2005 request is $30.0 million.
1 This report does not consider nuclear-armed missile defense interceptors. The
Administration requested no funds for them for FY2004, and the Senate passed, by voice
vote with no debate, an amendment by Senators Levin, Feinstein, and Stevens to bar use of
FY2004 funds for work on such interceptors. U.S. Congress. Congressional Record, May
These initiatives are controversial. Many on both sides of the issue see them as
a gateway to new weapons. They collectively give new impetus to the decades-long
debate over the role of nuclear weapons in promoting U.S. security.
The four initiatives will likely be at issue for some years to come because they
could lead to increased requests in the future. While it costs nothing to lift the ban
on low-yield R&D, doing so would presumably lead to low-yield R&D that would
have a cost. The FY2004 request for ACI was $6.0 million, but there were calls to
increase its funding substantially. The FY2004 RNEP request was $15.0 million to
continue a study, but after the study ends in FY2006, the request would grow sharply
if Congress and the Administration decide to develop the weapon. NNSA estimates
that it would cost $83 million over three years to implement an 18-month test
readiness posture, and $25 million to $30 million a year to sustain it thereafter,
compared to the $10 million or so spent each year to maintain the current 24- to 36-
This report discusses these initiatives and the issues they raise. This section
presents background and key arguments; subsequent sections provide a history and
issues for Congress for each initiative. The report is designed for those who want a
detailed introduction to the debate, including its history; those seeking arguments and
counterarguments; and those looking for answers to specific questions. It will track
congressional and executive actions on these initiatives through regular updates as
The Broader Context for the Four Initiatives
During the Cold War, the focus of U.S. nuclear policy was deterring or
retaliating in response to a Soviet nuclear attack. U.S. nuclear forces were designed
and sized for that contingency. Other contingencies in which the United States might
have used nuclear weapons were much more remote, and would have required far
fewer nuclear weapons, than responding to a Soviet attack. Defense planners dealt
with them, but as “lesser included cases” not requiring new weapons.
The end of the Cold War led to a reduction of the U.S. nuclear weapons
development effort. The first Bush Administration halted nuclear testing except for
safety and reliability of the nuclear stockpile, canceled weapons that were under
development, and withdrew most U.S. battlefield nuclear weapons. Numbers of
personnel in the nuclear weapons complex — the facilities managed by the
Department of Energy (DOE) to develop, test, manufacture, and maintain nuclear
weapons — dropped sharply. At congressional direction, the United States began a
moratorium on nuclear testing in October 1992 that is still in effect and worked to
negotiate the Comprehensive Test Ban Treaty (CTBT), a ban on all nuclear
explosions. Before the moratorium, the United States conducted nuclear tests on an
ongoing basis; with the end of testing, the Clinton Administration changed the test
readiness posture to 24 to 36 months.
2 U.S. Department of Energy, National Nuclear Security Administration, Report to
Congress: Nuclear Test Readiness, April 2003, p. 1.
The George W. Bush Administration came into office with a different stance on
nuclear weapons policy, which the September 11 attacks stiffened. It delivered a
Nuclear Posture Review (NPR) to Congress in December 2001.3 This classified
study, mandated by legislation,4 spelled out the Administration’s views on the role
of nuclear weapons. A strategic relationship with Russia based on the threat of
mutual annihilation had become “very inappropriate,”5 according to J.D. Crouch,
Assistant Secretary of Defense for International Security Policy, and the threat had
become more diffuse. The NPR focused on nuclear capabilities deemed needed for
various contingencies rather than mainly to counter the Soviet nuclear threat. That
is, the Administration perceived a wider range of possible uses for nuclear weapons
quite different than the U.S.-Soviet nuclear war scenarios that dominated Cold War
strategy. Crouch said, “The capabilities-based approach argues that there may be
multiple contingencies and new threats that we have to deal with. We’re focusing
on how we will fight ... not who or when.”6 To support this approach, the NPR
envisioned a “new triad” — strike forces (nuclear and nonnuclear); defenses; and an
infrastructure better able to respond to military needs — tied together by command
and control, intelligence, and planning. The goal was to “assure allies and friends”
that the United States had “credible non-nuclear and nuclear response options,”
“dissuade competitors” through a “diverse portfolio of capabilities [that] denies
payoff from competition,” “deter aggressors,” and “defeat enemies.”7
The NPR called for retaining the capability to deter or respond to a variety of
contingencies, and for retaining Cold War-era nuclear weapons, at lower levels, to
do so. With the demise of the Soviet Union and the September 11 attacks, a subset
of targets — terrorist-related facilities, some of which had existed for decades —
assumed greater prominence. One example is hardened and deeply buried facilities.
Rogue (and many other) states have built such facilities, which might house leaders
and key communications facilities. Another example is facilities — some deeply
buried, others not — for producing or storing weapons of mass destruction (WMD).
The end of the Cold War also led to changed constraints on use of nuclear
weapons. In a nuclear war with the Soviet Union, assuring the immediate destruction
of thousands of targets of all sorts would have been critical to national survival.
3 For further information on U.S. nuclear policy, see CRS Report RL31623, U.S. Nuclear
Weapons: Changes in Policy and Force Structure, and CRS Report RS21619, Nuclear
Weapons and U.S. National Security: A Need for New Weapons Programs?, both by Amy
4 The requirement for the NPR is in P.L. 106-398, Floyd D. Spence National Defense
Authorization Act for Fiscal Year 2001, Section 1041.
5 U.S. Department of Defense, “Special Briefing on the Nuclear Posture Review,” Presented
by J.D. Crouch, Assistant Secretary of Defense for International Security Policy, January
7 U.S. Department of Defense. “Findings of the Nuclear Posture Review,” briefing slides,
January 9, 2002, slide titled “QDR: Defense Policy Goals,” available at
[http://www.defenselink.mil/news/Jan2002/g020109-D-6570C.html], and U.S. Department
of Defense. “Special Briefing on the Nuclear Posture Review,” January 9, 2002.
Reducing fallout and “collateral damage,” or damage to people and things that were
not the intended targets, was secondary in importance to achieving required
probabilities of damage on targets. Warheads were typically designed to achieve that
primary goal. Now, nuclear weapons to hold at risk specific, limited targets in rogue
states in a credible manner are judged to need several characteristics:
!They would need to be able to counter threats that rogue states might
pose, such as hard and deeply buried facilities or production
facilities for WMD. Weapons to attack buried facilities would have
different characteristics than weapons to defeat chemical and
!They would be better able to minimize (not eliminate) collateral
damage consistent with military objectives. While this has long
been a goal of U.S. nuclear policy, earth penetration and increased
accuracy sharply reduce the yield needed to destroy a target, making
it more feasible to attain that goal.
!They would need to be available readily, if not immediately, because
the specific threat might become known with little warning.
!They could be few in number. While an all-out nuclear attack on the
Soviet Union would have struck thousands of targets, the option of
an attack on a rogue state might involve a handful of targets.
According to the National Nuclear Security Administration (NNSA),8 “The
Administration believes the broader range of capabilities of a nuclear stockpile with
these weapons will serve as a more credible, and hence more effective, deterrent than
the Cold War stockpile we have today. This more effective deterrent will make the
use of nuclear weapons less likely.”9
Some critics believe that the only use for nuclear weapons should be to deter
and, if necessary, respond to the use of nuclear weapons against the United States.10
They oppose attempts to develop nuclear weapons with lower yield and reduced
collateral damage as blurring the distinction between conventional and nuclear
weapons. Even U.S. use of a low-yield atomic bomb, they assert, would result in a
firestorm of protest and worldwide opprobrium.
8 NNSA is a semiautonomous agency within the Department of Energy (DOE) that has
managed the nuclear weapons complex since 2001. This complex is composed of
laboratories, manufacturing plants, and a test site that design, develop, manufacture,
maintain, and dismantle nuclear weapons, and that also used to conduct nuclear explosions
to test them. (Before 2001, DOE managed this complex directly through its Office of
9 Comments provided to the author by the National Nuclear Security Administration, August
10 See, for example, Union of Concerned Scientists, “Global Security: Nuclear Weapons,”
[http://www.ucsusa.org/ gl obal_security/nuclear_weapons/index.cfm?pageID=30].
Further, they believe that the U.S. nuclear arsenal has long been sufficient
against a range of threats. Hardened and deeply buried targets (HDBTs), for
example, have existed for decades, yet the first Bush Administration halted work on
nuclear weapons then under development and halted nuclear testing except for safety
and reliability, effectively bringing work on new weapon types to a close.
Presumably that administration would not have taken these steps had it envisioned
a need for new weapon types.
The United States needs the active support of the entire international community
in the war against terrorism, many observers believe. Assistance may take many
forms: providing information to the CIA on suspicious individuals, safeguarding
shipping containers, securing radioactive material, or closing bank accounts of
suspected terrorists. Other nations will be more likely to cooperate on issues
important to the United States, it is argued, if the United States cooperates with other
nations on issues important to them. One such issue is nuclear nonproliferation. The
United States pledged in the Nuclear Nonproliferation Treaty (Article VI) “to pursue
negotiations in good faith on effective measures relating to cessation of the nuclear
arms race at an early date and to nuclear disarmament.” In addition, as an
inducement in securing international support for extending the treaty indefinitely, the
United States and other nuclear weapon states pledged in 1995 to support the
CTBT.11 They made a similar statement at the 2000 NPT review conference.12
Critics maintain that the Administration’s nuclear initiatives are at odds with
these pledges. They argue that developing new or modified weapons runs counter
to the NPT pledge and that such weapons will ultimately require nuclear testing to
ensure that they work. Reducing the time to conduct a nuclear test supports a future
decision to test and, to the critics, implies an intent to test. These positions may harm
U.S. security by undermining efforts to make worldwide cooperation on
nonproliferation of nuclear and other WMD more effective. Even if some steps
might be innocuous by themselves, critics believe that many around the world will
see the confluence of them as pieces of a single policy and will infer that the United
States is now much more willing to consider using nuclear weapons. As an example
of international concern over the new U.S. policies, the foreign ministers of the New
Agenda Coalition (Brazil, Egypt, Ireland, Mexico, New Zealand, South Africa, and
Sweden) issued a statement that said in part,
4.The Ministers stressed that each article of the NPT is binding on the
respective States parties, at all times and in all circumstances ...
11 For further information on this treaty, including subsequent Senate action, see CRS Issue
Brief IB92099, Nuclear Weapons: Comprehensive Test Ban Treaty, CRS Report 97-1007,
Nuclear Testing and Comprehensive Test Ban: Chronology Starting September 1992,and
CRS Report RS20351, Comprehensive Test Ban Treaty: Pro and Con, all by Jonathan
12 Statement by the Delegations of France, the People’s Republic of China, the Russian
Federation, the United Kingdom of Great Britain and Northern Ireland, and the United
States of America to the 2000 NPT Review Conference, May 1, 2000. Available at
[ ht t p: / / www.cei p.or g/ f i l e s/ pr oj ect s/ npp/ r e sour ces/ npt 2000p5.ht m] .
7.The Ministers reiterated their deep concern at emerging approaches to the
broader role of nuclear weapons as part of security strategies, including
rationalizations for the use of, and the development of new types of nuclear13
These positions may undermine U.S. security, critics fear, by spurring nations
that might be targets of a U.S. nuclear attack to step up their efforts to acquire nuclear
weapons. Of the three “Axis of Evil” nations, Iraq, the first target of a U.S. attack,
did not have nuclear weapons as far as is currently known. Having nuclear weapons
to deter the United States may be a consideration for North Korea and Iran, both of
which apparently have active programs to produce fissile material. North Korea has
stated that it has several nuclear weapons; while Iran has declared that its nuclear
program is for peaceful purposes, few in the United States believe that assertion.
Even in military terms, critics argue, nuclear use would offer marginal to
negative value. Deeply buried targets could be defeated (not necessarily destroyed)
with conventional weapons, destruction of entrances and air shafts, or demolition by
special forces. The effectiveness of low-yield nuclear weapons for destroying
biological or chemical agents is uncertain, as discussed below, and intelligence might
be unable to locate targets in underground complexes, as recent efforts to find Iraqi
WMD showed, rendering nuclear weapons of any yield of little use. Nations could
counter earth penetrators by burying facilities more deeply. Even low-yield or earth
penetrating weapons would throw a large amount of radioactive debris into the
atmosphere. Substituting radioactive materials for biological or chemical agents may
offer little advantage to the local population or U.S. troops.
R&D on Low-Yield Nuclear Weapons
A provision in the FY1994 National Defense Authorization Act barred R&D
that could lead to production by the United States of a nuclear weapon of less than
5 kilotons (kt)14 that had not entered production by November 1993. In its FY2004
legislative proposals, the Administration asked Congress to rescind this provision on
grounds that it undercut U.S. ability both to explore technical options that could deter
or respond to emerging threats, and to revitalize the nuclear weapons enterprise. This
provision is often referred to as the Spratt-Furse provision, after its legislative
authors, Representatives John Spratt, Jr., and Elizabeth Furse, or as the PLYWD
provision, for Precision Low-Yield Weapon Design. The Armed Services
Committees considered the proposed rescission; the Appropriations Committees did
not do so because it involved no money. The FY2004 NDAA, Section 3116,
repealed the Spratt-Furse provision but (1) made clear that the repeal did not
13 “Deep Concern At The lack Of Progress,” declaration, issued by the Foreign Ministers of
the New Agenda Coalition, United Nations Headquarters, New York, 23 September 2003,
[ ht t p: / / www.acr onym.or g.uk/ docs/ 0309/ doc20.ht m] .
14 One kiloton is equivalent to the explosive force of 1,000 tons of TNT; the yield of the
Hiroshima bomb was 15 kilotons.
“authoriz[e] the testing, acquisition, or deployment of a low-yield nuclear weapon”
and (2) barred the Secretary of Energy from beginning engineering development or
subsequent phases of work on a low-yield weapon without specific congressional
authorization. With the provision lifted, there is no parallel request for FY2005.
Military Utility of 5-kt Weapons. The debate over the Spratt-Furse
provision treated 5 kt as critical because that was the threshold in the law. Those
who would lift the ban argued that the ban made it difficult to train weapon designers
and impossible to develop essential new weapons; their opponents were concerned
that lifting the ban would signal U.S. interest in developing capabilities that would
make the use of nuclear weapons more likely. Yet the threshold was of only slight
relevance to the military or the weapons labs because very low yield nuclear weapons
have uncertain military utility. Substantial advances would be needed in the
understanding of low-yield weapon effects to assess the military utility and collateral
damage of such weapons.
Military utility derives from nuclear-weapon effects that are prompt and that can
be measured with reasonable precision. Damage from the blast wave falls into this
category; fallout and secondary fires (e.g., fuel set afire) do not. For a 500-kt
weapon, the blast effects are highly predictable: almost anything on the Earth’s
surface over a large, known area will be destroyed. (For comparison, the Hiroshima
bomb had a yield of 15 kt.) At the other end of the scale, the blast effects of a 500-
pound conventional bomb are strongly influenced by accuracy, height of burst, local
features, etc. For example, a hill or building can block or redirect the blast wave.
Such factors also shape the effects of a low-yield nuclear weapon, and their influence
increases as yield decreases. Modeling such factors is more complicated than for
high-yield weapons, has been done to a much lesser extent, and depends more on
unknown details such as a building’s construction. It may be that detailed
engineering studies and computer models of the prompt effects of existing low-yield
nuclear weapons could provide as much military value as designing new low-yield
A 1979 report by the Office of Technology Assessment elaborated on the
problem of uncertain weapon effects in discussing a 1-kt terrorist nuclear weapon
detonated at ground level in a city:
the highly built-up urban structure in which the weapon is placed will
significantly modify the resulting nuclear environment. This occurs when the
lethal range of effects shrink to such an extent that they are comparable to the
size of urban structures. It is indeed reasonable to expect that the blast effects
of a small weapon ... will be severely influenced by nearby structures having
comparable dimensions. Preliminary calculations have confirmed this. ... In
summary, the ranges of nuclear effects from a low-yield explosion in the
confined space of an urban environment will differ significantly from large yield15
effects, but in ways that are very difficult to estimate.
15 U.S. Congress, Office of Technology Assessment. The Effects of Nuclear War. OTA-NS-
If the Warsaw Pact had invaded Western Europe during the Cold War, such
uncertainties of low-yield weapons would have been immaterial for some missions
(an atomic demolition munition could destroy a bridge, and an artillery-fired atomic
projectile could destroy troop concentrations, more reliably than conventional
munitions) and in some situations (as a last resort to keep front-line troops from
being overrun). Now, though, extreme accuracy may enable conventional weapons
to destroy small targets, and massive conventional bombing can attack troop
concentrations. At the same time, this same accuracy may give low-yield nuclear
weapons new military missions by enabling them to destroy targets that in the past
would have required much larger weapons. On the other hand, low-yield weapons
face uncertainties that render them of questionable value in going after key targets,
such as buried facilities for which details of the structure and its overlying earth and
rock are poorly known. This uncertainty increases as yield drops. For reasons such
as these, George Miller, formerly Associate Director for Defense and Nuclear
Technologies at Lawrence Livermore National Laboratory, observed, “The vast
majority of the ‘low yield’ concepts that would provide substantial reductions in
collateral damage have yields greater than 5 kt. Concepts in the less than 5 kt regime
are at an early R&D stage.”16
History Through the FY2003 Budget Cycle
The United States has had many low-yield weapons in the stockpile for
decades.17 In the wake of the 1991 Persian Gulf War, the U.S. military became more
interested in them for attacking underground structures. One news report stated that
“[p]enetrating Saddam’s hardened bunkers proved to be one of the most daunting
tasks that faced the Air Force.”18 In late 1991, the Air Force reportedly asked Los
Alamos National Laboratory to study a very low yield warhead that could destroy
underground bunkers.19 A 1992 Lawrence Livermore National Laboratory journal
stated, “Several possibilities for new types of weapons are being evaluated for
prototype development. These include ... low-yield earth-penetrating warheads for
attacking underground command posts with minimum collateral damage...”20 DOE
16 Personal communication, August 15, 2003.
17 Thomas Dowler and Joseph Howard III, “Countering the Threat of the Well-Armed
Tyrant: A Modest Proposal for Small Nuclear Weapons,” Strategic Review, Fall 1991,
“Countering the Threat of the Well-Armed Tyrant,” p. 34-40, discuss low-yield nuclear
weapons and their military advantages. William Arkin, “Nuclear Junkies: Those Lovable
Little Bombs,” Bulletin of the Atomic Scientists, July-August 1993: 22-27, discusses the
bureaucratic progress of such weapons and their drawbacks.
18 Tony Capaccio, “Mini-Nukes Weighed for Third World Contingencies,” Defense Week,
September 8, 1992: 12.
19 “Lab Proposes Smaller Nuclear Weapons,” Associated Press (newswire), October 8, 1992.
20 “Future Directions for the Nuclear Weapons Program at LLNL [Lawrence Livermore
National Laboratory],” Energy and Technology Review, January-February 1992: 18.
began a concept definition study for an Aircraft Delivered Precision Low-Yield
Weapon in FY1992, which it planned to continue in FY1993 and FY1994.21
At the same time, many in Congress and elsewhere sought to restrain U.S.
nuclear weapons development. Legislation mandated a U.S. nuclear testing
moratorium, which started in October 1992 and continues to the present, and directed
the President to provide a plan for achieving a multilateral CTBT by September 30,
1996.22 Preparations were underway in the U.N. in 1993 to begin CTBT negotiations
in 1994. The NPT review and extension conference, which would decide whether
to extend the treaty indefinitely, was planned for April 1995. Some saw testing, and
weapons development that could lead to testing, as unhelpful to these efforts.
In 1993, the House Armed Services Committee, concerned over efforts by DOE
to study low-yield nuclear weapons, included a provision barring R&D on these
weapons in its version of the FY1994 National Defense Authorization Act, H.R.
The committee is aware of recent efforts by the department [of Energy] to
perform concept and feasibility studies for designing very low yield nuclear
weapons. The committee opposes these efforts. Very low yield nuclear
warheads threaten to blur the distinction between conventional and nuclear
conflict, and could thus increase the chances of nuclear weapons use by another
nation. In addition, the committee believes that the development of very low
yield nuclear weapons undermines U.S. efforts to discourage nuclear weapons
development by other nations, and would undercut U.S. efforts to negotiate an
extension of the Non-Proliferation Treaty or a Comprehensive Test Ban. Finally,
the utility of very low yield nuclear weapons is questionable given the increasing
effectiveness and availability of precision guided conventional munitions.
The committee therefore recommends a provision (sec. 3105 (e)) that
would direct the Secretary of Energy to discontinue the ongoing concept design
work within the department’s nuclear weapons laboratories and to refrain from
any future feasibility, engineering, development, or production work associated
with very low yield nuclear weapons. The committee further directs the
Secretary to work with the President, and other interested agencies in23
discouraging the development of similar weapons in other countries.
The conference report summarized the House position, noted that the Senate bill
had no similar provision, and stated:
The Senate recedes with an amendment that would clarify that the
prohibition applies to activities that could lead to production of new low-yield
21 U.S. Department of Energy, Office of Chief Financial Officer, FY1994 Congressional
Budget Request, Volume 1: Atomic Energy Defense Activities. DOE/CR-0012, April 1993,
22 Sec. 507 of P.L. 102-377, Energy and Water Development Appropriations Act, FY1993,
contains these requirements.
23 U.S. Congress, House Committee on Armed Services, National Defense Authorization Act
for Fiscal Year 1994. H.Rept. 103-200, (Washington:GPO, 1993) p. 427.
weapons. While the conferees agree that the provision is intended to prohibit
research and development geared toward the production of any low-yield nuclear
weapons by the United States, the conferees recognize that there are instances
where the Department of Energy may need to conduct research on these types of
weapons for other purposes. This would include research, in the interest of
counter-proliferation, on the designs of low-yield weapons as a way to: (1)
understand others’ activities, including potential terrorist threats; (2) provide
information for export control activities; and (3) understand the potential damage
that could be inflicted by the use of these types of weapons. In addition, the
conferees agree that nothing in this section would prohibit the Department of
Energy from performing the research and development necessary for
modifications to existing weapons in order to address safety or reliability
The conferees direct the Secretary to work with the President and
interested agencies in discouraging the development of similar weapons in
The final legislation (P.L. 103-160, November 30, 1993, 107 stat 1946) follows:
SEC. 3136. PROHIBITION ON RESEARCH AND DEVELOPMENT OF
LOW-YIELD NUCLEAR WEAPONS.
(a) UNITED STATES POLICY- It shall be the policy of the United States
not to conduct research and development which could lead to the production by
the United States of a new low-yield nuclear weapon, including a precision
(b) LIMITATION- The Secretary of Energy may not conduct, or provide
for the conduct of, research and development which could lead to the production
by the United States of a low-yield nuclear weapon which, as of the date of the
enactment of this act, has not entered production.
(c) EFFECT ON OTHER RESEARCH AND DEVELOPMENT- Nothing
in this section shall prohibit the Secretary of Energy from conducting, or
providing for the conduct of, research and development necessary —
(1) to design a testing device that has a yield of less than five
(2) to modify an existing weapon for the purpose of addressing safety
and reliability concerns; or
(3) to address proliferation concerns.
(d) DEFINITION- In this section, the term `low-yield nuclear weapon’ means a
nuclear weapon that has a yield of less than five kilotons.
24 U.S. Congress, Committee of Conference, National Defense Authorization Act for Fiscal
Year 1994. H.Rept. 103-357, USGPO, 1993, p. 840-841.
Several years later, the Office of Defense Programs asked DOE’s General
Counsel about the legal meaning of this prohibition. In a memorandum of March
Section 3136 would appear to prohibit the Department from taking or
supporting any action that could result in producing a weapon with a yield of less
than 5 kilotons, unless one of the following four exemptions applies: (1) the
activities are related to a weapon that entered production prior to December ,
1993; (2) the activities are supporting design of a testing device; (3) the activities
are directed to modifying an existing weapon to address safety and reliability
concerns; or (4) the activities are supporting counterproliferation work.
As you can see, the question of legality on any course of action involving25
low-yield nuclear weapons is highly fact dependent.
In its report on the FY2001 defense authorization bill, S. 2549, the Senate
Armed Services Committee made clear the issue prompting DOE’s request to the
General Counsel and sought to redress that issue legislatively:
The committee recommends a provision that would require the Secretaries
of Defense and Energy to assess requirements and options for defeating hardened
and deeply buried targets. The provision would expressly authorize the
Department of Energy (DOE) to conduct any limited research and development
that may be necessary to complete such assessments.
The committee notes that a recent legal interpretation of existing law raised
questions regarding whether DOE could participate in or otherwise support
certain Department of Defense (DOD) studies and options assessments for
defeating hardened and deeply buried targets. This provision removes any
uncertainty and expressly allows DOE to assist the DOD with a review of these
targets and the options for defeating such targets. The committee believes that
DOE should provide information and all other assistance required to help DOD
make informed decisions on whether: (1) to proceed with a new method of
defeating hardened and deeply buried targets and; (2) to seek any necessary
modifications to existing law.
The committee is concerned that the ability to defeat hardened and deeply
buried targets will continue to be a significant challenge for the foreseeable26
Although the House bill did not include a similar provision, the conference bill,
H.R. 4205, did. Section 1044, “Report on the defeat of hardened and deeply buried
targets,” required the Secretary of Defense, in conjunction with the Secretary of
25 Memorandum from Mary Anne Sullivan, General Counsel, to Thomas F. Gioconda,
Brigadier General, USAF, Principal Deputy Assistant Secretary for Military Applications,
Office of Defense Programs, “Statutory Prohibition on Research and Development of Low-
Yield Nuclear Weapons,” March 19, 1999.
26 U.S. Congress, Senate Committee on Armed Services, National Defense Authorization Act
for Fiscal Year 2001. S.Rept. 106-292. 106th Congress, 2nd Session. USGPO, 2000, p. 348-
Energy, to conduct a study on this topic, due by July 1, 2001.27 That section directed
that the study review U.S. requirements to defeat such targets “and stockpiles of
chemical and biological agents and related capabilities,” review plans to meet the
requirements and determine the adequacy of the plans, “identify potential future
hardened and deeply buried targets,” determine what is needed to defeat them and to
defeat stockpiles of chemical and biological agents, assess options to defeat such
targets, and “determine the capability and cost of each option...” Section 1044 gave
the Secretaries authority to “conduct any limited research and development that may
be necessary to perform those assessments.” Further, “The conferees believe that
DOE should provide information and other assistance required to help DOD make
informed decisions on whether: (1) to proceed with a new method of defeating
hardened and deeply buried targets; and (2) to seek any necessary modifications to
The DOD-DOE report required by Section 1044, dated July 2001, contained a
rationale for low-yield nuclear earth penetrator weapons.29 It stressed the problem
that hardened and deeply buried targets (HDBTs) posed to the United States:
Our potential adversary’s weapons of mass destruction (WMD), long-range
missiles, modern air defenses, most sophisticated command and control systems,
national leadership in wartime, and a variety of tactical arms are increasingly
concealed and protected by networks of hard and deeply buried facilities. If the
United States does not have the means to defeat these facilities and the
threatening assets they protect, adversaries may perceive that they have a
sanctuary from which to coerce or attack the United States, its allies, or its
coalition partners with threats much more powerful than in past conflicts. (p. 3)
It discussed the potential problem posed by buried facilities containing chemical
or biological weapons (CBW):
Ordnance employing fragmentation and blast effects will not accomplish
this objective [destroying agents within munitions or containers] and may further
worsen the situation by releasing agents into the atmosphere and surrounding
environment. In some situations, there may be a need for multiple types of
payloads to accomplish several objectives. For example, in the case of CBW
located within a hardened facility, the goals might be in situ neutralization of the
agents plus access-denial that prevents adversaries from recovering and using
agents or production equipment not destroyed. This class of problems is the
most vexing challenge to defeat of HDBTs. (p. 14)
27 For text of Section 1044, see U.S. Congress. Committee of Conference. Enactment of
Provisions of H.R. 5408, the Floyd D. Spence National Defense Authorization Act for Fiscal
Year 2001. H.Rept. 106-945. 106th Congress, 2nd Session. USGPO, 2000, p. 277-278.
28 Ibid., p. 851. The President signed the measure into law on October 30, 2000 (P.L. 106-
29 U.S. Departments of Defense and Energy. Report to Congress on the Defeat of Hard and
Deeply Buried Targets, Submitted by the Secretary of Defense in Conjunction with the
Secretary of Energy in Response to Section 1044 of the Floyd D. Spence National Defense
Authorization Act for Fiscal Year 2001, PL 106-398, July 2001, 24 p.
While the DOD-DOE report focused on conventional weapons, special
operations forces, intelligence, etc., for defeating HDBTs, it also discussed the
potential value of nuclear weapons for this mission, especially low-yield penetrators:
DoD and DOE have completed initial studies on how existing nuclear weapons
can be modified to defeat those HDBTs that cannot be held at risk with
conventional high-explosive weapons or current nuclear weapons. (p. 4)
... we also must prepare for those unique and emerging strategic threats that are
critical and well protected ... This will require additional investment in
intelligence, special weapons, and counter-WMD capabilities, including nuclear
weapons. (p. 6)
Nuclear weapons have a unique ability to destroy both agent containers and
CBW agents. Lethality is optimized if the fireball is proximate to the target.
This requires high accuracy; for buried targets, it also may require a penetrating
weapon system. Given improved accuracy and the ability to penetrate the
material layers overlying a facility, it is possible to employ a much lower-yield
weapon to achieve the needed neutralization. The ability to use a lower yield
would reduce weapon-produced collateral effects. The current nuclear weapons
stockpile, while possessing some limited ground penetration capability and lower
yield options (not yet certified), was not developed with this mission in mind.
Despite the above, the DOD-DOE report ended by noting, “DoD has not defined
a requirement for a nuclear weapon for WMD Agent Defeat missions.” (p. 24)
The FY2002 National Defense Authorization Act, and its predecessor versions
in House and Senate, did not address the issue of low-yield nuclear weapons or
nuclear earth penetrators.
As noted, the Administration completed its congressionally-mandated Nuclear
Posture Review (NPR) in December 2001. A text purported to consist of leaked30
excerpts of the NPR surfaced in March 2002 and was widely quoted in the press.
Among other things, this text stated that, compared to a surface-burst nuclear weapon
of a given yield, a nuclear earth penetrator weapon could destroy many buried
facilities with much lower yield and thereby reduce fallout by a factor of 10 to 20.31
In action on the FY2003 National Defense Authorization Act, Representative
Weldon initially offered an amendment on the House floor that, among other things,
30 Nuclear Posture Review [Excerpts], available at
[http://www.globalsecurity.org/wmd/library/policy/dod/npr.htm]. This is not an official
source. Secretary of Defense Donald Rumsfeld confirmed that the NPR was leaked. When
asked about the leak at a press conference, he said, “I am disturbed by the leak because I
think it’s enormously unprofessional.” U.S. Department of Defense. News Transcript:
“DoD News Briefing — Secretary Rumsfeld and Gen. Pace,” March 15, 2002.
[http://www.defenselink.mil/news/Mar2002/t03152002_t0315sd.html]. Michael Gordon,
“U.S. Nuclear Plan Sees New Targets and New Weapons,” New York Times, March 10,
31 Nuclear Posture Review [Excerpts], p. 47.
would have repealed the Spratt-Furse provision upon presidential certification that
another nation conducted a nuclear test for new or improved nuclear weapons, or that
another nation was developing, in underground facilities, WMD that could pose an
imminent risk to the United States, or that the repeal “is in the national security
interest of the United States.”32 Before the amendment was debated, Representatives
Spratt and Weldon subsequently modified it. While the 1993 provision barred
“research and development which could lead to the production by the United States
of a new low-yield nuclear weapon,” the modification barred such development but
not research, and clarified “development” as follows: “The term ‘development’ does
not include concept definition studies, feasibility studies, or detailed engineering
design work.”33 The amendment passed, 362-53. The Senate bill did not have a
similar provision and the House provision was dropped in conference.
A description of the “phases” of weapon development is needed to understand
the subsequent discussion of legislation on low-yield R&D and the other three
nuclear initiatives. The development of new nuclear weapons proceeds through a
series of defined “phases.” DOD and the Atomic Energy Commission, a predecessor
of DOE, entered an agreement on March 21, 1953, defining these phases as follows:
Phase 1, weapon conception; phase 2, program study; phase 2a, design definition and
cost study; phase 3, development engineering; phase 4, production engineering; phase
retirement.34 (Note that phase 2 involves some prototype and subsystem testing as
well as paper and computer studies.) The key dividing line is between phase 2a
(design definition) and phase 3 (full-scale development). In phase 2a, Los Alamos
and Lawrence Livermore National Laboratories, the two laboratories that design the
nuclear explosive components of nuclear weapons, would generate one or more
designs each for a particular warhead; personnel from Sandia National Laboratories,
which design nonnuclear components of a warhead, would participate in the two
design teams. At the end of phase 2a, one option and laboratory would be selected.
(No new warhead has been designed since the 1980s.) In phases 3 and 4, which now
would occur concurrently, the phase 2a design would be turned into a detailed design
for a producible weapon through computer simulation, nuclear and nonnuclear
testing, etc. For alterations and modifications of existing weapons, those in phase 6,
the phase structure above is repeated as the “X” in “phase 6.X,” though with some
different names: phase 6.1, concept assessment; phase 6.2, feasibility study and
option down-select; phase 6.2a, design definition and cost study; phase 3,
development engineering; phase 4, production engineering; phase 6.5, first
production; and phase 6.6, full-scale production. 35
32 U.S. Congress, Congressional Record, May 9, 2002: H2334.
33 Ibid., p. H2335.
34 U.S. Atomic Energy Commission. “An Agreement Between the AEC and the DOD for
the Development, Production, and Standardization of Atomic Weapons,” March 21, 1953,
35 U.S. Department of Defense, Department of Energy, and Nuclear Weapons Council,
“Procedural Guideline For The Phase 6.X Process,” April 19, 2000, 13 p., defines these
phases and explains the roles of DOD and DOE in weapons development.
With that as background, Section 3143 of P.L. 107-314, FY2003 National
Defense Authorization Act, “Requirements for Specific Request for New or Modified
Nuclear Weapons,” required DOE (1) to have a single line item for all funds
requested for R&D that could lead to U.S. production of a new nuclear weapon in
phase 1, 2, or 2a, or a modified nuclear weapon in phase 6.1, 6.2, or 6.2a, or concept
work occurring before phase 1 or 6.1; and (2) to request funds for each weapon
activity in phase 3 or 6.3 or higher as a separate line item. Further, the legislation
provided a clear definition of a key term:
(3) The term “new nuclear weapon” means a nuclear weapon that contains a pit
or canned subassembly, either of which is neither —
(A) in the nuclear weapons stockpile on the date of the enactment of
this act; nor
(B) in production as of that date.36
While the move from phase 2a or 6.2a to phase 3 or 6.3 typically involves a
large increase in cost, requiring each weapon in phase 3 or 6.3 to have its own line
item enhances the visibility of these programs and prevents them from being lumped
with other programs, aiding congressional oversight and control of the process.
The explanatory language of this section of the conference report referenced the
Spratt-Furse provision: “The conferees agree that nothing in this section may be
construed to modify, repeal, or in any way affect the provisions of section 3136 of the37
National Defense Authorization Act for Fiscal Year 1994 ....”
Legislative Actions in the FY2004 Budget Cycle
The Administration, in DOD legislative proposals for FY2004, requests repeal
of the Spratt-Furse provision restricting low-yield nuclear weapon R&D. Its draft
language states simply, “Section 3136 of the National Defense Authorization Act for
Fiscal Year 1994 (P.L. 103-160; 107 Stat. 1946) is repealed.”38 The Administration
offered the following rationale:
Section 3136, the so-called PLYWD legislation, prohibits the Secretary of
Energy from conducting any research and development which could potentially
lead to the production by the United States of a new low-yield nuclear weapon,
including a precision low yield warhead.
36 A pit is the fissile core of a nuclear weapon. The pit and associated materials are called
the primary stage. Detonating the primary provides the energy to detonate the canned
subassembly, a term used to refer to the secondary stage. That stage provides most of a
weapon’s energy through fission and fusion.
37 U.S. Congress, Committee of Conference, National Defense Authorization Act for Fiscal
Year 2003. 107th Congress, 2nd Session, H.Rept. 107-772, USGPO, 2002, p. 786.
38 For the proposed bill, see [http://www.defenselink.mil/dodgc/lrs/docs/March3-bill.pdf].
This legislation has negatively affected U.S. Government efforts to support
the national strategy to counter WMD and undercuts efforts that could strengthen
our ability to deter, or respond to, new or emerging threats.
A revitalized nuclear weapons advanced concepts effort is essential to: (1)
train the next generation of nuclear weapons scientists and engineers; and (2)
restore a nuclear weapons enterprise able to respond rapidly and decisively to
changes in the international security environment or unforeseen technical
problems in the stockpile. PLYWD has had a “chilling effect” on this effort by
impeding the ability of our scientists and engineers to explore the full range of
technical options. It does not simply prohibit research on new, low-yield
warheads, but prohibits any activities “which could potentially lead to production
by the United States” of such a warhead.
It is prudent national security policy not to foreclose exploration of
technical options that could strengthen our ability to deter, or respond to, new or
emerging threats. In this regard, the Congressionally-mandated Nuclear Posture
Review urged exploration of weapons concepts that could offer greater
capabilities for precision, earth penetration (to hold at risk deeply buried and
hardened bunkers), defeat of chemical and biological agents, and reduced
collateral damage. The PLYWD legislation impedes this effort.
Repeal of PLYWD, however, falls far short of committing the United States to
developing, producing, and deploying new, low-yield warheads. Such warhead
concepts could not proceed to full-scale development, much less production and
deployment, unless Congress authorizes and appropriates the substantial funds39
required to do this.
The Senate bill, S. 1050, as reported from the Armed Services Committee,
contained this repeal (Section 3131) and added that nothing in the repeal “shall be
construed as authorizing the testing, acquisition, or deployment of a low-yield
nuclear weapon.” On May 20, the Senate considered an amendment by Senators40
Feinstein and Kennedy to rescind Section 3131. The debate, in which more than
repealing the low-yield ban “will ... begin a new era of nuclear proliferation.”
Senator Durbin said, “This bill is about to discard 50 years of American foreign42
policy and 50 years of American nuclear policy.” Senator Kyl said, “The reason
low-yield weapons research is being sought is because the world has changed since
the time we developed these huge megaton nuclear weapons ... the United States
39 U.S. Department of Defense, “DoD Legislative Proposals: National Defense
Authorization Act for Fiscal Year 2004 (Sent to Congress on March 3, 2003) —
Section-by-Section Analysis.” [http://www.defenselink.mil/dodgc/lrs/docs/March3-
40 For further details on congressional action on the low-yield ban, ACI, and RNEP in 2003,
see John Isaacs, “Congress Jumps off the Ban-Wagon,” Bulletin of the Atomic Scientists,
July-August 2003: 20, 21, 69.
41 U.S. Congress, Congressional Record, May 20, 2003: S6663.
42 Ibid., p. S6669.
would prefer, if it had to, to use a much smaller weapon, a low-yield weapon.”43 The
amendment was tabled, 51-43.44 When that amendment failed, Senator Jack Reed
offered an amendment to modify Section 3131 by replacing “research and
development” with “development engineering.” Thus, the amendment would have
barred development engineering that could lead to U.S. production of a new nuclear
warhead. Senator Warner offered an amendment to the Reed amendment, replacing
the latter with the original committee amendment and adding a section, “The
Secretary of Energy may not commence the engineering development phase or any
subsequent phase of a low-yield nuclear weapon unless specifically authorized by
Congress.”45 The difference between the two amendments was summed up as
Mr. Reed. Essentially, the functional difference between my amendment and
your second degree is, at this point, under my amendment the administration
would have to come and lift the prohibition; under your amendment, they would
have to come and get an authorization. I think that is the functional difference.
Mr. Warner. I think the Senator is correct.46
On May 21, the Senate agreed to the Warner amendment, 59-38, and agreed to
the Reed amendment, as amended by the Warner amendment, 96-0. Subsequently,
on May 22, the Senate passed S. 1050, as amended, 98-1, and on June 4 passed H.R.
striking all after the enacting clause and inserting the text of S. 1050.
The House Armed Services Committee tackled the same subject. As reported
by the committee, Section 3111 of H.R. 1588 modified the low-yield provision
substantially from the language the Administration had requested, adopting by voice
vote an amendment by Representative Spratt.47 The FY1994 provision is as follows:
(b) LIMITATION- The Secretary of Energy may not conduct, or provide
for the conduct of, research and development which could lead to the
production by the United States of a low-yield nuclear weapon which, as
of the date of the enactment of this act, has not entered production.
43 Ibid., p. S6676.
44 For amendment text, debate, and vote, see U.S. Congress. Congressional Record, May
45 For text, debate, and vote on the two amendments, see U.S. Congress. Congressional
Record, May 20, 2003: S6690-S6691; May 20, 2003: S6691-S6696 and May 21, 2003:
S6789-S6792; and May 21, 2003: S6792, respectively.
46 U.S. Congress, Congressional Record, May 20, 2003: S6695.
47 David Morris, “Markup Reports: Panel Extends Pentagon Power over Personnel,
Wildlife,” CongressDaily, May 14, 2003: 2.
Section 3111 would revise that provision:
(b) LIMITATION- The Secretary of Energy may not develop, produce, or
provide for the development or production of a low-yield nuclear weapon
which, as of November 30, 1993, has not entered production. ...
(d) Effect on Studies and Design Work — Nothing in this section shall
prohibit the Secretary of Energy from conducting, or providing for the
conduct of, concept definition studies, feasibility studies, or detailed
engineering design work.
The amendment apparently drew the line separating permitted from prohibited
activities between phase 2a/6.2a and phase 3/6.3. The explanatory statement of the
committee was somewhat clearer on this point: “The amendment would maintain the
prohibition on development of new nuclear weapons with yields less than five
kilotons, but would allow research on such weapons, including concept definition
studies, feasibility studies, and detailed engineering design.”48 A statement of
additional views by Representative Spratt and 24 other Democrats, in the
committee’s report on the bill, elaborated: “the amendment permits research of such
[sub-5-kt] weapons, it prohibits development engineering (referred to as Phase 6.3
activities by the Department of Energy) and later stages of development.”49
The amendment is more permissive than the original provision but less so than
the one the Administration had proposed, apparently permitting work through phase
2a/6.2a but barring subsequent development. It also narrows the potentially broad
applicability of the phrase “which could lead to the production.” That phrase raises
the prospect that a paper study, a computer model, or a metallurgical experiment
might be illegal if it could somehow be construed as leading to production. With the
new language, that possibility would no longer be enough to block such work; rather,
the test would be whether the development or production is of an actual low-yield
weapon. The rule for H.R. 1588 did not provide for amendments to this provision.
The House passed the bill, as amended, on May 22, 361-68.
Conference Issues and Outcomes
Several aspects of the low-yield ban were potentially at issue in conference.
Both bills sought to define which activities were permitted and which were barred
in order to remove an ambiguity of the original legislation: the issue of R&D that
“could lead to the production by the United States of a new low-yield nuclear
weapon, including a precision low-yield warhead.” They approached the problem
differently, though. The Senate would have rescinded the low-yield R&D provision
while barring “engineering development” or subsequent phases of a low-yield
weapon without congressional approval; the House would have retained the low-
yield ban while modifying it substantially. The Senate bill, as amended, clearly drew
the line separating permitted activity and activity requiring explicit congressional
48 House Armed Services Committee, National Defense Authorization Act for Fiscal Year
49 Ibid., p. 519.
authorization between phase 2a (or 6.2a) and phase 3 (or 6.3). The House language
appeared to do the same thing, but it was not completely clear that “develop” and
“detailed engineering design work” referred to that same line. It was unclear whether
there were a difference between the Senate’s requirement for “explicit authorization”
of engineering development or beyond and the requirement in Section 3143 of P.L.
107-314 for DOE to include a line-item request for each new or modified nuclear
weapon in phase 3 or 6.3 or beyond. A senior NNSA staff member held that while
committee or conference report language would typically comment on weapons in
phase 3 or 6.3 or beyond, NNSA would view authorization and appropriations of
funds for such weapons, typically hundreds of millions of dollars, as constituting the
“explicit authorization” that the Senate’s bill required, but that in any event it would
be up to Congress to define what form of authorization sufficed.50
The House agreed to the conference report,51 362-40, on November 7, 2003.
The Senate agreed to the conference report, 95-3, on November 12. The President
signed the measure into law (P.L. 108-136) on November 24.
The conference bill, section 3116, adopted the Senate provision. Section 3116
repealed the Spratt-Furse provision, stated that the repeal shall not “be construed as
authorizing the testing, acquisition, or deployment of a low-yield nuclear weapon,”
and barred the Secretary of Energy from beginning engineering development, or
subsequent phases, of work on a low-yield weapon without specific congressional
authorization. The bill further required a report by the Secretaries of State, Defense,
and Energy assessing whether or not the repeal of Spratt-Furse “will affect the ability
of the United States to achieve its nonproliferation objectives and whether or not any
changes in programs and activities would be required to achieve those objectives.”
The legislation thus clearly delineates what R&D on low-yield nuclear weapons
is prohibited, eliminating a point of confusion in the original provision, and sets the
requirement (congressional approval) for overriding that prohibition. At the same
time, the requirement for an assessment of “any changes in programs and activities”
needed to achieve U.S. nonproliferation objectives is unclear both in terms of what
those objectives are and how broadly the phrase “any changes” is to be interpreted.
Such changes could range from developing new low-yield nuclear weapons, to
training more specialists in Arabic, to strengthening the International Atomic Energy
Agency. And what definition of nonproliferation objectives is the report to use?52
50 Telephone conversation, June 10, 2003.
51 U.S. Congress. Committee of Conference. National Defense Authorization Act for Fiscal
Year 2004. H.Rept. 108-354, 108th Congress, 1st Session, USGPO, 2003.
52 The National Strategy to Combat Weapons of Mass Destruction (U.S. White House.
December 2002), for example, lists such nonproliferation tools as active nonproliferation
diplomacy, multilateral regimes, nonproliferation and threat reduction cooperation, and
controls on nuclear materials (p. 4-5), but does not address goals except in the most general
Developments in the FY2005 Budget Cycle
It appears that low-yield R&D will be a much less salient issue in the FY2005
cycle than in the previous one. Lifting the Spratt-Furse provision was a policy issue
that was resolved last year, and not a budget issue. Some critics expressed concern
that NNSA would use the lifting of the provision to develop new weapons. In an
interview of December 2003, however, NNSA Administrator Linton Brooks stated,
“there is no list of low-yield weapons we’re thirsting to develop — that’s a
misconception.”53 An NNSA manager with responsibilities in this area elaborated
in March 2004:
We have no requirements from DoD for low yield weapon research. ... We
definitely do not have funds identified specifically for that area of research. ...
We may get other requests between now and the start of FY2005. One or more
could be for concepts that would delve into the low yield area. So I can’t and
won’t claim that none will be done in FY2005, just that as of today, we aren’t54
Issues for Congress
Would lifting the ban lead to acquisition and use of nuclear
weapons? Supporters of the ban argue that new weapons would make acquisition
and use of nuclear weapons more likely. Any new weapon would broaden the
President’s range of options. A President might find options made available by
current nuclear weapons to be unacceptable, but a new weapon might offer a tradeoff
between costs and benefits that tilts in favor of nuclear use. Critics further believe
that the Administration is seeking to lift the ban with the intention of building low-
yield weapons. Senator Feinstein asked Secretary of Defense Donald Rumsfeld
about it, “and he said it’s just a study .... They did the same thing with the nuclear
posture review: Oh, it’s just an intellectual exercise. I don’t believe either of those,
not one whit, and I think there’s a very clear march on to develop these weapons.”55
Critics further argue that even research on low-yield weapons would lead down a
slippery slope to testing, production, deployment, and use. Senator Feinstein said,
“The repeal of Spratt-Furse opens the door for America to begin to develop nuclear
weapons again.”56 Senator Biden said that many supporters of the ban are concerned
because “we believe very much that if one of these weapons ... is developed, it will
ultimately be fielded,” and that low-yield weapons send a “dangerous signal ... to57
other countries, whether intentional or not, that we intend to fight a nuclear war.”
Senator Kennedy said, “a mini-nuke is still a nuke. ... If we build it, we will use it.
53 “The Bush Administration’s Views on the Future of Nuclear Weapons: An Interview with
NNSA Administrator Linton Brooks,” Arms Control Today, January/February 2004: 3.
54 Information provided by NNSA staff, March 4, 2004.
55 Senators Edward Kennedy and Dianne Feinstein, press conference, “A New Proliferation
of Nuclear Weapons,” September 16, 2003, transcript by Federal News Service.
56 U.S. Congress, Congressional Record, May 20, 2003: S6665.
57 Ibid., p. S6673 and S6682.
... it is a one-way street that can lead only to nuclear war.”58 Critics prefer to stop the
momentum in this direction at the earliest opportunity.
Those who would lift the ban see “lowering the nuclear threshold” and a
“slippery slope to nuclear use” as a misreading of U.S. policy and practice. Congress
will decide if a weapon is to move to phase 3/6.3 through the authorization and
appropriations process; approval would be far from automatic. The provision
repealing Spratt-Furse in the Senate Armed Services Committee’s bill (S. 1050, Sec.
3131) stated explicitly, “Nothing in the repeal ... shall be construed as authorizing the
testing, acquisition, or deployment of low-yield nuclear weapons.” Senator Kyl
stated, “this is not an authorization. All we are doing is removing a self-imposed
restriction on thinking about this, on doing research.”59
Further, they note, use of nuclear weapons has always been a presidential
decision, and one that Presidents treat as arguably the most weighty decision that they
can make. Having more weapons, or low-yield weapons, does not mean that the
United States will use them. As Senator Warner observed,
the threshold for using nuclear weapons remains very high indeed. ... the United
States had a large number of low-yield nuclear weapons in our inventory during
the ‘50s, ‘60s, and ‘70s which have now been removed from the inventory.
During each of these decades there were significant national security challenges
to the United States. None of those challenges came close to reaching the60
nuclear threshold ...
Would low-yield weapons make nuclear proliferation more likely?
Those who would retain the ban assert that resuming low-yield R&D could spur
nuclear proliferation. If the world’s only superpower requires for its security new
types of nuclear weapons in addition to the ones it already has, then this implies that
other nations need nuclear weapons for their security as well. Building weapons that
might be used against rogue states will not deter these states from building nuclear
weapons, it is argued, but will instead lead them to develop such weapons of their
own to deter U.S. attack. They maintain that although nonproliferation efforts have
not been 100 percent successful, they have served as an important restraint.
Resuming low-yield R&D could lead to nuclear testing, critics argue, likely
leading to resumed nuclear testing by others. The head of the nuclear directorate of
Russia’s defense ministry reportedly stated in January 2002, “if any of the five
countries officially possessing nuclear weapons starts testing nuclear munitions
again, and we consider our nuclear stockpile to be in a critical state, we too will carry
out nuclear tests.”61 Beyond that, the NPT is the cornerstone of the nuclear
nonproliferation regime, which incorporates other agreements, treaties, groups, and
58 Ibid., p. S6687.
59 Ibid., p. S6677.
60 Ibid., p. S6687.
61 Letter from Representative Edward Markey and 39 other Representatives to
Representatives Bob Stump and Ike Skelton on Robust Nuclear Earth Penetrator and nuclear
test readiness, May 1, 2002.
arrangements as well.62 The NPT faced a key decision in 1995: states party had to
decide whether or not to extend the treaty indefinitely. They decided to do so, but
this extension was linked to a set of nonproliferation objectives, one of which was
completing negotiations on a CTBT no later than 1996. Accordingly, as Sidney Drell
and others argue, “A decision to resume [nuclear] testing to build low-yield nuclear
weapons could deal the [nonproliferation] regime a fatal blow ...”63
Those opposed to the ban reject the argument that studying low-yield weapons
would lead to nuclear proliferation. No other nation has such a prohibition, so lifting
it would simply move the United States to a position of parity in that regard, they
assert. Nations will act in their own interests on decisions involving their security
or ambitions. The United States signed the CTBT in 1996, yet India and Pakistan
tested nuclear weapons in 1998. These critics question the NPT’s utility. North
Korea and (apparently) Iran made substantial progress on their nuclear programs even
while being party to that treaty. Nuclear programs in all four nations were underway
decades ago, long before the U.S. ban on low-yield R&D. The ban’s opponents
argue that U.S. actions — especially on something as remote as preliminary studies
of possible weapons — will not make other nations more likely to develop WMD.
To the contrary, they believe, U.S. ability to respond to WMD proliferation efforts
by rogue states may dissuade them from undertaking such actions in the first place.
As Senator Domenici said, “to permit [weapons scientists] to work in this area [low-
yield weapons] is part of the deterrent.”64
The ban’s opponents reject the contention that R&D on new weapons will lead
to nuclear testing. As C. Paul Robinson, Director of Sandia National Laboratories,
said, “I can categorically state that no one is proposing returning to nuclear testing.”65
The United States has decades of experience with low-yield nuclear weapons,
including those with altered radiation outputs, such as the neutron bomb of the 1970s.
Use of existing, tested designs, perhaps with modifications not requiring testing,
would provide high confidence. A better understanding of the effects of low-yield
weapons is crucial for understanding their utility, yet key variables are how a blast
wave would interact with structures; computer models and engineering studies would
arguably provide the needed data.
Would low-yield weapons offer military value? Some supporting the
ban argue there is little meaningful use for low-yield weapons. They note a statement
by Linton Brooks, at the time Acting Administrator of NNSA: “we have no
62 For further information on the nuclear nonproliferation regime, see CRS Report RL31559,
Proliferation Control Regimes: Background and Status, by Sharon Squassoni, Steven
Bowman, and Carl Behrens.
63 Sidney Drell, James Goodby, Raymond Jeanloz, and Robert Peurifoy, “A Strategic
Choice: New Bunker Busters Versus Nonproliferation,” Arms Control Today, March 2003:
64 U.S. Congress, Congressional Record, May 20, 2003: S6674.
65 James Kitfield, “The Pros and Cons of New Nuclear Weapons,” National Journal, August
requirement to actually develop any new weapons at this time.”66 As the Afghanistan
and Iraq Wars showed, it is often difficult to know what to target. Al Qaeda presents
few if any targets that would be suitable for nuclear weapons. Another lesson of
these wars is that if targets can be located, they can often be defeated with
conventional weapons. Deep burial, large tunnel complexes with multiple barriers,
deception, etc., might make targets in rogue states difficult to destroy with nuclear
weapons even if they can be located. Many potential targets are in cities, but
detonating a low-yield nuclear weapon in or even near a city could cause much
collateral damage. By one estimate, a 5-kiloton weapon detonated near and upwind
from Damascus, Syria, at a depth of 30 feet would cause 230,000 fatalities and
another 280,000 casualties within two years.67 Use of a low-yield earth penetrator
against the bunkers thought to house Saddam in Baghdad, a city of nearly 5 million
people, could have caused casualties on a similar scale.
The ban’s opponents claim that low-yield weapons may offer important military
capabilities, such as for destroying biological munitions (discussed under ACI,
below) or attacking underground bunkers. There is military interest in low-yield
weapons. Admiral James Ellis, Commander, U.S. Strategic Command, wrote that
“US Strategic Command is interested in conducting rigorous studies of all new
technologies, and examining the merits of precision, increased penetration, and
reduced yields for our nuclear weapons.”68 At present, in the view of those who
would lift the ban on low-yield R&D, the ban makes it impossible to assess the value
of the capabilities sub-5-kt weapons might bring to the stockpile, or how they might
be designed for specific missions. The ban, they assert, imposes broader constraints
on nuclear weapons designers as they seek to stay well away from any activities that
could be construed as violating the Spratt-Furse provision. Senator Sessions
concluded, “it would be irrational to prohibit research that could inform future
decisions as to whether such weapons would enhance the national security of our
66 “Statement of Linton F. Brooks, Acting Under Secretary of Energy and Administrator for
National Security, National Nuclear Security Administration, U. S. Department of Energy
Before the Subcommittee on Strategic Forces, Committee on Armed Services, U. S. Senate,
Apr.il 8, 2003,” p. 6. Available on the Senate Armed Services Committee website at
67 U.S. Congress, Congressional Record, May 20, 2003: S6666. Senator Kennedy cited this
estimate. Matthew McKinzie, staff scientist on the nuclear program of the Natural
Resources Defense Council, performed the analysis in April 2003 using the HPAC (“Hazard
Prediction Assessment Capability”) computer code produced by SAIC, Inc., for the Defense
Threat Reduction Agency. Telephone conversation with Matthew McKinzie, September 23,
2003. For further information on the code and calculations using it, see McKinzie’s
statement in “New Nuclear Weapons vs. Nonproliferation: The Choice Before Congress,”
press briefing held by the Arms Control Association, Washington, DC, April 29, 2003:
[ ht t p: / / www.ar ms cont r ol .or g/ event s / newnucl e ar weapons_apr 03.asp] .
68 Letter from Admiral Ellis to Senator John Warner, (no date), in U.S. Congress.
Congressional Record, May 20, 2003: S6680.
country.”69 Senator Domenici noted that the ban increases the leadtime to develop
a low-yield weapon should the need arise.70
Advanced Concepts Initiative
For FY2004, the Administration requested $21.0 million for the Advanced
Concepts Initiative (ACI). Of this amount, $15.0 million was for the Robust Nuclear
Earth Penetrator (RNEP) weapon, discussed in the next section, and $6.0 million
($2.0 million for each of the three weapons labs) was for other advanced weapon
concepts, discussed here. This latter program involves studies through phase 2a/6.2a
by small groups of weapons personnel at Los Alamos, Livermore, and Sandia, as well
as liaison with DOD commands. A Los Alamos publication stated:
The ACI program will conduct new weapon studies and will explore concepts for
new warhead designs and modifications to meet DoD needs that are not met by
the current stockpile. The ACI is a program for developing and exercising71
capability and for applying that capability to examine options.
In its FY2004 legislative proposals, the Administration argued for ACI:
A revitalized nuclear weapons advanced concepts effort is essential to: (1)
train the next generation of nuclear weapons scientists and engineers; and (2)
restore a nuclear weapons enterprise able to respond rapidly and decisively to
changes in the international security environment or unforeseen technical72
problems in the stockpile.
The FY2004 NDAA and EWDAA both provided $6.0 million, as requested, but
conferees on the latter made $4.0 million available for authorization only after
delivery of a DOE-DOD report on the nuclear stockpile and a 90-day congressional
review period. As of early March 2004, the report had not been delivered, and
NNSA could offer no indication as to when that might occur. As a result, ACI may
be limited to a budget of $2 million for much if not all of FY2004. The FY2005
request for ACI is $9.0 million.
69 U.S. Congress, Congressional Record, May 20, 2003: S6667.
70 Ibid., p. S6675.
71 John Immele, “Los Alamos and the New Triad,” Nuclear Weapons Journal, May/June
72 U.S. Department of Defense, DoD Legislative Proposals, National Defense Authorization
Act for Fiscal Year 2004. Sent to Congress March 3, 2003. Section-by-Section Analysis, p.
History Through the FY2003 Budget Cycle
During much of the Cold War, the three weapons laboratories had teams of
scientists and engineers studying advanced weapons concepts. Studies ranged from
modifications of existing weapons, to improvements for next-generation weapons,
to exploration of new weapons technologies and weapons for new missions. Tools
used included laboratory experiments, computer modeling, data from past and
ongoing nuclear tests, and data from experiments the teams added to nuclear tests.
These teams were small, perhaps a half-dozen full-time professional staff, but during
the Cold War they drew on the large advanced and exploratory R&D program that
was a major part of the core nuclear weapons programs at the laboratories. Team
members, new staff as well as experienced designers, would typically work on
advanced concepts for a year or two. Team members would also interact with DOE
headquarters, DOD elements, contractors, and others.
With the end of the Cold War and the end of nuclear weapons development, the
laboratories wound down their advanced concepts programs in the middle 1990s.
Since then, there have been statements emphasizing the value of the type of work that
advanced concepts programs performed, and subsequent calls for an ACI to
reestablish advanced concepts programs at the labs. ACI would recreate the small
teams, but would not reorient the bulk of the nuclear weapons program, which is
focused on sustaining current weapons through the Stockpile Stewardship Program.
The Chiles report of 1999 stated, “DOE should also encourage the laboratories to
continue their decades old practice of exploratory development programs since these
programs have allowed experienced engineers and scientists to maintain their
systems engineering skills and train new employees.”73 The Foster panel’s FY1999
report stated: “The nuclear weapons complex should work on a range of design and
development tasks that exercise and sustain the capability to produce new weapon
designs. This provides both a broader set of technical options to meet future needs,
and a program for training new generations of stewards.”74
The purportedly leaked version of the Nuclear Posture Review stated:
There are several nuclear weapon options that might provide important
advantages for enhancing the nation’s deterrence posture: possible modifications
to existing weapons to provide additional yield flexibility in the stockpile;
improved earth penetrating weapons (EPWs) to counter the increased use by
potential adversaries of hardened and deeply buried facilities; and warheads that
reduce collateral damage.
To further assess these and other nuclear weapons options in connection with
meeting new or emerging military requirements, the NNSA will reestablish
advanced warhead concepts teams at each of the national laboratories and at
headquarters in Washington. This will provide unique opportunities to train our
73 U.S. Commission on Maintaining United States Nuclear Weapons Expertise, Report to
Congress and the Secretary of Energy. March 1, 1999, p. 32.
[ ht t p: / / www.f a s.or g/ nuke/ gui de/ usa/ doct r i ne/ doe/ c hi l e sr pt .pdf ]
74 John Foster, Chairman, et al., FY 1999 Report of the Panel to Assess the Reliability,
Safety, and Security of the United States Nuclear Stockpile, November 8, 1999, p. 12.
next generation of weapon designers and engineers. DoD and NNSA will also
jointly review potential programs to provide nuclear capabilities, and identify
opportunities for further study, including assessments of whether nuclear testing75
would be required to field such warheads.
The Administration now proposes to re-create these teams.
In July 2003, NNSA provided details on the FY2004 studies that the non-RNEP76
part of ACI would conduct. Perhaps $200,000 would go for participation by the
weapons labs in an Air Force concept study of enhancing and modernizing a current
warhead, such as examining if any additional military characteristics are useful and
feasible. The balance would go for other laboratory studies. Specifically, Los
Alamos would spend its funds on enhancing computer modeling and simulation
capability and for mechanical testing. Sandia would continue a study of “mission
end-to-end command and control,” which would enable military personnel to know
where a warhead was at all times. This would be applicable to conventional as well
as nuclear weapons. For example, if a laser-guided bomb lost track of its laser signal
because of fog or smoke, end-to-end command and control would indicate its
location. Livermore would undertake an early study on “design to effect” (i.e.
designing nuclear weapons to obtain desired output characteristics). These projects
may or may not support research directly on low-yield nuclear weapons.
Legislative Actions in the FY2004 Budget Cycle
For FY2004, the Senate Armed Services Committee recommended authorizing77
the full $21.0 million requested for Advanced Concepts. The House Armed
Services Committee stated that ACI had several important purposes, such as
exercising the weapons design process, training the next generation of nuclear
weapons scientists and engineers, and understanding what adversaries might do in
the area of nuclear weapon design. Accordingly, the committee expressed its
“[belief] that NNSA should consider more significant future investment in these
[Advanced Concepts] activities” and “recommends that NNSA proceed with its
advanced concepts initiative forthwith.”78 No amendments were offered on non-
RNEP ACI in the Senate. The House rejected an amendment by Representative
Tauscher to transfer all funds from ACI (including funds for RNEP) to conventional
programs to defeat hardened and deeply buried targets, as discussed under RNEP,
below. As a result, non-RNEP ACI was not at issue in the conference.
In its report on H.R. 2754, the FY2004 Energy and Water Development
Appropriations Bill, the House Appropriations Committee delivered a scathing
75 Nuclear Posture Review [Excerpts], p. 34-35.
76 Information provided by a DOE official, July 23, 2003.
77 U.S. Congress, Senate Committee on Armed Services, National Defense Authorization Act
for Fiscal Year 2004, S.Rept. 108-106, p. 442.
78 U.S. Congress, House Committee on Armed Services, National Defense Authorization Act
for Fiscal Year 2004, H.Rept. 108-106, p. 429-430.
criticism of the nuclear weapons program. The committee found that a flawed
budget process raised questions about the legitimacy of nuclear weapon requests:
Unfortunately, the country possesses neither a modern stockpile nor a modern
nuclear weapons complex. Instead, both are largely carryovers from the Cold
War era. After careful consideration, the Committee has concluded that much of
the current situation results from a flawed budget process. ... the weapons
activities portion of the NNSA budget is effectively insulated from any such
[intra-agency] tradeoffs — DoD sets requirements that another agency has to
fund, and DOE treats the weapons activities budget as untouchable because DoD
set the requirements.
There needs to be a serious debate about whether the approximately $6
billion spent annually on DOE’s nuclear weapons complex is a sound national
security investment. Until that debate occurs and the DOE weapons budget
request is subject to meaningful budget trade-offs, this Committee will not
assume that all of the proposed nuclear weapons requests are legitimate79
This disconnect, the committee stated, led to new weapon-related initiatives,
such as ACI, RNEP, and enhanced test readiness, that overstretched NNSA’s
It appears to the Committee the Department [of Energy] is proposing to rebuild,
restart, and redo and otherwise exercise every capability that was used over the
past forty years of the Cold War and at the same time prepare for a future with
an expanded mission for nuclear weapons. Nothing in the past performance of
the NNSA convinces this Committee that the successful implementation of
Stockpile Stewardship program is a foregone conclusion, which makes the
pursuit of a broad range of new initiatives premature. Until the NNSA has
demonstrated to the Congress that it can successfully meet its primary mission
of maintaining the safety, security, and viability of the existing stockpile by
executing the Stockpile Life Extension Program and Science-based Stewardship
activities on time and within budget, this Committee will not support redirecting80
the management resources and attention to a series of new initiatives.
For these reasons, it recommended eliminating the $6.0 million requested for
ACI. The House passed H.R. 2754 on May 18, 2003, 377-26; no amendments were
offered on ACI, RNEP, test readiness, or any other Weapons Activities provisions.
(Weapons Activities is the part of NNSA’s budget that funds stockpile stewardship.)
The Senate Appropriations Committee recommended providing the full amount
requested for ACI.81 The Senate considered several amendments on ACI on
79 U.S. Congress, House Committee on Appropriations, Energy and Water Development
Appropriations Bill, 2004. H.Rept. 108-212, 108th Congress, 1st Session, USGPO, 2003, p.
80 Ibid., p. 145.
81 U.S. Congress. Senate. Committee on Appropriations. Energy and Water Development
Appropriation Bill, 2004. S.Rept. 108-105, 108th Congress, 1st Session, USGPO, 2003, p.
September 16. It tabled, 53-41, an amendment by Senator Feinstein to eliminate the
full $21.0 million for RNEP and other ACI, and for other purposes.82 It adopted on
voice vote an amendment by Senator Jack Reed that barred use of funds provided by
H.R. 2754 for phase 3 or 6.3 or beyond for advanced nuclear weapon concepts
including RNEP.83 Senator Reed stated that his amendment “would assure that the
appropriations bill is consistent with the language [on RNEP by Senator Nelson of
Florida] that is included in the FY2004 Defense authorization bill.”84 The Senate
agreed to various amendments en bloc, including one by Senator Bingaman that
barred spending funds provided by the bill on additional and exploratory studies (a
category that excludes RNEP) under ACI until 30 days after NNSA gives Congress
a detailed report on activities planned in that category for FY2004.85 The Senate
passed the bill, H.R. 2754, on September 16, 92-0.
Conference Issues and Outcomes
Non-RNEP ACI was not expected to be at issue in the defense authorization
conference because each House provided the same amount for it. Senate conferees
could, however, have considered whether to endorse the House language regarding
the program’s future budget and proceeding, modify it, or be silent on it. Because of
congressional interest in nuclear weapon programs, conferees could also have
considered directing DOE to specify in detail the projects that future-year ACI
requests will support.
Non-RNEP ACI was at issue in the energy and water conference. A point of
possible contention was whether conferees could arrive at a compromise somewhere
between the House and Senate figures, or if the funding request was so small relative
to other NNSA projects that a cut of, say, 50 percent would prevent one or two of the
three labs from performing its planned study. Also at issue were the two
amendments the Senate passed.
An issue that implicitly faced energy and water conferees was the gulf between
the House Appropriations Committee report, which argued that “pursuit of a broad
range of new [nuclear] initiatives [was] premature,” and the House Armed Services
Committee report, which urged NNSA to “consider more significant future
82 For amendment text, debate, and vote, see U.S. Congress. Congressional Record,
September 15, 2003: S11436; September 15, 2003: S11436-S11451 and September 16,
83 For amendment text, debate, and vote, see U.S. Congress, Congressional Record,
September 16, 2003: S11534, S11534-S11538, and S11538, respectively.
84 U.S. Congress, Congressional Record, September 16, 2003: S11535.
85 For amendment text and en bloc agreement, see U.S. Congress. Congressional Record,
September 16, 2003: S11541 and S11539, respectively.
investment” in ACI (including RNEP).86 Further, the authorization bill as passed by
the House included the $6.0 million requested, while the appropriations bill as passed
by the House eliminated those funds.
The section on the low-yield R&D ban, above, gave details on final passage of
the NDAA. Regarding the EWDAA, the House agreed to the conference report, 387-
36, on November 18, and the Senate agreed to it on the same day by unanimous
consent. The President signed the measure into law (P.L. 108-137) on December 1.
The NDAA provided the amount requested for non-RNEP ACI, $6.0 million.
It was not at issue in conference, and the conference report did not comment on it.
The EWDAA provided $6.0 million for Advanced Concepts (excluding RNEP),
but the conferees made $4.0 million available for obligation only after the Secretaries
of Energy and Defense deliver to Congress a revised Nuclear Weapons Stockpile
plan detailing a plan and schedule for achieving the President’s proposed adjustments
to the strategic weapons stockpile (including a reduction in operationally deployed
weapons to 1,700-2,200 by 2012), and 90 days elapse for review by the Armed
Services and Appropriations Committees.87 The DOE-DOD report was due with the
submission of the FY2005 budget request in early February 2004, but the two
departments had not submitted it a month later.88 This provision therefore limits ACI
funds to $2.0 million until the report is submitted. Further, conferees included as
section 313 the amendment by Senator Bingaman barring spending on ACI until 30
days after NNSA submits a report detailing certain planned ACI activities. NNSA
submitted the report on March 3, 2004, so the activities are suspended until April 2.89
While the EWDAA provided the Senate amount for ACI and not the House
amount, the conferees’ language requiring a DOE-DOD report on the link between
the stockpile plan and the President’s planned nuclear force structure addressed a
central concern expressed in the House Appropriations Committee’s report:
While the conventional forces in the Defense Department go through a 21st-
Century transformation to meet the challenges of a new era, the NNSA is forced,
through inertia and indecision, to maintain all contingencies regardless of how
unlikely the threat. The Department of Defense needs to determine the
composition of the stockpile required to support the President’s announced
stockpile reductions, and then coordinate with DOE to establish the nuclear
86 U.S. Congress, House Committee on Appropriations, Energy and Water Development
Appropriations Bill, 2004, H.Rept. 108-212, p. 145; and U.S. Congress. House. Committee
on Armed Services. National Defense Authorization Act for Fiscal Year 2004, S.Rept. 108-
87 U.S. Congress, Committee of Conference, Making Appropriations for Energy and Water
Development for the Fiscal Year Ending September 30, 2004, and for Other Purposes.thst
H.Rept. 108-357, 108 Congress, 1 Session, USGPO, 2003, p. 154, 156.
88 Information provided by NNSA staff, March 4, 2004.
89 Information provided by NNSA staff, March 4, 2004.
weapons complex requirements based on deliberate, timely, well-justified90
decisions supported by Congress.
Regarding the amendment by Senator Reed on RNEP, the conference report
contained the following language: “The conferees remind the Administration that
none of the funds provided may be used for activities at the engineering development
phases, phase 3 or 6.3, or beyond, in support of advanced nuclear weapons concepts,91
including the Robust Nuclear Earth Penetrator.”
The FY2005 Budget Cycle
The FY2005 request is $9.0 million. ACI activities include
developing advanced concepts which could be applied to the stockpile of the
future, code development for system-specific nuclear effects, phenomenology,
and exercise of design skills; conducting pre-conceptual, conceptual, feasibility,92
design and costing studies of options.
NNSA Administrator Brooks said that ideas for ACI include the ability to
destroy biological agents, improved safety and security, and more capable or robust
weapon designs that might be developed without testing. He felt that the effort
probably “will be focused on safety, security flexibility, greater margins than on93
fundamental new capabilities.”
NNSA’s FY2005 budget request document includes, for the first time, budgets
for out years (in this case FY2006-FY2009) as well as for the current year. The
requests projected for ACI are: FY2006: $14.4 million; FY2007, $14.9 million;
FY2008, $14.6 million; and FY2009, $29.5 million.94
Issues for Congress
The Administration, DOD, NNSA, the nuclear weapons laboratories, the Foster
panel, and others see many benefits for ACI. Many are fine-grained in their detail.
The labs in particular have a detailed sense of what they would undertake through
ACI; many of the arguments for ACI presented here are based on discussions with
90 House Appropriations Committee, Energy and Water Development Appropriations Bill,
91 Committee of Conference, Making Appropriations for Energy and Water Development
for the Fiscal Year Ending September 30, 2004 ..., p. 156.
92 U.S. Department of Energy, Office of Management, Budget and Evaluation/CFO, FY 2005
Congressional Budget Request, vol. 1, National Nuclear Security Administration. DOE/ME-
93 “The Bush Administration’s Views on the Future of Nuclear Weapons: An Interview with
NNSA Administrator Linton Brooks,” Arms Control Today, January/February 2004: 3.
94 U.S. Department of Energy, Office of Management, Budget and Evaluation/CFO, FY 2005
Congressional Budget Request, vol. 1, National Nuclear Security Administration. DOE/ME-
weapons lab personnel. Whether because of the relatively small amount of money
involved, uncertainty over which projects ACI would undertake, or the attention
drawn to RNEP, test readiness, and the ban on low-yield R&D, opponents of those
three issues made few comments on ACI as a whole. Consequently, due to the
paucity of public discourse on ACI, some counterarguments come from critics of the
program, but others are presented as they might be made by the critics.
Does deterrence require new types of nuclear weapons? A key
aspect of the debate over ACI — and over low-yield weapons and RNEP as well —
is the relationship between deterrence and new weapons. George Miller of
Deterrence is a dynamic concept — it has to continually evolve and be able to
respond to changes that our adversaries will make in order to be relevant and
effective. That’s what happened in the Cold War. The real question that is being
debated is whether nuclear weapons will be allowed to develop in response to the
changing world situation. If they are to evolve, modifications to the Cold War
arsenal will be required if for no other reason than to limit collateral damage. If
they are not to evolve, the nuclear arsenal will become irrelevant and will die95
away. Standing still is not stable.
Critics assert that the only role of nuclear weapons is to deter other nations from
using nuclear weapons against the United States, or perhaps against its friends and
allies. For this purpose, existing weapons are more than sufficient and their use is
fully credible. In this view, there would be no need for new weapons.
With the threat of a U.S.-Russian nuclear war very low, the Administration is
seeking something from nuclear weapons beyond deterrence. A goal set forth in the
Quadrennial Defense Review is that “[w]ell targeted strategy and policy can ...96
dissuade other countries from initiating future military competitions.” The Nuclear
Posture Review, amplifying on this point, stated that a “[d]iverse portfolio of97
capabilities denies payoff from competition.” The Administration further argues
that preemption may be needed to counter threats of WMD, and that the United
States may use “all of our options” in responding to an attack using WMD. While
these two statements do not state that the United States would use nuclear weapons
preemptively, neither do they rule out such use.
Another goal of the current Administration involves having potential adversaries
take certain actions, such as having Iran open its nuclear facilities to inspection,
having North Korea verifiably halt all work related to nuclear weapons, or (until
recently) having Iraq cooperate fully with inspectors. Ideally, these goals would be
achieved through diplomatic, economic, or similar means, but if those means do not
suffice, preemptive force might be threatened or, as in Iraq, used. Coercion — the
95 Personal communication, September 5, 2003.
96 U.S. Department of Defense, Quadrennial Defense Review Report. September 30, 2001,
p. 12. [http://www.defenselink.mil/pubs/qdr2001.pdf]
97 U.S. Department of Defense. Findings of the Nuclear Posture Review, January 9, 2002,
slide, “QDR: Defense Policy Goals.”
threat to use force against states if they do not take certain actions — goes beyond
deterrence, which threatens to use force against states if they do take certain actions.
Coercion works earlier in the threat chain, with the intention of eliminating (through
diplomacy, threat, or force) activities that could create threats rather than forestalling
the use of existing capabilities, one argument being that it is hard to know what is
needed to deter a rogue state.
A shift from deterrence to something more, whatever form that may take, could
lead to a change in weapons desired to implement the new policy. Conventional or
nuclear forces credible for deterrence might not be credible under the new approach.
Would any regime believe a U.S. threat to use nuclear weapons? New nuclear
weapons — especially lower-yield weapons tailored to destroy specific targets —
render such threats more credible, supporters argue.
ACI’s supporters believe that the ability to respond to potential threats, as ACI
is intended to do, would promote U.S. security. John Gordon, former NNSA
Administrator, said, “a demonstrated ability to design, develop, and produce new
warheads, including small builds of special purpose weapons, could be an important
element in our overall deterrent posture. Such capabilities could act to convince an
adversary that it could not expect to negate U.S. nuclear forces, for example, by
seeking to house vital command and control functions in hard, deeply-buried
installations.”98 U.S. nuclear weapons able to hold at risk such targets could be used
not only for deterrence but also for coercion and preemption.
In the view of critics, deterrence and something more come from threat of
regime change, and that threat comes from conventional, forces, not nuclear forces.
New weapons, in this assessment, might not enhance deterrence even if tailored for
specific missions, as ACI might do, because rogue states would see use of new
weapons as no more credible than use of current weapons. Critics cite a 1992 letter
by the first Bush Administration setting forth a presidential decision “to modify U.S.
nuclear testing policy immediately, to impose limits on the number, purpose and
yield of our tests. The purpose of all U.S. nuclear tests of our weapons will
henceforth be for the safety and reliability of our deterrent forces.”99 Omitted from
this statement was weapons development as a purpose for testing. As developing
new weapons typically required testing, critics interpret the statement as indicating
that there was no need for new nuclear weapons. If existing weapons were sufficient
for U.S. security in 1992, they assert, these same weapons should suffice now. At
that time, U.S. nuclear weapons addressed as lesser included cases targets that are of
current concern, such as deeply buried targets and biological weapon storage sites.
Additionally, critics oppose the Administration’s policy of preemption and see
98 Letter from John Gordon, Administrator, National Nuclear Security Administration, to the
Honorable John W. Warner, Chairman, Committee on Armed Services, United States
Senate, May 17, 2001, p. 2.
99 Letter from [National Security Advisor] Brent Scowcroft, [Secretary of Defense] Dick
Cheney, and [Secretary of Energy] James Watkins to Honorable J. Bennett Johnston,
Chairman, Subcommittee on Energy and Water Development, Committee on
Appropriations, United States Senate, July 10, 1992.
concurrent requests for buttressing U.S. nuclear capabilities as a dangerous part of
Do ACI programs offer significant military value? There have been
several suggestions for new types of nuclear weapons in addition to earth penetrator
weapons (EPWs). General Richard Myers, Chairman, Joint Chiefs of Staff, stated,
“In terms of anthrax, it’s said that gamma rays can, you know, destroy the anthrax
spores, which is something we need to look at.”100 Kathleen Bailey, former Assistant
Director for Nonproliferation, Arms Control and Disarmament Agency and Robert
Barker, former Assistant to the Secretary of Defense for Atomic Energy, suggest
several new warheads: “[w]arheads with suppressed electromagnetic pulse for more
effective missile defense; [r]educed residual radiation warheads for low collateral101
damage; and [r]obust warheads for longer shelf-life.” Another possibility would
be warheads for intercepting missiles armed with chemical or biological agents.102
If these ideas are of mutual interest to DOE and DOD, projects of this sort would
appear to fall within the purview of ACI.
Critics question the value or feasibility of such weapons. Peter Zimmerman, a
physicist, consultant to the Senate Foreign Relations Committee, and formerly
science adviser for arms control at the State Department, writes: “Reduced residual
radiation weapons would be similar to the ‘neutron’ bombs, made in the 1970s, with
very low fission yield. For classified reasons they cannot be readily made today
without compromise to higher priority programs. Suppressing electromagnetic pulse
is very difficult because that would require capturing the many high-energy electrons
liberated by a nuclear explosion. Ultra-robust warheads could have been developed
and tested decades ago, but weren’t, perhaps because the nuclear weapons103
community feared that that could have ended the need to test.” The Administration
and Congress appear to have little if any interest in nuclear-armed interceptors, as
In a 1989 report, DOE explained its low interest in more robust warheads. It
stated that the nuclear weapon laboratories were studying the feasibility of such
warheads, which would have reduced sensitivity to aging, to material properties that
might change slightly during remanufacture, to tolerances of manufacture, etc. “We
cannot predict the success of this research program,” DOE said, and noted that “such
warheads would be less operationally efficient than the optimized warheads in
100 U.S. Department of Defense. “DoD News Briefing — Secretary Rumsfeld and Gen.
Myers.” May 20, 2003. [http://www.defenselink.mil/transcripts/2003/tr20030520-
101 Kathleen Bailey and Robert Barker, “Why the United States Should Unsign the
Comprehensive Test Ban Treaty and Resume Nuclear Testing,” Comparative Strategy,
April-June 2003: 134.
102 See Hans Kruger and Edgar Mendelsohn, “Neutralization of Chemical/Biological
Ballistic Warheads by Low-Yield Nuclear Interceptors,” Lawrence Livermore National
Laboratory publication UCRL-ID-110403, August 1992, and Hans Kruger, “Defense
Against Biological or Chemical Bomblet Warheads with Nuclear Interceptors,” UCRL-ID-
103 Personal communication, September 2, 2003.
current use and have some size and weight penalties with attendant cost and
performance penalties to weapon delivery systems.”104 Now, though, such penalties
are arguably less important than they were during the Cold War — and ultra-robust
warheads accordingly more desirable — because the United States plans to reduce
the number of warheads carried by operationally deployed ballistic missiles,105 and
bombers can be loaded with fewer bombs and missiles, making extra carrying
capacity available on these systems.
Is it technically feasible to use nuclear weapons to destroy stored
bioweapon stockpiles? If so, is such use advisable? One concept that
ACI might explore would be nuclear weapons for destroying chemical or biological
warfare agents. ACI supporters advance it as a main reason for studying new or
modified weapons, yet there is much less understanding of this topic than of RNEP.
This section deals with attacks on biological warfare (BW) agents because some of
them, such as anthrax spores, are more lethal than the most toxic chemical warfare
agents on a weight-for-weight basis. It first addresses the technical feasibility of such106107
attacks, drawing heavily on the work of Hans Kruger and Jonathan B. Tucker.
Kruger is a physicist and an authority on the effects of nuclear weapons on chemical
and biological agents. He led Livermore’s weapons effects group from around 1973
to 1999, when he retired, and is currently working at Livermore on these issues.
Tucker is a Senior Researcher at the Center for Nonproliferation Studies, Monterey
Institute of International Studies. He has written extensively on BW and was a U.N.108
biological weapons inspector in Iraq. After considering feasibility, this section
addresses the advisability of such attacks, taking into account political, intelligence,
and collateral-damage issues, and then offers several concluding observations.
Feasibility. In calculating the effects of nuclear weapons on BW agent,
Kruger makes several assumptions about the target.109 The agent would be stacked
in steel barrels or larger containers in various configurations, stored in the open or in
a typical warehouse. It would be in the form of a water slurry,110 for which he used
104 U.S. Department of Energy. “Annual Report to Congress, Volume I: Program Status of
Preparations for Further Limitations on Nuclear Testing, Required by Section 1436 of the
National Defense Authorization Act, FY1989,” December 1989, p. 16.
105 U.S. Department of Defense, “Findings of the Nuclear Posture Review,” January 9, 2002,
slide titled, “NPR Decisions Made.”
106 Personal communication, September 16, 19, 23, and 24, and October 7, 2003.
107 Personal communication, September 30, October 1, and October 4, 2003.
108 See also Robert Nelson, “Nuclear ‘Bunker Busters’ Would More Likely Disperse than
Destroy Buried Stockpiles of Biological and Chemical Agents,” c. 2003, 20 p., available at
109 These assumptions are detailed in H. Kruger, Radiation-Neutralization of Stored
Biological Warfare Agents with Low-Yield Nuclear Warheads, UCRL-ID-140193, August
110 Personal communication, October 7, 2003.
water as a surrogate for the purpose of computer modeling.111 He stated that
“[b]iological agent is typically stored in barrels or larger storage containers.”112 He
further stated that BW agent in a water slurry is a conservative assumption because
less radiation would reach the bottom of a stack of barrels filled with water slurry
than with, e.g., dry anthrax powder.113
While there may be various ways to neutralize BW agent without the
cooperation of the country possessing such agent, Kruger argues that using a nuclear
weapon to neutralize BW agents requires applying a lethal dose of radiation to all
parts of the storage site. Nuclear weapons produce various forms of radiation. X-
rays are stopped near the top of the material because they do not have enough energy
to penetrate further. Gamma rays (electromagnetic radiation of higher energy than
x-rays) penetrate more deeply. One study finds gamma rays of limited value: In two
cases studied (attacking BW agent in an underground bunker, one case with a nuclear
explosion inside the bunker and the other with the explosion in the ground but
outside the bunker), “the amount of bio-agents effectively irradiated by this process
[gamma rays] will in all likelihood be a small fraction of the total.”114 Neutrons are
the most effective mechanism. They are more penetrating than gammas. In addition,
neutrons generate gamma rays when they strike the target material; gammas
generated far down in the stack of stored agent can penetrate to the bottom of the
stack. The key to calculating the effects of a nuclear weapon on BW agents, then,
according to this argument, is the number of neutrons and their energy level; a dose
of neutrons and gamma rays sufficient to kill BW agent at the bottom of the stack
would expose the rest of the stack to a much greater dose.
The fireball of a nuclear explosion would probably incinerate BW agents, but
radiation is a more useful kill mechanism from a military point of view because it can
be calculated much more precisely if one knows the conditions under which the agent
is stored. Currently-deployed nuclear weapons, which use a combination of fission
and fusion to produce an explosion, must be detonated close to the ground —
creating much fallout — in order to kill BW agents; for a greater height of burst, the
number and energy level of neutrons produced would be insufficient for this task.115
111 Kruger, Radiation-Neutralization of Stored Biological Warfare Agents ..., p. 3.
112 Ibid., p. 2.
113 Personal communication, October 7, 2003.
114 Michael May and Zachary Haldeman, "Effectiveness of Nuclear Weapons against Buried
Biological Agents Targets," Center for International Security and Cooperation, Stanford
University, May 2003, p.14. Available at [http://www.ciaonet.org/wps/mam09/mam09.pdf].
115 Kruger provided the following details: “most of the fallout is from the radioactive fission
products (the two pieces into which a uranium or plutonium atom splits when it fissions)
that is initially vaporized and then solidifies on the solid matter ‘scooped up’ from the
surface. ... think of it as a vapor plating type of process. This results in relatively large
particulates that drop sufficiently fast to reach the ground within a few to a few tens of
kilometers from ground zero. The same amount of radioactive fission products is present
when the detonation occurs at such a height that no ground material is scooped up.
However, now the fission products form much smaller particles that have a much lower
Nuclear weapons can be made, however, that generate more neutrons. Such a
weapon is a reduced blast/enhanced radiation weapon, such as the “neutron bomb”
of the 1970s. (Neutron bombs were controversial and are discussed further below.)
For a given yield, these weapons produce an order of magnitude more neutrons than
do standard low-yield nuclear weapons, and the neutrons have a higher energy level.
Kruger calculates that a 10-kt weapon of this type burst at a height of 10 meters
would neutralize BW agents to a radius of about 50 meters, vs. 10 meters for a 10-kt
fission weapon at the same height of burst.116 (For comparison, the Hiroshima bomb
had a yield of 15 kt.) He notes several consequences of the increased neutron flux:
a fusion warhead [e.g., a neutron bomb] has a larger sterilization area than a
fission warhead of equal yield. Another (to me: significant) consequence is that
I can detonate a fusion warhead at an altitude above the ground (higher than the
“optimum” altitude) sufficiently high that the fireball does not touch the ground,
and still produce a sterilizing dose at the very bottom of the barrel stack over a
militarily significant area. Avoiding fireball contact with the ground eliminates117
most of the local fallout. This is not feasible for a fission warhead.
If an adversary stored BW agent in an underground bunker, the surrounding
earth, steel, and reinforced concrete would significantly attenuate the neutrons from
a weapon that burst far enough from the bunker (vertically or horizontally) that the
fireball did not enter it. To destroy the agent, an earth penetrator weapon (EPW)
would have to place the fireball inside the bunker. The chief kill mechanism would
be the heat of the fireball, the effectiveness of which would be increased because the
fireball would (at first) be concentrated within the structure. The delayed radiation
resulting from the fission debris would expose any remaining agent to a further
Critics question Kruger’s postulated target as optimized for destruction by a
nuclear weapon. The next few paragraphs focus on anthrax because it is one of the
tougher BW agents to kill and is also one of the most militarily useful agents.
!Would a nation make anthrax in aqueous form? It is easier to make
anthrax as a water-based slurry than a fine powder. According to
Tucker, “It appears that at the time of the 1991 Gulf War, Iraq only
had the capability to produce wet anthrax. Iraqi BW scientists
appear to have worked during the 1990’s to acquire the know-how
to produce dry, powdered anthrax, using a closely related bacterium
settling velocity. Now the fission debris is spread over a very much larger distance and as
a result the radioactive dose on the ground is very much lower.” Personal correspondence,
September 24, 2003.
116 Kruger, Radiation-Neutralization of Stored Biological Warfare Agents..., p. 7.
117 Personal communication, September 16, 2003.
118 For a detailed technical analysis of the mechanisms by which a nuclear explosion would
destroy BW agent stored underground, see May and Haldeman, "Effectiveness of Nuclear
Weapons against Buried Biological Agents Targets," [http://www.ciaonet.org/wps/mam09/
(the biopesticide Bt) as a model system. If a country had the
capability to produce dried, powdered anthrax, it would presumably
store it in that form because of its much longer shelf-life and the fact
that dry, powdered anthrax is much easier to disseminate as an
aerosol from an aircraft sprayer system or a missile warhead.” An
Office of Technology Assessment (OTA) report of 1993 provides
further information: “when anthrax bacteria are incubated under
particular conditions, they transform themselves into the rugged
spore form, which has a long shelf-life ... This spore-forming ability
makes anthrax particularly well suited for delivery by missiles or
bombs. The spores are stable when suspended in air, can survive
explosive dissemination from a bomb or shell, and ... will live for
several days if direct sunlight is avoided.”119
!Would a nation need huge quantities of anthrax? The OTA report
states that “a gram of anthrax [in spore form] theoretically contains
some 10 million lethal doses.”120 The ability of a single envelope
with a small quantity of powdered anthrax to shut down the Capitol
complex in October 2001 is a case in point. Tucker states, “A
bioterrorist attack in a confined space, such as a subway station,
would require a few ounces of dry anthrax. Military applications or
strategic attacks against cities would require much more material,
perhaps hundreds of pounds to attack a large city. As a rough rule
of thumb, about a kilogram of dry, concentrated anthrax spores
would be required to infect 50 percent of the people in a square
kilometer, assuming efficient aerosol dissemination under optimal
conditions. A kilo of dry anthrax is equivalent to a much larger
volume of wet slurry.”
!Would a nation make and store anthrax long before it intended to
use it? According to Tucker, “A nation would probably not make
anthrax far in advance of use unless it had mastered the technology
to produce dry powdered anthrax, which has a shelf life of several
years. A wet slurry of anthrax will tend to coagulate after several
months, rendering it unusable as a weapon. Thus, a country that
possessed only wet-anthrax technology would probably produce the
agent shortly before use.”
!Would a nation store large quantities of anthrax aboveground?
Attacking anthrax stored in shallow underground bunkers would
require the use of EPWs to explode in the bunker; otherwise, the
ground would shield the bunker from neutrons. An EPW would
create much fallout and could cause other problems, as detailed in
the section on RNEP, below. Storing anthrax in a bunker at a depth
considerably greater than that which EPWs could reach would
119 U.S. Congress, Office of Technology Assessment, Technologies Underlying Weapons of
Mass Destruction. OTA-BP-ISC-115. USGPO, December 1993, p. 79.
120 Ibid., p. 78.
prevent neutrons, heat, and delayed radiation (from radioactive
material left by the nuclear explosion) from reaching the bunker.
Tucker states that “[s]ince anthrax spores are quite rugged and are
in a state of suspended animation, it’s reasonable to assume that
[ground] shock alone would be unlikely to kill them reliably.” The
statement from the OTA report that anthrax spores can survive
explosive dissemination, cited above, supports this point. For
further protection from neutrons, an adversary could shield anthrax
with material that reflects neutrons (e.g., beryllium) or that absorbs
them (e.g., polyethylene mixed with boron). Alternatively, dry
powdered anthrax could be hidden anywhere in small lots, in which
case its destruction would depend mainly on locating the agent; once
that is done, various types of weapons and tactics might be used to
Advisability. Critics point to a number of intelligence hurdles. A successful
attack would require detailed knowledge of the target, such as storage configuration.
Knowing the target’s precise location would be crucial because of the very short
range at which even an enhanced radiation weapon would sterilize BW agent. Yet as
Tucker notes, “During the 1991 Gulf War, many suspect BW production and storage
facilities were targeted in error, and several real facilities such as the al-Hakam
factory — which UNSCOM121 later determined was Iraq’s largest anthrax production
facility — were totally unknown to U.S. intelligence.” It would be important to
know the type of agent because, as Kruger states, “the neutralizing dose [of radiation]
for different bio agents varies by a couple of factors of two.” Compensating for
uncertainty as to which agent was being attacked by using a weapon able to
neutralize the most radiation-resistant agent could result in using a weapon with a
yield several times larger than needed.
Critics also note political problems with nuclear weapon development and use.
An attack that sought to minimize fallout would have to use reduced blast/enhanced
radiation weapons. Neutron bombs, which the United States developed in the 1970s,
received intense domestic and foreign criticism as the “capitalist bomb” that spared
property while killing people. Similar criticism would doubtless reemerge if the
United States began a program to develop such weapons, especially if the intended
targets were perceived to be rogue states in the Third World. The development
program could require nuclear testing, fueling further criticism.
Critics assert that there are nonnuclear approaches to destroying BW agents.
According to one report, the Defense Department is working on nonnuclear weapons
for destroying chemical or biological agents.
Destroying the actual chemical or biological material [in Iraq] is a task that may
fall to agent defeat weapons being developed by the U.S. Navy and Lockheed
Martin under a program originally dubbed Vulcan Fire and now spearheaded by
the secretive U.S. Defense Threat Reduction Agency. The HTI-J-1000, as it is
121 The U.N. established UNSCOM, the United Nations Special Commission, in 1991 to
supervise the elimination of chemical and biological weapons and certain missiles in Iraq
following the Gulf War. See [http://www.un.org/Depts/unscom/] for details.
called, would be the fill inside the penetrating warhead used on the massive
Munition (JDAM) used to attack underground bunkers.
That report states that the chemical reactions would generate intense heat that would
destroy the biological and chemical agents, and would generate other chemicals that
would neutralize these agents.123
Michael Levi, director of the Strategic Studies Project at the Federation of
American Scientists, suggested another approach to defeating a hypothetical
underground anthrax storage bunker in Iraq:
U.S. planners may not want to directly attack the bunker. Instead, a watch
could be placed on the facility using satellite imagery coupled with armed
unmanned aerial vehicles. Anyone or anything attempting to enter or leave the124
bunker would be destroyed, making the anthrax inside unusable.
Kruger rejects the idea of using conventional high explosives to neutralize BW
agents for two reasons. First, he calculates that it would take a huge quantity of these
explosives, perhaps hundreds of 500-pound bombs, to raise the temperature of the
quantity of agent in a typical storage site sufficiently, assuming the agent was in a
water slurry. Similarly, if acid were used instead of explosives to kill BW agent, it
would take a very large quantity of acid. Second, he argues that attacking a BW
storage site with conventional weapons would cause much more collateral damage
than would an attack with nuclear weapons, even standard low-yield weapons:
Conventional weapons are likely to only kill a small fraction of the agent while
dispersing a significant fraction of the remaining live agent. My past dispersal
calculations for a generic target in a country of current interest have shown
collateral damage areas from dispersed bio agent that were very many orders-of-
magnitude larger than the fallout area for the ground burst of a low-yield fission125
It might be possible to deploy reduced blast/enhanced radiation weapons
without testing them, such as by adapting designs made and tested several decades
ago. Optimizing these weapons for defeating specific BW agents, however, could
involve changes that would eventually lead to testing.126
An attempt to seal off a facility encounters at least two problems. First, a U.S.
attack on a rogue state with no warning would be unlikely. With warning, that state
122 Frank Vizard, “Extinguishing the Threat,” Scientific American.com, February 18, 2003.
[ h t t p : / / www.sci e nt i f i camer i can.com/ a r t i c l e .cf m?ar t i c l e ID=00000CB6-18E9-1E4 D -
124 Michael Levi, “The Case against New Nuclear Weapons,” Issues in Science and
Technology, Spring 2003: 65-66.
125 Personal communication, September 16, 2003.
126 Information provided by Livermore staff, September 25, 2003.
could easily move materials and equipment to other sites before the attack. Indeed,
leaders who believed that their nation was at risk from a U.S. attack would probably
make plans well in advance to do just that. Second, some nations, such as
Afghanistan and North Korea, have extensive tunnel networks that might provide
undetected access, at a distance, to military facilities connected to such networks.
Conclusion. It appears that, under certain specific circumstances, a nuclear
blast is technically capable of sterilizing BW agents. One example would be a large
quantity of agent in aqueous solution stored at or near the Earth’s surface, with the
location pinpointed within meters and with accurate intelligence as to the agent
present and how it is stored. (The agent would have to be stored long enough that
intelligence assets could locate it, but not so long that it would deteriorate
significantly.) Another example would be a depot with a ton of powdered anthrax,
in which case a relatively small amount (as compared with hundreds of barrels) of
agent could do immense harm and it would be imperative to minimize the risk of
anthrax escaping. Few would suggest, though, that nuclear weapons would
automatically be the means of first choice for attacking BW agent. Nuclear-weapon
use entails clear political disadvantages, the effectiveness of nuclear weapons in
particular situations may be uncertain, and nonnuclear means may be available.
Whatever the case, it appears likely that Congress would examine closely any request
to proceed to development engineering or production of nuclear weapons intended
to destroy BW agents. Such examination might include the following questions:
!How probable is the military scenario put forth to justify
development and acquisition of these weapons?
!How plausible is the scenario? What are its flaws? How broad or
narrow are the circumstances under which the scenario exists?
!Would nuclear use be the only option in that scenario, or could
nonnuclear means suffice? What is the state of progress on
!What are the military and political pros and cons of developing and
acquiring these weapons?
Can ACI help in the fight against nuclear proliferation? According to
the FY2003 report of the Foster Panel,
A second requirement defined in the NPR is for advanced concept exploration.
Work must proceed on future concepts and technologies to avoid technical
surprises, to attract and train future stockpile stewards, and to assess intelligence
information on the continuing development and proliferation of WMD and their
delivery methods. The Panel emphasizes that it is essential for Congress to be
kept apprised of developments in foreign weapons programs and their potential127
implications for U.S. security.
127 FY 2003 Report to Congress of the Panel to Assess the Reliability, Safety, and Security
of the United States Nuclear Stockpile, p. 3.
ACI could improve U.S. understanding of potential terrorist nuclear weapons,
which could improve U.S. ability to interpret fragments of intelligence, to detect
evidence of nations or groups working to develop nuclear weapons, to detect the
transport of such weapons and attempts to bring them into the United States, and to
disable such weapons. It could also help avoid technical surprise by improving U.S.
ability to understand foreign weapon developments. For example, creating nuclear
weapon designs consistent with the Chinese or North Korean weapons program may
help the United States interpret intelligence clues to gain insight into the pace of the
program, its technological successes and failures, and warhead characteristics.
ACI’s supporters reject the claim that the program will encourage nuclear
proliferation. They argue that nations will decide to go nuclear for reasons far larger
than ACI. Even if ACI encouraged proliferation, that would not necessarily mean
that the United States should halt the program. It is likely, supporters claim, that
North Korea and Iran are pursuing nuclear weapons out of fear of U.S. conventional
forces. Does this mean, supporters ask, that the United States should halt military
modernization and shrink its conventional forces to keep those nations from going
nuclear? At the same time, it is argued, India and Pakistan are pursuing nuclear
weapons not because of U.S. conventional or nuclear forces but because of threats
that each sees from the other, and that India sees from China; these dynamics have
nothing to do with ACI.
ACI’s critics maintain that designing new nuclear weapons tailored for use
against targets in nonnuclear weapon states would undercut U.S. nuclear
nonproliferation efforts by in effect telling the world to do what we say, not what we
These critics assert that the program is scarcely needed to improve U.S. ability
to understand nuclear weapons developments made by rogue states or terrorists. U.S.
nuclear weapons experience accumulated over the past six decades provides ample
information for interpreting intelligence clues connected with the design and
fabrication of such weapons. Nor is ACI needed to improve U.S. ability to
understand Russian or Chinese nuclear weapons programs. The Stockpile
Stewardship Program has greatly increased U.S. weapons knowledge, arguably
placing this nation in a better position now to interpret intelligence data on Russian
and Chinese nuclear programs than it was a decade ago.
Another argument is that rogue states want nuclear weapons to deter U.S.
nuclear forces. A North Korean statement of August 2003 makes this point:
The Bush administration openly disclosed its attempt to use nuclear weapons
after listing the DPRK as part of ‘an axis of evil’ and a target of a ‘preemptive
This prompted us to judge that the Bush administration is going to stifle our
system by force and decide to build a strong deterrent force to cope with it. ...128
Our deterrent force ... is a means for self-defence to protect our sovereignty.
128 Kim Yong Il, Vice Minister of Foreign Affairs for Democratic People’s Republic of
Thus, it is argued, nuclear weapons designed to threaten key targets in rogue states
will only accelerate development of nuclear weapons by those states.
Does ACI offer unique value for the nuclear weapons enterprise?
ACI’s proponents believe that the United States must retain weapons design
capability because new weapons might be needed in the future, such as to attack
specific targets or to replace current weapons that deteriorate beyond the point of
repair. To that end, it is further argued, the United States should avoid a situation in
which it needs a new weapon yet no one at the labs has had the experience of
Supporters see ACI as complementing the Stockpile Stewardship Program.
Maintaining existing weapons involves predicting, detecting, and fixing problems
with existing weapons. Designing new weapons, however, arguably involves
different skills and thought processes, just as editing differs from writing. A designer
might be given a target and required warhead characteristics (yield, weight,
compatibility with a specific delivery vehicle, etc.) and be tasked to design a warhead
that produces the required military effectiveness while minimizing collateral damage.
Experienced weapon designers from all three labs have stated to NNSA headquarters
that the current Stockpile Stewardship Program (maintenance of the enduring
stockpile) is not sufficient to provide the range of opportunities necessary to train and
mentor the next generation of designers.129
ACI, it is argued, will provide such training. The purportedly leaked NPR stated
that ACI “will provide unique opportunities to train our next generation of weapon
designers and engineers.”130 Los Alamos Director John Browne stated that ACI
articulates a strategy for ... transferring nuclear warhead design knowledge, and
exercising design skills. This initiative provides an outstanding opportunity for
the nuclear weapons complex to ensure that existing expertise is transferred to
a future generation of stockpile stewards, and to extend the front-line weapons131
lifetimes beyond that of the designers who designed and tested them.
ACI will also develop new links between the weapons community and DOD
agencies. It is helpful, proponents believe, for some members of the weapons
community to have personal contacts with members of the U.S. Strategic Command
and the Defense Threat Reduction Agency, among others, to gain an understanding
Korea, text of speech delivered at six-country nuclear talks in Beijing, as published by
KCNA, the DPRK’s official news agency; posted by Reuters, August 29, 2003.
[http://www.cnn.com/2003/WORLD/ asiapcf/east/08/29/nkor ea.text.reut/index.html ]
129 Personal correspondence, NNSA official, September 3, 2003.
130 Nuclear Posture Review [Excerpts], p. 35.
131 “Statement of John C. Browne, Director, Los Alamos National Laboratory, Submitted
to United States House of Representatives, Committee on Armed Services, Subcommittee
on Military Procurement, June 12, 2002.”
[ h t t p : / / www.l a nl .gov/ o r gs/ pa/ News/ t e st i mony061202.sht ml ]
of what weapons might be of value to DOD, and to give DOD a sense of what
weapons technologies may be available.
The harshest critics of nuclear weapons believe that there is no need to keep
alive the skills to design new weapons. Current weapons, they argue, are more than
sufficient, and their designs have been proven. Until nuclear weapons are abolished,
these critics would maintain the existing stockpile; in their view, the multibillion-
dollar stockpile stewardship program is excessive for this purpose. This school of
thought would reject ACI completely.
Some critics of ACI see a value in maintaining design skills so as to improve
U.S. ability to understand warhead problems that may emerge, to understand how
changes made during refurbishment may affect weapon performance, and to design
replacement weapons should that be necessary. Such skills can be maintained, they
would argue, by designing weapons similar to current ones rather than designing new
types of weapons. That approach would not require ACI.
Critics would question training as a justification for ACI. The number of people
trained by ACI is very small, and there are already training programs for junior staff.
In October 1998, Sandia started an Intern Program, a two-year course for new
employees on the nuclear weapons program and Sandia’s role in it. In October 1996,
Los Alamos started the Theoretical Institute for Thermonuclear and Nuclear Studies
on nuclear weapons design for new personnel in that field.132
Some question whether ACI, as proposed for FY2004, offers a useful plan for
maintaining weapon design capabilities. For example, Raymond Jeanloz, Professor
of Earth and Planetary Science at University of California, Berkeley, and a member
of the University of California President’s Council,133 argues that repackaging an
existing bomb with no work on the nuclear explosive component (RNEP), working
on low-yield designs (an area extensively studied in the past, including with
underground tests), or other ideas proposed for ACI amount more to reworking old
ground than to significantly exercising design activity.134
There is no need for new links between DOD and the nuclear weapons
establishment, in the critics’ view, because several DOD-DOE groups have been in
existence for many years. These include the Nuclear Weapons Council, program
officer groups, and the Stockpile Assessment Team of Strategic Command’s
Strategic Advisory Group.
132 U.S. Commission on Maintaining United States Nuclear Weapons Expertise, Report to
Congress and the Secretary of Energy, p. 20.
133 The University of California operates three DOE laboratories: Los Alamos, Lawrence
Livermore, and Lawrence Berkeley National Laboratories. This council provides oversight
of these labs for the university.
134 Personal correspondence, September 2, 2003.
Robust Nuclear Earth Penetrator
For FY2004, the Administration requested $15.0 million to continue a study,
which began in May 2003, of the Robust Nuclear Earth Penetrator, or RNEP. The
NDAA provided the amount requested, while the EWDAA provided $7.5 million.
The FY2005 request is $27.6 million.
The RNEP study, part of the Advanced Concepts Initiative, is examining the
desirability and feasibility of modifying an existing B61 or B83 bomb to enable it to
penetrate into the ground before detonating. By so doing, a weapon of specified yield
would have a much greater effect against hardened and deeply buried targets — such
as are used in some rogue states to protect key assets — than would a nuclear weapon
of the same yield burst on the Earth’s surface. Modification would mainly involve
strengthening the weapon’s case, rather than changing the nuclear explosive package.
In the 1990s, DOE conducted a similar modification, converting some B61 bombs
into B61-11 earth penetrators, a modification that did not require nuclear testing. If
NNSA converts B61s or B83s into RNEPs, current plans envision that the conversion
would not require nuclear testing.
While the histories of earth penetrator weapons and the Spratt-Furse ban on low-
yield R&D are intertwined, repeal of that ban does not necessarily open the door to
RNEP. The ban applied only to weapons under 5 kt; RNEP, if it proceeds, would be
a modification of the B61 bomb or the B83 bomb, both of which are available in a
number of yields, with the maximum said to be far above 5 kt. Further, both have
been in the stockpile for many years. Even if they had yields only below 5 kt, the ban
applied only to weapons that had not entered production by November 30, 1993.135
Structures can be hardened and deeply buried in an attempt to withstand an
attack by conventional munitions. Such structures may be used to protect leadership,
biological agent production facilities, nuclear weapon storage sites, and other
strategic targets. Reports reference such facilities in Iraq, North Korea, and Iran.136
A DOD-DOE report indicates that thousands of such structures exist worldwide,
many in nations that do not pose a threat to the United States.137
135 The definition of a new weapon in P.L. 107-314, Section 3143, cited above, appears to
preclude the argument that modifications of the B61 or B83 would constitute a “new”
136 Glenn Kessler, “Nuclear Sites in Iran Worry U.S. Officials,” Washington Post, December
Aviation Week & Space Technology, December 16, 2002: 44; and Joby Warrick, “Iran, N.
Korea Nuclear Plans Pose New Risks,” Washington Post, December 24, 2002: 1, 11.
137 Departments of Defense and Energy. Report to Congress on the Defeat of Hard and
Deeply Buried Targets ..., p. 8.
While special operations forces or precision-guided conventional bombs might
defeat deeply buried structures by attacking power supplies, ventilation systems, and
exits, these structures would probably have backup power supplies, escape tunnels,
etc. Accordingly, destruction would be a surer means of completely eliminating the
threat they pose. Facilities at shallow depths, perhaps 10 meters, are vulnerable to
destruction by conventional munitions that penetrate the ground. Destruction of
facilities buried at greater depths may require a nuclear weapon. During the Cold
War, the United States deployed the B53 nuclear bomb, a weapon said to have a yield
of some megatons (thousands of kilotons), for such missions. A weapon of such high
yield would create blast and fallout that would destroy surface structures and kill
inhabitants out to many miles; in the current environment, a President would be
unlikely to order its use against a target in a rogue state so as to avoid killing very
large numbers of civilians. As C. Paul Robinson, President of Sandia National
Laboratories, said, we would be self-deterred from using high-yield weapons.138
Figure 1. Earth penetration reduces yield
needed to destroy buried targets
At issue, insofar as the nuclear option is concerned, is how to destroy hardened
and deeply buried facilities with lower-yield weapons. Much of the energy of a
ground-burst weapon is reflected back into the air. Scientists on both sides of the
larger debate agree that a nuclear weapon that penetrates even a short distance into
the ground will transfer much more of its energy into ground shock, which can
destroy buried targets. For example, Sandia reports that a 1-megaton (1,000-kiloton)
weapon burst on the Earth’s surface has about the same effect on buried targets as a
33-kiloton weapon detonated at 5 meters depth. Most of the payoff comes from
burying the weapon several meters in the ground, especially for lower-yield weapons;
138 C. Paul Robinson, “Maintaining a Viable Nuclear Weapons Program in a Test Ban
Environment: A Strong Technical Foundation in the Laboratories,” address presented at the
Nuclear Security Decisionmakers Forum, March 28, 2000, Albuquerque, NM:
[ ht t p: / / www.sandi a.gov/ me di a/ speeches/ NSDMF2000.doc]
burial beyond several meters produces diminishing returns.139 Figure 1 illustrates
these points. Robert Nelson reports similar data.140 Thus a lower-yield earth
penetrator weapon (EPW) can have the same effect on buried targets as a much
higher-yield surface-burst weapon. This large reduction in yield, in turn, reduces
fallout and blast damage. A common misconception is that detonating a weapon a
few meters underground reduces fallout; in fact, a weapon of specified yield
detonated at shallow depth in wet soil will create more fallout than if it were
detonated on the surface. The reduction in fallout from lower yield, though, can
offset the increase in fallout from a shallow subsurface burst. Continuing the
example just noted, fallout with a specific intensity (150 rads in 24 hours for an
unsheltered person) would, by one calculation, be generated over 2,700 square
kilometers by a 1-megaton surface-burst weapon, or over 150 square kilometers by
a 33-kiloton weapon detonated at a depth of 5 meters.141 Figure 2 illustrates these
points (though it assumes individuals would be sheltered).
Figure 2. Lower yield reduces fallout;
shallow depth of burst increases it
Lower-yield weapons, even EPWs, may not suffice to destroy all buried targets.
Stephen Younger, then Associate Laboratory Director for Nuclear Weapons, Los
Alamos National Laboratory, wrote, “Some very hard targets require high yield to
destroy them. No application of conventional explosives or even lower-yield nuclear
explosives will destroy such targets, which might include hardened structures buried
139 Memorandum, “Comparison of Surface and Sub-Surface Nuclear Bursts,” from Steven
Hatch, Sandia National Laboratories, to Jonathan Medalia, October 30, 2000.
140 Robert Nelson, “Low-Yield Earth-Penetrating Nuclear Weapons,” Science and Global
Security, 10: 3-4, 2002.
141 Personal communication from Steven Hatch, Sandia National Laboratories, September
4, 2003. The level of fallout intensity used is 150 rad over 24 hours for a person in a
basement shelter. (A dose of 100-200 rem produces minimal symptoms in humans, while
a dose of 300-600 rem may kill half of those exposed. Rads and rems are measures of
energy absorbed by the body; for most (but not all) forms of radiation, one rad equals one
rem. Dale Moeller, Environmental Health, Cambridge, Harvard University Press, 1997, p.
beneath hundreds of feet of earth or rock.” Indeed, “Superhard targets, such as those
found under certain Russian mountains, may not be able to be defeated reliably by
even high-yield nuclear weapons.”142 Similarly, Robert Nelson wrote, “A low-yield
nuclear EPW would still only be able to destroy facilities relatively close to the
surface ... Very large yield (>~100 kt) weapons are still required to destroy facilities
buried under the equivalent of 100 m of concrete.”143
Even low-yield EPWs would generate much fallout. For example, photographs
of two nuclear tests show the fallout cloud generated by low-yield weapons buried
hundreds of feet with the emplacement hole sealed. A test named “Des Moines,”a
2.9-kiloton test conducted June 13, 1962, at the Nevada Test Site, was buried in a
mountain. Pictures show a huge cloud of radioactive dust, with cars and pickups of
spectators rushing away from it. Another test, “Baneberry,” with a yield of 10 kt,
was conducted December 18, 1970, in a vertical shaft some 900 feet deep at the
Nevada Test Site.144 The fallout cloud rose about 10,000 feet into the air,145 and was
tracked to the Canadian border.146 According to DOE, “Following Baneberry, new
containment procedures were adopted to prevent a similar occurrence.”147 The
minimum depth of burial for even the smallest test is 600 feet.148 An EPW, in
contrast, would be buried perhaps a few tens of feet and the path by which it
penetrated the earth would not be sealed. Fallout from such weapons would not be
contained at all, and standard-yield EPWs under consideration would produce more
fallout than low-yield EPWs.
Improvements in accuracy also permit a reduction in warhead yield for attacking
buried targets. Four Los Alamos scientists calculate that a moderately hard structure
buried at shallow depth could be destroyed by a 15-kiloton weapon with an accuracy
of 100 meters, or by a 0.5-kiloton weapon with an accuracy of 5 meters, with both
weapons burst on the surface. Earth penetration further reduces the yield needed.
142 Stephen Younger, “Nuclear Weapons in the Twenty-First Century,” Los Alamos National
Laboratory, LAUR-00-2850, June 27, 2000, p. 10.
143 Nelson, “Low-Yield Earth-Penetrating Nuclear Weapons,” p. 18.
144 The yield of these tests is from U.S. Department of Energy. Nevada Operations Office.
Office of External Affairs. United States Nuclear Tests: July 1945 Through September
145 U.S. Department of Energy, Nevada Operations Office website:
[http://www.nv.doe.gov/news&pubs/photos&films/general.htm ]. This site also provides
depth of burial and a photograph of the test.
146 U.S. Congress, Office of Technology Assessment, The Containment of Underground
Nuclear Explosions. OTA-ISC-414, October 1989, p. 31.
147 U.S. Department of Energy, Nevada Operations Office website: [http://www.nv.doe.gov/
148 Office of Technology Assessment, The Containment of Underground Nuclear
Explosions, p. 36.
The combination of improved accuracy and earth penetration, they argue, sharply
reduces the yield required and the collateral damage.149
While the foregoing discussion has focused on technical aspects of nuclear
weapon effects, it must be kept in mind that even low-yield earth penetrator weapons
could, depending on the circumstances, kill large numbers of people. The calculation
for such a weapon burst near Damascus, cited above, illustrates this point.
History Through the FY2003 Budget Cycle
Interest in destroying buried targets traces back to the early days of the Cold
War when the Soviet Union first built underground command posts and missile silos.
Current interest in EPWs for use against HDBTs traces back to the Persian Gulf War,
as is detailed in the earlier section on low-yield R&D.
This latter experience led to the development of the current EPW, the B61 mod
11, or B61-11. (B61 is the basic weapon, while the mod number indicates a
modification, converting the bomb to a specific purpose, delivery by a particular
aircraft, etc.) The modification involved a strong case and some internal
strengthening. Because it entailed only minimal changes to the nuclear explosive
package, DOE did not conduct nuclear tests of the weapon.150 DOE converted a
small number of older B61-7 bombs to B61-11s in the mid- to late 1990s. The B61-
11, however, has its limitations. The Nuclear Posture Review stated, according to
purportedly leaked excerpts, that the weapon “has a very limited ground penetration
capability” and “cannot survive penetration into many types of terrain in which
hardened underground facilities are located.”151
The growth in number of buried targets and the weaknesses of the B61-11 led
DOD and DOE to study RNEP. According to a U.S. Strategic Command officer,
Our current arsenal, developed in the Cold War, was not designed to address this
growing worldwide threat [of buried targets]. There are facilities today which
we either cannot defeat, even with existing nuclear weapons, or must hold at risk
using a large number of weapons. As a result, both the Department of Defense
and the Department of Energy, through the Nuclear Weapons Council, have
approved a study of how to effectively counter this threat. This study of a
149 Bryan Fearey, Paul White, John St. Ledger, and John Immele, “An Analysis of Reduced
Collateral Damage Nuclear Weapons,” Comparative Strategy, October/November 2003, p.
150 According to a Los Alamos publication, “The essence of the modification is a field
changeout of the weapon’s case to provide an earth-penetration capability.” “B61-11
Update,” Weapons Insider, August 1996: 2. Further, “modifications to the warhead’s
electrical system and physics package were minimal,” “B61-11 NESE Completed,” Weapons
Insider, June/July/August 1998: 5.
151 Nuclear Posture Review [Excerpts], p. 47.
Robust Nuclear Earth Penetrator (RNEP) will evaluate modifications to existing152
nuclear weapons that do not require nuclear testing.
Linton Brooks, then Acting Administrator of NNSA, provided further details on
this study will examine whether or not two existing warheads in the stockpile —
the B61 and the B83 — can be sufficiently hardened through case modifications
and other work to allow the weapons to survive penetration into various
geologies before detonating. This would enhance the Nation’s ability to hold153
hard and deeply buried targets at risk.
Thus RNEP would not necessarily be a low-yield weapon. John Gordon, then
Director of the National Nuclear Security Administration, stated in 2002 that the
emphasis is on “a more standard yield system called an enhanced penetrator ...154
There’s no design work going on low-yield nuclear weapons.” The cited advantage
of the higher yield is that while EPWs can destroy a number of buried targets with
less yield than a surface-burst weapon, increasing yield increases the radius of
damage to buried targets, some of which may be too hard, or too deep, to destroy
with a low-yield weapon. In July 2003, NNSA provided further detail on this topic.
The ongoing RNEP study will select one of the two candidate bombs. The study
aims to retain all existing capabilities. It is not part of the study to create new, lower
yi elds. 155
NNSA had planned to begin the RNEP study in FY2002 by redirecting $10
million from Directed Stockpile Work Research and Development, a program within
stockpile stewardship.156 As discussed below, however, FY2003 legislation had the
effect of delaying this study into FY2003. For FY2003, the Administration requested
funds to begin this RNEP study. The planned budget for the study was $15.0 million
a year for FY2003-FY2005, inclusive. Brooks said, “The RNEP feasibility and cost
152 “Statement of RADM John T. Byrd, USN, Director of Plans and Policy, United States
Strategic Command, before the House Armed Services Committee, Procurement
Subcommittee, June 12, 2002.” The Joint Nuclear Weapons Council is a small DOD-DOE
agency established by 10 USC 179 that coordinates nuclear weapons activities between the
two departments. Statement is available at [http://armedservices.house.gov/opening
statementsandpressreleases/ 107thcongress/02-06-12byrd.html ].
153 “Statement of Linton F. Brooks, Acting Under Secretary of Energy and Administrator
for National Security, National Nuclear Security Administration, U. S. Department of
Energy Before the Subcommittee on Strategic Forces, Committee on Armed Services, U. S.
Senate, April 8, 2003”: [http://www.senate.gov/~armed_services/statemnt/2003/April/
154 U.S. Congress. Senate. Committee on Armed Services. Hearing: Nuclear Posture
Review. February 14, 2002, n.p. Transcript prepared by eMediaMillWorks, Inc.
155 Information provided by NNSA staff, July 25, 2003.
156 Information provided by NNSA staff, August 11, 2003. See also Walter Pincus,
“Nuclear Warhead Study Aims at Buried Targets,” Washington Post, March 15, 2002: 16.
study is currently scheduled for completion in 2006; however, we are looking at
opportunities to reduce study time.”157
The House Armed Services Committee supported the FY2003 request. It
included a provision calling for a National Academy of Sciences study on the short-
and long-term effects on civilians and military forces of an EPW, an above-ground
nuclear explosion to destroy buried targets, and use of conventional weapons to
destroy WMD storage or production facilities.158 Representative Markey offered an
amendment on the House floor to bar permanently the use of funds to develop or test
a nuclear EPW, and to bar FY2003 funds for conducting a feasibility study of that
weapon. The amendment was defeated, 172-243, on May 9, 2002.159
The Senate Armed Services Committee recommended eliminating funds for
RNEP and requiring the Secretaries of Defense and Energy to report on RNEP,
including military requirements, employment policy, targets, and ability of
conventional weapons to “address” types of targets that RNEP would hold at risk.160
The conference bill included the House provision for a National Academy of
Sciences study; as of August 2003, it appears that the study will be completed around
April 2004.161 The conference bill fully funded RNEP but barred obligation of
FY2003 funds for it until 30 days after the report called for by the Senate Armed
Services Committee had been submitted to the Armed Services Committees. DOD
submitted that report in classified form on March 19, 2003; the study began on May
1, 2003, with a meeting of NNSA, DOD, and nuclear weapons laboratory personnel
in Washington. The FY2003 Energy and Water Development Appropriations Act
included the amount requested for Stockpile R&D, the category within Directed
Stockpile Work that includes RNEP.162 NNSA indicated that this measure provided
the full $15.0 million requested for RNEP.163
Legislative Actions in the FY2004 Budget Cycle
For FY2004, the Armed Services Committees and their respective Houses
included the full $15.0 million request in the defense authorization bills. The House
rejected, 199-226, an amendment by Representative Tauscher to transfer all $15.0
million from RNEP (and $6.0 million from ACI) to conventional technology for use
157 Statement of Linton F. Brooks, April 8, 2003.
158 U.S. Congress. House. Committee on Armed Services. Bob Stump National Defense
Authorization Act for Fiscal Year 2003. H.Rept. 107-436, 107th Congress, 2nd Session,
USGPO, 2002, p. 343-344.
159 U.S. Congress, Congressional Record, May 9, 2002: H2339-H2343.
160 U.S. Congress, Senate Committee on Armed Services, National Defense Authorization
Act for Fiscal Year 2003. S.Rept. 107-151, 107th Congress, 2nd Session, USGPO, 2002, p.
161 Information provided by NNSA staff, August 11, 2003.
162 Conference Report on H.J.Res. 2, Consolidated Appropriations Resolution, 2003. In U.S.
Congress. Congressional Record, February 12, 2003: H1035.
163 Information provided by NNSA staff, August 1, 2003.
against hard and deeply buried targets. The Senate tabled, 56-41, an amendment by
Senator Dorgan to bar use of funds for RNEP, and adopted an amendment by Senator
Nelson (FL) to require congressional authorization to start engineering development
or subsequent phases of RNEP.
In action on the FY2004 energy and water development appropriations bill, the
House Appropriations Committee recommended reducing RNEP funds to $5.0
million, for reasons noted earlier in the “Legislative Actions in the FY2004 Budget
Cycle” section under ACI. The House Appropriations Committee recommended
having RNEP funds support penetration by conventional as well as nuclear weapons:
The Committee directs that funding provided for the Robust Nuclear Earth
Penetrator (RNEP) be used for research on the problem of deep earth penetration
through hard or hardened surfaces, including modeling and simulation of the use
of advanced materials, and varied trajectories and speeds. The Committee further
directs that the National Nuclear Security Administration (NNSA) coordinate the
RNEP research program with ongoing programs at the Department of Defense
relating to research on earth penetration to maximize the dual-use applicability
for both conventional and nuclear weapons.164
The House agreed to this bill without amending the Weapons Activities account,
which includes RNEP.
The Senate Appropriations Committee recommended providing $15 million for
RNEP. On September 16, as detailed under ACI, the Senate tabled an amendment
by Senator Feinstein to eliminate all FY2004 funds requested for RNEP and adopted
an amendment by Senator Jack Reed that barred use of funds provided by H.R. 2754
for phase 3 or 6.3 or beyond for advanced nuclear weapon concepts including RNEP.
Conference Issues and Outcomes
RNEP was at issue in the Energy and Water Development Appropriations Act
conference. Action on it was of concern to many because, of the four initiatives,
RNEP is the one closest to deploying a weapon in response to NPR guidance. Non-
RNEP ACI studies are in early stages, and enhanced test readiness and rescinding the
low-yield ban are not weapons programs. In contrast, RNEP, if it proceeds, would
involve conversion of an existing weapon. Conversion, under the current plan,
would not involve nuclear testing, and would require a much smaller production
effort than would production of complete new weapons. Even if the conference
decided to proceed with the RNEP study, and if Congress later decided to proceed
with the actual modification, deployment would not likely be quick; it took six years
to develop the B61-11 and certify it for acceptance into the nuclear stockpile,165
though lessons learned from that process might expedite deployment of RNEP.
164 U.S. Congress, House Committee on Appropriations. Energy and Water Development
Appropriations Bill, 2004. H.Rept. 108-212, p. 145.
165 “B61-11 Enters the Stockpile,” Weapons Insider [published by the Nuclear Weapons
Program of Los Alamos National Laboratory], September/October 2001: 1-2.
The NDAA provided the funds requested for RNEP; conferees did not comment
on it. THE EWDAA provided $7.5 million. Further,
The conferees remind the Administration that none of the funds provided may
be used for activities at the engineering development phases, phase 3 or 6.3, or
beyond, in support of advanced nuclear weapons concepts, including the Robust166
Nuclear Earth Penetrator.
The FY2005 Budget Cycle
The FY2005 request document seems to cast serious doubt on assertions that
RNEP is only a study. Beginning with the FY2005 budget cycle, NNSA presented
a detailed four-year projection along with the current request; for RNEP, the figures
are: FY2005, $27.6 million; FY2006, $95.0 million; FY2007, $145.4 million;
FY2008, $128.4 million; and FY2009, $88.4 million, for a five-year total of $484.7167
million. Further, the document states the performance targets for RNEP as
!FY2005: “Complete 56% of scheduled RNEP Phase 6.2/6.2A activity.”
Further, “In FY2005, subsystem tests and a full system test of the
proposed design will be completed.”
!FY2006: “Complete 100% of scheduled RNEP Phase 6.2/6.2A activity.”
!FY2007: “Report results of RNEP Phase 6.2/6.2A to Nuclear Weapons
Council [a joint DOD-DOE agency that coordinates nuclear weapon
requirements, production schedules, etc.] Obtain, if applicable, RNEP
Phase 6.3 appropriate authorization. Complete initial 25% of scheduled
RNEP Phase 6.3 activity (if authorized).”
!FY2008: “Complete 65% of RNEP Phase 6.3 activity (if appropriately
!FY2009: “Complete 100% of scheduled RNEP Phase 6.3 activity (if
authorized). Complete 15% of scheduled RNEP Phase 6.4 activity (if168
An NNSA manager responsible for the program maintained that the budget
increase beyond FY2005 is an artifact of the budget process. He stated that the
money was inserted in the out years as a “placeholder” to protect the option of
proceeding with RNEP. Were this not done, it is argued that NNSA would face two
choices that it deems unsatisfactory: (1) By the time the budget for one fiscal year
166 Committee of Conference, Making Appropriations for Energy and Water Development
for the Fiscal Year Ending September 30, 2004 ..., H.Rept. 108-357, p. 156.
167 U.S. Department of Energy, Office of Management, Budget, and Administration/CFO.
FY 2005 Congressional Budget Request. volume 1, National Nuclear Security
Administration. DOE/ME-0032, February 2004, p. 63. The RNEP budget is available under
“Directed Stockpile Work” at [http://www.mbe.doe.gov/budget/05budget/index.htm].
168 Department of Energy, FY 2005 Congressional Budget Request, volume 1, p. 69, 70, 76.
is submitted, the budget for the next fiscal year is largely fixed; without the
placeholder, a decision to proceed with RNEP could therefore not be implemented
until the second fiscal year. (2) Alternatively, without the placeholder, a decision to
proceed with RNEP could be implemented promptly only by taking the needed funds
out of other programs. Similarly, the move to Phases 6.3 and 6.4 reflects how the
program might be expected to advance if it proceeds. The official, however,
indicated that no decision has been made on whether or not to proceed with RNEP
pending completion of the study.169
The RNEP study was initially projected to cost $45 million — $15 million a
year for FY2003-FY2005. The numbers, however, have changed for each year. For
FY2003, delay in submission of a DOD study required by the FY2003 National
Defense Authorization Act (P.L. 107-314, Sec. 3146) delayed the start of NNSA’s
RNEP study; as a result, $6.0 million was spent of the $15.0 million appropriated.
For FY2004, Congress cut the RNEP appropriation to $7.5 million. For FY2005, the
request is $27.6 million, vs. the $15.0 million originally planned. Finally, FY2006,
not FY2005, will be the last year of the RNEP study; NNSA estimates the FY2006
request at $30 million. The four-year total is about $71 million. Owing to the
uncertainties of the program, NNSA could not, as of early March 2004, project an
RNEP budget for FY2007-FY2009.
According to NNSA, the study’s cost has grown for a number of reasons. The
$45 million did not take into account participation in the study by Y-12 Plant, which
would make components of RNEP, or of Pantex Plant, which would convert existing
weapons into RNEPs; their participation adds some $2 million. DOE has imposed
additional project management requirements that add $2 million. The rest of the
increase comes from a better definition of the requirements of the study, refinement
of cost estimates, and an increase in surety (safety, security, and use control) of the
proposed weapon. On the latter point, DOE requires that any modifications of a
nuclear weapon look for ways to increase its surety.170 NNSA says it has found ways
to increase RNEP surety, and plans to do so.
Critics are likely to view the outyear plan in the FY2005 budget request
document as indicating an intent to proceed with RNEP. They may also question
how the cost of the RNEP study grew some 58%. Congress, the media, and others,
however, have not generally had access to the information presented here on the
RNEP budget. This report will add views of critics and others when they become
Issues for Congress
Would RNEP promote deterrence? In the post-9/11 world, a new set of
threats has come to the fore, including threats from rogue states seeking WMD. To
deter rogue states, RNEP supporters maintain that the United States must hold at risk
those assets that the leaders value most, and that in some instances RNEP may be the
169 Telephone interview, February 10, 2004.
170 U.S. Department of Energy, Order DOE O 452.1B, “Nuclear Explosive and Weapon
Surety Program,” approved August 6, 2001, Section 4(f).
only way to do so. Everet Beckner, NNSA Deputy Administrator for Defense
Programs, reportedly opined: “If we’ve got serious enemies in the world, I would like
them to be deterred as much as possible by our military capabilities. ... I want them
to be afraid of us.”171 Similarly, the House Policy Committee argued:
Credible deterrence requires that the President be able to hold at risk the
things most important to an adversary who would seek to attack America. Deep
underground facilities, including hardened bunkers and hard-rock tunnels,
provide effective haven from attack. ...
The President should have options — the options of conventional forces,
of precision conventional weapons, and of nuclear weapons that are capable of172
holding all targets at risk.
Critics question whether it is correct to apply the Cold War model of deterrence
to rogue states, whose leaders might not be overly concerned about a U.S. nuclear
attack on a few facilities in their nation or about the casualties it would cause. They
might even expect to benefit from worldwide condemnation of the United States that
an attack would likely generate. Furthermore, some RNEP critics argue that the use
of any nuclear weapon would not be credible because of the massive damage and
casualties that would result, and thus would not enhance deterrence.
Would RNEP provide added military value? RNEP’s supporters note
that the Nuclear Weapons Council, a DOD-DOE organization chaired by DOD to
coordinate nuclear weapons activities between the two agencies, approved the
conduct of the RNEP study, and that DOD developed an Operational Requirements
Document for RNEP that gives the desired specifications and capabilities.173 They
assert that RNEP is not redundant with the B61-11 because of the latter weapon’s
Critics assert that nuclear EPWs are not needed. They note Brooks’s statement,174
“we have no requirement to actually develop any new weapons at this time.”
Representative Skelton stated, “the key to defeating hard deeply buried targets lies175
more in accuracy and penetration rather than the inherent explosive capability.”
Many conventional options are available for defeating such targets.176 U.S. forces
demonstrated the ability of ground troops to attack tunnel complexes in Afghanistan
171 John Fleck, “Should Nuclear Research Resume?,” Albuquerque Journal, March 17, 2003.
[ h t t p : / / www.abqj our nal . com/ paper boy/ i a / s ci t e ch/ 846058news03-17-03.ht m]
172 U.S. Congress, House Policy Committee, Subcommittee on National Security and
Foreign Affairs, Differentiation and Defense: An Agenda for the Nuclear Weapons
Program. February 2003, p. 7. [http://wilson.house.gov/media/photos/nuclearreport.pdf].
173 Information provided by NNSA staff, August 18, 2003.
174 Statement of Linton F. Brooks, April 8, 2003, p. 6.
175 U.S. Congress, Congressional Record, May 21, 2003: H4505.
176 Michael Levi, Fire in the Hole: Nuclear and Non-Nuclear Options for
Counterproliferation. Non-Proliferation Project, Global Policy Program, Carnegie
Endowment for International Peace, working paper no. 31, November 2002, p. 17-25.
and the ability of precision conventional ordnance to destroy underground bunkers
in Iraq. It would be better, in this view, to spend funds on improving the ability to
destroy these targets with conventional means rather than on nuclear weapons.
Would fallout from RNEP bar use of the weapon? Critics of RNEP
point to at least one claim by EPW supporters that low-yield EPWs would limit177
collateral damage sufficiently to contemplate their use. (RNEP could well have
significantly higher yields.) Critics challenge this view. They argue that no EPW
could penetrate deeply enough to be fully contained, so that “the goal of a benign
earth-penetrating nuclear weapon is physically impossible,”178 as the descriptions of
nuclear tests Des Moines and Baneberry make clear. For example, Nelson calculates
A one kiloton earth-penetrating “mininuke” used in a typical third-world urban
environment would spread a lethal dose of radioactive fallout over several square
kilometers, resulting in tens of thousands of civilian fatalities.179
Under these conditions, it is argued, RNEP would be unusable:
the vast resulting collateral damage to noncombatant populations would
presumably limit employment of the weapon to retaliatory, or “intra-war”
preemptive use in scenarios of all-out warfare, involving another nuclear
weapon-state with the resources to both construct such deeply buried targets and
threaten the survival of the U.S. as a nation-state — that is, China and Russia. ...
the weapon is strategically, legally, and morally unsuitable for preemptive or180
retaliatory counter-proliferation warfare.
Similarly, Representative Frank stated:
... we have said in Afghanistan, in Iraq, we are these days likely to be ... in an
effort to rescue a people from an oppressive government. How welcome will our
wagon be when it comes to nuclear arms? Do we tell the people of Afghanistan,
do not worry, we will free you from the Taliban by using nuclear weapons within
your country. ... I think you undercut the whole notion that America can be181
coming to the rescue of the victims of oppression.
177 Nelson, “Low-Yield Earth-Penetrating Nuclear Weapons,” p. 2, references Dowler and
Howard, “Countering the Threat of the Well-Armed Tyrant.” The latter article, p. 36,
argues that blast, crater radioactivity, and fallout would be confined to a small area. See
also Robert Nelson, “Nuclear Bunker Busters, Mini-Nukes, and the US Nuclear Stockpile,”
Physics Today, November 2003: 32-37.
178 Nelson, “Low-Yield Earth-Penetrating Nuclear Weapons,” p. 2.
179 Ibid., p. 1.
180 Christopher Paine with Thomas Cochran, Matthew McKinzie, and Robert Norris,
Countering Proliferation, or Compounding It?: The Bush Administration’s Quest for Earth-
Penetrating and Low-Yield Nuclear Weapons, Natural Resources Defense Council, May
181 Representative Barney Frank, colloquy on the Markey amendment to H.R. 4546, the Bob
Stump National Defense Authorization Act for Fiscal Year 2003, Congressional Record,
May 9, 2002: H2340.
The argument over the extent of fallout RNEP would cause is, in the view of
RNEP supporters, secondary to the key point, which, they contend, is that
circumstances might call for the use of this weapon. A President always has various
military options, and no President since 1945 has chosen to use nuclear weapons.
Nonetheless, nuclear weapons might at some point be the only way to achieve a
critical military objective, supporters argue. In that case, EPWs would be preferable
in this view because they could achieve the desired result at a lower yield, and with
less fallout, than surface-burst weapons. For some such uses, the B61-11 might
suffice, but RNEP may be needed for others.
Would RNEP have an adverse effect on nuclear nonproliferation?
On the issue of whether the United States should avoid researching EPWs in order
to lead the rest of the world by example, Representative Thornberry argued:
It is an interesting line of argument to say that we make the world safer
when we tie our hands behind our back, that the problem is with the United
States, and that if we would just set a good example, the Saddam Husseins and
the Kim Jong Ils and even the Putins would fall right in line ... The problem is
not American strength. ... Peace comes when America is strong and when182
America has additional options.
The United States, in the view of RNEP’s supporters, will not cause other nations to
abandon WMD programs by abandoning RNEP. Nor, they suspect, would study,
development, and deployment of RNEP cause nations to pursue WMD programs;
they assert that deployment of the B61-11 had no such effect, so they doubt that a
second EPW would, either.
Critics see grave proliferation risks arising from RNEP. They ask how the
United States can preach to rogue states to restrain their nuclear programs while this
nation is exploring new nuclear weapon capabilities. They see RNEP as tailored for
use against rogue states, and fear that that implies a U.S. willingness to use the
weapon for that purpose. They foresee serious consequences from such use. As
Representative Markey said, using one nuclear weapon “will destroy our moral and
political credibility to end the spread of weapons of mass destruction, especially
nuclear weapons.”183 In this view, it is inconsistent for the United States to fight Iraq
because of WMD and then to develop WMD itself. Even studying RNEP, critics
contend, would imply that the United States is considering its use, which could make
nonnuclear weapon states accelerate their efforts to obtain nuclear weapons to deter
U.S. attack on them, or to build new underground facilities at much greater depth.
Will Targets Be Available for RNEP? Fewer nations hold potential future
targets for RNEP in early 2004 than was the case a year earlier. Any government that
emerges in Iraq would in all probability be unable and unwilling to develop WMD
for many years. Libya is in the process of giving up its WMD; indeed, President
182 U.S. Congress, Congressional Record, May 21, 2003: H4506.
Bush praised Colonel Muammar Ghadafi of Libya for making “the right decision”
in ending its WMD programs.184
Moving from the strategic to the tactical level, a successful attack on a deeply
buried facility by an earth penetrator weapon would require precise targeting
information. Yet events of the last year raise troubling questions on this score.
Judging from press reports, U.S. intelligence failed to detect entire WMD programs
in Libya, Iran, and North Korea, and detected chemical, biological, and nuclear
weapon programs in Iraq that apparently did not exist. Given that record, critics may
well question whether intelligence can be sufficiently precise and reliable to pinpoint
underground bunkers to attack.
RNEP’s supporters would note that Iran and North Korea pose a much more
serious nuclear threat than did Iraq and Libya. Iran agreed in December 2003, under
pressure from Britain, France, Germany, and Russia, to permit more intrusive
inspections by the International Atomic Energy Agency (IAEA). It did so in order
“to prove that the Iranian nuclear program is for civilian purposes,” Gholam Reza
Aghazadeh, Iran’s Vice President, was quoted as saying.185 The inspections,
however, discovered equipment and materials that could be linked to nuclear
weapons.186 Shortly after these discoveries, Iran appeared to begin pulling back from
full disclosure. A senior Iranian official reportedly said that his country “has given
enough answers to the [IAEA’s] questions. ... We have other research projects which
we haven’t announced to the agency and we don’t think it is necessary to announce
to the agency.”187 North Korea’s statements have been contradictory and confusing
for some time, but press reports have presented evidence that that nation has a
program to produce highly enriched uranium, plutonium, and nuclear weapons, and
that it received aid from Pakistan.188
Regarding the ability of tactical intelligence to provide the basis for nuclear
weapon use, RNEP supporters would note that the bar against such use is extremely
high, but that many intelligence assets, whether satellite imagery, communication
intercepts, or information from defectors, would be brought to bear to ensure that the
target was in fact connected with WMD.
184 U.S. White House, “President Announces New Measures to Counter the Threat of
WMD.” Remarks by the President on Weapons of Mass Destruction Proliferation, Fort
Lesley J. McNair, National Defense University, Washington, DC, February 11, 2004.
185 Christine Hauser and Nazila Fathi, “Iran Signs Pact Allowing Inspection of Its Nuclear
Sites,” New York Times, December 18, 2003.
186 See, for example, David Sanger, “In Face of Report, Iran Acknowledges Buying Nuclear
Components,” New York Times, February 23, 2004: 6; and Karl Vick, “Another Nuclear
Program Found in Iran,” Washington Post, February 24, 2004: 1.
187 “Iran Digs in Heels over Nuclear Secrets,” Reuters, February 25,2004.
188 On the latter point, see David Rohde and David Sanger, “Key Pakistani Is Said to Admit
Atom Transfers,” New York Times, February 2, 2004: 1.
Nuclear Test Readiness
During the Cold War, when the United States conducted nuclear tests on an
ongoing basis, it took 18-24 months to prepare a typical test. With the end of the
Cold War and the start of a moratorium on nuclear tests, the Clinton Administration
required a U.S. ability to conduct a nuclear test in at most 24 to 36 months of a
presidential order to conduct the test. DOD and NNSA, however, found that it would
take close to 36 months to test, and they are uncertain as to their continued ability to
test within even this length of time. Accordingly, they recommended “enhanced test
readiness,” or moving to and maintaining a posture in which tests could be conducted
in less time. They supported an 18-month test readiness posture.
For FY2004, the Administration requested $24.9 million to maintain the
existing posture and transition to the 18-month posture.189 This figure is in addition
to funds to maintain the Nevada Test Site, where NNSA would conduct any nuclear
tests. NNSA indicated that it has tightly integrated the test readiness and enhanced
test readiness programs into one program called “test readiness,” so that it would be
difficult to split them into the two components. Of the requested $24.9 million,
however, it would take at least $15.0 million to maintain the current 24- to 36-month
posture.190 The NDAA provided the amount requested for test readiness; the
EWDAA did as well, but conferees stated their expectation that NNSA would focus
on restoring a 24-month posture before requesting “significant additional funds to
pursue” an 18-month posture.
Consistent with the legislation, NNSA plans to move toward a 24-month
readiness posture in FY2004 and achieve it in early FY2005. It plans to reach an 18-
month posture later in FY2005, and requests $30.0 million for FY2005 for test
Between 1951 and 1992, the United States conducted nuclear tests on a routine
basis, with the numbers ranging from 96 in 1962, to 0 in 1959 and 1960 due to a
moratorium, to 6 in 1992, the most recent year of U.S. testing, for a total of 1,030
tests.191 As a result, the United States was always prepared to test, although “18
189 U.S. Department of Energy, Office of Management, Budget and Evaluation/CFO,
Department of Energy FY 2004 Congressional Budget Request. Volume 1, National
Nuclear Security Administration. DOE/ME-0016, February 2003, p. 293.
190 Personal communication, Kerry Webb, Manager, Test Readiness Program, NNSA,
September 29, 2003.
191 Department of Energy, United States Nuclear Tests, July 1945 Through September
1992, p. viii. In addition to the 1,030 U.S. tests, 24 joint U.S.-U.K. tests were held at the
Nevada Test Site
months was a minimal time to design and prepare most tests.”192 All U.S. tests since
In October 1992, the United States began a moratorium on nuclear testing that
continues to the present. Ambassador Linton Brooks, NNSA Administrator, said in
December 2003, “The policy of the United States is that there is not now a need for
testing.” He further stated that he did not see a reason to resume testing in the next
two or three years.193
During the era of testing, DOE and its predecessors conducted several types of
nuclear tests, each with its own goals, schedules, and characteristics. Some are
explained here; it is useful to understand the differences in order to place test
readiness in context.
!Weapons modification and development. New-design warheads ( i.e.
those with a new primary or secondary stage, typically required
nuclear testing. Most U.S. nuclear tests were for weapons
development. In addition, nuclear testing was occasionally used to
understand a problem in a stockpiled warhead, or to check a fix
made to correct the problem, or for both purposes.194 To conduct
such tests, the labs had to design the test, fabricate a one-of-a-kind
test device, fabricate the “diagnostics” (the elaborate set of
instruments needed to gather the data), and emplace and seal the
device and diagnostics in a shaft. These tests typically required 18
to 24 months,195 counting from when the test was placed on the test
calendar until it was conducted.196 Most tests used standard
diagnostics; designing non-standard diagnostics added time.
Developing a new warhead typically required several tests; some
fixes required one test, others needed several. Most modification
192 U.S. Department of Energy, National Nuclear Security Administration, Report to
Congress: Nuclear Test Readiness, April 2003, p. 6.
193 “The Bush Administration’s Views on the Future of Nuclear Weapons: An Interview
with NNSA Administrator Linton Brooks,” Arms Control Today, January/February 2004:
194 Information provided by Livermore staff, September 17, 2003. A Livermore study states,
“Fifteen U.S. nuclear weapon systems have required post-deployment nuclear testing to
identify or resolve problems.” George Miller, Paul Brown, and Carol Alonso. Report to
Congress on Stockpile Reliability, Weapon Remanufacture, and the Role of Nuclear Testing,
report UCRL-53822, Lawrence Livermore National Laboratory, October 1987, p. 19. Note
that many of the 15 were older systems.
195 U.S. Department of Energy. National Nuclear Security Administration. Nevada
Operations Office. Enhanced Test Readiness Cost Study, July 1, 2002, DOE/NV-828, p.
196 Information provided by NNSA staff, September 11, 2003.
and development tests were conducted in vertical shafts in the
ground; drilling them typically took six to eight weeks.197
!Effects tests. Other tests examined how a nuclear explosion would
affect military equipment, materials, electronics, nuclear warheads,
etc., by exposing them to radiation from a nuclear explosion. These
tests typically used a low-yield nuclear device designed to be a
reasonably predictable radiation source; in these cases, design and
instrumentation of the device itself were not at issue. The
experiments required elaborate preparation, such as creating a
vacuum chamber to test how x-rays would affect satellites. Effects
tests typically used horizontal tunnels and experimental chambers
bored into a mountain; excavation of the site for one such test, using
three shifts a day, might take a year.198 Planning the excavation,
emplacing the experiments, and sealing the tunnels took added time.
Such tests were “inherently a two to three year process.”199 A small
fraction of U.S. tests were of this type.
!Stockpile confidence tests. Beginning in the 1970s, DOE would pull
a warhead from the field (e.g., a bomb deployed at a bomber base),
modify it minimally as needed to make certain that its yield would
not exceed the 150-kt ceiling of the Threshold Test Ban Treaty and
to make it compatible with testing procedures, and test it. The
purpose was to assure that weapons produced in quantity and then
deployed — as distinct from one-of-a-kind devices built for weapons
development tests — had the expected yield and met other basic
performance criteria. These tests typically used diagnostics
comparable to weapon development tests but less complex than for
effects tests. They required less time for preparation than weapon
development tests, typically about a year. They were conducted in
vertical shafts, a little less than once a year.200
!Tests intended as political statements. When the Soviet Union
terminated the 1958-1961 nuclear test moratorium in September
1961, the United States responded with a series of tests beginning
two weeks later,201 even though, by one analysis, “The United States
197 Ibid., p. 18.
198 Office of Technology Assessment, The Containment of Underground Nuclear
Explosions, p. 20.
199 Department of Energy, Enhanced Test Readiness Cost Study, p. 21.
200 Information in this paragraph and note provided by Lawrence Livermore National
Laboratory staff, September 10 and 17, 2003. The first stockpile confidence test (SCT)
using a format agreed to by DOD and DOE was held in 1979. Another four tests — not
called SCTs but that were for all practical purposes — were held beginning in 1972 before
the first SCT. Counting these four tests, DOE conducted a total of 17 SCTs.
201 Department of Energy, United States Nuclear Tests, July 1945 Through September 1992,
was not prepared to conduct major experiments!”202 All these tests
had yields below 20 kt.203 Such tests might be done in the future to
respond to a nuclear test by Russia or China. They would surely use
existing instruments and nuclear devices, and would probably
involve experiments planned in advance to maximize their technical
value. NNSA finds that it may be feasible to conduct a test of this
sort in as little as six months, though that posture would entail added
costs compared to an 18-month posture, would divert personnel
from other tasks, and would require considerable advance
preparat i on.204
At present, the only nuclear-related tests that NNSA is conducting are
“subcritical experiments” (SCEs). CRS offers the following definition based on
documents and on discussions with DOE and laboratory staff: “Subcritical
experiments involve chemical high explosives and fissile materials in configurations
and quantities such that no self-sustaining nuclear fission chain reaction can result.
In these experiments, the chemical high explosives are used to generate high
pressures that are applied to the fissile materials.” Twenty SCEs were held between
July 1997 and September 2003. SCEs study the behavior of plutonium.205 One
common purpose is to determine if material changes induced by radioactive decay
of aged plutonium would degrade weapon performance. All have been held in a
tunnel complex about 1,000 feet underground at Nevada Test Site. The United States
sees them as consistent with the moratorium because they produce no fission chain
reaction. They are also of value for maintaining the test site and exercising test-
NNSA plans to conduct three SCEs in FY2004 and two or more in FY2005.206
One of the FY2004 experiments, named “Armando,” and two FY2005 tests,
“Unicorn” and “Krakatau,” will be more complex than many of the earlier
experiments. Unicorn, for example, will be the first conducted in a vertical shaft dug
from the surface. According to NNSA,
Initial site preparation for this experiment is underway. This activity, and the
means for emplacement of the experimental hardware into the vertical hole, will
202 Harold Karan Jacobson and Eric Stein, Diplomats, Scientists, and Politicians: The United
States and the Nuclear Test Ban Negotiations, Ann Arbor, University of Michigan Press,
203 Department of Energy, United States Nuclear Tests, July 1945 Through September 1992,
p. ii, 10-12.
204 Department of Energy, Nuclear Test Readiness, p. 8-9.
205 See ibid., p. 15, regarding the first 19 tests. Regarding the 20th test, see U.S. Department
of Energy. Nevada Site Office. “National Nuclear Security Administration Scientists to
Conduct Piano Subcritical Experiment,” news release, September 16, 2003, 1 p. Available
206 Information provided by NNSA staff, February 26, 2004.
appear visually similar to those employed in underground nuclear tests conducted
prior to the 1992 moratorium. ... In addition to providing important information
for stockpile stewardship, the Unicorn subcritical experiment will exercise key
NTS capabilities not otherwise exercised in experiments carried out at the U1a207
Some of the smaller, simpler SCEs held at the U1a complex may be replaced
by the Joint Actinide Shock Physics Experimental Research Facility, a gas gun that
accelerates a projectile to high speed and sends it into a mass of plutonium.208 As a
result, the schedule of subcritical experiments for FY2005 and beyond is uncertain.209
In the future, the most probable type of nuclear test (as distinct from SCEs)
would be to resolve a defect that emerged in a stockpiled weapon to remedy a safety
or reliability problem. Such tests would probably be similar in form to weapon
modification or development tests. Regardless of purpose, any future tests would be
the first done in conjunction with the stockpile stewardship program, which came
into existence after the moratorium began. Preparation for testing would involve
three parallel tracks:
!(1): Technical work involving the nuclear explosive. To fix
a warhead problem, NNSA would “assess the problem,
develop and implement a solution, and plan and execute a test
that would provide the precise information needed to certify
the ‘fix.’”210 NNSA estimates that it would take around 18
months to prepare this type of test.211 As discussed below,
some question this estimate. Technical preparation for other
types of tests would take different lengths of time. Tests to
make a political statement could be done in less than 18
months if the needed support activities were in place, and
effects tests would take more than 18 months.
!(2): Support activities that can be done in advance to prepare
for a generic test, such as conducting analyses of test safety,
training personnel who can prepare the diagnostic equipment
for a test, updating test-related agreements with federal and
state agencies to ensure compliance with federal regulations,
207 U.S. Department of Energy. National Nuclear Security Administration. Nevada Site
Office. “‘Unicorn’ Subcritical Experiment Planned,” news release, August 23, 2003, 1 p.
Available at [http://www.nv.doe.gov/news%26pubs/newsreleases/Pdfs/Unicorn
208 For more detail, see U.S. Department of Energy. National Nuclear Security
Administration. Nevada Site Office. “Nevada Test Site: JASPER.” DOE/NV-709, rev. 1,
February 2003, 2 p.
209 Information provided by NNSA staff, February 26, 2004.
210 Department of Energy, Nuclear Test Readiness, p. 1.
and obtaining or locating heavy equipment needed to prepare
for a test.212
!(3): Support activities for a specific test, such as designing
and building instruments to acquire the needed data, and
analyzing the safety of the particular test in its emplacement
Test readiness involves bringing the generic steps to the point where completing
them and support activities for a specific test would take a specified length of time.
An NNSA source indicated that the need to spend 18 months to complete the first
track is what led NNSA to select 18 months as the test readiness goal.213
History Through the FY2003 Budget Cycle
When the 1992 moratorium started, it was unclear how long it would last. The
legislation creating the moratorium permitted it to end no sooner than July 1, 1993,
but President Clinton continued it several times in the next several years. At the
same time, he considered the possibility of a return to testing. His decision in August
1995 to press for a true zero-yield CTBT (as opposed to a treaty permitting nuclear
tests of very low yield) was conditioned on several “safeguards,” actions that the
United States would take consistent with the treaty in order to protect U.S. interests
despite the loss of testing. Safeguard F was:
The understanding that if the President of the United States is informed by the
Secretary of Defense and the Secretary of Energy ... that a high level of
confidence in the safety or reliability of a nuclear weapon type which the two
Secretaries consider to be critical to our nuclear deterrent could no longer be
certified, the President, in consultation with Congress, would be prepared to
withdraw from the CTBT under the standard “supreme national interests” clause214
in order to conduct whatever testing might be required.
As a result, it was unclear how ready the nuclear weapons laboratories and the
Nevada Test Site should be to conduct a nuclear test. Maintaining readiness to
conduct a test promptly would increase cost at the test site and the labs, and require
personnel to spend time on readiness rather than on ongoing tasks. In a November
1993 Presidential Decision Directive, President Clinton decided to maintain the
ability to conduct a nuclear test in 24 to 36 months of a presidential order to test.
This so-called test readiness posture was first established in FY1996.215
Congress, the Administration, and others have expressed concern over the
length of time needed to test. In January 1996, the Senate passed, 87-4, the
212 Ibid., p. 7.
213 Information provided by NNSA staff, February 27, 2004.
214 U.S. White House. Office of the Press Secretary. “Fact Sheet: Comprehensive Test Ban
Treaty Safeguards,” 1 p., August 11, 1995.
215 Department of Energy. Nuclear Test Readiness, p. 4.
resolution of ratification of the START II Treaty. The resolution included a number
of declarations expressing the sense of the Senate. Declaration 12(E) included the
following language: “The United States is committed to maintaining the Nevada
Test Site at a level in which the United States will be able to resume testing within
one year following a national decision to do so.”216 In its FY2001 report, the Foster
panel argued that the current posture of 24 to 36 months was “excessive” and that the
United States should not be in a position where increasing test readiness was viewed
as indicating a major problem with the nuclear arsenal. It noted some reluctance by
NNSA and DOD to increasing test readiness because the test moratorium seemed
unlikely to end, so that funds spent on test readiness would not be available for other
projects. It recommended a test readiness posture of 3-12 months, and having
devices for nuclear testing available on short notice.217 The purportedly leaked
December 2001 Nuclear Posture Review asserted that “the current 2-3 year test
readiness posture will not be sustainable as more and more experienced test
personnel retire” and “the 2-3 year posture may be too long to address any serious
defect that might be discovered.”218
There was some uncertainty over the state of test readiness. In October 1999,
Secretary of Defense William Cohen, in a statement before the Senate Armed
Services Committee, maintained, “We would be able to conduct a nuclear test within
18 months to two years of a decision to do so.”219 The next day, Secretary of Energy
Bill Richardson testified, “The test site is up and running and ready — while we are
capable of fielding a well instrumented test in 18 to 24 months, my scientists tell me
we, if pressed, could conduct a simple test in less than one year.”220
DOD and NNSA conducted several studies between 1999 and 2003 to
determine the time needed to conduct a nuclear test and what the optimum posture
should be. Based on these studies,
NNSA concluded that because of a loss of expertise and degradation of some
specific capabilities, the U.S. would more likely require about 36 months to test,
with less confidence in being able to achieve the 24-month end of the range.
Furthermore, as time passes without further action, the 36-month posture is221
viewed as increasingly at risk.
216 U.S. Congress, Congressional Record, “Treaty with the Russian Federation on Further
Reduction and Limitation of Strategic Offensive Arms (The START II Treaty),” resolution
of ratification, p. S461.
217 John Foster, Chairman, et al. FY2001 Report of the Panel to Assess the Reliability,
Safety, and Security of the United States Nuclear Stockpile. March 2002, p. 28-29.
218 Nuclear Posture Review [Excerpts], p. 35-36.
219 U.S. Congress, Senate Committee on Armed Services, Comprehensive Test Ban Treaty.
Hearings, 106th Congress, 1st Session, [hearings held October 6 and 7, 1999], S. Hrg. 106-
220 Ibid., p. 109.
221 U.S. White House, Office of the Press Secretary. “Fact Sheet: Comprehensive Test Ban
Treaty Safeguards,” 1 p., August 11, 1995.
While NNSA’s Nevada Site Office, which manages the Nevada Test Site,
received funds in 2001 and 2002 to buttress the 24- to 36-month test readiness
posture, NNSA and DOD favored enhancing test readiness. In 2002, they decided
to plan to move to an 18-month posture. The Nuclear Weapons Council endorsed
this transition in September 2002, and on March 13, 2003, NNSA provided Congress
a notification of intent to make this transition.222 Congress provided $15 million, the
requested amount for enhanced test readiness, in H.J.Res. 2 (P.L. 108-7), the
Consolidated Appropriations Resolution for FY2003.223
In action on the FY2003 National Defense Authorization Act, P.L. 107-314, a
provision in the House bill required the Secretary of Energy to report, with the
FY2004 budget submission, on a plan and budget to enhance test readiness,
specifically the time and budget for a one-year test readiness posture. The Senate bill
had no similar provision. The conference bill contained a provision, Section 3142,
SEC. 3142. Plans for Achieving Enhanced Readiness Posture for Resumption by
the United States of Underground Nuclear Weapons Tests.
(a) Plans Required. — The Secretary of Energy, in consultation with the
Administrator for Nuclear Security, shall prepare plans for achieving, not later
than one year after the date on which the plans are submitted under subsection
(c), readiness postures of six months, 12 months, 18 months, and 24 months for
resumption by the United States of underground nuclear weapons tests.
(b) Readiness Posture. — For purposes of this section, a readiness posture of a
specified number of months for resumption by the United States of underground
nuclear weapons tests is achieved when the Department of Energy has the
capability to resume such tests, if directed by the President to resume such tests,
not later than the specified number of months after the date on which the
President so directs.
(c) Report. — The Secretary shall include with the budget justification materials
submitted to Congress in support of the Department of Energy budget for
FY2004 (as submitted with the budget of the President under section 1105(a) of
title 31, United States Code) a report on the plans required by subsection (a). The
report shall include —
(1) an assessment of the current readiness posture for resumption by
the United States of underground nuclear weapons tests;
(2) the plans required by subsection (a) and, for each such plan, the
estimated cost for implementing such plan and an estimate of the
annual cost of maintaining the readiness posture to which the plan
222 Ibid., p. 1, 2.
223 U.S. Congress, Committee of Conference, Making Further Continuing Appropriations
for the Fiscal Year 2003, and for Other Purposes. H.Rept. 108-10, 108th Congress, 1st
Session, USGPO, 2003, p. 904.
(3) the recommendation of the Secretary, developed in consultation
with the Secretary of Defense, as to the optimal readiness posture for
resumption by the United States of underground nuclear weapons
tests, including the basis for that recommendation.
The conference report further “[encouraged] the Secretary of Energy to submit plans
for achieving and the cost of achieving and maintaining the recommended test
readiness posture with, or as part of, the report required by the provision.”224
The required report, with a cover date of April 2003, recommended a test
readiness posture of 18 months; estimated that achieving this posture by the end of
FY2005 would cost $83 million for the three-year transition period and that
sustaining it would cost $25-$30 million a year thereafter for resources unique to
nuclear testing, vs. about $15 million a year for this purpose for the current posture;
assessed that the erosion of expertise over time due to retirements and the long time
since the last nuclear test make the 24-36 month posture more likely a 36-month
posture, with even that “viewed as increasingly at risk”; and argued that a 6-12 month
posture would require “a substantial diversion of personnel and facilities at the
laboratories and production plants,” and at the shorter end of that posture “would
represent a major redirection of the stockpile stewardship program.”225 NNSA also
noted that it costs some $225 million a year to support general requirements at the
Nevada Test Site such as infrastructure, personnel, and equipment.
Legislative Actions in the FY2004 Budget Cycle
In S. 1050, Section 3132 required the Secretary of Energy to achieve an 18-
month nuclear test readiness posture by October 1, 2006, and to maintain that posture
thereafter. If as a result of the review for the report required by P.L. 107-314, Section
3142, the Secretary, in consultation with the Administrator for Nuclear Security (the
head of NNSA), determined that a posture other than 18 months was preferable, the
Secretary “may, and is encouraged to, achieve and thereafter maintain” that posture,
and shall submit a report stating the preferred posture and the reasons for it. NNSA’s
report on test readiness, however, indicated a preference for 18 months. In H.R.
1588, Section 3116 permitted DOE to obligate not more than 40 percent of FY2004
funds for nuclear test readiness until 30 days after DOE submitted the report required
by P.L. 107-314, Section 3142. NNSA submitted that report, dated April 2003.
There were no amendments to these provisions, or any others on enhanced test
readiness, in the House or Senate.
The House Appropriations Committee recommended eliminating the $24.9
million requested for “enhanced test readiness” on grounds of inadequate budget
justification and dubious program credibility. It was “concerned with the open-ended
commitment to increase significantly funding for the purpose of Enhanced Test
Readiness without any budget analysis or program plan to evaluate the efficiency or
224 U.S. Congress, Committee of Conference, National Defense Authorization Act for Fiscal
Year 2003. H.Rept. 107-772, 107th Congress, 2nd Session, USGPO, 2002, p. 786.
225 Department of Energy, Nuclear Test Readiness, p. 1, 2, 5, 6.
effectiveness of this funding increase.”226 It questioned NNSA’s ability to maintain
an 18-month readiness posture during a test moratorium when that was the minimum
time to prepare a test when testing was ongoing. It linked budget and program issues:
“the Committee will not spend money on a perceived problem when the Department
[of Energy] has not provided a rationale or a plan that addresses the underlying
problems inherent in maintaining a testing capability during a testing moratorium.”227
The House passed H.R. 2754 as reported from the committee without amending the
Weapons Activities section, which includes test readiness.
The Senate Appropriations Committee recommended providing the full $24.9
million requested, but did not comment on the readiness program. On September 16,
the Senate rejected an amendment by Senator Feinstein on various nuclear programs,
as detailed under ACI. One provision of the amendment would have barred use of
funds in the bill for modifying the test readiness posture of the Nevada Test Site to
a readiness posture of less than 24 months.
Conference Issues and Outcomes
With the House having eliminated the $24.9 million requested for test readiness
and the Senate having provided the full request, the appropriation amount was at
issue. Complicating a decision, the House eliminated $24.9 million for “enhanced
test readiness,” while the Administration requested $24.9 million for “test readiness.”
Some of the requested funds were to maintain the 24- to 36-month posture and some
were to transition to an 18-month posture. Yet the House Appropriations Committee
expressed concern over the latter even as it “supports the continued maintenance of
the Nevada Test Site.”228 Eliminating the $24.9 million would have eliminated funds
for both postures. If conferees decided to reduce funds for test readiness, would they
have eliminated the full request, as the House voted to do, or eliminated only the
funds for moving to an 18-month posture (perhaps $9.9 million), or settled on some
other amount? (Note that the Nevada Test Site receives other funds. For example,
the budget category “Nevada Site Readiness,” for which DOE requested $39.6
million for FY2004, “[i]ncludes activities required to maintain the Nevada Test Site
(NTS) that are not unique to the test readiness mission, but do support the stockpile
DOE’s submission of the test readiness report rendered moot the limit on
obligations in the House defense authorization bill and the provision in the Senate
defense authorization bill regarding a test readiness posture of other than 18 months.
Those provisions were not at issue in conference. While not a matter in dispute
between House and Senate, authorization conferees could have considered whether
an 18-month test readiness posture should be achieved by October 1, 2006, as per the
Senate bill, or by the end of FY2005, the date to which DOD and NNSA agreed.
226 U.S. Congress, House Committee on Appropriations, Energy and Water Development
Appropriations Bill, 2004. H.Rept. 108-212, p. 148-149.
227 Ibid., p. 149.
228 Ibid., p. 148-150.
229 Department of Energy, FT 2004 Congressional Budget Request, p. 291.
They could also have followed the House Appropriations Committee’s lead and
required DOE to provide a more comprehensive justification before endorsing any
The NDAA conference report did not address test readiness. The EWDAA
conference report provided the full funds requested, but changed the test readiness
The conferees recognize that test readiness activities in Nevada were allowed to
atrophy during the last decade under the current nuclear test moratorium ...
However, the conferees expect the NNSA to focus on restoring a rigorous test
readiness program that is capable of meeting the current 24-month requirement
before requesting significant additional funds to pursue a more aggressive goal230
of an 18-month readiness posture.
Thus the conferees, by basing their decision on the need for NNSA to restore a
24-month posture before proceeding to an 18-month posture, simultaneously avoided
the issue of what aspects of test readiness would have been cut by reducing funds
below the request; expressed their support for reversing the “atrophy” of the test
program; pressed NNSA to “restor[e] a rigorous test readiness program”; implied that
it would be technically difficult to restore an 18-month posture; reflected concerns
expressed by the House Appropriations Committee, Senator Feinstein, and others
about moving to an 18-month posture; and left the door open to moving to that
The FY2005 Budget Cycle
The FY2005 request for test readiness is $30.0 million.231 As noted, conferees
on the FY2004 energy and water development appropriations conference report
“expect” NNSA to move to a 24-month posture before requesting “significant
additional funds” to move to an 18-month posture. NNSA indicates that that
provision does not restrict its actions on test readiness. It anticipates reaching a 24-
month posture in early to mid FY2005, and hopes to achieve an 18-month posture by
the end of FY2005.232 Unlike ACI and RNEP, the budget document does not provide
out-year budgets for test readiness. NNSA estimates, however, that it can maintain
an 18-month readiness posture for $25 million a year once it has established that
posture, which the FY2005 request would enable it to do.
Issues for Congress
Is an 18-month test readiness posture desirable? Some see the current
24- to 36-month posture as adequate. In this view, enhanced test readiness would
divert resources, as well as the time of experts, to a task that is unlikely to be carried
out, to the detriment of the ongoing stockpile stewardship program that has so far
230 Committee of Conference, Making Appropriations for Energy and Water Development
for the Fiscal Year Ending September 30, 2004 ..., H.Rept. 108-357, p. 159-160.
231 Department of Energy, FY 2005 Congressional Budget Request, vol. 1, p. 86.
232 Telephone conversation with NNSA staff, February 26, 2004.
enabled the United States to maintain its nuclear stockpile without testing. They
believe that the current test readiness program has sufficient means to maintain skills,
such as subcritical experiments, simulations, and exercises. They point to a National
Academy of Sciences panel that stated:
We judge that the United States has the technical capabilities to maintain
confidence in the safety and reliability of its existing nuclear-weapon stockpile
under the CTBT, provided that adequate resources are made available to the
Department of Energy’s (DOE) nuclear-weapon complex and are properly233
focused on this task.
Some fear that enhancing test readiness could have adverse international
consequences. One advocacy group was “alarmed that the Administration and
Congress would take steps toward resuming nuclear testing, which would undo
decades of arms control and set the stage for a new arms race.”234 Further, “U.S.
resumption of full-scale nuclear testing would threaten the viability of the Nuclear
Non-Proliferation Treaty.” Representative Markey and 55 House colleagues wrote
to the House Appropriations Committee that they “are concerned that the proposed
enhanced test readiness program will renew interest in testing among other
Stewardship may not require an 18-month readiness posture, in the view of
those opposed to shortening the posture, because test readiness could be improved
on an ad hoc basis if needed. It would in all likelihood take months to convince the
President and Congress that a test was needed. The interval between discovery of a
problem and a political decision to test could be used for preliminary planning for the
test, reducing the time between the decision to test and the conduct of the test.
The other main purpose for testing beside stockpile stewardship is development
of new weapons. Those favoring the current readiness posture see an 18-month
posture as irrelevant for that purpose. The idea that the United States can only
respond to a particular threat by testing a new weapon in 18 months, and then
producing and using it, is not credible in this view, given the massive U.S.
conventional and nuclear forces in being.
Those who would enhance test readiness argue that the current posture is simply
too long. Secretary of Energy Spencer Abraham said, “Should our scientists decide
233 National Academy of Sciences. Committee on Technical Issues Related to Ratification
of the Comprehensive Nuclear Test Ban Treaty. Technical Issues Related to the
Comprehensive Nuclear Test Ban Treaty. Washington, National Academy Press, 2002, p.
234 Alliance for Nuclear Accountability, letter to defense authorization conferees, September
235 Letter from Representative Edward Markey and 55 other Representatives to Honorable
Sonny Callahan, Chairman, House Appropriations Committee, Energy and Water
Development [Subcommittee], and Honorable Peter J. Visclosky, Ranking Member, House
Appropriations Committee, Energy and Water Development [Subcommittee], July 9, 2002,
p. 2, on RNEP and test readiness, [http://www.house.gov/markey/Issues/iss_nonprolif
we cannot certify the reliability of our nuclear stockpile, we must be capable of
conducting a nuclear test in a much shorter time frame than the current three
years.”236 NNSA says that the 24- to 36-month posture has become closer to a 36-
month posture, and even that is at risk as people formerly involved in the test
program retire.237 It is difficult to maintain skills that would be needed for a
resumption of testing under a 24- to 36-month posture, it is argued, because that
posture implies that test-related activities would only be a minor part of the workload
for staff who would be responsible for conducting nuclear tests. A more active test
readiness program would sharpen their skills and help recruit people with skills
needed for testing. It may even prove infeasible to stabilize the posture at its present
unsatisfactory level as skills will erode. In this view, the choice would be between
continuing current trends and reversing them. They dismiss charges that enhancing
test readiness would lead automatically to testing; a decision to test would be decided
separately by the President and Congress.
Those who would improve test readiness divide into two groups. NNSA claims
that 18 months is the optimal time:
An 18-month posture is appropriate because this is the minimum time we would
expect it would take, once a problem was identified, to assess the problem,
develop and implement a solution, and plan and execute a test that would provide
the information needed to certify the ‘fix.’ An 18-month posture also reflects
what is cost effective while continuing to support other stockpile stewardship238
Others feel that 18 months is too long. As noted, the Senate called for a one-
year test readiness posture and the Foster panel recommended a 3-12 month posture.
Still others believe that changing the test readiness posture would have little
practical significance because technical considerations determine the time needed to
conduct various types of tests. A press report quotes Raymond Jeanloz, a member
of the former NNSA Advisory Committee, on this point:
The committee was told by the national nuclear laboratories that “the nation
would be able to perform a test in 3 to 6 months” if the goal was simply to
produce an explosion, he said. “From the labs’ point of view, until they know
why they would have to have a test to address some hypothetical technical
problem, they don’t know how long it would take them. So this whole business
of a three-year, or a one-and-a-half year, or a half-year delay before they can test239
is incredibly artificial,” he said.
236 Spencer Abraham, “Facing a New Nuclear Reality,” Washington Post, July 21, 2003: 21.
237 Department of Energy, Nuclear Test Readiness, p. 5.
238 Ibid., p. 6.
239 David Ruppe, “Energy Department might not release disbanded committee’s report,”
GovExec.com, August 13, 2003: [http://www.govexec.com/dailyfed/0803/081303gsn2.htm]
The advisory committee was created pursuant to the Federal Advisory Council Act (5 USC
App.), and was reportedly disbanded in June 2003; ibid.
Is an 18-month posture feasible for stockpile stewardship or
weapons development tests? As discussed earlier, tests to meet a political
need could be conducted rapidly, while some complex tests, such as effects tests,
would probably take longer than 18 months even with enhanced test readiness.
Judging from history, such tests would be rare. At issue would be the time to
conduct the most likely types of tests, notably those to remedy a problem in an
existing warhead discovered through the stockpile stewardship program, or those to
design new warheads.
NNSA claims that an 18-month posture should be feasible.
At the time of an active underground test program, 18 months was a minimal
time to design and prepare most tests, and is therefore a reasonable indicator of
the time that would be required to prepare a device for testing if the details of the240
planned test are not known in advance.
Critics respond that this argument is illogical. If it took a minimum of 18
months to prepare a test when the test program was in full swing, they ask, how can
the nuclear test program expect to establish that as a maximum after more than a
decade without testing? In the intervening years, test expertise has eroded through
disuse, many people formerly in the test program have shifted to other areas or have
retired or died, fewer new staff have been trained with these skills than have left the
test program, construction crews would have to be found and trained, new diagnostic
instruments would have to be designed to replace technology from 1992. Further,
NNSA would have to meet regulatory standards for environment, safety, and health
added since 1992.241 It may be possible to reach 18 months eventually, but given the
uncertainties, costs, and personnel issues, it is uncertain when that could occur.
Debate over the four nuclear initiatives will in all likelihood continue for some
years. This report concludes by discussing some substantive and logical aspects of
the debate that cut across the four initiatives. The topics below speak to the nature
of the debate in 2003 and may bear on it in future years.
Four small issues, or one large one? Supporters of the four initiatives
argue that opponents are reading too much into the initiatives. They are, in this view,
unexceptional elements of a nuclear weapons program, each with its own
justification. Supporters argue: lifting the ban on low-yield R&D simply permits
scientists to study the full range of weapon issues, as is the case for scientists in all
other nuclear weapon states; ACI is for early-stage studies; and RNEP is just a study
that at most could lead merely to adding a second type of earth penetrator weapon to
the U.S. inventory. A three-year test readiness posture is much too long, and testing
may be needed for other purposes, such as to check on a fix to a warhead type that
developed a defect. As Secretary of Energy Spencer Abraham said,
240 Department of Energy, Nuclear Test Readiness, p. 6.
241 Ibid., p. 10.
We are not planning to resume testing; nor are we improving test readiness in
order to develop new nuclear weapons. In fact, we are not planning to develop
any new nuclear weapons at all. Our goal is designed to explore the full range
of weapons concepts that could offer a credible deterrence and response to new
and emerging threats as well as allow us to continue to assess the reliability of242
our stockpile without testing.
Domestic and foreign critics see these measures as parts of a whole — a military
policy with a lowered threshold against nuclear use. They contend: lifting the low-
yield ban will enable pursuit of more-usable weapons; and, ACI will present the
military with new weapon options. In turn, military requirements for these new
weapons will lead to testing, and enhanced test readiness will permit expedited
testing and deployment of these weapons. They view RNEP as redundant, since the
United States has another nuclear penetrator (the B61-11), conventional penetrators,
and other nonnuclear options, but believe RNEP will make nuclear use that much
more likely by giving the United States yet another nuclear option. From this
perspective, these initiatives are consistent with other policy decisions, such as the
Administration’s policy of preemption, its willingness to use “all of our options,” its
disregard for negative security assurances, and its refusal to pursue the CTBT.
Critics see this policy as extremely insensitive to world opinion at best, and
Nuclear preemption and use: ambiguities and uncertainties. The
Administration’s policy statements leave a crucial uncertainty. In September 2002,
The National Security Strategy of the United States of America stated, “To forestall
or prevent ... [attacks with WMD] by our adversaries, the United States will, if
necessary, act preemptively.”243 And in December 2002, the National Strategy to
Combat Weapons of Mass Destruction stated, “The United States will continue to
make clear that it reserves the right to respond with overwhelming force — including
through resort to all of our options — to the use of WMD against the United States,
our forces abroad, and friends and allies.”244 These two statements do not specifically
declare that the United States will use nuclear weapons preemptively against possible
WMD attacks. Nonetheless, critics take them together as a sign that the
Administration is contemplating just such a course. Is the Administration willing to
rule out preemptive use of nuclear weapons? It may prefer ambiguity on this issue,
whether because it has not reached a decision or as a tactic to extend deterrence. But
by removing ambiguity, declaring that it would not use nuclear weapons
preemptively, the Administration could reduce fears worldwide about possible
nuclear war. At issue for the Administration is deciding whether the benefits of
resolving this ambiguity outweigh the costs.
Will the new weapons deter? The Administration, expressing concern
about the terrorist WMD threat, discusses a purpose of new or modified nuclear
242 Spencer Abraham, “Facing a New Nuclear Reality,” Washington Post, July 21, 2003: 21.
243 U.S. President, The National Security Strategy of the United States of America,
September 2002, p. 15.
244 U.S. White House, National Strategy to Combat Weapons of Mass Destruction,
December 2002, p. 3.
weapons as being to counter WMD facilities. The Administration may wish to
clarify how the nuclear initiatives — or any military means, for that matter — could
deter or retaliate against a terrorist WMD threat. If the perpetrator is unknown, or is
known but has no known address, what targets would be attacked, and would new
weapons offer any advantage for such attacks? Are any assets of sufficient value to
terrorists that holding them at risk might deter terrorist attacks? If so, would current
U.S. nuclear and conventional forces suffice for that purpose? Could the United
States perhaps deter terrorist use of WMD by threatening to overthrow (using
conventional forces) regimes that support terrorism, or by threatening to destroy
(using nuclear or other forces) facilities that these regimes see as critical?
There is a more fundamental deterrence issue. Deterrence depends on holding
at risk assets of great value to the leadership. But while the United States was able
to calculate what assets needed to be held at risk to deter the Soviet Union (and vice
versa), the United States cannot simply assume that what deterred the Soviet Union
would deter rogue states. The leaders of a nuclear-armed rogue state might calculate
that the United States would not use nuclear weapons against them because of the
risk of international opprobrium, or that U.S. nuclear weapons would not have a
decisive effect on the regime because it could hide or bury its own WMD and the
facilities of highest value to the leaders. They might be willing to accept the use of
several nuclear weapons against their country, especially if they follow the U.S.
debates and come to believe that the United States would only use a few low-yield
weapons of limited effectiveness.
Will research lead to testing, acquisition, and use? A key concern of
opponents of nuclear testing is the prospect that ACI, RNEP, and low-yield weapons
development could lead to nuclear tests. One could imagine various ways in which
these programs could lead to tests. Lifting the ban on low-yield R&D, for example,
might lead to ACI developing new types of low-yield warheads: earth penetrators,
chemical and biological agent neutralizers, warheads with reduced neutron or
enhanced gamma radiation output, and warheads with reduced electromagnetic pulse.
Future military interest in these warheads would, in this view, lead inexorably to
Yet there are plausible arguments that testing would not be needed. The B61-11
penetrator was converted from a B61-7 bomb without testing, and the RNEP
conversion anticipates converting an existing B61 or B83 without testing. Those
warheads are available in a number of yields; it may be that a lower-yield option
exists that would eliminate the need to develop a new-design EPW. The
Administration has enough confidence in nonnuclear interceptors to withdraw from
the ABM Treaty in order to deploy a system using them. Nuclear weapons tailored
to destroy chemical and biological agents may not be needed. The ability to destroy
these agents is arguably much more dependent on precise intelligence than on
weapon characteristics, some doubt the efficacy of nuclear weapons to destroy such
targets, and some doubt that the advantages of avoiding contamination by chemical
or biological agents outweigh the disadvantages of contaminating an area with
radiation. Beyond that, whether enhancing nuclear test readiness will lead to testing
depends on policy rather than technical judgments on future weapons R&D.
Some critics of the Administration’s nuclear initiatives fear that studying
nuclear weapon concepts or new weapons will lead to their use, but that result is far
from inevitable, supporters reply. Nuclear weapons are an option available to the
United States, but so are conventional forces, diplomacy, and economic leverage.
The United States has had low-yield nuclear weapons continuously since the 1950s,
and might have benefitted tactically from using them in the Korean War, Cuba, and
the Vietnam War. Yet despite these military benefits and options, Presidents
refrained from using these weapons. It appears, then, that nuclear weapons would
only be used as a last resort. On the other hand, any new weapon would provide a
President with a wider range of options for nuclear use. A President might find
options made available by current nuclear weapons to be unacceptable, but a new
weapon might tilt the tradeoff between costs and benefits in favor of nuclear use.
Any new weapon has limits to its military value. The claim that EPWs
may be the only way to destroy certain hardened and deeply buried targets does not
mean that any nuclear EPW will destroy any such target. Rather, nuclear EPWs
would make a difference only against some specified range of targets. A bunker
hardened to a given level and buried to depth X could be destroyed by a conventional
EPW, and buried more deeply at depth Y could not be destroyed by a nuclear EPW
of 5 kilotons yield. Only if the facility is buried between depths X and Y would that
EPW make a difference. The United States could increase the weapon’s yield,
permitting it to destroy the bunker at greater depth, but the target nation could
counter by increasing the hardness of the bunker, burying it deeper, or camouflaging
its location. Further, the U.S. intelligence may not know details of the geology above
the bunker or of the layout of a tunnel complex, both of which may greatly affect the
weapon’s effectiveness. Thus the weapon adds military value over only a specific
range of targets.