NUCLEAR COMMAND

AND STATUTORY CONTROL

DAKOTA S. RUDESILL*
Almost 50 years after Congress voted on limiting the President’s power to
initiate nuclear war, a half century after an intoxicated Commander in Chief
reportedly called for nuclear strikes, and 30 years after the Cold War and its
conversation about nuclear command and control ended, today the nation is
again discussing nuclear launch authority. This article emphasizes that this
interdisciplinary conversation is importantly legal, due in part to claims and
assumptions that nuclear weapons are constitutionally special –
unconstitutional, reserved only to the President, or usable only if Congress has
formally declared war. This article recommends instead that Congress make
nuclear weapons statutorily special. That is, Congress should recognize the
nightmarish risks associated with unfettered Executive power over nuclear
launch, acknowledge the importance of good process in decision-making, and
write specially tailored rules informed by the covert action statute and other
national security frameworks. This article moves the nuclear command and
control conversation forward, analyzing developments over the past 30 years in
military planning, international security, Supreme Court doctrine, and the
national security statutory regime that have strengthened the case for Congress
to craft guardrails to prevent abuse of the Commander in Chief power and foster
careful inter-agency deliberation. To catalyze further discussion, this article
includes the text of a model statute: a Nuclear Forces Control Act.

* Associate Professor, Moritz College of Law, and Senior Fellow, Mershon Center for
International Security Studies, The Ohio State University. This article draws on insights from
several scholarly fields as well as the expertise of military leaders and policymakers. I thank
colleagues with each of these perspectives for stimulating conversations or commenting on
drafts: Justin Anderson, James E. Baker, Robert Chesney, Dan Chiu, Geoffrey Corn, Mary
DeRosa, Brian Egan, Ned Foley, Steve Huefner, General Robert Kehler (ret.), David Koplow,
David Kris, Hans Kristensen, Major General Tim McMahon (ret.), Debby Merritt, Mark
Nevitt, Deborah Pearlstein, Brad Roberts, Peter Shane, Daniel Tokaji, Rachel
VanLandingham, Matthew Waxman, Amy Woolf, and other Ohio State colleagues. I also
thank participants at the Duke-Virginia Foreign Relations Roundtable, at the Legislation
Roundtable held at Yale Law School, and at conferences and workshops hosted by Stanford
University’s Center for International Security and Cooperation, Georgetown University Law
Center, the University of Nebraska School of Law, the Journal of National Security Law and
Policy, the U.S. Strategic Command, and the Center for Strategic and International Studies.
For support for this research project I thank the Moritz College of Law and its Center for
Interdisciplinary Law and Policy Studies (CILPS). For research assistance I thank Billy Braff,
Krystina Garabis, Alyson Houk, Brandon Miller, Patrick Ojeil, Erin Reinke, Thomas Rovito,
Alex Thierer, and Michael Walsh. I thank the editors of the Journal for their diligent work
amid all the disruptions of a global pandemic. I am responsible for all content and any errors.
The views expressed here are mine and do not imply endorsement by any U.S. Government
agency.

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

INTRODUCTION ...........................................................................................1
I. THREE ENDURING NUCLEAR NIGHTMARES...........................................6
A. Nuclear Nightmare Number One: Initiated or Temporally
Imminent Attack .........................................................................7
B. Nuclear Nightmares Two and Three: Rogue President, and
Precipitous President.................................................................13
II. NUCLEAR-CONVENTIONAL CONVERGENCE & THE NATIONAL
LEADERSHIP-LEVEL REVIEW GAP ......................................................19
A. The Most Likely Nuclear Employment Scenario .....................21
B. Operational Planning and Lawyering Up .................................24
III. THE CASE FOR PROCESS – AND A PROCESS STATUTE ..........................27
A. The Value of Deliberation and Inter-Agency Process ..............28
B. Implementation: The Need for a Statute ...................................33
IV. THE STRENGTHENED CASE FOR THE CONSTITUTIONALITY OF
STATUTORY RULES FOR NUCLEAR WEAPONS .....................................35
A. Reform Should Not be Dissuaded by the Constitutional
Conversation to Date ................................................................37
1. Cold War-era Statutory Proposals .....................................37
2. Claims that Nuclear Weapons are Constitutionally Special
40
B. Separated and Shared Powers Over Nuclear Weapons.............42
C. Firmer Footing for Nuclear Rule Writing .................................50
1. Change in the International Security Environment ............50
2. Post-9/11 Wartime Supreme Court Precedents ..................51
3. Special Statutes: Covert Action and FISA ..........................54
V. MAKING NUCLEAR WEAPONS STATUTORILY SPECIAL .......................59
A. The Nuclear Forces Control Act (NFCA).................................59
1. Purposes and Projects ........................................................59
2. Restrictions on Force and Funding ....................................61
3. Reporting Requirements and Norm Building......................64
4. Protecting the Statute..........................................................68
B. Potential Objections and Alternatives .......................................70
1. Potential Objections ...........................................................70
2. Potential Alternatives .........................................................72
CONCLUSION .............................................................................................75
APPENDIX: DRAFT STATUTE .....................................................................77

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 1

INTRODUCTION

Acquiescence to a claim of expansive executive authority, U.S.

Supreme Court Justice Robert Jackson famously warned, allows that
power to become “a loaded weapon” available for use or abuse. 1 The
nuclear weapons over which the President has all but complete power are,
however, not metaphorical. More than 500 global-range U.S. ballistic
missiles are continually in alert postures, cocked and loaded with over
1,000 thermonuclear warheads. 2 Hundreds more warheads can be loaded
onto aircraft and other missiles.3 As all three branches of government and
the public have allowed for 75 years, the President need only reach for the
phone or the nuclear “football” to order a nuclear attack at any time,
without anyone’s authorization or second vote. 4
After decades of inertia, the nation has resumed a conversation about
nuclear command and control that has been dormant since the Cold War’s
end 30 years ago.5 The Defense Department, Congress, the nuclear policy

1 See Korematsu v. United States, 323 U.S. 214, 246 (1944) (Jackson, J., dissenting).
All three branches of government have repudiated the race-based deprivation of liberty at issue
in Korematsu and ratified at that time by the Court majority. See Trump v. Hawaii, 138 S.Ct.
2392, 2423 (2018) (“Korematsu was gravely wrong the day it was decided…”).
2 The “cocked and loaded” terminology is that of President Trump regarding U.S.
conventional capabilities for a potential strike on Iran. See Donald J. Trump
(@realDonaldTrump), TWITTER (June 21, 2019, 9:03 AM), https://perma.cc/H7QP-T9XB. It
applies equally well to the U.S. nuclear posture: nuclear-armed global range ballistic missiles
on land and at sea in alert postures. The numbers above are based on Kristensen & Korda,
adjusted downward due to a portion of the sea-based ballistic missile force being on
submarines in port or otherwise not in firing position. See Hans M. Kristensen & Matt Korda,
United States Nuclear Forces 2019, 75 BULL. ATOMIC SCIENTISTS 122 (2019). U.S.
submarine-launched ballistic missiles carry many warheads – multiple independently targeted
re-entry vehicles (MIRVs) – and U.S. land-based missiles could re-upload them. For
discussion, see Dakota S. Rudesill, MIRVs Matter: Banning Hydra-Headed Missiles in a New
START II Treaty, 54 STAN. J. INT’L L. 83 (2018) (arguing for treaty limiting MIRVs).
3 See U.S. DEP’T OF DEF., NUCLEAR POSTURE REVIEW 41–48 (2018) [hereinafter 2018
NPR].
4 The nuclear “football,” containing nuclear weapons information and secure
communications equipment, has been at the President’s side in one form or another since the
1960s.
5 At the end of the Cold War scholarly, practitioner, and public attention to nuclear
command and control sharply declined. For engagement near the end of the Cold War, see,
e.g., MANAGING NUCLEAR OPERATIONS (Ashton B. Carter, John D. Steinbruner & Charles A.
Zraket, eds., 1987) (non-legal scholars, policy experts, and civilian and military practitioners
analyze policy, military, technology, arms control, and psychological issues); FIRST USE OF
NUCLEAR WEAPONS: UNDER THE CONSTITUTION, WHO DECIDES? (Peter Raven-Hansen, ed.
1987) (legal scholarship on law and nuclear weapons). After the Soviet Union’s demise, law
professors stopped writing on the subject. Cf., ELAINE SCARRY, THERMONUCLEAR

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

community, legal scholars, and the public are engaged. 6 Presidential

candidates have debated a U.S. “no first use” policy.7 Frequently, the
nuclear command and control debate returns to a dilemma. A primary
virtue of the system’s legal, decision process, and technology architecture
is also a potential liability: concentration of discretion to use the world’s
most destructive weapons in one person. What could enable timely
presidential decision in the classic nuclear nightmare of an adversary
nuclear attack may permit other nightmares: an unwarranted launch order,
or a precipitous order where the necessity and legality of the strike are
questionable and the President has bypassed advisors and ignored pertinent
fact and law.
Nuclear command and control is getting renewed attention because the
possibility of U.S. use of nuclear weapons may be rising. 8 Russia has

MONARCHY: CHOOSING BETWEEN DEMOCRACY AND DOOM (2014) (professor of English

argues that nuclear command and control system is unconstitutional). Several law students
wrote related Notes.
6 See Sandra Erwin, U.S. STRATCOM to Take Over Responsibility for Nuclear
Command, Control and Communications, SPACENEWS (July 23, 2018),
https://perma.cc/ZPZ9-DWLD (describing changes in system’s organization regarding
communications); Jon. B. Wolfstahl (@JBWolfstahl), TWITTER (Dec. 10, 2018, 5:49 PM),
https://perma.cc/K6ZH-N45T (Secretary of Defense James Mattis reportedly “inserted
himself into the nuclear weapons chain of command”); Bruce G. Blair, Strengthening Checks
on Presidential Nuclear Launch Authority, ARMS CONTROL TODAY (Jan. / Feb. 2018),
https://perma.cc/9FGF-W4KD (calling for checks on presidential launch authority); S. 272,
116th Cong. (2019) (legislation introduced by Sen. Elizabeth Warren (D-MA) stating “It is
the policy of the United States to not use nuclear weapons first”); Restricting First Use of
Nuclear Weapons Act of 2017, H.R. 669, 115th Cong. (2017) (legislation introduced by Rep.
Ted Lieu (D-CA)); Restricting First Use of Nuclear Weapons Act of 2017, S. 200, 115th Cong.
(2017) (legislation introduced by Sen. Ted Markey (D-MA)); Authority to Order the Use of
Nuclear Weapons: Hearing Before the S. Comm. on Foreign Relations, 115th Cong. 22 (2017)
[hereinafter SFRC 2017 Hearing] (first congressional hearing on nuclear command and
control since 1976); Anthony J. Colangelo & Peter Hayes, An International Tribunal for the
Use of Nuclear Weapons, 2 J. PEACE & NUCLEAR DISARMAMENT 219 (2019) (arguing for
international tribunal to hold military officers accountable for illegal nuclear launch orders);
Richard K. Betts & Matthew C. Waxman, The President and the Bomb: Reforming the
Nuclear Launch Process, 97 FOR. AFF. 119 (2018) (recommending the Secretary of Defense
and Attorney General must confirm a nuclear launch order and its legality). See also the
interdisciplinary conferences mentioned supra note *, which have included legal scholars.
7 See S. 272, supra note 6 (Sen. Warren legislation); Bruce Blair & Jon Wolfsthal, We
Still Can’t ‘Win’ a Nuclear War. Pretending We Could Is a Dangerous Fantasy, WASH. POST
(Aug. 1, 2019, 1:55 PM), https://perma.cc/N7TX-FACE (Sen. Warren argues for “no first
use”; former Vice President, then-presidential candidate Joe Biden proposed that nuclear
deterrence be the sole purpose of U.S. nuclear arms).
8 See Ernest Moniz & Sam Nunn, The Return of Doomsday: The New Nuclear Arms
Race – and How Washington and Moscow Can Stop It, 98 FOR. AFF. 150 (2019) (risk of
nuclear employment highest since Cuban Missile Crisis in 1962).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 3

revived as a geopolitical adversary of the United States, which withdrew

from a landmark nuclear arms treaty in August 2019 after Russian
violations.9 China and North Korea have enhanced their nuclear
capabilities, and tensions with Iran are spiking. 10 The risk of trans-national
terrorist networks acquiring weapons of mass destruction endures. The
United States has also elected an unconventional personality to the
presidency, one who has more explicitly threatened the “fire and fury” of
nuclear war than any predecessor. 11
The design of the nuclear command and control system is, however,
an urgent question that transcends the moment’s geopolitics or leadership.
It demands great care. Because nuclear deterrence directly depends on the
responsiveness of the system, any changes must be well justified. 12 Any
changes must facilitate good decision-making in the most perilous
circumstances by both the most and least prudent presidential
personalities.
This is not an easy conversation. The matter is bedeviled by its
importance and complexity. A cloak of classification shrouds many of the
system’s most important details. The status quo reflects decades of inertia
behind the notion that, for reasons of national security necessity, nuclear
weapons are the President’s weapons. The fact that the current system has
never been used operationally is both cause for celebration and a challenge

9 See Hans M. Kristensen & Matt Korda, Russian Nuclear Forces 2019, 75 BULL.
ATOMIC SCIENTISTS 73, 82 (2019) (Russia modernizing nuclear forces); Ilya Arkhipov, Putin
Warns U.S. of New Arms Race After Nuclear Deal’s Collapse, BLOOMBERG (Aug. 5, 2019,
9:23 AM), https://perma.cc/MT9D-H5WE (United States withdrawal from 1987 Intermediate-
range Nuclear Forces (INF) Treaty with Russia effective August 2019 in response to alleged
Russian violation spurs warnings and concerns of new nuclear arms race).
10 See Hans M. Kristensen & Matt Korda, Chinese Nuclear Forces 2019, 75 BULL.
ATOMIC SCIENTISTS 171, 171 (2019); Hans M. Kristensen & Robert S. Norris, North Korean
Nuclear Capabilities 2018, 74 BULL. ATOMIC SCIENTISTS 41, 41 (2018); Babak
Dehghanpisheh & Tuqa Khalid, Iran Makes New Nuclear Threats that Would Reverse Steps
in Pact, REUTERS (July 8, 2019, 7:01 AM), https://perma.cc/GN64-KXGK (Iran has enriched
uranium in violation of 2015 agreement, and threatened to restart deactivated centrifuges and
further enrich).
11 See Peter Baker & Chloe Sang-Hun, Trump Threatens ‘Fire and Fury’ Against North
Korea if It Endangers U.S., N.Y. TIMES (Aug. 8, 2017), https://perma.cc/B8HK-JTFV. See
also Donald J. Trump (@realDonaldTrump), TWITTER (Jan. 2, 2018, 7:49 PM),
https://perma.cc/EU9T-ACEH (President says he has “a Nuclear Button, but it is a much
bigger & more powerful one than [North Korea’s], and my Button works!”).
12 See Charlie Dunlap, The Danger of Tampering with America’s Nuclear Command and
Control System, LAWFIRE (Nov. 22, 2017), https://perma.cc/VK6Z-SRAC [hereinafter
Dunlap, The Danger of Tampering].

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

to analysis and reform.13 On the public record, we simply do not know

precisely how the decision process would unfold. We do, however, have
the opportunity and responsibility to think about how it should.
This article speaks to an interdisciplinary audience. For the legal
community, this article explains how policy, military, technology, and
other factual aspects of nuclear operations powerfully shape how the law
operates – currently and potentially – regarding “the bomb.” Meanwhile,
this article’s message to the policy, military, and legal communities is that
new approaches are possible. Nuclear weapons need not be assumed to be
constitutionally special: left entirely to one branch of government.
Instead, nuclear weapons should join a short list of critical national
security capabilities that are statutorily special: governed by
congressionally-written tailored decision process rules. Nuclear weapons
should join covert action and foreign intelligence surveillance on this list. 14
Like nuclear weapons, these matters were left too long to near-total
executive discretion. As process statutes, the covert action statute and
Foreign Intelligence Surveillance Act (FISA) run over the top of any
underlying legal authority for use of force. They allow effective and urgent
defense of the country, while promoting deliberation and discouraging
abuse of authority.
In terms of organization, this article begins by analyzing important
developments in the nearly 50 years since Congress voted on nuclear
command and control, in the half century since a drunken President
reportedly called for nuclear use, and in the 30 years since the Cold War
ended and the conversation among legal scholars about nuclear launch
authority went on hiatus.15 Part I explains that nightmarish risks remain
unaddressed in the nuclear command and control system at the national
leadership level. Part II explores convergence – how the nuclear and
conventional worlds have become more similar, in ways that augur toward
creation of decision process rules for nuclear use. Today, a nuclear use
scenario is likely to have more in common with the 2011 raid that killed
Usama bin Laden than with the Cold War worst-case scenario for which

13 President Truman’s 1945 decision to employ the atomic bomb against Japan and the
1962 Cuban Missile Crisis both predated the advent of the nuclear command and control
system we have inherited.
14 See 50 U.S.C. § 3093 (2019) (covert action); Foreign Intelligence Surveillance Act
(FISA), 50 U.S.C. § 1801 et seq (1978).
15 Before a 2018 piece on illegal nuclear orders, almost exactly three decades passed
since publication of the last law review article by a law professor on nuclear launch authority.
See Anthony J. Colangelo, The Duty to Disobey Illegal Nuclear Strike Orders, 9 HARV. NAT’L
SEC. J. 84 (2018); Peter Raven-Hansen, Nuclear War Powers, 83 AM. J. INT’L L. 786 (1989).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 5

the nuclear command and control system is optimized. In Part III, this
article sets out the value of robust decision process at the inter-agency
National Security Council (NSC) level of the Executive Branch, a forum
used effectively to make decisions about bin Laden and other complex,
highly classified national security matters. Because the President can
waive self-imposed rules, they should be provided by statute.
Part IV explains that the constitutional case for congressional authority
to legislate regarding nuclear use has grown stronger since the nuclear
constitutional conversation halted at the Cold War’s end. Thirty years
later, the renewed nuclear command and control conversation is often
distracted by problematic claims and assumptions. Changes in the
international security environment, developments in constitutional
doctrine, and entrenchment of process statutes regarding covert action and
surveillance all run against viewing nuclear weapons as constitutionally
special and left entirely to the President. These developments would
provide more authority for current legislation mandating “no first use.”
That, however, is a separate question from what this article recommends
in Part V: a decision process statute. Such a law would make decisions
about nuclear use, like covert action and surveillance, statutorily special.
Part V also addresses potential objections and outlines possible
alternatives. Finally, this article’s appendix includes a draft statute – a
prompt for the renewed dialogue to continue, for new thinking, and for
Congress to end its acquiescence.
Ultimately, this proposal seeks to provide rules and build norms that
would help ensure that nuclear weapons could only be employed in two
situations, both in which law other than the proposed statute provides
authority for the use of force. One use situation is imminent or initiated
adversary nuclear attack. The second is where adversary nuclear attack is
not imminent, the available time is used by the relevant national security
stakeholders to evaluate a range of nuclear and non-nuclear options,
preferably using the NSC process, and non-nuclear weapons would not be
sufficient. The statute this article recommends would be informed by the
covert action statute and other frameworks.
This proposal challenges longstanding habit and received wisdom. It
will face resistance – both reflexive and thoughtful – from individuals and
institutions invested in our risk-riven nuclear command and control
system. This new approach may also be criticized by those who accept
minority legal views of nearly unlimited presidential power generally, or
apply that view specifically to nuclear weapons. But making certain
critical national security capabilities statutorily special is now
longstanding elsewhere in the core of the national security legal regime.

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

And it is amply warranted here. Importantly, the statute this article

recommends would do nothing to undermine nuclear deterrence. It would
do nothing to impede proper use of the Commander in Chief power. It
would do nothing to frustrate necessary and legal use of nuclear weapons.
This proposal would, in contrast, help ensure that the government operates
as it should at the national leadership level regarding nuclear arms.
There is no guarantee that in our constitutional system a committed
rogue president could be stopped. But in minimally invasive fashion, this
proposal would provide the keepers of our nation’s loaded nuclear
weapons new grounds to resist impulsive or illegal orders. 16

I. THREE ENDURING NUCLEAR NIGHTMARES

In November 2017, the Senate Foreign Relations Committee held the

first congressional hearing on nuclear command and control in 41 years. 17
General Robert Kehler, the former chief of the nation’s nuclear command,
the U.S. Strategic Command, testified about three potential nuclear use
scenarios: an imminent or ongoing attack against the United States or its
allies, presidential consideration of a nuclear strike absent an apparent
threat, and strategic warning that nuclear attack against the United States
may be actively contemplated by an adversary. 18 All three are nuclear
nightmares.19 All three were discussed at the last hearings in 1976.20 Yet
four decades later the nuclear command and control system – in its legal,
process, and technology aspects – remains optimized only for the first
nuclear nightmare. Of course, the law prohibits unnecessary,
disproportionate, or otherwise illegal use of force of any kind. 21 But on the

16 For a thoughtful treatment of the larger phenomenon of resistance, see Rebecca Ingber,
Bureaucratic Resistance and the National Security State, 104 IOWA L. REV. 139 (2018)
(critiquing both antipathy for and too much faith in bureaucratic resistance to check a
misguided President).
17 See SFRC 2017 hearing, supra note 6.
18 See SFRC 2017 hearing, supra note 6.
19 Nuclear weapons are deeply terrifying, and the “nuclear nightmare” meme has been
around for some time. See, e.g., JOSEPH CIRINCIONE, NUCLEAR NIGHTMARES (2015) (policy-
focused book about current nuclear threats and arms control); NUCLEAR NIGHTMARES: THE
WARS THAT MUST NEVER HAPPEN (British Broad. Corp. Television 1979) (documentary film
on nuclear weaponry and deterrence).
20 See First Use of Nuclear Weapons: Preserving Responsible Control: Hearings before
the Subcomm. on Int’l Sec. & Scientific Affairs of the Comm. on Int’l Relations of the House
of Rep., 94th Cong. (1976) [hereinafter HIRC 1976 hearings].
21 See U.N. Charter art. 51 (inherent right of self-defense); U.S. DEP’T OF DEF., OFFICE
OF GEN. COUNSEL, LAW OF WAR MANUAL (2016) [hereinafter U.S. LAW OF WAR MANUAL],
specifically §§ 1.11.5 (under jus ad bellum “To constitute legitimate self-defense under

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 7

public record the precise contours of the nuclear command and control
system’s process norms are unclear, and are untested since a disturbed
President’s reported drunken nuclear instructions a half century ago. One
simply cannot be confident that the law of armed conflict or these uncertain
norms could restrain a President who is committed to ordering a nuclear
attack either without justification (what this article conceives as a Rogue
President) or before its legality, implications, and alternatives are fully
evaluated (a Precipitous President).

A. Nuclear Nightmare Number One: Initiated or Temporally

Imminent Attack

The Nuclear Command and Control System (NCCS) is, in the words
of the Defense Department, “a legacy of the Cold War.” 22 It was designed
with the unforgiving realities of missile flight times in mind. If the Soviet
Union or China were to be deterred from shooting first in a crisis –
launching a first strike that would decapitate the U.S. government or
destroy U.S. forces on the ground – a complex array of machines,
organizations, and people distributed across thousands of miles would
have to do a series of complicated things in a matter of minutes while under
nuclear attack.23 The system would have to detect and analyze the threat,
transmit that information to national leadership, evaluate the threat and
potential responses, make a decision, and communicate that decision
promptly, securely, and reliably to personnel in the field operating
missiles, submarines, and bombers in time for them to launch and inflict
unacceptable damage on the adversary before U.S. forces were destroyed
or otherwise rendered inoperative.24 The timelines are almost impossibly

customary international law, it is generally understood that the defending State’s actions must
be necessary” and “[P]roportionate to the nature of the threat being addressed”); Id. §§ 2.2,
2.4 (necessity and proportionality principles of the jus in bello international law of armed
conflict (LOAC), also referenced as International Humanitarian Law).
22 2018 NPR, supra note 3, at 56.
23 The Defense Department recently defined deterrence as ensuring that an adversary
understands that attack “will fail and result in intolerable costs for them.” See 2018 NPR,
supra note 3, at 20.
24 See OFFICE OF THE ASSISTANT SEC’Y OF DEF. FOR NUCLEAR, CHEM., AND BIOLOGICAL
DEF. PROGRAMS, NUCLEAR MATTERS HANDBOOK, 85-105 (2016), https://perma.cc/CL5E-
PMPC (describing system). Nuclear planners have been long concerned about the system’s
ability to operate under attack despite nuclear weapon effects that include computer-
destroying electromagnetic pulse (EMP). See Ashton P. Carter, Communications
Technologies and Vulnerabilities, in MANAGING NUCLEAR OPERATIONS, supra note 5, at 273,
273-78 (EMP effects); Walter Slocombe, Preplanned Operations, in MANAGING NUCLEAR
OPERATIONS, supra note 5, at 121, 137 (even if “dire predictions of near total C3 collapse

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

compressed. By some public estimates, U.S. national leadership would

have perhaps 10 minutes to evaluate an incoming ballistic missile attack
from the Eurasian landmass, make a decision about launch, and transmit
orders to the field. If an adversary launched ballistic or cruise missiles at
Washington, D.C., from a submarine close to the U.S. coastline in
depressed flight trajectories, according to some public estimates it could
be challenging for the national leadership to have even 200 seconds before
the first adversary warheads detonated. 25 Were an adversary nuclear
device clandestinely transported to within range of the national leadership
on the ground, warning time could also be zero. 26
The inherited system, designed during the Cold War, focuses on the
President. It does this for the constitutional reason of the President’s
position as Commander in Chief, and to reflect civilian control of the
armed forces. Presidential focus and the unity of command – rather than,
for example, vesting launch authority with a committee or involving
Congress – in theory also provides the U.S. government its most credible
possibilities of having meaningful decision time before nuclear war
begins, of shooting first in a pre-emptive strike if it looks like an adversary
is readying attack, of “launch under attack” with adversary missiles in the
air, or of surviving an adversary first strike and still being able to command
the U.S. arsenal.27 The credibility of the system working under sudden,

prove too pessimistic, something . . . will go wrong”). Concern about cyber threats is growing
and one driver of a multi-billion dollar modernization. See 2018 NPR, supra note 3, at viii.
25 See MARC AMBINDER, THE BRINK: PRESIDENT REAGAN AND THE NUCLEAR WAR
SCARE OF 1983 at 25-26 (2018) (less than three minutes if Soviets used submarine-launched
ballistic missiles); Slocombe, supra note 24, at 132-37 (other decision timelines).
26 One commonly discussed decapitation scenario involves an adversary nuclear weapon
secreted into the nation’s capital and detonated blocks from the White House. At that point,
the system’s decision focus would shift from the President to the next-in-line surviving
presidential successor, as provided by the Constitution and statute and organized by the federal
government’s Continuity of Operations (COOP), Continuity of Government (COG), and
Enduring Constitutional Government (ECG) plans. See U.S. CONST. art. II, § 1, cl. 6; U.S.
CONST. amend. XXV; 3 U.S.C. § 19 (2006); Presidential Policy Directive 40 (PPD-40),
National Continuity Policy, July 15, 2016 (unpublished document); Federal Continuity
Directive 1: Federal Executive Branch National Continuity Program and Requirements, Jan.
17, 2017, https://www.gpo.gov/docs/default-source/accessibility-privacy-coop-
files/January2017FCD1-2.pdf (continuity plan issued pursuant to and referencing PPD-40).
27 See U.S. CONST., art. II, § 2, cl. 1. The President can, and at least during the Cold War
did delegate authority to lower level officials. See, e.g., Slocombe, supra note 24, at 133-34;
Paul Bracken, Delegation of Nuclear Command Authority, in MANAGING NUCLEAR
OPERATIONS, supra note 5, at 352, 352-72. The defense doctrine of the North Atlantic Treaty
Organization (NATO) has long involved processes for delegating authority for use of U.S.
nuclear weapons. See, e.g., Catherine McArdle Kellher, NATO Nuclear Operations, in
MANAGING NUCLEAR OPERATIONS, supra note 5, at 445, 445-69.

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 9

massive, catastrophic nuclear attack, or during a limited or even protracted

nuclear war, is integral to nuclear deterrence as a theory and as practiced
for over half a century.28
Current procedures provide for a “threat conference” at which the
President would confer, to the extent they or their deputies are available,
with the Secretary of Defense, the Chairman of the Joint Chiefs of Staff,
and the Commanders of the U.S. Strategic Command and the North
American Aerospace Defense Command (NORAD).29 Other military
commanders (for example, the NATO Commander or regional combatant
commanders) or civilian officials, such as the Director of National
Intelligence or Secretary of State, may be added depending on the threat
and time available. These advisors would inform and advise. Selecting
among options and ordering use of nuclear weapons are decisions that are
exclusively the President’s. It is also up to the President about whether to
participate in a threat conference – or whether to talk with advisors at all
before giving an order. Because having two officials who have to agree
doubles the chances of a successful decapitation strike, the theory goes,
there is no second vote.30
The system’s technology supports presidential decision where speed,
unity of decision, verification of the President’s identity, authentication of

28 The NCCS also provides the President information about damage inflicted by
adversary weapons, the status of U.S. forces under attack, and the damage inflicted by U.S.
weapons, to inform as best as is possible ongoing presidential decision-making and
communications with the adversary about war termination. These additional tasks would be
especially important if nuclear use was not in the context of a full superpower exchange.
Limited and even “protracted” nuclear war options were theorized and prepared from the onset
of the nuclear age and became important parts of the NCCS and U.S. nuclear war planning
during the Cold War, thanks to concerns about the unwinnable nature of a full superpower
nuclear exchange and the ascendance of game theory in U.S. nuclear theology. See, e.g.,
HENRY A. KISSINGER, NUCLEAR WEAPONS AND FOREIGN POLICY 132-202 (1957) (theorizing
limited nuclear war). The practical plausibility of these ideas, and their implications, were
criticized. Limited nuclear war options continue to be theorized. See, e.g., ON LIMITED
NUCLEAR WAR IN THE 21ST CENTURY (Jeffrey A. Larsen & Kerry M. Kartchner, eds., 2014)
(edited volume by scholars and practitioners on history and on current challenges).
29 See Amy F. Woolf, CONG. RES. SERV., IF10521, DEFENSE PRIMER: COMMAND AND
CONTROL OF NUCLEAR FORCES 1 (2018), https://perma.cc/9MLN-XAMT. The NATO
Commander’s title, referencing the Supreme Allied Commander title of the Western allies’
top military commander in Europe in World War II, is the Supreme Allied Commander –
Europe (SACEUR).
30 See Woolf, supra note 29, at 1 (discussing sole decision power of President). Bomber,
land-based ICBM, and submarine crews, even after receiving an authenticated launch order,
follow a “two person rule” – they cannot fire without two officers acting. See Blair, supra note
6, at n.1. Such a rule for the national leadership level has been periodically discussed but never
adopted. See, e.g., Herb Lin, A Two-Person Rule for Ordering the Use of Nuclear Weapons,
Even for POTUS?, LAWFARE (Nov. 9, 2016, 2:54 PM), https://perma.cc/BT2S-VYBJ.

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

orders, and communications security are imperative. It enables the most

current intelligence to be transmitted to the President rapidly. Open
sources indicate that the nuclear “football” carried by a military aide
assigned to the President contains the highly classified Nuclear Decision
Handbook – the “black book” – with a summary of standing strike options.
The football contains communications equipment to verify the President’s
identity, and to encode and promptly transmit the President’s order to the
Pentagon and U.S. Strategic Command.31 The public record is ambiguous
as to whether current procedures provide for the Secretary of Defense to
confirm that the order came from the President.32 Open sources include
references to the President’s potential ability to bypass top advisors and
issue orders directly to lower level officers staffing the Pentagon’s
National Military Command Center.33 The public record indicates that
after the President’s orders are transmitted to bombers, missile officers, or
submarines, the factual and legal basis for the President’s orders are not
reviewed by forces in the field before weapons are released (and once
fired, nuclear-armed missiles cannot be recalled). 34 From the top down, the
inherited system implicitly assumes that the President’s orders are legal.
Thanks to the system’s security and authentication capabilities, orders are
understood to be issued by a President or Acting President successor

31 See Woolf, supra note 29, at 1.

32 For such a reference, see id. There is nothing on the public record to suggest that the
Secretary’s involvement is legally required.
33 President Kennedy raised exactly this possibility in a now-declassified January 1962
memorandum, “ALERT PROCEDURES and JCS EMERGENCY ACTIONS FILE.”
Kennedy asked whether the system at that time would allow him to order nuclear use “without
first consulting with the Secretary of Defense and/or the Joint Chiefs of Staff” by calling the
Pentagon’s Joint War Room, evidently a predecessor to the National Military Command
Center. President Kennedy asked “What would I need to say” to order “an immediate nuclear
strike?” Memorandum from President John F. Kennedy (Jan. 16, 1962), reprinted in SCARRY,
supra note 5, at 409 n.17.
34 See Woolf, supra note 29, at 1. According to open sources, U.S. Minuteman missiles
fire within two minutes, and submarine-based ballistic missiles fire within 15 minutes. See Id.
Depending on the location of the missiles and their targets, flight times would range from
several minutes to several dozen. Bomber aircraft would presumably release their weapons at
a release point once within range of their target. Depending on the location and readiness of a
bomber at the time it receives a launch order, whether and how long it takes to equip
(“generate”) and prepare the bomber for its mission, flight time to the release point (potentially
from bases in the continental United States to the other side of the planet), and potential flight
time of a cruise missile-carried warhead, a launch order could take minutes to dozens of hours
to be executed using bomber aircraft. Unlike the land-based and sea-based legs of the U.S.
nuclear triad, U.S. bomber aircraft were taken off alert by President George H.W. Bush in
1991.

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 11

invested with authority by the people and the Constitution, and in our
republic uniquely accountable to all of them.35
As inherited from the Cold War, the theory and practice of nuclear
deterrence rest on legal theories of practical necessity and self-defense,
focused on the President. Neither Congress nor any international body
have provided specific standing legal authority to use nuclear weapons. 36
In the archetypical, worst nuclear nightmare of initiated adversary nuclear
attack, there would not be time for Congress or the United Nations to
convene and act. As a question of U.S. law, the credibility of a timely U.S.
nuclear launch decision therefore came to rest on the President’s authority
as Commander in Chief under Article II of the Constitution.37 Indeed, the
importance to nuclear deterrence of the President’s authority to order
nuclear use both benefitted from and helped drive growth in theories of
executive power during the Cold War. Of course, the notion that the
President acting without Congress could initiate use of force unilaterally
has its critics generally, and in the nuclear context, as well. But it is well
grounded in constitutional law doctrine.38 Under international law, a
nuclear response or preemptive nuclear launch in the face of attack would
rest on jus ad bellum theories of the inherent right of self-defense, and the
right to act in self-defense before an imminent attack (sometimes called
preemptive or anticipatory self-defense).39

35 As Akhil Amar notes, all other senior officials in the U.S. government are chosen or
confirmed by officials who are selected by a subset of the people. See AKIL AMAR, AMERICA’S
CONSTITUTION: A BIOGRAPHY (2005).
36 Enacted at the dawn of the atomic age, the Atomic Energy Act of 1946, 42 U.S.C. §
2121(b)(1), provides the President authority to “direct the [Atomic Energy] Commission to
deliver such quantities of special nuclear material or atomic weapons to the Department of
Defense for such use as he deems necessary in the interest of national defense.” An expansive
interpretation is implicit authorization for launch. Professor Peter Raven-Hansen emphasizes
that the legislative history disputes that reading. See Raven-Hansen, supra note 15, at 790-91.
Congress has never explicitly authorized or prohibited combat use of nuclear weapons. Under
international law, threat or use of force is presumed to be illegal unless self-defense or other
exceptions are operative. See U.N. Charter art. 2(4).
37 U.S. CONST. art. II, § 2.
38 See, e.g., The Prize Cases, 67 U.S. 635, 665 (1862) (authority to repel attacks).
39 See U.N. Charter art. 51 (right of self-defense). The International Court of Justice
wrestled with nuclear deterrence in a landmark 1996 decision. The Court found the
catastrophic and hard-to-contain effects of nuclear weapons difficult to reconcile with
international law. But the Court could not rule use or threat of use per se illegal because of
“the fundamental right of every State to survival” and self-defense, the successful practice
precedent of nuclear deterrence, and the reservations many states have attached to multilateral
nuclear treaties. See Legality of the Threat or Use of Nuclear Weapons, Advisory Opinion,
1996 I.C.J. Rep. 226, ¶¶ 95-97, 102 (July 8). For analysis of pre-attack self-defense, see Ashley
Deeks, Taming the Doctrine of Preemption, in THE OXFORD HANDBOOK OF THE USE OF FORCE

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

The President has become so empowered legally regarding nuclear

weapons, and the potential decision timeline is so unforgivably
constrained in the classic nuclear nightmare, that at the national leadership
level there are no clear indications that the system anticipates involvement
of the Attorney General or any other legal advisor. Unless the President or
a subordinate acts affirmatively to include a lawyer, 40 any legal advice
during a nuclear threat conference would therefore be operative by virtue
of having been provided earlier to the participants, provided in the moment
by lawyers fortunate enough to be at the elbow of their principal, 41 or else
having been “baked in” – that is, reflected in nuclear operational plans
thanks to legal review of strike options during the earlier standard war
plan-writing process. This planning-stage legal review includes jus ad
bellum international law regarding resort to use of force, certainly involves
jus in bello principles of the international law of armed conflict (LOAC),
and should also include other relevant U.S. law. 42 Legal review during
preparation of standing nuclear war plans includes assumptions about the
presence of non-combatant persons and property, the configuration of
adversary forces and assets, and potentially too some consideration of
weather conditions that can influence radioactive fallout patterns.
Intelligence about these “facts on the ground” can change rapidly and
dictate reassessment. Such legal reassessment ought to include legal
advisors. Based on the public record, however, today we cannot be
confident that legal advisors actually would be available to the President
at the moment of nuclear decision.

IN INTERNATIONAL LAW (Marc Weller, ed., 2015) (analyzing anticipatory, preemptive, and
preventive theories of self-defense); U.S. LAW OF WAR MANUAL, supra note 21, § 1.11.5.1
(right of self-defense against imminent attack).
40 See JAMES E. BAKER, IN THE COMMON DEFENSE: NATIONAL SECURITY LAW FOR
PERILOUS TIMES 226 (2007) (“the civilian lawyers that will engage on operational questions
involving the means and methods of conflict (as opposed to the development of doctrine) are
the lawyers the president, the secretary of defense, and their immediate advisors designate” at
their discretion).
41 For example, the Staff Judge Advocate – the senior military lawyer – at the U.S.
Strategic Command. General Kehler, the former head of the U.S. Strategic Command, has
written that he would want a military lawyer on hand at any point employment of nuclear
weapons would be contemplated. See C. Robert Kehler, Nuclear Weapons & Nuclear Use,
145 DAEDALUS 50 (2016); see also Lt. Col. Theodore T. Richard, U.S. Air Force, Nuclear
Weapons Targeting: The Evolution of Law and U.S. Policy, 224 MIL. L. REV. 862 (2016)
(recent Strategic Command Staff Judge Advocate’s analysis of U.S. nuclear targeting and the
law). Of course, the operational commander would stand between the military lawyer and the
President.
42 For discussion, see U.S. LAW OF WAR MANUAL, supra note 21, §§ 1.11, 2.2 – 2.6
(discussing jus ad bellum, and jus in bello LOAC principles of necessity, distinction,
proportionality, humanity, and honor, the last of which is an addition).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 13

B. Nuclear Nightmares Two and Three: Rogue President, and

Precipitous President

The current nuclear command and control system may be the best that
can be crafted to deal with the classic nuclear nightmare of launch in the
face of imminent or initiated adversary attack. That risk is resurgent thanks
to the advancing nuclear capabilities of Russia, China, and North Korea,
and their long-term trajectories of increasingly confrontational relations
with the United States.43 The inherited system, however, creates serious
hazards in the hands of a President intent on first use of nuclear weapons
where adversary attack is not temporally imminent, and the necessity and
otherwise legality of nuclear use have not been established. One variant of
this nightmare is a Rogue President who orders nuclear use without evident
factual predicate or legal basis. A second related but distinct risk – one
where there is no imminent threat of adversary nuclear attack – is that of a
Precipitous President. That is, a Commander in Chief who resorts to
nuclear weapons where a crisis or conventional conflict is underway and
U.S. nuclear use at some point might be necessary and legal, but nuclear
use’s implications, legality, or alternatives have not yet been carefully
evaluated. A Precipitous President is an impulsive leader who reaches for
“the button” too quickly.
The President would be able to rely neither on international law nor on
Article II constitutional authority for a non-necessary use of force. Use of
force is illegal if it is unnecessary, and under U.S. law any use of force
rising to the level of “war” (which any nuclear strike certainly would due
to its effects or escalation risk) that is not a response to an armed attack
would require congressional authorization. 44 Even where the United States
is involved in an armed conflict that is authorized under international and
U.S. law, particular uses of force are still illegal if they do not comply with
the international law of armed conflict (LOAC) and its jus in bello
principles of necessity, distinction, proportionality, and humanity.
Compliance with LOAC requires careful analysis of intelligence and

43 Five other states are understood to possess nuclear weapons (the United Kingdom,
France, India, Pakistan, and Israel). Iran is not believed to be a nuclear weapons state but as a
2015 international agreement limiting Iran’s nuclear program collapses there are indications
Iran may be renewing its drive for a nuclear weapon. See Dehghanpisheh & Khalid, supra
note 10.
44 See U.N. Charter arts. 2(4) (general prohibition on threat or use of force) and 51
(necessity exception for self-defense); Mary B. DeRosa & Ashley Nicolas, The President and
Nuclear Weapons: Authorities, Limits, and Process 6-7, Dec. 2019,
https://media.nti.org/documents/The_President_and_Nuclear_Weapons_Authorities_Limits_
and_Process.pdf (discussing Executive Branch precedents regarding use of force).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

tailoring of the use of force in terms of target selection, choice of weapon,

angle of attack, and other respects, a process that involves military
personnel, lawyers, and sometimes national leadership. 45 Ensuring the
lawfulness of nuclear operations is especially challenging because the
effects of nuclear weapons are so powerful (including heat, blast, and
prompt radiation) and hard to contain (especially radioactive fallout and
computer-destroying electromagnetic pulse (EMP)).46 The inherited
nuclear command and control system structurally anticipates that
presidential nuclear orders will be legal, however, and the President will
never misuse the nation’s nuclear loaded weapons.
As Professor and former NSC legal advisor James E. Baker observes,
presidents get the process they choose “within the constitutional and
statutory framework of decision-making.” 47 The Constitution and statute
provide a general chain of command.48 However, there is no statute or
publicly known executive order governing nuclear launch specifically. 49
In a legitimate crisis the President may be content to consult the anticipated
line-up of threat conference civilian and military leaders. The President
could additionally involve other civilian officials, the Attorney General, or
other lawyers. Or, an impulsive President could open the nuclear football
and give an order without consulting anyone.50 As senators worried at the
recent congressional hearing, the President could awaken senior civilian
and military officials with a strike order, rather than their waking the
President with a threat warning.51
Senior officials have only bad options in the face of a Rogue President
or Precipitous President. There is no legal rule or known framework of
norms to apply short of asking whether a strike order is illegal under

45 For discussion, see James E. Baker, LBJ’s Ghost: A Contextual Approach to Targeting
Decisions and the Commander in Chief, 4 CHI. J. INT’L L. 407 (2003).
46 See Legality of the Threat or Use of Nuclear Weapons, Advisory Opinion, 1996 I.C.J.
Rep. 226, ¶ 92 (July 8) (describing view that nuclear weapon effects are too difficult to contain
to meet distinction and humanity requirements of the law of armed conflict).
47 See BAKER, supra note 40, at 99-100, 106.
48 See U.S. CONST. art. II, § 2 (Commander in Chief Clause); 10 U.S.C. § 162(b) (military
chain of command).
49 The universe of classified or otherwise unreleased presidential and agency-level
directives with legal force is of unknown size but is certainly vast. See Dakota S. Rudesill,
Coming to Terms with Secret Law, 7 HARV. NAT’L, SEC. J. 241, 283-99 (2015).
50 “There is no capability to directly launch nuclear weapons from the ‘football.’ Upon
presidential direction, military command center personnel would transmit an order that would
be issued over multiple systems to the fielded forces.” See E-mail from Karen Singer, U.S.
Strategic Command Pub. Aff. Off. (March 11, 2019, 2:59 PM) (on file with author).
51 See SFRC 2017 hearing, supra note 6. (testimony of Peter D. Feaver).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 15

LOAC, and how that body of law is interpreted and applied by the United
States. General Kehler testified that “the military does not blindly follow
orders” and executes only legal orders. 52 Here, an impulsive presidential
launch order might fall into a worrisome grey zone: ambiguously legal,
potentially illegal if carefully analyzed, but in the moment perhaps not
“clearly illegal” or “manifestly unlawful.”53 Yet whether the President’s
order is legally ambiguous or clearly illegal, the complete list of recourses
for the Secretary of Defense, the Commander of the U.S. Strategic
Command, or subordinates in the chain of command is short. Their only
options at that point would be verbal dissuasion of the President, refusal,
or resignation. If the President persisted, General Kehler testified that “I
do not know exactly” what happens. 54
While those in the chain of command would benefit from time to
consult lawyers and analyze the legal issues and intelligence, the President
in contrast could relieve an objecting official immediately. If an official
tried to continue in office despite dismissal, the official could rely only on
their knowledge of the nuclear command and control system and on their
own power of persuasion with other personnel in a desperate,
insubordinate attempt to thwart transmission of the President’s order to the
field. With nuclear war and countless lives in the balance, a rogue secretary
or relieved general or admiral would bureaucratically battle a Rogue
President or Precipitous President. The nation and world could get to this
nightmarish point stunningly fast.55
The only options remaining would be complicated, fraught processes
provided by the Constitution: removal of the President by the Cabinet
under the Constitution’s 25th Amendment, or by Congress via
impeachment by the House and trial and then conviction by the Senate. 56

52 See SFRC 2017 Hearing, supra note 6, at 40 (testimony of Gen. C. Robert Kehler
(USAF, Ret.)); see also Michael Collins, Retired General, Others Urge Caution in Limiting
President’s Power to Order Nuclear Strike, USA TODAY (Nov. 14, 2017, 6:00AM),
https://perma.cc/X7J3-GAH6.
53 See U.S. LAW OF WAR MANUAL, supra note 21, §18.3; Rome Statute of the
International Criminal Court, art. 33(1)(c), entered into force, July 1, 2002, 2187 U.N.T.S. 90.
For discussion, see Colangelo, supra note 15, at 91.
54 See SFRC 2017 Hearing, supra note 6, at 20 (testimony of Gen. C. Robert Kehler).
55 A similar point could be reached if there were a temporally imminent threat justifying
the use of nuclear weapons but the President seemed intent on choosing a nuclear option that
was not necessary or legal under the circumstances. For a fictionalized depiction of such a
situation, see JEFFREY LEWIS: THE 2020 COMMISSION REPORT ON THE NORTH KOREAN
NUCLEAR ATTACKS AGAINST THE UNITED STATES (2018) (President seeks nuclear strike on
North Korea and China in response to North Korean attack).
56 U.S. CONST. amend. XXV; U.S. CONST. art. II, § 4.

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

These processes involve a large number of officials who are typically

scattered across the capital or country (or world) at any given moment,
surely a considerable amount of informal process and politicking behind
the scenes, and multiple formal process steps. Both removal efforts could
be contested by the President and their most loyal aides and partisans. 57
Accordingly, these removal processes could be expected to take at the very
least many hours. Removal could also take days, weeks, or months. In
contrast, a nuclear launch order can be executed in minutes. The
Constitution’s solutions for removing a Rogue or Precipitous President
are, for this reason, probably best thought of in the nuclear context as ex
post processes – as first steps along a long road of correction and national
reflection on an atomic atrocity. The President can push “the button” faster
than Executive or Legislative Branch officials can constitutionally oust the
Commander in Chief.
For over half a century the nation and the world have been fortunate
not to have a U.S. President who successfully reaches for the nuclear
“loaded weapon” and orders plainly or ambiguously illegal use of the
bomb. But the historical record suggests we came close. President Richard
Nixon reportedly called for nuclear use while drunk and while battling
depression during the Watergate scandal. A President who boasted that “I
can go into my office and pick up the telephone and in 25 minutes 70
million people will be dead” dangerously combined alcohol and
psychoactive drugs.58 It was simple historical good luck that President

57 Even with ample time the process did not result in removal of the two Presidents
impeached and tried in the Senate to date, Andrew Johnson and Bill Clinton.
58 For Nixon’s statement, reportedly made to legislators during Watergate, see HIRC
1976 hearings, supra note 20. Accounts of Nixon’s nuclear instructions may be apocryphal.
There are, however, multiple accounts, and they have some consistency:

The CIA’s top Vietnam specialist, George Carver, reportedly said that in
1969, when the North Koreans shot down a US spy plane, “Nixon became
incensed and ordered a tactical nuclear strike . . . . The Joint Chiefs were
alerted and asked to recommend targets, but [national security advisor]
Kissinger got on the phone to them. They agreed not to do anything until
Nixon sobered up in the morning.”

[On another occasion, a White House aide] told a colleague “of the time
he was on the phone [listening] when Nixon and Kissinger were talking.
Nixon was drunk, and he said, ‘Henry, we’ve got to nuke [Vietnam].’”

Anthony Summers & Robbyn Swan, Drunk in Charge (part two), GUARDIAN (Sept. 2, 2000),
https://perma.cc/LT3X-LG89, excerpt from ANTHONY SUMMERS, THE ARROGANCE OF
POWER: THE SECRET WORLD OF RICHARD NIXON (2000); TIM WEINER, ONE MAN AGAINST
THE WORLD 89–91 (2015); ELIZABETH DREW, RICHARD M. NIXON 55–57 (2007). For anxiety,

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 17

Nixon’s impairment was obvious to his aides. It was good luck that they –
lacking evident norms and violating the chain of command – prevailed
upon the alerted Joint Chiefs of Staff to sit on, and ultimately ignore, the
instructions.59 Nixon’s Saturday Night Massacre of Justice Department
officials was a terrible day for our nation, but it was still good luck that
when President Nixon fired Executive Branch officials seriatim it was to
frustrate the Watergate investigation rather than to find a subordinate who
would transmit an unjustified nuclear launch order.60
It may be emotionally reassuring simply to decide to believe – as we
implicitly do now – that there will never be a nuclear Saturday Night
Massacre. It is comforting to postulate that no President will ever again
impulsively order a nuclear strike – and next time press those orders to
execution. But loading so many innocent lives on wishful thinking is not
rational. A sizable portion of humanity regularly becomes mentally
compromised due to illness, alcohol, or drugs. Normal-appearing people
regularly commit horrifying crimes, awful errors in judgment, and other
anti-social acts. Nixon’s well documented abuses of power demonstrate
that high performing leaders are not immune. On the contrary, research
suggests that psychopathic personality traits could be four or more times
as common in top corporate positions compared to the population
generally.61 A reasonable inference is that psychopathy’s incidence is at
least as high among politicians. Additionally, we now know that Nixon
was not alone in being high or otherwise mentally compromised in high
office. Kennedy reportedly took a number of drugs that had psychoactive

Nixon took medication that caused “slurred speech, mental confusion, and irritability. Mixed
with alcohol, it enhances alcohol’s effects.” DREW, supra, at 27. Whether or not reports of a
drunk President dialing up a nuclear strike are true, they are plausible – based on Nixon’s
known condition, his bragging about his ability to kill with nuclear weapons, and in view of
what we know about human nature.
59 See supra note. The President is the only White House official in the military chain of
command. See 10 U.S.C. § 162(b).
60 The historical record shows concern about President Nixon’s mental state reportedly
resulting in multiple deviations from the chain of command regarding nuclear weapons.
Secretary of Defense James R. Schlesinger reportedly told the military that any nuclear launch
orders had to be cleared with him. See Garrett M. Graff, The Madman and the Bomb, POLITICO
(Aug. 11, 2017), https://perma.cc/ZK5N-7TWU. With the President apparently distraught and
intoxicated, the Secretary of State and other senior officials put U.S. nuclear forces on higher
alert to deter Soviet intervention in the 1973 Yom Kippur War. See WALTER ISAACSON,
KISSINGER: A BIOGRAPHY 530–33 (1992). Whether or not these second-hand accounts are
accurate, their plausibility indicts the nuclear command and control system. Change the
President’s personality, the Soviet response, or the conduct of the President’s advisors, and
the consequences could have been catastrophic.
61 See Paul Babiak, Craig S. Neumann & Robert D. Hare, Corporate Psychopathy:
Talking the Walk, 28 BEHAV. SCI. & L. 174, 184 (2010).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

effects, and several other presidents had physical and mental afflictions
that reasonably could impact their judgment. 62 Heavy drinker and future
President Andrew Johnson was reportedly so ill and belligerently
intoxicated in the U.S. Senate Chamber at his swearing-in as Vice
President that he barely succeeded in executing the oath of office. 63
Accounts have Johnson impaired when Lincoln was assassinated and he
had to assume the presidency. Looking beyond American history, one does
not need to include mad monarchs, bad emperors, or 20 th century
totalitarians to compile a long list of leaders who have suffered serious
physical and mental illnesses, substance abuse problems, and clinical
defects of character that compromised their judgment. 64 One especially

62 See Rose McDermott, The Politics of Presidential Medical Care: The Case of John F.
Kennedy, 33 POLITICS & LIFE SCI. 77, 84-85 (2014); Visar Berisha, Shuai Wang, Amy
Lacross & Julie M. Liss, Tracking Discourse Complexity Preceding Alzheimer’s Disease
Diagnosis: A Case Study Comparing the Press Conferences of Presidents Ronald Reagan
and George Herbert Walker Bush, 45 J. ALZHEIMER’S DISEASE 959, 961-63 (2015),
https://perma.cc/M8H8-95KV (questions about Reagan’s mental functioning); Joshua Wolf
Shenk, Lincoln’s Great Depression, ATLANTIC (Oct. 2005), https://perma.cc/9P83-36SW
(Lincoln suffered depression and suicidal thoughts as President); Edwin A. Weinstein,
Woodrow Wilson’s Neurological Illness, 57 J. AM. HIST. 324, 336–46 (1970) (President
Wilson suffered a stroke and other illnesses that caused delusions and incapacity). Two
Presidents became ill and died while in office, two others were shot but recovered, and two
lingered after assassination attempts before dying in office.
63 See Jonathan R. Allen, Andrew Johnson Drunk at Lincoln’s Second Inaugural, CIVIL
WAR HIST. & STORIES, https://perma.cc/M6UP-DGMR (Johnson was ill with typhoid and “too
drunk to perform his duties & disgraced himself & the Senate by making a drunken foolish
speech,” observed one Senator); see also Vice President Andrew Johnson, Inaugural Address
(Mar. 4, 1865), https://perma.cc/4NVB-HQBA (rambling, angry remarks). Many other senior
officials have had serious drinking problems and impairment, including: House Speaker Carl
Albert (see GARRETT M. GRAFF, RAVEN ROCK 220–21 (2017) (in-patient treatment for
alcoholism while first in line for the presidency after Vice President’s resignation)); Senator
and Defense Secretary nominee John Tower, see Andrew Rosenthal, F.B.I. Document on
Tower Cited ‘Pattern of Alcohol Abuse’, N.Y. TIMES (Mar. 17, 1989), https://perma.cc/4T7N-
UV5W; Supreme Court Justice William O. Douglas, see Robert W. Mull, Yakima and Justice
Douglas: The Curious Story of a Famous but not a Favorite Son, 1 COLUMBIA MAG. 2 (1987),
https://perma.cc/9LG2-ZGPG; Vice President Daniel Tompkins, see RAY W. IRWIN, DANIEL
D. TOMPKINS: GOVERNOR OF NEW YORK AND VICE PRESIDENT OF THE UNITED STATES 309 n.
55 (1968). Speaker Albert, Senator Tower, and Vice President Tompkins were blocked from
ascending further. President Nixon, President Andrew Johnson, and Justice Douglas were not.
64 See, e.g., Marvin Rintala, Family Portrait: Churchills at Drink, 21 BIOGRAPHY 1, 2–3
(1998) (British Prime Minister Winston Churchill suffered depression and drank heavily,
which in the assessment of British naval officers at times impacted his decision-making);
Michael White, So Tony Blair Turned to Alcohol – It’s a Rare PM that Doesn’t, GUARDIAN,
(Sept. 2, 2010), https://perma.cc/R7GU-C8NL (H.H. Asquith, the Prime Minister who took
Britain into World War I, known as “Squiffy” for drunkenness at Parliament); Craig Wallace,
Only Human – Disability in Australian Politics (Part 1: Human Leaders), MUSEUM OF

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 19

horrifying example is the South American chief of state who in the 1800s
prolonged a hopeless war that killed much of his nation’s population,
driven in part by delusions of glory.65
In the 1800s it took months or years to kill on the scale a Rogue or
Precipitous President could murder in minutes. An unnecessary or
otherwise illegal nuclear strike could also cause genetic defects and
suffering for generations, do extensive harm to civilian property and the
global economy, and damage the environment. If it precipitated a general
nuclear exchange the strike could end civilization. At the least, an
unwarranted U.S. nuclear attack would do catastrophic and irreparable
damage to the moral authority of the United States. For our failure to
design a nuclear command and control system reasonably able to impede
its entirely plausible misuse, the country would bear heavy moral
responsibility. With the great power represented by nuclear weapons
comes the overdue responsibility of crafting a decision process reliably
able to frustrate a leader whose mental faculties fail in the ways that those
of humans regularly do.

II. NUCLEAR-CONVENTIONAL CONVERGENCE & THE NATIONAL

LEADERSHIP-LEVEL REVIEW GAP

After the Cold War, the United State dramatically reduced nuclear
weapons but did nothing known to counter the command and control
system’s perilous over-reliance on the President’s mental state. The United
States has left nuclear weapons constitutionally special – reserved for one
person in one branch. But legal, policy, and technological inertia has not
stopped a slate of contextual tectonic shifts that suggest that it is time for
reform. We can group these under the heading of convergence between
nuclear and conventional operations.
The Department of Defense has grappled with several aspects of
convergence. The U.S. military has in conventional conflicts relied heavily
on sensors, communications systems, and forces (bombers, cruise missiles,

AUSTRALIAN DEMOCRACY (Dec. 2, 2016), https://perma.cc/9XUK-USRJ (Australian Prime

Ministers suffered heart attacks and other ailments that impacted their work).
65 Estimates of Paraguay’s total population losses during its war with the triple alliance
of Argentina, Brazil, and Uruguay have varied widely but probably total five to 18 percent.
See Vera Blinn Reber, The Demographics of Paraguay: A Reinterpretation of the Great War,
1864-70, 68 HISP. AM. HIST. REV. 289, 310 (1988). For comparison, the Soviet Union’s
population declined by four percent during a Second World War rightly regarded as
cataclysmic for that state. See id. at 308. Scholars disagree about the causes of the war but
based on the historical record one may reasonably lay a significant share of the blame for
continuing the war – against powers with 38 times Paraguay’s population and larger militaries
– with Paraguayan leader Francisco Solano Lopez. See id. at 289 n.1, 319.

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

and submarines) built for the nuclear mission. The Pentagon has also
explored the use of high-precision conventional weapons to destroy
adversary nuclear forces.66 The Defense Department has in response made
a number of adjustments to policy, doctrine, and its organization, and
broadened the range of strike options available to the President.67
Recently, the Pentagon’s 2018 Nuclear Posture Review warned of
convergence abroad: advancing nuclear capabilities of Russia, China, and
North Korea, together with their development of “hybrid war” doctrines
that integrate nuclear and conventional weapons.68
Above the departmental level, however, the U.S. government has not
come to terms with convergence. This Part will analyze important aspects
of this sea change that augur toward revision of the nuclear command and
control system at the national leadership level. One powerful convergence
driver is a change in threat and mission: the most likely nuclear
employment scenario now has more in common with a sensitive counter-

66 Convergence is an outgrowth in part of the high-tech Revolution in Military Affairs

(RMA), the synergistic use of advanced military systems for seeing, talking, and striking. See
WILLIAM A. OWENS, LIFTING THE FOG OF WAR (2000) (former Vice Chairman of the Joint
Chiefs of Staff analyzes RMA). Meanwhile, the United States, Russia, and China use
components of their nuclear command and control systems to support both conventional and
nuclear operations. These developments raise escalation concerns: that a non-nuclear conflict
that included attacks on entangled systems could incentivize a nuclear response before the
capability is lost. See James M. Acton, Escalation Through Entanglement: How the
Vulnerability of Command-and-Control Systems Raises the Risks of an Inadvertent Nuclear
War, 43 INT’L SEC. 56 (2018); see also 2018 NPR, supra note 3, at 21 (Pentagon warning that
United States would consider using nuclear weapons if faced with “significant nonnuclear
attacks” on “U.S. or allied nuclear forces, their command and control, or warning and attack
assessment capabilities”).
67 Nuclear Posture Reviews (NPRs) by the Department of Defense often emphasize
conventional options alongside nuclear options. See KURT GUTHE, THE NUCLEAR POSTURE
REVIEW: HOW IS THE “NEW TRIAD” NEW?, CTR. FOR STRATEGIC AND BUDGETARY
ASSESSMENTS (2002) 1-2, https://perma.cc/BXR7-4JRW (George W. Bush Administration in
classified 2001 NPR envisioned conventional global strike capabilities together with nuclear
weapons as one third of a “new triad”); U.S. DEP’T OF DEF., NUCLEAR POSTURE REVIEW vii–
viii (2010) [hereinafter 2010 NPR] (Obama Administration emphasizes greater role for
conventional weapons in deterring conventional threats deterred during Cold War by nuclear
weapons). Additionally, the Defense Department created Global Strike Command in the late
2000s to provide a single command for the service focused on worldwide employment of long-
range conventional and nuclear force. See AIR FORCE GLOBAL STRIKE COMMAND,
https://perma.cc/SS3Y-MG3K.
68 See 2018 NPR, supra note 3, at v; see also Robert Peters, Justin Anderson & Harrison
Menke, Deterrence in the 21st Century: Integrating Nuclear and Conventional Force, 12
STRATEGIC STUD. Q. 15, 16, 18–25 (2018) (evidence and implications of hybrid strategies);
BRAD ROBERTS, THE CASE FOR U.S. NUCLEAR WEAPONS IN THE 21ST CENTURY 245 (2016)
(“theory of victory” involving nuclear use to damage the U.S. will to fight).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 21

terror raid than a global thermonuclear war.69 Other key elements of

convergence are changes in how the military plans, and the expanding
operational role of law and lawyers. Convergence undermines old
assumptions about the special nature of nuclear weapons. It suggests
reform of the nuclear command and control system to mandate an
appropriately robust decision process at the national leadership level, in
situations in which adversary nuclear attack is not imminent, with
particular emphasis on lawyer-provided legal review at the presidential
moment of decision.

A. The Most Likely Nuclear Employment Scenario

During the Cold War, the primary design loadstars for U.S. nuclear
forces and the command and control system were deterrence of a general
strategic nuclear exchange and deterrence of an overwhelming Warsaw
Pact conventional attack in Europe. The United States also developed a
series of lower intensity, flexible response options to deal with a slate of
contingencies. In part because a primary rationale for U.S. nuclear
weapons during the Cold War was deterring attack on Western Europe by
larger Warsaw Pact conventional forces, nuclear and conventional war
planning were in some ways linked. Shorter range tactical (non-strategic)
nuclear weapons had especially important roles in NATO war plans. 70
Even so, there has long been a general conceptual distinction between the
nuclear and non-nuclear realms. The norms that developed and the nuclear
command and control system made nuclear weapons special. War plans
regarding the intercontinental-range and more powerful strategic nuclear
forces based in the United States were for the most part separately
prepared. The conceptual, operational, and bureaucratic “firebreak” at the
conventional/nuclear “threshold” has resonated with an international norm
against use of nuclear weapons since 1945, and has endured. 71
Although deterrence of resurgent Russia and China remains the highest
priority of U.S. nuclear forces, an important change has occurred. Today,
the most likely nuclear employment scenario is no longer a nuclear

69 The term was made famous by the Cold War movie WAR GAMES (1982).
70 See Dakota S. Rudesill, Regulating Tactical Nuclear Weapons, 102 GEO. L.J. 99, 114–
15 (2013) (discussing tactical nuclear weapons in U.S. and NATO doctrine); H ANS M.
KRISTENSEN, U.S. NUCLEAR WEAPONS IN EUROPE: A REVIEW OF POST-COLD WAR POLICY,
FORCE LEVELS, AND WAR PLANNING (2005), https://perma.cc/WVF9-GVW5.
71 See Peters, Anderson & Menke, supra note 68, at 27–32 (analyzing six enduring
challenges to integration of U.S. convention and nuclear plans).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

exchange with a major nuclear power. 72 Nor is it a Rogue President

ordering a strike out of the blue. Rather, the most probable use of nuclear
weapons today by the United States would be use of one or a handful of
weapons, in combination with non-nuclear forces, against a limited target
set in service of limited objectives, potentially where adversary nuclear
attack is not underway or temporally imminent. 73
With alternatives on the table, and with lack of a “no first use” policy,
senior officials could consider use of a nuclear weapon because of its
unique destructive and signaling capabilities. One commonly mentioned
mission, for example, is use of one or more lower yield warheads to reach
a hardened or deeply buried target, such as a bunker used by terrorists or a
rogue state that intelligence suggests is contemplating a catastrophic
attack. This scenario was central to the 2000s debate about building a
Robust Nuclear Earth Penetrator (RNEP) version of the Air Force’s B61
air-delivered nuclear bomb.74 Even though RNEP went nowhere, the
scenario returns with some frequency in connection with ongoing
modernization of the B61 bomb and the Trump Administration’s drive for
a low-yield warhead for the Trident II submarine-launched ballistic
missile.75 Other hypothetical limited nuclear use scenarios involve striking
a rogue state’s weapons of mass destruction, or ensuring destruction of
North Korean nuclear-armed mobile missiles before they can deploy from
their bases.76
These most likely uses of nuclear weapons in important respects
resemble conventional operations against counter-terrorism (CT) high
value targets (HVTs). There may be strategic warning of days to weeks. A
variety of nuclear and non-nuclear force employment options would be

72 The most recent Nuclear Posture Review (NPR) identifies deterrence of nuclear attack
as the primary mission of nuclear forces. See 2018 NPR, supra note 3, at vii. Among nuclear
powers, Russia and China pose the greatest potential nuclear threat to the United States and
its allies. Deterrence of non-nuclear attack is another mission of U.S. nuclear forces. See 2018
NPR, supra note 3, at vii.
73 See Woolf, supra note 29, at 2 (mentioning scenario).
74 See JONATHAN MEDALIA, CONG. RES. SERV., RS21762, ROBUST NUCLEAR EARTH
PENETRATOR BUDGET REQUEST AND PLAN, FY2005-FY2009 (2004), https://perma.cc/RA3P-
TRFL.
75 See 2018 NPR, supra note 3, at xii, 33, 44 (modernization informed by North Korean
reliance on “hardened and deeply buried facilities;” ability to strike them is key U.S. nuclear
force attribute).
76 This article mentions the kind of hypothetical scenarios that often get discussed in
nuclear and policy conversations, without endorsement. Another limited nuclear use scenario
of rising concern is use in the context of a conventional conflict with Russia or China. The
risks of escalation there are significant, however, and for that reason such a possibility is both
important and still probably less likely than the scenarios mentioned in the main text above.

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 23

available. Evaluating and tailoring the options, and preparing for a range
of potential outcomes (including success, failure, and other consequences),
would require analysis in advance of a complex mix of intelligence,
diplomatic, policy, military, and legal questions. The stakeholders
represented in the National Security Council (NSC) process bring a range
of information, advice, and capabilities.
The 2011 strike against Al Qaeda leader Usama bin Laden is an
instructive case study.77 Quality but questionable intelligence collected by
the CIA suggested that bin Laden had been found. The issue was elevated
to the NSC, leading to extensive review of the intelligence by a select
group of cabinet-level members of the NSC and staff from multiple
agencies. The NSC staff coordinated weeks of work by the White House,
military, intelligence agencies, and multiple cabinet departments as the
Council developed, evaluated, and prepared a range of options for
presidential decision. These included waiting and collecting more
intelligence, a massive conventional strike by Air Force bombers, a drone-
delivered conventional strike, and a complex air/ground assault involving
special operations forces from multiple military services under CIA
authority. The inter-agency process was also used to evaluate a slate of
important, complex, and in some respects novel questions under
international and U.S. law.78 These questions reportedly included the
strike’s legality in view of Pakistan’s sovereignty and the international law
of armed conflict, whether to kill or capture bin Laden, as a question of
U.S. law whether the strike would be conducted under statutory or Article
II presidential authority, and whether the strike would be conducted under
the covert action statute or regarded as a traditional military activity in the
armed conflict with Al Qaeda.79 The work of the inter-agency team was
urgent, and concerned matters of life and death. It involved the most

77 For accounts of the decision process, see LEON PANETTA, WORTHY FIGHTS: A MEMOIR
OF LEADERSHIP IN WAR AND PEACE 306–21 (2016) (then-CIA Director); CHARLIE SAVAGE,
POWER WARS 257–71 (2015) (investigative journalist).
78 Scholars who served in government have since provided cogent insights into the
complexity of several issues with which the NSC grappled. See Jennifer C. Daskal, The
Geography of the Battlefield: A Framework for Detention and Targeting Outside the “Hot”
Conflict Zone, 161 U. PA. L. REV. 1165 (2013) (challenges of defining the battlefield); Ashley
Deeks, “Unwilling or Unable”: Toward a Normative Framework for Extra-Territorial Self-
Defense, 52 VA. J. INT’L L. 483 (2012) (use of force against non-state actor on another state’s
territory).
79 Traditional military activity falls outside the statutory definition of covert action. See
50 U.S.C. § 3093(e)(2). See also Authorization for the Use of Military Force, Pub. L. No. 107-
40, 115 Stat. 224 (2001) (post-9/11 force authorization).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

sensitive intelligence sources and highly classified military capabilities. 80

The process succeeded in involving multiple government stakeholders in
careful deliberation regarding a multitude of issues, presented well-crafted
and distinct options to the President, got a clear and timely presidential
decision, securely transmitted that order to the field, and facilitated a
successful operation. There were no leaks before the operation (including
to Pakistan, where the operation was conducted) that could have
compromised intelligence sources and methods, endangered U.S.
personnel, or tipped off the target. Operation Neptune’s Spear achieved
complete tactical surprise, without U.S. loss of life.81
Ultimately, the mastermind of the 9/11 terrorist attacks was found at
an above-ground compound and incapacitated at President Obama’s orders
by a special operations team. But one could imagine nuclear weapons as a
suggested option had bin Laden been reported instead to be hiding in a
heavily defended cave deep beneath a mountain’s rocky slopes and
preparing another 9/11-scale attack, or if the United States had good
intelligence that a rogue state such as North Korea were moving to provide
nuclear arms to terrorists.82 There is no obvious reason why the same
tailored inter-agency process could not have accommodated review of one
or more nuclear options.

B. Operational Planning and Lawyering Up

The plans used in sensitive counter-terrorism operations in recent years

were developed with the benefit of an operational planning process
conducted and refined by the U.S. military’s geographic combatant
commands and the Joint Staff over the past several decades. Among other
things, modern military planning involves objective-based planning,
driven by the idea of targeting for effect rather than destruction. These
ideas were fairly novel in military doctrine when the nuclear command and
control conversation began its long hiatus 30 years ago. Today, the nuclear
operational planning process has been informed by a conventional
operational planning process that fully embraced these doctrinal

80 The latter included a stealth helicopter previously unknown to the public. See Tom
Geoghegan & Sarah Shenker, Stealth Helicopters Used in Bin Laden Raid, BBC (May 6,
2011), https://perma.cc/PLU4-SME7.
81 Although operationally effective and generally reflecting good process, the tailored
NSC process used for the bin Laden decision has been faulted for relying too heavily on White
House lawyers to the exclusion of the Attorney General, State Department Legal Advisor, and
other Senate-confirmed senior lawyers. See SAVAGE, supra note 77, at 258–60.
82 See Andrew J. Coe, North Korea’s New Cash Crop, 28 WASH. Q. 73 (2005) (nuclear
weapon transfer concern).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 25

innovations in the 1980s and 1990s and employed and refined them in the
long years of war since the 9/11 attacks.83 This revised nuclear planning
process has facilitated a reduction in the number of U.S. nuclear targets
and weapons.
This greater planning similarity is a second decades-long convergence
driver. The revised process now provides more nuclear and non-nuclear
options, more closely tailored to specific objectives, with higher fidelity to
the law of armed conflict. The trend will only continue: President Trump’s
Defense Department announced that “the United States will sustain and
replace its nuclear capabilities, modernize [command, control, and
communications technology], and strengthen the integration of nuclear and
non-nuclear military planning.”84
A related contributor to convergence has been a general “lawyering
up” throughout the national security enterprise. 85 More lawyers are more
involved in more issues than ever, including the most highly classified
operations. Operational law (“op-law”) – application of the principles of
the LOAC – has greatly expanded as a practice field, the most extensive
version of the Pentagon’s Law of War Manual was released in recent years,
and there has been explosive growth in commentary on op-law issues from
non-government experts.86 Within the military, the regional combatant
commands facilitated the process via increasing incorporation of op-law
in their near-continual combat operations since the Cold War’s end. 87 The
U.S. Strategic Command followed suit, expanding the roles of op-law and
lawyers in planning and exercises.88 This was important: the first joint

83 See Kehler, supra note 41, at 57-58.

84 See 2018 NPR, supra note 3, at viii.
85 For discussion of the general phenomenon, see JACK GOLDSMITH, POWER AND
CONSTRAINT: THE ACCOUNTABLE Presidency After 9/11 xi-xiii, 122-201 (2012); Margo
Schlanger, Intelligence Legalism and the National Security Agency’s Civil Liberties Gap, 6
HARV. NAT’L SEC. J. 112 (2015); Laura K. Donohue, National Security Pedagogy and the Role
of Simulations, 6 J. NAT’L SEC. L. & POL’Y 489, 492-94 (2013).
86 See U.S. LAW OF WAR MANUAL, supra note 21. Examples of engagement by non-
government commenters regarding operational law include the Lawfare and Just Security law
blogs and the University of Nebraska School of Law’s annual advanced operational law
conference.
87 The regional combatant commands have used lethal force in Panama (Operation Just
Cause, 1989), Iraq (Operation Desert Storm, 1991; Operation Southern Watch, 1992-2003;
Operation Desert Strike, 1996; Operation Northern Watch, 1997-2003; Operation Desert Fox,
1998; Operation Iraqi Freedom, 2003-onward), Serbia (Operation Allied Force, 1999);
Afghanistan (Operation Enduring Freedom, 2001-onward); Libya (Operation Odyssey Dawn,
2011); and Iraq and Syria (Operation Inherent Resolve, 2014-onward).
88 See Col. Charles J. Dunlap, Jr., U.S. Air Force, Taming Shiva: Applying International
Law to Nuclear Operations, AIR FORCE L. REV. 157, 167–69 (1997) (then senior lawyer at

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

nuclear command, founded at the Cold War’s end, turned away from
decades of deficient adherence to the law.89 The government nuclear
community also took a turn away from a Cold War-era culture of soft
disdain for lawyers and law. This culture owed much to the apocalyptic
overkill of Cold War nuclear plans, expectation that little to no time would
be available for legal advice at the moment of a “launch under attack”
decision, and the centrality of the President in the nuclear command and
control system.90
Reflection on the three decades since the Cold War’s end illuminates
the interwoven nature of the military planning and lawyering threads of
convergence. Conventional and nuclear operations are now more alike
than ever. Options are more easily considered alongside one another and
integrated into common plans that reflect an unprecedented and still-
growing role for lawyers and law.
* * *
At least this is the new state of affairs in the Defense Department.
Because the current nuclear command and control system has never been
used operationally, there is a gap in our understanding of what national
leadership level nuclear decision-making would involve.91

U.S. Strategic Command describes post-Cold War evolution); Kehler, supra note 41, 54-60
(recent Commander emphasizes importance of law and lawyers); see also Richard, supra note
41.
89 The U.S. Strategic Command was created in 1992 as a joint combatant command,
assuming operational responsibility from Strategic Air Command (SAC) and its naval
counterpart.
90 See Richard, supra note 41, at 930 (with massive target sets including economic
targets, Cold War nuclear war plans called into serious question theoretical legal protection
for civilian populations); Slocombe, supra note 24, at 135-36 (Cold War-era analysis notes
that missile flight time from the Soviet Union plus timelines for attack detection, threat
communication, and transmission of a presidential launch order “leaves almost no reserve time
for actual decisionmaking;” if Soviet missiles were instead launched from submarines off the
U.S. east coast there could be no warning or decision time at all for the President in
Washington, D.C., before the Soviet warheads detonated). Reflecting different threat
circumstances and changed attitudes about law and lawyers, the Obama Administration’s 2013
nuclear employment strategy stated that all plans must “be consistent with the fundamental
principles” of the Law of Armed Conflict (LOAC). See U.S. DEP’T OF DEF., REPORT ON
NUCLEAR EMPLOYMENT STRATEGY OF THE UNITED STATES SPECIFIED IN SECTION 491 of 10
U.S.C. 4–5 (2013); see also 2018 NPR, supra note 3, at 23 (Trump Administration states
“nuclear operations would adhere” to LOAC). But see Jeffrey G. Lewis & Scott D. Sagan, The
Nuclear Necessity Principle: Making U.S. Targeting Policy Conform with Ethics & the Laws
of War, 145 DAEDALUS 62 (2016) (criticizing the adherence of nuclear plans to LOAC).
91 During the Cuban Missile Crisis the nuclear command and control system was
comparatively primitive in its communications capabilities and in the forces it commanded.
The fastest-launching U.S. forces, land-based Minuteman missiles, were just being fielded.

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 27

The best the public can do today is hope: hope that any President would
order the kind of careful analysis of a full slate of options through the NSC
process that preceded the bin Laden raid, to include legal review by
lawyers.92 The public can only hope, too, that the right lawyers will be
included. Military lawyers certainly ought to be involved, but the legal
questions at the national leadership level will extend beyond op-law.
Contested questions of constitutional separation of powers, statutory
interpretation, and international law may well present. Civilian lawyers at
NSC and at the State, Defense, and Justice Departments will have
particular competence in sorting a potentially dense mixture of law, legal
policy, and constitutional norms.93
The inherited nuclear command and control system remains a special
process built around the President and Defense Department actors, for
nuclear operations that the convergence phenomenon has in important
ways made more like the conventional operations that have received robust
review at the NSC level, be they the bin Laden raid or other war plans
developed via modern, lawyered-up combatant command planning
methods. Nuclear-conventional convergence is, in short, a major
contextual development that begs focused thought about the benefits,
means, and contours of NSC-level inter-agency review of nuclear strike
decisions.

III. THE CASE FOR PROCESS – AND A PROCESS STATUTE

General Michael Hayden, formerly the second-ranking U.S.

intelligence official, observes that the inherited nuclear command and
control system “is designed for speed and decisiveness. It’s not designed

MICHAEL DOBBS, ONE MINUTE TO MIDNIGHT: KENNEDY, KHRUSHCHEV, AND CASTRO ON

BRINK OF NUCLEAR WAR 276–79 (2008).
92 There have been hints in policy documents in recent years, for example regarding an
intention to retain the ability to launch under attack but also plan for more likely 21st century
scenarios, use of “adaptive planning,” and a commitment not to allow adversaries to escalate
successfully beyond U.S. conventional capabilities. See U.S. DEP’T OF DEF., supra note 90;
U.S. DEP’T OF DEF., QUADRENNIAL DEFENSE REVIEW: REPORT TO CONGRESS 13 (2014). But
these bread crumbs fall short of providing public confidence that the Executive Branch would
employ an inter-agency process as robust as was used in the bin Laden raid, an decision
potentially as complex and ramified as nuclear use.
93 The questions presented by a potential strike on North Korea are a good example. See,
e.g., Tensions Rise Between the United States and North Korea, 112 AM. J. INT’L L. 95 (2018)
(presenting conflicting views on whether the President could authorize a first strike); Marty
Lederman, No, the President Cannot Strike North Korea Without Congressional Approval,
JUST SEC. (Aug. 10, 2017), https://perma.cc/9BZ7-2FUW (preemptive strike would violate
international and U.S. law).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

to debate the decision”94 – even where there is time for deliberation that
would improve the decision. To mitigate the Rogue and Precipitous
President risks described in Part I, and update the nuclear command and
control system at the national leadership level for the nuclear-conventional
convergence described in Part II, this Part argues for process rules. Here,
this article sets out the benefits of good process in decision-making, and
particularly the value of inter-agency review through the NSC. This Part
then disputes the suggestion that the problems now inherent in the system
can be effectively mitigated without legislation. A statute is the best way
to ensure that a President committed to impulsive nuclear button-pushing
will hit legal rules and process norms designed to help ensure that any
nuclear use is necessary and legal and the President’s authorities are not
abused.

A. The Value of Deliberation and Inter-Agency Process

Ideally, the President would have the time, inclination, and logistical
ability to confer with senior officials from multiple agencies (and
Congress) even in a situation in which an adversary nuclear attack were
temporally imminent or underway.95 Where an adversary attack is not
imminent but nuclear use is contemplated, there is no compelling reason
to avoid review through the inter-agency NSC process. There is good
reason it ought to be obligatory.
Congress created the NSC in the National Security Act of 1947 to
advise the President regarding the integrated use of the classic instruments
of national power – military, diplomatic, intelligence, and economic. 96
Congress has frequently amended the statute to perfect the NSC’s
structure.97 The NSC’s members under the statute as amended are the

94 See Woolf, supra note 29, at 1(quoting Hayden, who served as Principal Deputy
Director of National Intelligence, CIA Director, and NSA Director).
95 NATO’s decision-making process would require consultation with many stakeholders
even where a threat is imminent. During the Cold War it was also common to hear the claim
that Congress could be consulted and even take legislative action even if there were just days
before a conventional war in Europe went nuclear. See, e.g., Raven-Hansen, supra note 15.
96 See National Security Act of 1947, § 101, 61 Stat. 496 (1947), amended by 50 U.S.C.
§ 3021 (2014). See generally IVO H. DAALDER & I.M. DESTLER, IN THE SHADOW OF THE OVAL
OFFICE (2009) (analysis of NSC functioning). Law enforcement and the rule of law together
are an additional instrument of national power. See BAKER, supra note 40, at 20-31.
97 Congress has amended the statute more than a dozen times. In recent decades Congress
has re-enacted the statute and re-organized the NSC, directed the Defense Department to
explain how it will be a better participant, and replaced the head of the CIA with the Director
of National Intelligence as senior intelligence advisor to the President and the Council. See
National Defense Authorization Act for Fiscal Year 2017, Pub. L. 114-328, 130 Stat. 2422-

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 29

President, Vice President, and the Secretaries of State, Defense, Energy,

and Treasury, plus “other officers of the United States Government as the
President may designate.”98 Presidents by executive order typically
include other members (such as the White House Chief of Staff, Assistant
to the President for National Security Affairs (APNSA, often called the
national security advisor), and Attorney General), and other advisors
(particularly the Chairman of the Joint Chiefs of Staff and the Director of
National Intelligence, the Executive Branch’s senior military and
intelligence advisors, respectively).99 Every President has used the NSC
process and its White House staff to develop and appraise options in view
of the threat, frame issues for presidential decision, and coordinate
implementation across the many agencies of the U.S. national security
apparatus.100 The statute and executive orders together provide the
President and NSC staff ample latitude to tailor the NSC process.
Use of a nuclear weapon plainly meets the essential criteria for
consideration by the NSC.101 First, nuclear use would implicate multiple
instruments of national power (in fact all of them), and therefore benefit
from the coordinated input of the agencies represented on the NSC.
Second, any nuclear use would be a matter of great importance to the
nation and world. It would violate the 75 year-old norm against nuclear
use, detonate a weapon with effects that are difficult to contain and predict,
and inflict harm that would be brought to every corner of the world through
the internet and television. Even a low-yield tactical nuclear bomb would
have strategic – that is, major – implications in military, diplomatic,
intelligence, economic, and legal respects. 102 Using “the bomb” is
inevitably a big deal. And third, use of a nuclear weapon would require the

23, § 1085 (2016) (re-enactment and streamlining); National Defense Authorization Act for
Fiscal Year 2008, Pub. L. 110-181, 122 Stat. 291, § 952 (2008) (Pentagon participation);
Intelligence Reform and Terrorism Prevention Act of 2004, Pub. L. 108-458, 118 Stat. 3689,
§§ 1071-72 (2004) (senior intelligence advisor).
98 See 50 U.S.C. § 3021(c)(1).
99 See, e.g., President Donald J. Trump, National Security Presidential Memorandum
(NSPM) 4: Organization of the National Security Council, the Homeland Security Council,
and Supporting Staff (April 4, 2017) (Trump administration’s directive on NSC organization
generally). Although styled an “NSPM” the order has the same effect as an executive order.
100 See Jon J. Rosenwasser & Michael Warner, History of the Interagency Process for
Foreign Relations in the United States: Murphy’s Law?, in THE NATIONAL SECURITY
ENTERPRISE: NAVIGATING THE LABYRINTH 11 (Roger Z. George & Harvey Rishikof, eds.,
2010) (history and role of staff).
101 These principles can be discerned from scholarly and practitioner analyses of the NSC.
See, e.g., BAKER, supra note 40, at 105-19, 122-23; DAALDER & DESTLER, supra note 96.
102 See Rudesill, supra note 70, at 157–59 (discussing strategic effects of all nuclear
weapons and arguing for the end of the strategic/tactical distinction).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

knowledge and decision of the President. As Commander in Chief and

under the design of the nuclear command and control system, only the
President can direct employment of nuclear weapons. 103 Because nuclear
use so plainly qualifies for NSC review it is problematic that no publicly
known law, directive, or aspect of the nuclear command and control
system provides for it.
NSC review offers the attributes of “good process.”104 Better decisions
tend to come from including the right people at the right time, evaluating
the best information and advice, allowing a variety of experts and
institutional perspectives to be heard, identifying and testing assumptions,
evaluating and refining alternatives, presenting several high-quality
options for decision, and implementing a clear decision. Process can be
tailored for the particular timelines, operational details, information
sensitivities, and personalities involved.105 These attributes of good
process were for the most part reflected in NSC-level consideration of the
2011 bin Laden raid (discussed in Part II above), the 1990-91 Gulf War
(robust NSC review analyzed a range of options), and the 1962 Cuban
Missile Crisis (a tailored “Ex Com” NSC process with a large role for the
Attorney General reviewed a full slate of options including nuclear war). 106
The NSC is a proven process for considering all aspects of decisions as
consequential as nuclear use.

103 Pre-delegation and automation of the launch process creates dilemmas, especially in
the context of a crisis with a peer adversary such as the Soviet Union or Russia. Former
national security advisor Brent Scowcroft during the Cold War called this “the automatic
phase of war” as “the battle plan unfolds more of less automatically.” By one Cold War-era
estimate, due to pre-delegation and automation nearly half of U.S. strategic weapons could be
fired without presidential decision. See Raven-Hansen, supra note 15, at 786-87. During the
Cuban Missile Crisis, President John F. Kennedy was reportedly so concerned that U.S.
nuclear-armed missiles in Turkey would be fired without his authorization in the event of
Soviet conventional attack that he ordered their fuses removed and his personal authorization
required for their launch. See ROBERT F. KENNEDY, THIRTEEN DAYS: A MEMOIR OF THE
CUBAN MISSILE CRISIS 98 (1969).
104 For discussion of good process, particularly through the NSC, see BAKER, supra note
40, at 22-31, 99-125.
105 Process can “find the right balance between speed and strength, secrecy and input [and]
always meet deadlines.” See BAKER, supra note 40, at 124.
106 See ROBERT M. GATES, DUTY: MEMOIRS OF A SECRETARY AT WAR 538-43 (2014) (bin
Laden decision process described by Secretary of Defense); RICHARD N. HAASS, WAR OF
NECESSITY, WAR OF CHOICE (2010) (firsthand account and analysis of NSC process in advance
of 1991 and 2003 wars with Iraq); see DOBBS, supra note 91, at 38 (Ex Com met in complete
secrecy for nearly a week); GRAHAM T. ALLISON & PHILIP ZELIKOW, ESSENCE OF DECISION:
EXPLAINING THE CUBAN MISSILE CRISIS (1999) (analysis); KENNEDY, supra note 103, at 48,
52, 98 (consideration of use of nuclear weapons during Cuba crisis).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 31

The inherited nuclear command and control system, however, is

focused on the President plus officials from only one NSC player, the
Defense Department. That structure carries with it an implicit suggestion
that the options for the United States are the military options in the “black
book” prepared by the Defense Department inside the nuclear “football.”
The current nuclear command and control system has no publicly
known process rules that would prevent available time for deliberation to
be squandered by classic national security process maladies. These include
secrecy, speed, exclusion of key actors and information, and
personalization and ego.107 Consequentialist fears of “blood on your
hands” from failure to act also often interfere with good process, and tend
to favor acquiescence to executive action. The 2003 invasion of Iraq is
instructive on every point. By multiple accounts, President George W.
Bush made a “lonely decision” in secret for war. 108 He did so in the context
of great anxiety about additional 9/11-scale attacks and before many
agencies represented on the NSC understood that a decision had been
made.109 The President decided on war without focused NSC deliberation
on the question of going to war, and without skeptical analysis of
simplistic, assumption-laden Defense Department plans for Iraq after
regime change.110 The President’s precipitous decision left the U.S.
government unprepared when no meaningful stocks of weapons of mass
destruction (the primary stated rationale for war) were found in Iraq and
the country devolved into long years of chaos, insurgency, and civil war.
The region, the reputation of the United States government abroad and at
home, and thousands of American families who lost loved ones in the war
have not fully recovered. A poorly-made decision for war in Iraq in turn
facilitated the rise of the so-called Islamic State, years of civil war in Syria,
and a massive refugee flow that has destabilized U.S.-allied Europe. 111

107 See, e.g., BAKER, supra note 40, at 124.

108 See HAASS, supra note 106 (book contrasts poor process regarding Iraq in 2002–3 with
good process in 1990–91); DONALD RUMSFELD, KNOWN AND UNKNOWN: A MEMOIR 456-57
(2012) (many NSC discussions about Iraq generally and war preparations but President made
a “lonely decision” for war).
109 The former State Department Policy Planning Staff director recalls that Secretary of
State Colin Powell had to go around the NSC process and air his concerns with the President
at a private dinner. The President had already decided. See HAASS, supra note 106, at 233–37.
110 See PETER R. MANSOOR, SURGE 6-7 (2014) (historian and former Army officer
analyzes planning errors).
111 Congress did ratify the President’s decision and authorize force, investing a second
branch in the decision. Congress’s deliberations were shaped by arguments that Congress had
to back their Commander in Chief in a time of war with terrorists that the Bush Administration

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

Of course, additional process is not a panacea. Even the best process

cannot cure a no-win situation, or always guarantee that the objectively
best option will be selected. Process itself also has potential drawbacks. 112
These include delay, micromanagement, bureaucratic parochialism,
groupthink, and risk of leaks. Despite their success in running the bin
Laden decision, for example, the Obama NSC staff were at times criticized
for too much process – taking too long to decide and intruding into agency-
level details and authorities.113
Managing the NSC process is a task of the President and senior
advisors, not a reason not to use it. The bin Laden raid case study and other
successful examples of inter-agency process demonstrate that NSC review
can enhance decision-making about the most highly sensitive matters.
Secrets can be kept, operational surprise maintained, timely legal advice
can be provided, and thoughtful decisions made.
For a matter as complex and ramified as use of nuclear weapons, a
process that includes an array of actors responsible for different aspects of
the decision, carefully evaluates the intelligence, and which examines a
full range of options and their implications, increases the chances that an
ill-considered nuclear strike will not happen. In the event that nuclear use
is necessary and legal, more robust review would help identify the best
nuclear option. That is, one that is tailored in terms of weapon, yield,
target, and means of delivery, and thereby better able to serve its objectives
and minimize unnecessary harm – as is prudent from military and
diplomatic standpoints and is required by the law of armed conflict.
The high-value target (HVT) counter-terrorism “playbook” and its
companion for cyber operations, reflected in a series of NSC directives
during the Obama Administration, provide a modern NSC-level process
precedent.114 They need a companion that provides NSC process for
consideration of use of nuclear weapons, operative where decision time is
available and tailored for nuclear matters.

claimed were linked with Iraq’s Saddam Hussein regime. See Authorization for the Use of
Military Force Against Iraq Resolution of 2002, Pub. L. No. 107-243, 116 Stat. 1498 (2002).
112 See BAKER, supra note 40, at 124.
113 See, e.g., GATES, supra note 106, at 587 (Obama Administration Defense Secretary
complains of micromanagement).
114 See EXEC. OFF. OF THE PRESIDENT, PROCEDURES FOR APPROVING DIRECT ACTION
AGAINST TERRORIST TARGETS LOCATED OUTSIDE THE UNITED STATES AND AREAS OF ACTIVE
HOSTILITIES (2013) https://perma.cc/D9FY-DWPA (declassified playbook guidelines);
Dakota S. Rudesill, Trump’s Secret Order on Pulling the Cyber Trigger, LAWFARE (Aug. 29,
2018), https://perma.cc/NFV7-EL6F (discussing Obama-era NSC-level decision processes for
counter-terrorism and cyber operations).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 33

B. Implementation: The Need for a Statute

As in the case of inter-agency review of counter-terrorism and cyber

operations, NSC review could be implemented by executive order. But an
administrative-only solution, like other proposals offered during the Cold
War and during the revived conversation about nuclear command and
control, have shortcomings. This section evaluates several non-legislative
proposals, and explains why they are more problematic than a statute.
Changes to the nuclear command and control process could be
promulgated most quickly (and potentially in the greatest detail)
administratively. Ambassador Richard Betts and Professor Matthew
Waxman recommend that the President by executive order mandate that
any launch order be accompanied by Secretary of Defense attestation that
the order is valid and Attorney General agreement that it is legal. 115 The
President should issue an executive order that goes at least as far as Betts
and Waxman urge. However, as proposed it is limited to two cabinet
actors. The Betts-Waxman process order would not necessarily capture the
benefits of more fulsome NSC review, to include the State Department
(diplomatic equities) and U.S. Intelligence Community (intelligence
considerations). It also leaves constraints on the President to the discretion
of the President. A chief executive can cancel an executive order as easily
as promulgate one – including orally, and in secret. 116 This is not a
hypothetical. Via classified directives, President Trump reportedly vitiated
much of the NSC decision process he inherited regarding counter-terror
raids and cyber operations.117 The valuable expectation-setting and norm-
building a nuclear decision process executive order might do could be
largely canceled on day one of the term of a new president who prefers
maximum flexibility – or at any point thereafter.
If an executive order is the quickest but least resilient route to reform,
the procedurally most difficult but legally sturdiest is a constitutional

115 See Betts & Waxman, supra note 6. See also James M. Acton, Keynote Remarks at
the 2019 Project on Nuclear Issues Capstone Conference (Apr. 30, 2019) (physicist and policy
expert urges executive order to add other cabinet officials to nuclear launch decisions).
116 It is the Justice Department’s position that the President may revise or withdraw an
executive order at any time, in writing or orally, without public notice. See Rudesill, supra
note 49, at 291.
117 See Dustin Volz, Trump, Seeking to Relax Rules on U.S. Cyberattacks, Reverses
Obama’s Directive, WALL ST. J. (Aug. 15, 2018, 11:36 PM), https://perma.cc/BJ2P-ATEE;
Charlie Savage & Eric Schmitt, Trump Poised to Drop Some Limits on Drone Strikes and
Commando Raids, N.Y. TIMES (Sept. 21, 2017), https://perma.cc/N73N-A8DY; Rudesill,
Cyber Trigger, supra note 114.

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

amendment.118 An amendment could govern nuclear weapons specifically

and enduringly restructure the Executive Branch. 119 But the amendment
process has fallen into disuse. It is probably not a possibility in this hyper-
divided age.120
Some participants in the nuclear dialogue have suggested using judicial
process, but it could be equally unavailing.121 As discussed in Part IV
below, the courts sometimes do speak and speak powerfully regarding
national security and separation of powers.122 Federal courts, however,
prefer to avoid national security matters, often deploying avoidance and
justiciability doctrines – such as deference canons, the political question
doctrine, and standing, ripeness, and state secrets theories – to sidestep the
merits or toss suits entirely.123 Because abuse of the nuclear command and

118 See U.S. CONST. art. V.

119 See Yonkel Goldstein, Note, The Failure of Constitutional Controls over War Powers
in the Nuclear Age: The Argument for A Constitutional Amendment, 40 STAN. L. REV. 1543
(1988) (calling for amendment to define the roles of Congress and the President regarding
nuclear war).
120 The Constitution’s other path to amendment is through a constitutional convention
called by two-thirds of the states. See U.S. CONST. art. V. Balanced budget advocates are
getting closer to having enough states call for a constitutional convention. See Michael Wines,
Inside the Conservative Push for States to Amend the Constitution, N.Y. TIMES (Aug. 22,
2016), https://perma.cc/3R5N-7VY5. The convention to perfect the Articles of Confederation
teaches that such a process could spiral, pulling in many other issues or even again lead to an
effort to write a new Constitution. See id. Meeting the three-fourths-of-states ratification
threshold would be hard.
121 See, e.g., Paul A. Hemesath, Note, Who’s Got the Button - Nuclear War Powers
Uncertainty in the Post-Cold War Era, 88 GEO. L.J. 2473, 2502-03 (2000) (arguing for
amendment of the War Powers Resolution to allow court challenge and ruling on nuclear
weapons under the Constitution).
122 See, e.g., Hamdan v. Rumsfeld, 548 U.S. 557, 561 (2006) (presidential order regarding
suspected enemy fighters in post-9/11 armed conflict with Al Qaeda is contrary to statute and
invalid); Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 587-89 (1952) (presidential
order to seize steel mills during Korean War is contrary to will of Congress and invalid); The
Prize Cases, 67 U.S. 635, 665 (1862) (presidential order to blockade the South during Civil
War leading to seizure of ships is valid in absence of legislative authorization or restriction);
Little v. Barreme, 6 U.S. (2 Cranch) 170, 177-78 (1804) (presidential order regarding seizure
of ships during naval war with France is contrary to statute and invalid).
123 See, e.g., Clapper v. Amnesty Int’l USA, 568 U.S. 398, 400, 420-23 (2013) (petitioners
lack standing to challenge FISA Amendments Act because they cannot show they were
secretly surveilled); United States v. Richardson, 418 U.S. 166, 174-76 (1974) (taxpayer lacks
standing to challenge intelligence statute); Campbell v. Clinton, 203 F.3d 19, 19 (D.C. Cir.
2000) (suit by Members of Congress alleging War Powers Resolution violation dismissed for
lack standing and political question reasons); Dellums v. Bush, 752 F. Supp. 1141, 1149-52
(D.D.C. 1990) (suit during run-up to war with Iraq by Member of Congress dismissed on
ripeness). Cf., John Hart Ely, Suppose Congress Wanted a War Powers Act that Worked, 88
COLUM. L. REV. 1379, 1407-17 (1988) (criticizing justiciability doctrines).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 35

control system could precipitate a cataclysmic nuclear war, the system

does implicate life, liberty, and indeed everything else protected by the
Constitution.124 Courts, however, do not open themselves to adjudicating
claims of catastrophically poor judgment in use of force or maintaining
civilization-ending weapons.125 Even if nuclear command and control
were to get before the judiciary, courts generally refuse to impose process
or otherwise insist on limits on executive national security decision-
making unless individual rights or exercise of Congress’s powers are
implicated.126
For that reason, because other lawmaking routes are so problematic,
and because of the risks inherent in the status quo, it is to statutory
solutions that we now turn.

IV. THE STRENGTHENED CASE FOR THE CONSTITUTIONALITY OF

STATUTORY RULES FOR NUCLEAR WEAPONS

Since the dawn of the nuclear age, Congress has regularly legislated
regarding nuclear hardware. The nation’s legislature has authorized,
funded, structured, modernized, and overseen a truly massive array of
nuclear assets: a stockpile that at its height included some 30,000
warheads; the thousands of aircraft, missiles, and submarines that have
carried them; the dozens of military installations that have hosted them;
the nuclear laboratories and industrial enterprise that have designed and
created nuclear hardware, at an historical cost approaching a trillion
dollars; the world’s most advanced intelligence, surveillance, and
reconnaissance (ISR) and early warning (EW) capabilities, designed to

124 Cf., Elaine Scarry, War and the Social Contract: Nuclear Policy, Distribution, and the
Right to Bear Arms, 139 U. PA. L. REV. 1257, 1296-1301 (1991) (arguing that nuclear weapons
are unconstitutional and damage the social contract); Arthur S. Miller, Nuclear Weapons and
Constitutional Law, 7 NOVA L. REV. 21, 36 (1982) (arguing that nuclear weapons are
unconstitutional set against constitutional ethos and “so endanger the lives, liberties, and
property of all Americans that they should be considered to be a deprivation contrary to due
process”).
125 Courts “are not the only guardians of the Constitution. Their reluctance [to adjudicate
nuclear command] should not foreclose a growing dialogue.” Miller, supra note 124, at 36-
37.
126 See Hamdan, 548 U.S. 557, 624 (President’s Commission Order No. 1 regarding
enemy combatants invalid as violation of Uniform Code of Military Justice statute); Hamdi v.
Rumsfeld, 542 U.S. 507, 508 (2004) (adjudication must include constitutional due process
protections for U.S. citizen detained as enemy combatant); Youngstown, 343 U.S. at 587-89
(presidentially ordered seizure of steel mills for national defense purposes invalid because not
authorized by statute or Constitution). See also United States v. Curtiss-Wright Exp. Corp.,
299 U.S. 304, 320-22 (1936) (upholding statutory delegation of decision authority to President
regarding arms trade).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

detect and monitor nuclear threats worldwide; and a multi-layered globally

operational nuclear command, control, and communications system
crafted to enable timely and informed presidential decision even in the
throes of nuclear Armageddon.127 Additionally, Congress has a long record
of involvement in nuclear arms control. The Senate has offered its advice
and consent to a series of Washington-Moscow nuclear arms control
treaties that have capped and then reduced nuclear forces, and then
legislatively implemented them, starting in 1972.128 That was the year the
Congress also held its single vote on legislation governing operational
employment of nuclear weapons, the Fulbright Amendment. Although that
measure did not pass, other statutory proposals were frequently discussed
until the Cold War’s end.
A key question then as now is whether legislation governing nuclear
command and control would be constitutional. That debate needs to be
understood before considering new proposed legislation. Importantly, the
constitutional conversation also needs to be updated to account for
important developments since the nuclear command and control
conversation (at least outside the Defense Department) went on hiatus
when the Berlin Wall came down 30 year ago.
This article maintains here in Part IV.A that reformers should not be
dissuaded from statutory solutions by the nature of the constitutional
conversation to date. The fate of the Fulbright Amendment, like implicit
assertions that nuclear weapons are constitutionally special, are more red
herrings than they are instructive. Next, Part IV.B analyzes separation of
powers doctrine in relevant part. Congress has unused authority to govern
nuclear weapons legislatively in the face of expansive assertions of
presidential power. Ultimately, nuclear weapons need not be
constitutionally special – for either branch. Third, Part IV.C argues that
nuclear weapons can instead be made statutorily special. Thanks to a series
of international security, doctrinal, and legislative developments
subsequent to the Fulbright Amendment, the nuclear command and control
conversation is resuming with firmer footing for Congress to write
statutory rules. Especially salient is that the past half century has seen

127 Former Secretary of Defense Mark Esper wrote his dissertation on Congress and
strategic forces. See Mark T. Esper, The Role of Congress in the Development of the United
States’ Strategic Nuclear Forces, 1947-68, 405-08 (Aug. 31, 2008) (Ph.D. dissertation, George
Washington University) (Proquest) (arguing that Congress was extensively involved in
crafting of long-range nuclear forces).
128 For discussion, see Rudesill, 102 GEO. L.J., supra note 70, at 128–38 (analysis of
history of bilateral nuclear arms control agreements approved by Senate); David A. Koplow,
Eve of Destruction: Implementing Arms Control Treaty Obligations to Dismantle Weaponry,
8 HARV. NAT. SEC. J. 158 (2017) (Congress involved in implementation).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 37

enactment and entrenchment of tailored process statutes for covert action

and national security surveillance. Unlike the effective permission the War
Powers Resolution (WPR) provides for the President to initiate hostilities
for 60 to 90 days without congressional authorization, 129 the covert action
and surveillance regimes constrain the President’s ex ante access to
national security loaded weapons. They will be used as legislative
precedents in the next main section (Part V) for making nuclear weapons
statutorily special.

A. Reform Should Not be Dissuaded by the Constitutional

Conversation to Date

The new conversation about nuclear command and control sometimes

proceeds as if the Cold War-era debate resolved the constitutional
questions. Some, especially in the policy and military nuclear communities
that usually drive nuclear conversations, essentially believe that Congress
lost and whatever powers it could have asserted are permanently dormant.
Others with legal and non-legal backgrounds believe that nuclear weapons
are inevitably constitutionally special in one sense or another: uniquely
and per se unconstitutional, reserved exclusively to the President, or
available to the President for first use only if Congress formally declares
war. None of these claims are persuasive. They distract from Congress’s
opportunity to assert its under-utilized but still extant powers to govern the
nation’s nuclear loaded weapons.

1. Cold War-era Statutory Proposals

As discussed in Part III.B above, participants in the new nuclear

command and control conversation should not be dissuaded from
exploring reform due to the problems with executive order, constitutional
amendment, or litigation solutions. Similarly, reformers should not be
dissuaded by the failure of Congress to enact any one of several prominent
Cold War-era statutory proposals and the ensuing decades of acquiescence
to near-total Executive power.
Nearly a half century ago the Senate held Congress’s single vote to
date about whether to apply legislative rules to use of nuclear weapons. An
amendment to the WPR authored by Senator J. William Fulbright (D-AR)
would have barred nuclear use “without the prior explicit authorization of
the Congress” except in the most imminent threat situation imaginable:

129 See War Powers Resolution, Pub. L. No. 93-148, 87 Stat. 555, § 5(b) (1973) (60 to 90
day clock).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

where adversary nuclear weapons were in the air or had already detonated.
Arguments that the amendment would infringe on the President’s authority
as Commander in Chief played a role in the measure’s 68-10 defeat in
April 1972.130
It would be wrong, however, to interpret such overwhelming rejection
of the amendment as a compelling legislative precedent about the
constitutionality of writing rules for nuclear use. Of course, Congress is
not bound by its past rejection of a proposed law. And as is common in
legislative bodies, there was a lot else going on.
In the teeth of the Cold War, the Fulbright Amendment would have
challenged U.S. deterrence policy in two respects. First, absent “explicit
authorization of the Congress” – which could take hours to months – the
United States would have lost the deterrent power of the threat to shoot
first in anticipatory self-defense if it looked like the Soviet Union or China
were preparing a nuclear attack. Second, the Fulbright Amendment was
vulnerable to the charge that if enacted it could raise questions about the
credibility of NATO’s policy of relying on the threat of U.S. first use of
tactical nuclear weapons in Europe to deter attack by the Warsaw Pact’s
numerically superior conventional forces.131 Absent separate
congressional action, the Fulbright Amendment would have instituted a
“no first use” policy.132
Several Senators urged colleagues to reserve the nuclear and “no first
use” basket of questions to a separate bill in the future. 133 Senators also
argued that if the Fulbright Amendment passed it would open a Pandora’s
box of amendments.134 Additionally, critics assailed the amendment’s

130 See 118 CONG. REC. S12451 (1972) (statement of Sen. Javits); 118 CONG. REC. S12452
(statement of Sen. Eagleton). For discussion, see Stephen P. Mulligan, CONG. RES. SERV.,
LEGISLATION LIMITING THE PRESIDENT’S POWER TO USE NUCLEAR WEAPONS: SEPARATION OF
POWERS IMPLICATIONS 12 (Nov. 3, 2017). The vote on the Fulbright Amendment had an
unusually high number of non-voting Senators (22).
131 See McGeorge Bundy, George F. Kennan, Robert S. McNamara & Gerard Smith,
Nuclear Weapons and the Atlantic Alliance, 60 FOR. AFF. 753, 754 (1982) (willingness to use
nuclear weapons first was “major element in every doctrine” of NATO). As mentioned in
supra note 95, some experts and scholars believed during the Cold War that even in the event
of a Warsaw Pact conventional invasion of Western Europe there would probably have been
sufficient time for Congress to act before the United States used nuclear weapons.
132 That was Fulbright’s stated intent, but not U.S. policy. See 118 CONG. REC. S12450
(statement of Sen. Fulbright); 118 CONG. REC. S12451 (statement of Sen. Javits) (amendment
raised questions “fundamental to the whole strategic posture of the United States”).
133 See 118 CONG. REC. S12450 (statement of Sen. Spong), 118 CONG. REC. S12452
(statement of Sen. Eagleton), 118 CONG. REC. S12454 (statement of Sen. Cooper).
134 See 118 CONG. REC. S12450 (statement of Sen. Spong) (amendment would “open the
door for many other amendments”).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 39

scope. Fulbright’s proposal went beyond nuclear arms.135 Some Senators

worried that it gave the President more authority to use conventional force
than Congress intended. 136
The Fulbright Amendment’s fate was tied tightly to its Cold War
context, legislative strategy considerations, its reach into conventional
force questions, and concerns about drafting.137 Senator William B. Spong
(D-VA) was correct that a vote for or against the Fulbright Amendment
could create “misinterpretation” of Congress’s intent and confusion about
its legislative effort to interpret the Constitution. 138
Similarly, one should not read a general rejection of congressional
authority to govern nuclear command and control into concerns with
another prominent Cold War-era legislative proposal. The Fulbright
Amendment was inspired by a bill drafted by the Federation of American
Scientists that would, absent a declaration of war, require the President to
get the concurrence of a committee of congressional leaders before
ordering nuclear launch.139 To critics, the Federation was proposing an
impermissible legislative veto.140 Others disagreed and defended a de
minimis and warranted deviation from rigid formalism in separation of
powers.141

135 See Senate Passes Bill Defining Constitutional War Powers, in CQ ALMANAC 1972, at
05-842-05-851 (28th ed., 1972, 1973).
136 See 118 CONG. REC. S12450 (statement of Sen. Spong).
137 The Markey-Lieu bill of recent congresses is similar to the Fulbright Amendment but
simpler. Markey-Lieu requires a war declaration for first use of nuclear weapons unless the
President determines “that the enemy has first launched a nuclear strike against the United
States or an ally of the United States.” The bill does not concern conventional forces. See
Restricting First Use of Nuclear Weapons Act of 2017, H.R. 669, 115th Cong. (2017);
Restricting First Use of Nuclear Weapons Act of 2017, S. 200, 115th Cong. § 3 (2017).
138 See 118 CONG. REC. S12450 (statement of Sen. Spong).
139 See Peter Raven-Hansen, Introduction, in FIRST USE OF NUCLEAR WEAPONS: UNDER
THE CONSTITUTION, WHO DECIDES?, supra note 5, at ix (Fulbright Amendment grew out of
Federation of American Scientists proposal); Jeremy J. Stone, Presidential First Use is
Unlawful, in FIRST USE OF NUCLEAR WEAPONS: UNDER THE CONSTITUTION, WHO DECIDES?,
supra note 5, at 3, 11-12 (describing and arguing for Federation bill).
140 The Supreme Court later held the legislative veto unconstitutional. See I.N.S. v.
Chadha, 462 U.S. 919, 921 (1983) (invalidating legislative veto). See also Stephen L. Carter,
War Making Under the Constitution and the First Use of Nuclear Weapons, in FIRST USE OF
NUCLEAR WEAPONS: UNDER THE CONSTITUTION, WHO DECIDES?, supra note 5, at 109, 109-
28.
141 See William C. Banks, First Use of Nuclear Weapons: The Constitutional Role of a
Congressional Leadership Committee, in FIRST USE OF NUCLEAR WEAPONS: UNDER THE
CONSTITUTION, WHO DECIDES?, supra note 5, at 129, 129-42 (arguing that committee not a
Chadha violation); Charles Tiefer, The FAS Proposal: Valid Check or Unconstitutional Veto?,
in FIRST USE OF NUCLEAR WEAPONS: UNDER THE CONSTITUTION, WHO DECIDES?, supra note

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

Non-lawyer participants in the new nuclear command and control

conversation in particular need to understand that the fate of these
proposals, or any other introduced but unenacted to date, reflect at most
congressional acquiescence. As Justice Jackson observed, Congress can
make its powers meaningful through their use.142

2. Claims that Nuclear Weapons are Constitutionally Special

Just as controversies about past prominent legislative proposals should

not dissuade statutory reform, so ought not assertions – express or implied
– that nuclear weapons are constitutionally special.143
Some thinkers have argued that nuclear weapons are inherently
unconstitutional because they are so catastrophically destructive, carry
such unique escalation risk, and have concentrated so much authority in
one person. Their “unnatural monstrocity” means that they cannot be
reconciled with the Constitution’s ethos, popular rights including life and
liberty, and Congress’s role under the Constitution.144 However, the
Constitution’s text – beyond categorical distinctions among federal armies
and a navy, state troops and ships of war, the militia, and privateers
(Marque and Reprisal) – draws no distinctions among forces. 145 Today,
this constitutional speciality claim flies in the face of 75 unbroken years of
two-branch investment in, and public acceptance of, nuclear deterrence.
Finally, recall that the worst-case nuclear nightmare of imminent or
initiated adversary nuclear attack and the expansive presidential power that
comes with it persists as a remote possibility but is not the most likely
nuclear employment scenario.
A more common line of constitutional speciality thinking includes the
notion that “the bomb” must always be left entirely to the President.
Sometimes, a presidentialist view of separation of powers (see discussion

5, at 143, 143-65 (distinguishing Chadha as domestic case); Raven-Hansen, supra note 15, at
794 (deviations from formalist constitutional vision allowed by Supreme Court to promote
key congressional objectives and warranted here).
142 See Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 641 (1952) (Jackson, J.,
concurring) (discussing congressional acquiescence).
143 I thank a conversation with Professor Matthew Waxman at Stanford’s February 2019
conference on nuclear launch authority for the concept of nuclear weapons being
constitutionally special.
144 See Milner S. Ball, Nuclear War: The End of Law, 7 NOVA L. REV. 53, 57 (1982) (nuclear
weapons are “deconstitutionalizing or anti-constitutionalizing” because of their “blasphemy”
and “unnatural monstrocity” [sic]). For arguments against constitutionality, see Miller, supra
note 124, at 36-37; Scarry, supra note 124.
145 See U.S. CONST. art. I, § 8, cl. 10, 12, 13, 15, 16; U.S. CONST. art. I, § 10, cl. 3.

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 41

below) that embraces all modalities of force is invoked to put employment

of nuclear weapons beyond legislative control.146 This does not reflect
constitutional speciality of nuclear weapons. Where a presidentialist vision
is not deployed, however, discussions of nuclear force frequently will
nevertheless not admit room for legislative governance. This does reflect
constitutional speciality thinking (although commonly of a variety that is
more strong normative implication than explicit argument). Typically,
omissions of any potential legislative power over nuclear employment are
paired with reference to the time horizons and existential stakes of the
worst-case nuclear nightmare.147 But again: the scenario that has driven
design of the command and control system and most powerfully shaped
constitutional thinking about it is not the most likely scenario. In short,
when referencing the constitution, the nuclear use conversation – in print
and especially in the informal conversations in which constitutional law is
often most meaningful148 – goes too quickly to the most extreme case to
the detriment of a full discussion of the legislative regulatory possibilities.
A third variant of the speciality claim is that nuclear weapons are
constitutionally special for Congress. The Markey-Lieu bill, which builds
on ideas in the Fulbright and Federation legislative proposals, crisply
articulates this reasoning:

Sec. 2(a). FINDINGS. . . . (4) Nuclear weapons are uniquely powerful weapons
that have the capability to instantly kill millions of people, create long-term
health and environmental consequences throughout the world, directly
undermine global peace, and put the United States at existential risk from
retaliatory nuclear strikes.

(5) By any definition of war, a first-use nuclear strike from the United States
would constitute a major act of war.

(6) A first-use nuclear strike conducted absent a declaration of war by

Congress would violate the Constitution.

146 See, e.g., Robert F. Turner, Congressional Limits on the Commander in Chief: The
FAS Proposal, in FIRST USE OF NUCLEAR WEAPONS: UNDER THE CONSTITUTION, WHO
DECIDES?, supra note 5, at 37, 37-46.
147 See, e.g., Dunlap, The Danger of Tampering, supra note 12 (legal scholar rejects
legislative variation of Betts-Waxman proposal, without mention of room for other legislative
approaches, emphasizing adversary rapid attack risk that “leaves little time for the President
to exercise his Constitutional responsibility to provide for the common defense”); Woolf,
supra note 29, at 1 (policy expert rejects possibility of legislative action).
148 For the importance of the constitutional conversation in informal process and practice
settings, see BAKER, supra note 40.

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

Sec. 2(b). DECLARATION OF POLICY. – It is the policy of the United States

that no first-use nuclear strike should be conducted absent a declaration of war
by Congress.149

The Markey-Lieu bill has an exception for an initiated nuclear attack

against the United States or its allies. 150 The essential idea is, as Senator
Fulbright argued, that when nuclear weapons are used it is the “beginning
of a whole new war.”151 This claim of specialness and consequence has
some merit. Certainly, employment of a nuclear weapon against the forces
or territory of an adversary would be a major use of force and an act of
war. One would certainly hope the nation’s legislature would act before
nuclear weapons are used. A declaration of war would send powerful
messages of resolve and acceptance of responsibility by the elected
representatives of the people. But why must presidentialist claims of
constitutional specialness should be matched by equally rigid
congressionalist claims? Regular statutes have been used throughout U.S.
history to authorize wars that have often caused destruction on a scale one
or more nuclear weapons could inflict. Statutory force authorizations are
the only form of legislative war authorization used after World War II. 152
Statutes enacted in advance can also provide rules where Congress does
not or cannot act in a crisis.

B. Separated and Shared Powers Over Nuclear Weapons

The constitutional conversation about nuclear weapons has unfolded

in the context of separate, contested, and ultimately concurrent and shared
war powers of the Legislative and Executive Branches.
In their 1789 replacement for the Articles of Confederation, the
Framers’ project was one that we would recognize as balancing security
and liberty: to craft a federal central government strong enough to protect
a large country, but with powers sufficiently separated and checked that its
internal institutional and personal rivalries would prevent it from repeating

149 See Restricting First Use of Nuclear Weapons Act of 2017, H.R. 669, 115th Cong.
(2017); Restricting First Use of Nuclear Weapons Act of 2017, S. 200, 115th Cong. (2017).
150 See Restricting First Use of Nuclear Weapons Act of 2017, H.R. 669, 115th Cong.
(2017); Restricting First Use of Nuclear Weapons Act of 2017, S. 200, 115th Cong. (2017).
151 See 118 CONG. REC. S12450 (statement of Sen. Fulbright). See also SCARRY, supra
note 5, at 37-84 (“the Constitution requires a congressional declaration of war”).
152 See, e.g., Authorization for the Use of Military Force, Pub. L. No. 107-40, 115 Stat.
224 (2001) (authorizing force against 9/11 attackers and supporters) (authorizing force against
9/11 attackers and supporters); Act of June 5, 1942, ch. 325, 56 Stat. 307 (last U.S. war
declaration, against Nazi-controlled Romania).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 43

the predations against liberty that drove the American revolt against
British rule.153 Their project was equally defense-enabling and counter-
authoritarian.154 The Constitution created a President with responsibilities
as Chief Executive and Commander in Chief, 155 but as a check on military
dictatorship and to ensure accountability gave a refashioned bicameral
Congress an extensive slate of powers over the national security apparatus.
Via its powers to Raise and Support Armies and Provide and Maintain a
Navy, only the Congress could create a federal military.156 Via its
organizational, calling-forth, and governing powers over the state militias,
only the Congress could hand the new republic’s only extant military
forces of any consequence to the federal Commander in Chief.157 No funds
for these forces would be available without an act of Congress, and the
army’s funding expired after two years – a default fail-safe against an
oppressive military.158 Congress could statutorily “make Rules for the
Government and Regulation of the land and naval Forces,” controlling the
military establishment’s organization and justice system, and
operations.159 The nation’s legislature could authorize and control
privateers (Marque and Reprisal), make rules for captures on land and sea,
and define and punish infractions against international law. 160 Congress
was given power to Declare War, and tax and spend to “provide for the
common Defense.”161
The Constitution envisioned that the new federal government would
be better able to defend the country than the Articles’ confederal structure.
However, the Constitution did not give the power to “make war” to the
President nor to Congress.162 Similarly, the Constitution did not give either

153 See THE FEDERALIST NO. 51 (1788) (James Madison) (“Ambition must be made to
counteract ambition” and government must be structured “to control itself.”).
154 See Dakota S. Rudesill, The Land and Naval Forces Clause, 86 U. CIN. L. REV. 391,
394, 399, 416 (2018) (counter-authoritarian purposes); AMAR, supra note 35, at 114-121
(national defense-enabling powers and limitations).
155 See U.S. CONST. art. II, § 1-2.
156 See U.S. CONST. art. I, § 8, cl. 12-13.
157 See U.S. CONST. art. I, § 8, cl. 15-16.
158 See U.S. CONST. art. I, § 9, cl. 7; § 8, cl. 12.
159 See U.S. CONST. art. I, § 8, cl. 14. See Rudesill, The Land and Naval Forces Clause,
supra note 154, at 442, 480-81 (dual power theory of the Clause); AMAR, supra note 35, at
188 (Clause provides power over military justice and to proscribe President’s use of force, but
not direct it).
160 See U.S. CONST. art. I, § 8, cl. 11, 10.
161 See U.S. CONST. art. I, § 8, cl. 11, 1.
162 At the Constitutional Convention, the Committee of Detail’s draft accorded Congress
the power to “make war.” Its change by the delegates to “declare war” came in the context of

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

branch Congress’s power under the Articles of “directing” the “land and
naval Forces.”163 Giving command to the President and military-raising,
force-governing, funding-providing, and war-declaring to Congress was a
verdict for a capable federal government but one also set up for liberty-
protecting inter-branch tension.
The Constitution’s text, origins, and history of active interpretation by
the three branches put several aspects of the constitutional balance of
power largely beyond dispute. Only the Congress can create federal forces.
It can generally appropriate, terminate, or condition funding as it desires.
The President has what Professors Barron and Lederman term
“superintendence” – the Commander in Chief cannot be replaced at the
head of the military chain of command. 164 Using the forces Congress has
provided, the President can order them to repel sudden attacks, and has
some additional authority to use force absent congressional authorization
or restriction.165
If these constitutional four corners are clear, so too is that the
circumstances of the founding era were turned on their heads by the
nuclear age. The threat and its timelines, U.S. capabilities, and which
organs of government could be expected to take the greatest responsibility
for national defense could not have been more transformed. 166 The
Framers wrote the Constitution understanding that the President may need
to act – with what militia could be federalized or regulars were on hand –
before Congress could assemble. Similarly, the Framers gave the state

creation of the President as Commander in Chief but with frustratingly little explanatory
drafting history. See 2 MAX FARRAND, THE RECORDS OF THE FEDERAL CONVENTION OF 1787
at 182 (1911).
163 See ARTICLES OF CONFEDERATION OF 1781, art. IX.
164 See David J. Barron & Martin S. Lederman, The Commander in Chief at the Lowest
Ebb – Framing the Problem, Doctrine, and Original Understanding, 121 HARV. L. REV. 689,
767-70 (2008) [hereinafter Lowest Ebb Part I]; David J. Barron & Martin S. Lederman, The
Commander in Chief at the Lowest Ebb – A Constitutional History, 121 HARV. L. REV. 941
(2008) [hereinafter Lowest Ebb Part II]. See also DAVID J. BARRON, WAGING WAR: THE
CLASH BETWEEN PRESIDENTS AND CONGRESS 1776 TO ISIS (2016). Subsequent to authoring
the Harvard articles with Lederman, Barron was appointed to the U.S. Court of Appeals for
the First Circuit. See also Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 641 (1952)
(Jackson, J., concurring) (Constitution “undoubtedly puts the Nation’s armed forces under
presidential command” but its boundaries are uncertain). In contrast, the British could appoint
multiple Commanders in Chief, and the Congress under the Articles of Confederation could
do so as well.
165 See The Prize Cases, 67 U.S. 635, 665 (1863) (authority to repel attacks, in case where
Congress had not acted); Carter, supra note 140, at 118 & n. 58, 124 n. 20 (noting presidential
power to order use of force absent congressional authorization or restriction goes beyond
“repel attacks” power).
166 See Raven-Hansen, supra note 15, at 786 (making this observation).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 45

militias a primary role in national defense, and in the Constitution

explicitly gave states authority to “engage in War” if “actually invaded, or
in such imminent Danger as will not admit of delay.” 167 The Framers also
knew that the continent and Atlantic Ocean were vast and that sail ships,
horses, and human feet were slow. Communications with foreign powers,
intelligence collection about them, and the advance of adversary military
forces would all take time. Congress could convene and take a central role.
The Legislative Branch had temporal space to inquire, to deliberate in
consultation with the executive, and to decide about what kind of force to
employ (privateers, state forces, federalized militia, or federal regulars?),
whether and what kind of war to have, and what rules to write. “No
standing armies” was not just principle but reality: federal forces at the
Constitution’s ratification totaled a few hundred troops and no navy.168
James Wilson observed during the ratification debates that the
Constitution’s “system will not hurry us into war; it is calculated against
it. It will not be in the power of a single man, or a single body of men, to
involve us in such distress.”169 The nuclear age, in contrast, has been
characterized by standing U.S. nuclear forces of global reach and
immediate availability, armed with more thermonuclear bombs than
President Washington had federal soldiers, arrayed against nuclear
arsenals in Eurasia and under the seas that can devastate the country in
minutes. Reliance on the President’s urgent defense authority became
unavoidable.

167 See U.S. CONST. art. I, § 10, cl. 3 (conditional permission for states to wage war);
Youngstown, 343 U.S. at 644 (Jackson, J., concurring) (“the militia rather than a standing army
was contemplated as the military weapon of the Republic”); THE FEDERALIST NO. 29
(Alexander Hamilton) (national army “dangerous to liberty” is unnecessary if the federal
government could employ the state militias to defend the country). But see RICHARD H. KOHN,
EAGLE AND SWORD: THE BEGINNINGS OF THE MILITARY ESTABLISHMENT IN AMERICA 277-303
(1975) (efforts by some Framers to neuter militias, and controversy over Hamilton’s drive in
1790s for federal army).
168 Years before, the Continental Army had been demobilized and the Continental Navy
disbanded and its last ship sold. See 1 WILLIS J. ABBOT, I NAVAL HISTORY OF THE UNITED
STATES, ch. XV (1896). Alexander Hamilton successfully pushed for a few Treasury
Department revenue cutters in 1790, the predecessors to the U.S. Coast Guard. A small U.S.
Navy was established by the Naval Act of 1794. See 1 Stat. 350, ch. 12 (1794). During the
Washington and Adams Administrations, Congress blunted Executive drives for sizable
federal armies to deal with frontier and French threats. Congress did create a frontier force of
several thousand but to the frustration of Hamilton – then the second-highest ranking general
– abandoned work on a new army when the French threat fizzled. See BARRON, supra note
164, at 43-49; KOHN, supra note 167, at 277-88. The War Department in the mid-1790s had a
headquarters staff of seven, including the Secretary and doorkeeper. Working personnel at
times totaled two. See KOHN, supra note 167, at 290-92.
169 See 2 ELLIOT’S DEBATES 528 (J. Elliot, ed., 1832) (statement of Wilson).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

The legal aspect of the nuclear command and control conversation has
grappled with how the Founders’ constitutional vision is to be made
meaningful, and how the Congress can remain relevant, in such radically
changed and existentially perilous times. Debate often focused during the
Cold War, as it has again now that the conversation has resumed, on the
question of if and when the President may use nuclear weapons absent
congressional authorization, and on what ex ante controls Congress can or
should impose.
In the context of Justice Jackson’s canonical tripartite framework in
Youngstown Sheet & Tube v. Sawyer (the Steel Seizure Case) for
evaluating the intersection of congressional and presidential powers in
national security, these questions are ones of Categories Two and Three. 170
There is no question of the President’s authority under U.S. law to employ
nuclear force in Category One: where the President is acting pursuant to
congressional authority. The Cold War-era and Markey-Lieu statutory
proposals discussed in Part IV.A aspire to put any nuclear use in this
category. Category Two is “a zone of twilight” of “uncertain” power,
where the President is acting absent congressional authorization.171 A
presidential launch order in response to a sudden nuclear threat that does
not admit time for legislative action would be on firm ground in Category
Two, as acknowledged by virtually all participants in the nuclear command
and control debate, and the Fulbright and Markey-Lieu proposals. In
contrast, a Rogue President presents a terrible dilemma: practical control
of the nuclear trigger but highly-questionable-to-zero true constitutional
authority, because the threat that the Commander in Chief may repel
without congressional action is highly-questionable-to-zero. Like the
Rogue President, the Precipitous President who reaches for the nuclear
football in a developing crisis or ongoing conflict where adversary nuclear
attack is not imminent, and does so without evaluation of implications and
alternatives, could either be a Category Two or Three actor. If Congress
has not expressed its will, the President’s constitutional authority would
depend on the extent to which the threat is real, and if so whether nuclear
use is necessary and otherwise legal. It would also depend on whether one
agrees with the generally accepted but still contested proposition that the
President may employ force below the level of full war without

170 See Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 636-41 (1952) (Jackson,
J., concurring) (framework for analyzing collisions of the President and Congress regarding
national security). For discussion of the framework, see Heidi Kitrosser, It Came from Beneath
the Twilight Zone: Wiretapping and Article II Imperialism, 88 TEX. L. REV. 1401 (2010);
HAROLD HONGJU KOH, THE NATIONAL SECURITY CONSTITUTION 112, 134-36 (1990).
171 See Youngstown, 343 U.S. at 636-37 (Jackson, J., concurring).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 47

congressional authorization, beyond situations in which attacks are

imminent or underway – assuming nuclear use could ever fall below the
level of full war. Rounding out the Youngstown framework, a Rogue,
Precipitous, or even thoughtful President might instead order nuclear use
contrary to the express or implied will of Congress, for example in
violation of an enacted Markey-Lieu bill. Here, the Commander in Chief
would find their power at its “lowest ebb,” able to rely only on whatever
power Congress could not extinguish.172
The nuclear conversation about which branch wins in Youngstown’s
Category Three has fallen along the familiar lines of three general bodies
of theory.173 Rarely articulated in the nuclear context is the
congressionalist school of thought. It would allow Congress to legislate
virtually any rules for nuclear forces or any other, to include detailed
tactical direction of nuclear war.174 A second school of thought, what can
be termed the presidentialist view, reflects expansive claims of executive
power. Generally, in its most rigid and formalist incarnation, this vision
exalts the President’s supreme powers to act in defense of the country,
checked only via appropriations termination or impeachment and
removal.175 Otherwise, inter-branch clashes are left to elections and
politics.176 Presidentialism is unusually common inside the government’s
nuclear community, implicitly reflecting the view that nuclear weapons are
constitutionally special presidential weapons. Despite its adherents in
several presidential administrations, presidentialism like congressionalism
is a minority viewpoint among jurists and scholars. The majority school of
thought generally, with adherents among scholars in the nuclear command

172 See id. at 636-37 (Jackson, J., concurring).

173 For discussion of these schools of thought, see Rudesill, The Land and Naval Forces
Clause, supra note 154, at 426-31. These categories are by nature generalizations.
174 For a congressionalist view, see, e.g., Saikrishna Bangalore Prakash, The Separation
and Overlap of War and Military Powers, 87 TEX. L. REV. 299, 332 (2008).
175 See, e.g., JOHN YOO, THE POWERS OF WAR AND PEACE: THE CONSTITUTION AND
FOREIGN AFFAIRS AFTER 9/11, 155, 159-60 (2005) (non-appropriations statute cannot limit
presidential use of the military). A still smaller minority of thinkers go further and argue that
even the power of the purse cannot limit what the President can order the military to do, once
Congress raises forces. See, e.g., J. Terry Emerson, The War Powers Resolution Tested: The
President’s Independent Defense Power, 51 NOTRE DAME L. REV. 187, 210, 213 (1975)
(argument). The scholarly consensus is different. See WILLIAM C. BANKS & PETER RAVEN-
HANSEN, NATIONAL SECURITY LAW AND THE POWER OF THE PURSE 7, 181 (1994) (Congress
generally prevails regarding appropriations).
176 See, e.g., Turner, supra note 146, at 37-48.

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

and control conversation, is that of shared, concurrent power. 177 In

Youngstown Category Three the Commander in Chief remains what
Hamilton described as “the first general and admiral” of the nation, but
like the military commanders of Founding-era states and Great Britain is
bound by statute.178 This theory generally was reflected in Justice
Jackson’s analysis and the Supreme Court’s invalidation in Youngstown of
President Truman’s defiance of congressional will to seize steel mills for
national defense purposes in the teeth of the Korean War – a war some
feared would go nuclear. The majority, concurrent power view would
dictate a strong presumption that a Rogue or Precipitous President who
ordered a nuclear launch in defiance of statute would be acting illegally
under the Constitution.
Such a statute would not be congressional direction of tactical
maneuvers, analogous to the instructions on the movement of forces and
timing of attack typical of Congress’s micromanagement of the War for
Independence’s Commander in Chief, General George Washington.
Rather, a nuclear use statute would reflect Congress attaching conditions
to its raising-and-providing particular nuclear (air) armies and naval
forces; writing “Rules for the Government and Regulation of the land and
naval Forces;” deciding generally on the scope and intensity of war the
President would then command; and providing appropriations. 179
Congress would reasonably be exercising its checking legislative powers
to ensure that spending only “provide[s] for the common Defence and
general Welfare” and guards against potentially cataclysmic abuse of the
repel-attacks and superintendence presidential authorities.
The renewed nuclear command and control conversation should take
note of a significant body of new research since the Cold War that has
added to the balance of the Founding Era evidence and constitutional
history that lies against the presidentialist claim that in national security
Congress cannot bind the President via statute. Especially compelling is
the magisterial work of Professors Barron and Lederman, analyzing the
evidence regarding Youngstown Category Three clashes from the

177 See Banks, supra note 141 (assumption of executive-congressional shared war powers
underlying article’s argument); Carter, supra note 140, at 111-16; Jules Lobel, Conflicts
Between the Commander in Chief and Congress: Concurrent Power over the Conduct of War,
69 OHIO STATE L.J. 391 (2008); Raven-Hansen, supra note 15.
178 See THE FEDERALIST NO. 69 (Alexander Hamilton).
179 See Zachary S. Price, Funding Restrictions and Separation of Powers, 71 VAND. L.
REV. 357, 361-62, 426-37 (2018) (“near-complete” congressional power to withhold forces
and funding and attach conditions binding on the Commander in Chief).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 49

Founding to George W. Bush Administration.180 Presidentialist claims

were relatively rare from Washington’s Administration into the Twentieth
Century. Those assertions, and a more limited body of newer scholarship
by presidentialists, have stumbled against an extensive multi-century
record of Congress often governing the armed forces and presidential
employment of them in detail.181 Barron and Lederman, like a number of
other constitutional scholars writing after the Berlin Wall’s demise, fortify
a “well-developed understanding” from the Founding onward that the
Commander in Chief “could be subject to legislative control even as to
tactical matters of war.”182
This new scholarship has thrown instructive light on a series of
statutes, dating from the Founding Era to the present, that have conditioned
and restricted what the President can do with the armed forces. For
example, the Insurrection Act and Posse Comitatus Act have
circumscribed domestic use of force, and a series of statutes have regulated
detention and interrogation of the enemy (including barring torture). 183 As
discussed in more depth below, the covert action statute and Foreign
Intelligence Surveillance Act (FISA) bar direct (often lethal) action and
espionage by the military and intelligence agencies, respectively, subject
to legislated definitions and decision process stipulations.184 Like the
Washington Administration-era Calling Forth Act that would form the
basis of the Insurrection Act, FISA requires permission from a federal
judge.185 Inevitably, the precise contours of the indefeasible presidential

180 See BARRON, supra note 164; Barron & Lederman, Lowest Ebb Part I, supra note 164;
Barron & Lederman, Lowest Ebb Part II, supra note 164.
181 Compare YOO, supra note 174 (presidentialist claims), with Barron & Lederman,
Lowest Ebb Part I, supra note 164, & Barron & Lederman, Lowest Ebb Part II, supra note
164 (shared power and expansive congressional powers evident in constitutional record).
182 See Barron & Lederman, Lowest Ebb Part I, supra note 164, at 785-86. For other
scholarly studies that contest the presidentialist theory and have been published since the end
of the Cold War, see Barron & Lederman, Lowest Ebb Part II, supra note 164; Lobel, supra
note 177; Prakash, supra note 174; Rudesill, The Land and Naval Forces Clause, supra note
154.
183 For discussion, see Rudesill, The Land and Naval Forces Clause, supra note 154, at
450-67.
184 See Covert Action Statute, 50 U.S.C. § 3093; Foreign Intelligence Surveillance Act
(FISA), 50 U.S.C. 1801 et seq.
185 See Foreign Intelligence Surveillance Act, 50 U.S.C. §§ 1802-04 (1978); Calling
Forth Act, 1 Stat. 271 (1792). Statutorily-mandated judicial review of this use of force was
not controversial in Congress. See DAVID P. CURRIE, THE CONSTITUTION IN CONGRESS: THE
FEDERALIST PERIOD 1789-1801 253-56 (1997). The judicial role was dropped when the 1792
statute was amended. See 1 Stat. 424 (1795); The Insurrection Act, 2 Stat. 443 (1807). For

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

power in Youngstown’s Category Three that Congress cannot condition are

somewhat blurry. But the weight of the constitutional record shows that
the President’s “lowest ebb” is “lower and the blur is smaller than claimed
by presidentialists.”186

C. Firmer Footing for Nuclear Rule Writing

A deeper scholarly understanding of the full sweep of the originalist

evidence and constitutional history supporting congressional power to
legislate limits on use of the military instrument is only one development
since Congress last voted on nuclear command and control that provides
firmer footing now for Congress to act. The case for congressional rule
writing regarding nuclear weapons is also stronger because of change in
the international security environment, post-9/11 wartime Supreme Court
decisions on separation of powers, and entrenchment of framework
process statutes that provide useful legislative precedents for making
nuclear weapons statutorily special.

1. Change in the International Security Environment

As emphasized above and by Professor Peter Raven-Hanson at the

Cold War’s end, the factual realities of transportation methods,
communications infrastructure, and military forces informed the work of
the Framers.187 Tens of thousands of U.S. and Soviet nuclear warheads on
thousands of supersonic missiles pointed at each other, and growing fear
by the 1970s and 1980s that the Soviets might attempt a first strike,
constitutionally changed matters.188 Further factual change now allows for
readjustment regarding nuclear weapons of what Justice Jackson termed
the constitutional equilibrium.189 As described in Part II.A, the most likely
U.S. nuclear use scenario now probably looks in terms of strategic warning
and complexity like a sensitive counter-terrorism operation. There is
therefore more temporal room that amounts to more constitutional room
for Congress to assert its Article I powers and responsibility to manage the

discussion, see WILLIAM C. BANKS & STEPHEN DYCUS, SOLDIERS ON THE HOME FRONT 43-
45 (2016) (detailing early congressional action surrounding the Calling Forth Act).
186 See Rudesill, The Land and Naval Forces Clause, supra note 154, at 431.
187 See Raven-Hansen, supra note 15.
188 See AMBINDER, supra note 25, at 24 (concern Soviet Union would attempt a first
strike).
189 See Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 638 (1952) (Jackson, J.,
concurring).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 51

military instrument of power. Congress can write rules to mitigate the

Rogue and Precipitous President risks without as much worry that
legitimate exercise of the repel-attacks Commander in Chief power will be
meaningfully compromised. North Korea’s acquisition of the bomb and
global-range missiles and growing concern about conventional conflict
with Russia or China that could go nuclear further underscore the
importance of careful decision-making and guardrails against impulsive
nuclear trigger-pulling.

2. Post-9/11 Wartime Supreme Court Precedents

The George W. Bush Administration sought to expand the role of

nuclear weapons in U.S. strategy and repeatedly made some of the most
expansive assertions to date of executive power. 190 In landmark wartime
cases, the Supreme Court pushed the constitutional equilibrium back
toward the Youngstown vision of shared power, with implications for any
use of force.
In Hamdi v. Rumsfeld, the Court held in 2004 that a U.S. citizen
captured on a foreign battlefield and designated an enemy combatant by
the Commander in Chief could be detained and was entitled to due process
protections.191 The Court’s plurality construed the 1972 Non-Detention
Act’s bar on detention of a U.S. citizen absent an act of Congress to have
been satisfied implicitly by another statute, the post-9/11 Authorization for
the Use of Military Force (AUMF).192 A presidentialist view was
articulated by Justice Thomas in dissent, equating detention with other
“central functions of warmaking.”193 Hamdi stands for rejection of such
claims in wartime: the plurality emphasized the Youngstown vision of
shared power and the centrality of statutes, while a dissent by Justices
Scalia and Stevens maintained that the Treason Clause and its
prescriptions controlled instead.194 Note as well that the plurality was only
able to argue the Non-Detention Act’s satisfaction by implication because

190 See GUTHE, supra note 67, at 1-2 (nuclear weapons); PETER M. SHANE, MADISON’S
NIGHTMARE: HOW EXECUTIVE POWER THREATENS AMERICAN DEMOCRACY (2009) (analyzing
Bush Administration executive power claims)
191 See Hamdi v. Rumsfeld, 542 U.S. 507, 509 (2004).
192 See Non-Detention Act, 18 U.S.C. § 4001(a) (1948); Authorization for the Use of
Military Force, Pub. L. No. 107-40, 115 Stat. 224 (2001).
193 See Hamdi, 542 U.S. at 579, 596 (Thomas, J. dissenting). Thomas also agreed with the
plurality on detention authority on statutory grounds.
194 See Hamdi, 542 U.S. at 508 (O’Connor, J., plurality op.); Hamdi, 542 U.S. at 554
(Scalia & Stevens, JJ., dissenting).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

that statute did not include a requirement for explicit reference in later
congressional enactment or some other clear statement rule.
Hamdi informed the Court’s Hamdan decision two years later.195 In
this decision about a foreign national captured on the battlefield, the Court
upheld the Uniform Code of Military Justice statute over the President’s
order regarding detention of the enemy.196 The majority in Hamdan
grappled with the Executive Branch’s statutory argument for authority in
the 2001 AUMF, and recited the familiar constitutional principles that the
Commander in Chief has authority to repel attacks and has command the
armed forces without legislative direction. In ruling against the
Commander in Chief, the Court not merely distinguished the 2001 AUMF
statute but also cited by clause Congress’s legislative authorities over the
military. The majority again emphasized Youngstown, and wrote that the
President may not disregard congressional enactments. 197
Any suggestion that Hamdi and Hamdan were merely about detainees
founders against the plain doctrinal power of these cases in legal and
historical context. They continue a line of landmark decisions dating to the
Founding Era in which the Supreme Court has used cases about seizures
during war – of ships, people, and industry – to define separation of powers
doctrine.198 In these cases the Court has never sided with the President over
Congress, instead making clear that the Commander in Chief is subject to
statute even in times of war.199
The doctrinal reverberations of the post-9/11 cases were felt at the
Justice Department’s Office of Legal Counsel (OLC), the opinions of
which are precedent for the Executive Branch and tend to be friendly to
the President. In sweeping early 2000s presidentialist opinions, OLC had
gone as far as to assert that “Congress can no more interfere with the

195 See Hamdan v. Rumsfeld, 548 U.S. 557, 591-93 & n.23 (2006).
196 See Uniform Code of Military Justice, ch. 169, 64 Stat. 109 (1950), codified as
amended at 10 U.S.C. §§ 801- 946 (2016).
197 See Hamdan, 548 U.S. at 591-93 & n.23. See also Stephen I. Vladeck, Congress, the
Commander-in-Chief, and the Separation of Powers after Hamdan, 16 TRANSNAT’L L. &
CONTEMP. PROBS. 933, 960-61 (2007) (after Hamdan greater attention must be paid to
congressional powers).
198 See supra note 122 (Little, Prize Cases, Youngstown, Civil War, and Korean War); Ex
parte Milligan, 71 U.S. 2, 4 (1866) (captured alleged Confederate agent cannot be tried by
military commission if civil courts are open and operating). For an influential discussion of
Little, see Michael J. Glennon, Two Views of Presidential Foreign Affairs Power: Little v.
Barreme or Curtiss-Wright?, 13 YALE J. INT’L L. 4 (1988).
199 See supra note 122. Little and Youngstown, like Hamdan, held presidential orders
invalid in face of statute. The Prize Cases upheld presidential action where Congress had not
acted (under Youngstown doctrine, Category Two).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 53

President’s conduct of the interrogation of enemy combatants than it can

dictate strategic or tactical decisions on the battlefield.” 200 Post-Hamdi in
2004, and then post-Hamdan and before expiry of the George W. Bush
administration, in extraordinary moves OLC withdrew, modified, or
replaced a series of post-9/11 opinions on matters including interrogation,
detainees, surveillance, and use of force, and acknowledged congressional
authorities. OLC acknowledged that the President was not just subject to
statute regarding detention but interrogation as well. 201 Having essentially
argued that presidential orders about detainees (and other matters) and
battlefield maneuvers were constitutionally equivalent, OLC’s course
correction inevitably signaled that claims about the Commander in Chief
being beyond congressional regulation were suspect more generally.
To be sure, ambiguity remained about the precise contours of
congressional and presidential powers in Youngstown’s Category Three.
Clearly, however, Hamdi and Hamdan buttressed Congress’s
constitutional standing regarding war powers. It left the inter-branch
equilibrium in a sensible place: despite an attack on the homeland, an
atmosphere of fear (including of nuclear terrorism), and two ongoing wars,

200 See Memorandum from Jay S. Bybee, Assistant Att’y General, Office of Legal
Counsel, U.S. Dep’t of Justice, to Alberto R. Gonzales, Counsel to the President, Re:
Standards of Conduct for Interrogation Under 18 U.S.C. §§ 2340–2340A 39 (Aug. 1, 2002),
https://perma.cc/KL35-M9U5, superseded by, Definition of Torture Under 18 U.S.C. §§
2340–2340A, 28 Op. O.L.C. 297 (Dec. 30, 2004) [hereinafter Levin Memorandum].
201 See, e.g., Levin Memorandum, supra note 200 (withdrawing and replacing 2002
interrogation memorandum, one of several withdrawn or replaced); Memorandum from
Steven G. Bradbury, Principal Deputy Assistant Attorney General, Office of Legal Counsel,
U.S. Dep’t of Justice, Re: Memorandum for the Files, Status of Certain OLC Opinions Issued
in the Aftermath of the Terrorist Attacks of September 11, 2001, at 5 (Jan. 15, 2009),
https://perma.cc/447Q-NKFF (withdrawing or modifying prior memoranda, and admitting
congressional authority in several Article I clauses – the Land and Naval Forces Clause,
Captures Clause, and Define and Punish Clause – to write statutes binding on the President
regarding detainees). In addition to Supreme Court decisions, other factors were operative in
OLC’s remarkable doctrinal course corrections. One was new leadership at OLC that
recognized the overbroad and otherwise “deeply flawed” claims in post-9/11 OLC opinions.
See JACK GOLDSMITH, THE TERROR PRESIDENCY 152–60, 181 (2007) (account of lawyer who
took over OLC and administratively drove the first round of revisions). Another was President
Bush’s signature on legislation on detention of enemy combatants, interrogation (particularly
barring torture), and surveillance that restricted Executive power. See Detainee Treatment Act,
Pub. L. No. 109–148, 119 Stat. 2680 (2005) (after Hamdi governing detention of enemy
combatants); Id., §§ 1002, 1003, 119 Stat. 2680, 2739-44 (in the wake of revelations of abusive
interrogations, prohibiting “cruel, inhuman, and degrading treatment or punishment” of any
“individual in the custody or under the physical control” of the U.S. government and limiting
Department of Defense interrogation techniques to those in the Army Field Manual); FISA
Amendments Act, Pub. L. No. 110-261, 122 Stat. 2436 (2008) (bringing surveillance activities
initiated after 9/11 based on Article II authority under statutory regulation).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

the Commander in Chief generally may not defy enactments of the elected
representatives of the people.202

3. Special Statutes: Covert Action and FISA

Hamdan upheld a framework national security statute over presidential

order. This section explores two other major frameworks – for covert
action and national security surveillance – enacted after Congress last
voted on nuclear command and control nearly 50 years ago and further
entrenched after the Cold War’s end 30 years ago. These statutes regulate
highly sensitive activities that the executive claims are vital to national
security and implicate the Commander in Chief power. 203 The statutes
provide guardrails to prevent abuse of executive power, ensure
accountability, and facilitate good process by proscriptively defining
activities and providing decision process steps. They make “secret wars”
and monitoring of foreign agents inside the United States statutory special.
In so doing, by analogy and thick practice precedent, the covert action
statute and FISA have strengthened the ground for Congress to make
nuclear use statutorily special, as well.
These statutory frameworks are familiar to national security lawyers.
They are novel, however, to what can be an insular nuclear command and
control conversation.

202 For example of public concerns during litigation of Hamdan that terrorists may obtain
and use a nuclear weapon, see GRAHAM T. ALLISON, NUCLEAR TERRORISM: THE ULTIMATE
PREVENTABLE CATASTROPHE (2004).
In another post-Cold War case, the Supreme Court in Zivitofsky ruled for the President
over Congress regarding recognition of a foreign state. This foreign relations case was not
about the military or force. The majority emphasized the Ambassadors Clause of Article II.
See Zivitofsky ex rel. Zivitofsky v. Kerry, 576 U.S. 1, 2 (2015). The Court also dealt a
devastating blow to the gloss on presidential power generally provided by United States v.
Curtiss-Wright Exp. Corp. 299 U.S. 304 (1936). For nearly eight decades this simple
legislative delegation case’s unnecessary and over-read dicta provided Executive lawyers with
a talismanic citation in support of expansive presidentialist claims in any foreign or national
security context. See KOH, supra note 170, at 94 (terming phenomenon the “Curtiss-Wright,
so I’m right” cite); see also Marty Lederman, Thoughts on Zivitofksy, Part Seven: “Curtiss-
Wright-Out of Sight,” and the Fate of the Argument for an Exclusive Executive Diplomatic
Authority,” JUST SEC. (Jun. 14, 2015), https://perma.cc/N48H-3Z8R (Court undermined
Curtiss-Wright’s doctrinal power).
203 See, e.g., ALBERTO GONZALES, LEGAL AUTHORITIES SUPPORTING THE ACTIVITIES OF
THE NATIONAL SECURITY AGENCY DESCRIBED BY THE PRESIDENT 2 (Jan. 19, 2006) (Justice
Department white paper on warrantless surveillance in defiance of FISA); Intelligence
Authorization Act, Fiscal Year 1991, H.R. REP. NO. 102-166, at 27-28 (1991) (Conf. Rep.)
(discussion of Executive claims about constitutionality of covert action reporting provisions).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 55

The covert action statute has been assembled through a series of

enactments, most notably the 1974 Hughes-Ryan amendment and the 1991
Intelligence Authorization Act.204 Absent the President’s personal
authorization and reporting of a “finding” to Congress certifying that
criteria have been met, the statute as amended bans clandestine operations
to influence conditions abroad where the role of the United States will not
be acknowledged.205 The statute flatly bars covert actions targeting
Americans. The statutory definition extends beyond lethal operations, but
covert action has correctly been described as involving “secret wars” – or
at least quasi-deniable clandestine aspects of overt wars and foreign
policies. Open sources indicate that the bin Laden raid, CIA drone strikes,
CIA paramilitary operations, cyber attack on Iran’s nuclear program, and
assistance to the 1980s Afghan Mujaheddin fighters and other foreign
fighters have been conducted pursuant to the covert action statute. 206 The
modern statute requires the Executive Branch to explain any operation’s
legality.207
FISA bars another critical national security activity – surveillance of
the enemy and other foreign agents inside the country – subject to its
definitional restrictions and process stipulations.208 “Classic FISA,” the
original 1978 statute as amended, requires that before the Executive
Branch can surveil for foreign intelligence purposes U.S. persons
suspected of being agents of foreign powers, the Attorney General (or
lower level officials) must make a series of showings and convince a
federal judge to authorize electronic intelligence collection. 209 Section 702
of the modernized FISA, the FISA Amendments Act of 2008 as amended,
requires that the Attorney General and Director of National Intelligence
(DNI) make annual certifications to the FISA court before the U.S.
government can collect inside the United States data that is associated with

204 See 50 U.S.C. § 3093; Pub. L. 93-558, 88 Stat. 1795 (1974); Pub. L. No. 102-88, 105
Stat. 429 (1991). For discussion generally, see BAKER, supra note 40, at 148-58.
205 These criteria include negative definitions: exclusion of traditional diplomatic,
military, and law enforcement activities, mere collection of intelligence, and related support
activities. See 50 U.S.C. § 3093.
206 For discussion of the bin Laden raid as a covert action, for example, see SAVAGE,
supra note 77, at 257–71.
207 See 50 U.S.C. § 3093(b)(2).
208 See 50 U.S.C. § 1801 et seq. FISA is an example of a super-statute: an enactment that
powerfully shapes normative expectations in the public mind, beyond the black and white of
the U.S. Code. See William N. Eskridge, Jr., & John Ferejohn, Super-Statutes, 50 DUKE L.J.
1215 (2001).
209 See 50 U.S.C. § 1801, 1804.

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

non-U.S. persons located abroad.210 The statute includes many other

reporting requirements, as well.211
These statutes have a number of important things in common.
First, both statutes limit what the Commander in Chief can order the
armed forces and civilian intelligence officers to do. Both military and
civilian personnel have acted subject to the covert action statute.
Surveillance governed by FISA is carried out by both military and civilian
personnel as well, and most notably by the National Security Agency
(NSA) – a component of the Defense Department headed by a four-star
military officer. Both statutes bind the national security apparatus, and the
Commander in Chief. As I have argued, these statutes rest on several
constitutional authorities, including Congress’s power to enact “Rules for
the Government and Regulation of the land and naval Forces.” 212
Second, Congress enacted the covert action and surveillance
frameworks after revelations that unfettered Executive discretion had been
abused. The long decades when covert action had been left to executive
discretion were characterized by secret assassination attempts against
leaders of governments with which the United States was at peace,
clandestine efforts to influence U.S. politics and other conditions inside
the country, and other abuse of authority.213 Passage of FISA was
prompted in part by use of “national security” as a rationale for
wiretapping of civil rights leaders, dissidents, feminists, and students by
executive fiat, and suspicionless bulk collection of the Fourth
Amendment-protected communications of millions of Americans. 214
Third, both statutes were enacted over claims that Congress lacked the
authority to control the Commander in Chief, that Congress through
acquiescence had agreed, and that Congress could not after so long now
intrude on the Commander in Chief power and change the constitutional
equilibrium. Decades of thick practice have since transformed
unprecedented intrusions on Executive authority into centerpieces of the
national security legal regime.
Fourth, while these statutory frameworks are distinct in their details,
they both define terms in ways that limit presidential latitude, identify

210 See 50 U.S.C. § 1881(a).

211 See, e.g., 50 U.S.C. § 1802(a)(1)(C)-(2) (reporting to Congress regarding surveillance of
communications between or among foreign powers).

212 See Rudesill, The Land and Naval Forces Clause, supra note 154, at 465-73.
213 See S. Select Comm. to Study Governmental Operations with Respect to Intelligence
Activities, S. REP. NO. 94-755 (1976) (Church Committee Report).
214 See id.

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 57

responsible Executive Branch actors, and structure the decision process. 215
Through reporting requirements the statutes set expectations and foster
“good process” norms. They gather information for Congress, facilitating
oversight and policy formation by the Legislative Branch. The statutes also
manage secrecy – including legal secrecy – through reporting and
transparency provisions.
Fifth, both process statutes run over the top of underlying legal
authority to use force. This is reflected in the covert action statute’s
requirement for information on the legal basis for covert operations, in the
operation of both statutes during times of war or peace, and in specific
statutory language making FISA the exclusive authority for national
security surveillance.216
***
A few qualifications to this section’s analysis are in order. First, the
possibility of use of just one or a handful of nuclear weapons, for example
against a bin Laden-like target in a bunker or in a conventional conflict
with a nuclear state that escalates, has long existed. The need to deter large
nuclear attack by Russia and China also endures. Second, directing lethal
force is perhaps closer to the core of the Commander in Chief power than
capturing the enemy, authorizing “black ops,” or collecting intelligence.
Note too that the Supreme Court has not made clear the extent of
presidential authority regarding covert action and national security
surveillance that has survived enactment of the special statutes governing
those activities.217 Furthermore, deference to the President in national
security remains formidable, and successive administrations have used
force and made claims that arguably have reduced the salience of the

215 The covert action statute and FISA are analogous in some respects to decision process
statutes in the domestic realm. One example is the National Environmental Policy Act
(NEPA), 42 U.S.C. § 4321 et seq. NEPA governs the decision process used by federal agencies
when the government is contemplating action with potential environmental impact.
216 See 50 U.S.C. § 3093(b)(2) (2019) (requirement of statement of legal basis for covert
action); 50 U.S.C. §§ 1811-1812 (1811: FISA operation during war, 1812: FISA exclusive
authority). In contrast to FISA’s explicit language, the covert action statute by lacking any
mention of war applies at all times.
217 Critics of FISA’s constitutionality often cite a decision of FISA’s appellate court, the
Foreign Intelligence Surveillance Court of Review, which wrote in one of its only published
opinions that “We take for granted that the President does have [authority to conduct
warrantless searches for foreign intelligence] and, assuming that is so, FISA could not
encroach on the President’s constitutional power.” See In re Sealed Case, 310 F.3d 717, 742
(FISA Ct. Rev. 2002). The secretive court’s assumption remains untested at the Supreme
Court nearly two full decades later, as robust three-branch FISA practice continues into its
fifth decade. See also United States v. U.S. District Court (Keith), 407 U.S. 297, 298 (1972)
(warrant requirement for domestic security surveillance).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

framework statute that governs force generally, the WPR.218 Congress has
largely acquiesced.219
These are legitimate caveats, but important considerations lie against
the extent to which they argue against this article’s contention that the
constitutional footing has grown stronger for Congress to write statutory
rules for nuclear use. First, the range of nuclear employment scenarios has
not changed but what has is that the most likely scenario now involves
more deliberation time and less catastrophic stakes than the Cold War
nightmare upon which relegation of nuclear launch authority to the
President is predicated. Second, decades of statutory regulation of covert
action and surveillance activities once entirely left to the Executive – plus
extensive practice “gloss” – have like Hamdi and Hamdan inevitably
weakened presidentialist claims since the Cold War’s end. And arguments
based in congressional acquiescence to Executive claims regarding the
WPR or other matters only go so far. As Professor Baker has written, “that
Congress had not previously chosen to exercise [its] authorities did not
mean it did not possess the authority to do so, only that it had not found it
necessary and proper to do so….”220
Congress retains powerful authorities to control use of force.
International security, doctrinal, and statutory developments since the
Cold War’s end are reasonably and best read to provide firmer footing for
Congress to exercise those powers. Congress can reasonably write rules
for the most likely and most worrisome nuclear use scenarios, where
profound concern about impulsive presidential action endures and where
there is time for and much to be gained from Executive Branch “good
process.” The authorities Congress has exercised to write the covert action
law and other special statutes have generated now-longstanding
frameworks that, while not off-the-shelf perfect models for nuclear use,
provide tools that ought to be reconfigured for governing nuclear weapons.

218 For discussion, see Rudesill, The Land and Naval Forces Clause, supra note 154, at
465-66 nn. 318-19.
219 One partial exception was Congress’s legislation pursuant to the War Power Resolution
to withdraw U.S. military support to the Saudi air war in Yemen. President Trump vetoed it
and Congress did not over-ride. See Allie Malloy, Trump Vetoes Yemen War Powers
Resolution, His 2nd Veto Since Taking Office, CNN (April 17, 2019, 2:44 PM),
https://perma.cc/X8HT-SS42.
220 See BAKER, supra note 40, at 79; accord Carter, supra note 140, at 118-19
(conventional and nuclear weapons similar under Constitution, and Congress can act).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 59

V. MAKING NUCLEAR WEAPONS STATUTORILY SPECIAL

In a new essay, Professor Waxman suggests borrowing elements from

the covert action regime to govern overt uses of force generally.221 That is
the work of this Part, focused on nuclear weapons. This Part looks to other
statutory frameworks, as well: FISA, the Goldwater-Nichols Defense
Reform Act, and the WPR.222
To address the Rogue and Precipitous President risks, to manage
nuclear-conventional convergence, to ensure that the benefits of Executive
Branch inter-agency process and appraisal of options are captured where
time permits, and to build on international security, doctrinal, and statutory
developments that have strengthened Congress’s constitutional footing to
govern use of nuclear force, this Part proposes a process statute (see the
Appendix for a full draft), entitled the Nuclear Forces Control Act
(NFCA). Here, this article describes and analyzes the model statute and
addresses potential objections. This proposal’s objective is to prompt fresh
thinking about how Congress might end its three-quarter-century
acquiescence to all but complete presidential control over the nation’s
nuclear loaded weapons.

A. The Nuclear Forces Control Act (NFCA)

The NFCA balances constitutional equities and sets a new – or at least

somewhat clearer – equilibrium regarding nuclear command and control.
This proposal utilizes well-accepted congressional powers of restriction of
use of force and funds, definition of terms, structuring the national security
apparatus, creating criminal penalties for violation of the law, and
requiring reporting to Congress.

1. Purposes and Projects

221 See Matthew Waxman, Waging Covert War, Discussion Paper for Duke-Virginia
Foreign Relations Roundtable 2, 7 (Sept. 28, 2019) (unpublished manuscript on file with
author). Waxman observes that Congress is unlikely to enact a “blanket statutory framework
for overt warfare” akin to 50 U.S.C. § 3093, but suggests consideration of drawing “elements”
from the covert action regime to regulate overt warfare.
222 The idea of expanding the War Powers Resolution to include nuclear weapons is not
new; see, for example the congressional debate on the Fulbright Amendment, analyzed above.
Discussion of drawing elements from the covert action regime to govern overt use of force
has been less frequent. See id.

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

Like the WPR, the statute begins with the statute’s purposes. 223 Section
(a)(1) states that these are:
to exercise Congress’s constitutional authority to control use
of the forces it creates; to inform congressional oversight of
the international security environment and of the Executive
Branch; to prevent unnecessary or otherwise illegal use of
nuclear forces in any case and particularly where the United
States, its forces, or allies do not face temporally imminent
nuclear attack; to facilitate the most careful and thorough
decision-making process within the Executive Branch that is
practicable; and, to provide definitions and interpretive
guidance to facilitate compliance.
Unlike the WPR, the statute then in Section (a)(2) cites the full relevant
list of Congress’s authorities.224
The NFCA’s design reflects several meta-projects. First, through its
force limitations, decision-prompting reports, and other provisions, the
statute endeavors to clarify when nuclear weapons may be used and when
inter-agency process is in order. These rules, in turn, augur toward norms
and expectations that would help steel the resolve of presidential
subordinates to start pushing back on an impulsive President, ideally at an
earlier point than waiting for an order that is patently illegal. Second, the
draft statute makes no formal intrusion on the Commander in Chief power
beyond the NFCA’s well-grounded force limitations and suggestions for
use of NSC process. The statute does not direct forces, give Congress a
“second vote,” displace the President at the top of the chain of command,
require the President to get the assent of any subordinate for anything the
President can legally order, or give any other Executive Branch official
authority to order use nuclear weapons. Third, the NFCA leaves
undisturbed the nuclear command and control system’s flexibility and
responsiveness. Echoing emergency exceptions to reporting requirements
in the covert action statute and other laws, in a situation of extremis (here,
in the face of imminent or initiated adversary nuclear attack) the statute’s
reporting requirements could be met after executive action when time
permits.225

223 See War Powers Resolution, Pub. L. No. 93-148, 87 Stat. 555, § 2(a).
224 See War Powers Resolution, Pub. L. No. 93-148, 87 Stat. 555, § 2(b) (citing
Necessary and Proper Clause with unenumerated reference to Congress’s other powers).
225 See 50 U.S.C. § 3093 (2019) (exception of written finding and reporting requirements
allows the President to order covert action and later prepare written finding and report to
Congress); 50 U.S.C. § 1805 (2018) (Attorney General may authorize surveillance without
court order generally required by FISA in case of emergency); War Powers Resolution, Pub.

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 61

2. Restrictions on Force and Funding

Section (b) provides “Rules for Nuclear Forces,” absent compliance

with which forces “are not provided for operational employment.” This
language textually references Congress’s powers to “make Rules for the
Government and Regulation of the land and naval Forces” and to provide
such forces.226
The first rule is that legal authority to use force comes from some other
expression of Article I or II power. This is meant to frustrate argument that
this statute provides a standing authorization for use of nuclear weapons.
It also prompts analysis of whether a congressional enactment or
President’s Article II repel-attacks authority are operative in a particular
situation. Where time is available for Congress to deliberate, the rule
implicitly reserves the larger policy question of whether Congress should
authorize force – especially nuclear force – regarding a particular
adversary, one that is enormously consequential and ought to get deep
case-specific engagement. In short, this provision has the NFCA process
statute running over the top of some other authority for use of force, in the
same way that the covert action statute, FISA, and WPR operate and
prompt deliberation and reporting separate from legal authority for use of
force flowing from Article I or II of the Constitution.
Next, the statute in Section (b)(1)(B) codifies the nuclear declaratory
doctrine (policy) baseline of both the Obama and Trump Administrations:
nuclear weapons will only be employed in “extreme circumstances” to
defend the “vital interests” of the United States and its allies. 227 At present,
a Rogue or Precipitous President could waive this policy floor at will. The
statute also here stipulates that any use must be militarily necessary and
otherwise legal. This provision of the NFCA in these ways cements a
policy baseline absent congressional repeal, and underscores applicability
of a legal regime that includes the law of armed conflict. 228
The NFCA’s next criterion, Section (b)(1)(C), provides another part of
the statute’s proscriptive core. Subject to availability of some other
authority for use of force and to the declaratory doctrine just discussed,
Section (b)(1)(C) allows nuclear use in two circumstances only.

L. No. 93-148, 87 Stat. 555, § 4(a)(3) (President must report to Congress within 48 hours of
introduction of forces into hostilities).
226 See U.S. CONST. art. I, § 8, cl. 12-14.
227 See 2010 NPR, supra note 67, at ix; 2018 NPR, supra note 3, at ii.
228 See Part I supra regarding the law of armed conflict. It already applies to any nuclear
use, but underscoring its applicability has normative force.

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

One is where an adversary nuclear attack is imminent or initiated.

Here, the President’s repel-attacks Article II authority plainly would be
operative and “extreme circumstances” present. In order to tie the Article
II repel-attacks authority to its temporal rationale (action must be taken
before Congress can gather and deliberate), to foster careful inter-agency
review where there is time for it, and to make it harder for a Rogue or
Precipitous President to argue that mere possession of nuclear arms by a
potential adversary creates an imminent threat justifying a U.S. nuclear
attack, imminence is defined in the NFCA in specific terms. 229 Section
(d)(3) stipulates that imminence means that adversary nuclear attack is
“reasonably possible within the next 72 hours, based on assessment of
adversary capabilities and intentions. Adversary possession of the
technological capability of launching a nuclear attack, without other
significant indications of adversary intent, does not create imminence
under this Section.” If advisors and the President were discussing
reasonable questions as to the temporal and adversary capabilities and
intentions prongs, the President’s judgment would prevail. If subordinates
saw little to no factual predicate, on the other hand, then the NFCA
statutory provision would provide them new legal and normative grounds

229 This provision is meant to complicate for purposes of the NFCA the sweeping
understandings of imminence deployed by recent administrations. The George W. Bush
Administration argued that, after 9/11, the mere presence of some unknown number of Al
Qaeda members in Iraq, plus alleged Iraqi possession of weapons of mass destruction, created
an “urgent duty” for preventive war. See President George W. Bush, Remarks by the President
on Iraq (Oct. 7, 2002, 8:02 PM), https://perma.cc/FRS7-EBF4 (“urgent duty” because “Iraq
could decide on any given day to provide a biological or chemical weapon” to terrorists). The
Obama Administration argued that U.S. citizens alleged to be terrorist leaders pose a
“continued and imminent threat” to the extent they are “engaged in continual planning and
direction of attacks upon U.S. persons,” not because of their participation in any particular
plot nearing fruition, and therefore lethal force may be used against them. See Memorandum
from David J. Barron, Acting Assistant Att’y General, Office of Legal Counsel, U.S. Dep’t of
Justice, Memorandum for the Attorney General, Re: Applicability of Federal Criminal Laws
and the Constitution to Contemplated Lethal Operations Against Shaykh Anwar al-Aulaqi 40-
41 (July 16, 2010), https://perma.cc/2XZG-VCAW (redacted targeted killing memorandum).
The President might assert that her or his judgment about imminence and its meaning ought
to prevail. The NFCA thereby sets up a Youngstown Category Three confrontation rather than
one in more President-friendly Category Two. It is not clear which branch would win, but
there would be a strong argument for the nation’s legislature. Congress generally prevails in
Category Three (see Part IV.B) so long as it does not displace the Commander in Chief at the
head of the chain of command, which this draft provision does not. Judgment about national
security threats is not only a presidential responsibility, and on the contrary is inherently
reflected in Congress’s enactments. Congress can also define terms via statute, and thereby
limit government authorities. See 1 U.S.C. (first chapter of U.S. Code, providing definitions).
That is what the NFCA does here, in the course of writing rules for the forces Congress
provides.

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 63

for pushing back and ultimately refusing to comply – potentially starting a

series of firings and resignations that would slow an impulsive President
and signal to other legislative and Executive Branch actors (and hopefully
to the President, too) that something is horribly amiss.
The second circumstance in which the NFCA would not bar nuclear
use, found in Section (b)(1)(C)(ii), is where adversary attack is not
imminent and “the military, diplomatic, legal, intelligence, environmental,
and other implications of both nuclear and non-nuclear options have been
carefully evaluated, and non-nuclear weapons will not succeed in
defending the vital interests of the United States and its allies.” The next
provision, Section (b)(2), provides:

National Security Council and the Nuclear

Command, Control, and Communications System at
the National Leadership Level.
(A) National Security Council. Consistent with the
direction of the President, the National Security
Council established by 50 U.S.C. 3021 shall advise
the President in the evaluation of nuclear and non-
nuclear options, particularly as provided in
paragraph (b)(1)(C)(ii).
(B) The Vice President. The Vice President shall be
responsible for the coordination of National Security
Council interagency review of nuclear threats,
capabilities, and decisions.230
(C) Legal Advice. The nuclear command, control, and
communications system shall be configured to
facilitate the involvement of the Attorney General,
and other legal advisors as appropriate, to the extent
practicable.

These (b)(1)(C)(ii) and (b)(2) provisions, taken together, provide

additional criteria for executive decision where adversary nuclear attack is
not imminent. They avoid disruption of the chain of command. They send
clear signals from Congress that every effort ought to be made to evaluate
nuclear use decisions through the NSC process and integrate the Attorney
General or other legal advisors into threat conferences, NSC deliberations,
and other national leadership-level nuclear deliberations. While still

230 The most recent statutory designations of coordinators and other officials for the NSC
are the special advisor on international religious freedom and the coordinator for combating
foreign influence operations and campaigns. See 50 U.S.C. § 3021(g).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

allowing the President flexibility to design his or her decision process

(“Consistent with the direction of the President,” “to the extent
practicable”),231 the expectations generated by these provisions would
likely buttress the willingness of officials to question rogue or precipitous
presidential behavior and press for careful review via the NSC where time
plainly permits.
The last parts of the NFCA’s proscriptive core are provisions that wrap
the statute in a funding limitation (Section (b)(6)) and – borrowing FISA’s
criminal penalty – create criminal liability for its violation (Section (c)). 232
Congress’s powers of the purse and criminalization (particularly regarding
the armed forces) are among Congress’s strongest and well understood
within the national security apparatus. These provisions therefore could
have resolve-strengthening normative effects regarding the NFCA, over
and above the baseline responsibility of all personnel to refuse orders that
violate the law of armed conflict. This part of the NFCA may help
presidential subordinates conclude that the better part of honor is to resist
and if necessary resign or be relieved, in the process sending a message to
their colleagues, replacements, and others in positions of power – and
perhaps to a mentally clouded President too – that things are not right with
the President and “the bomb.”

3. Reporting Requirements and Norm Building

Congress could reasonably stop at this point. Or, Congress might

contemplate stronger provisions (such as outlined in Part V.B below) that
could present stronger impediments to a Rogue or Precipitous President
but simultaneously be more vulnerable on constitutional grounds. Another
path, reflected in the NFCA, would be additional expectation-creation and
norm-building, using Congress’s power to require Executive Branch
officials to report to the Legislative Branch that structures, authorizes,
limits, funds, and oversees its activities.
The set of reporting requirements in the NFCA is crafted to do several
things: help subordinates recognize when the President is in violation of
the NFCA’s (b)(1) imminent threat and careful evaluation requirements
for use of nuclear weapons; underscore that the President’s access to “the

231 This provision’s references to presidential discretion echo the existing NSC statute.
See 50 U.S.C. § 3021(b).
232 See 50 U.S.C. § 1809 (FISA criminal penalty). See also 18 U.S.C. § 1385 (posse
comitatus criminal penalty). To underscore that a presidential pardon would not fully relieve
subordinates of liability for disregarding the statute, Congress could rework FISA’s civil
liability provision for the nuclear context. See 50 U.S.C. § 1810 (FISA civil penalty).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 65

button” is subject the NFCA’s statutory framework; and further build

expectations around careful evaluation and NSC inter-agency good
process where adversary nuclear attack is not imminent.233
Simultaneously, reporting requirements would facilitate congressional
oversight of threats to the nation and of Executive Branch activities,
including potential presidential incapacity. 234
Section (b)(3) of the NFCA suggests four main provisions that serve
these objectives. These reporting provisions are tailored to the nuclear
context and well precedented in the national security legal regime. They
would not slow the nuclear command and control system in a crisis. Nor
would they intrude on the Commander in Chief’s position at the head of
the chain of command, or ability to act lawfully without the permission of
subordinates.
First, Section (b)(3)(A) of the NFCA requires specific senior officials
at the apex of the military and intelligence apparatus – the Secretary of
Defense, Director of National Intelligence, Chairman of the Joint Chiefs
of Staff, and Commander of the U.S. Strategic Command – to notify
Congress “as soon as possible” when in their judgment the nation, its
forces, or allies face temporally imminent nuclear attack. Such notice
would have obvious value to Congress, which conducts ongoing oversight
regarding threats.235 This NFCA provision is a tailored version of standing
provisions, enacted as part of framework statutes, that call for Executive
officials to provide risk assessments to Congress and exercise independent
judgment.236 The provision does not create a “second key” at the national

233 Executive Branch officials commonly complain there are too many reporting
requirements. Despite their numerousness, they are under-studied by scholars.
234 Statutorily-imposed reporting requirements gather information for Congress and
therefore reflect and enable Congress’s oversight powers. They also assign decisions to
particular actors, facilitating Congress’s management of the administrative apparatus. They
are abundant, including in annual legislation and standing national security framework
statutes. For discussion of work done by one of the annual national security acts, see e.g.,
Rudesill, The Land and Naval Forces Clause, supra note 154, at 452-53 (annual National
Defense Authorization Act finds constitutional footing inter alia in the Land and Naval Forces
Clause and structures the military portion of the national security apparatus).
235 Threats drive legislative changes to the organization, funding, and legal authorities of
the national security apparatus. Congress holds a slate of “worldwide threat assessment”
hearings. See, e.g., Cyber Threats are Biggest Concern for ODNI in Worldwide Threat
Assessment Report, HOMELAND SEC. TODAY (Mar. 12, 2018), https://perma.cc/C3PM-VF8B
(testimony to Senate Armed Services Committee).
236 See Goldwater-Nichols Defense Reform Act, Pub. L. 99-433, § 201, 100 Stat. 1005
(1986), codified as amended at 10 U.S.C. § 151(f) (with notice to Secretary of Defense, Joint
Chiefs of Staff may provide their own views to the President and Congress on any matter
concerning the Department of Defense), 153(b)(2) (Chairman of the Joint Chiefs of Staff shall
provide risk assessment to Congress regarding national military strategy, transmitted through

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

leadership level, of the kind present in missile launch control centers. 237 It
is not an ex ante or otherwise formal requirement for a subordinate to
assent to a president’s nuclear launch order. Instead, the threat notification
obligation engages the “human factor:” the integrity and conscience of
senior officials likely to be part of a nuclear threat conference (three of
whom who could transmit a nuclear launch order).238 Knowing that they
have obligations both to follow the law and to respect the Commander in
Chief’s judgments, if they were not in good conscience prepared to report
to Congress that attack is imminent, and saw no reasonable basis on which
to defer to the President’s claims of an imminent threat, an official of
integrity would infer that the President was on an ill-considered and very
well illegal path. The requirement to inform Congress of a nuclear threat
would in this way serve as a prompt for individual reflection, rather than a
formal process step added to the nuclear command and control system. In
anticipation of a Rogue or Precipitous President who relieves or evades the
usual threat conference participants, the draft statute could also impose the
imminent threat-reporting requirement on anyone in the chain of command
contacted directly by the President with a launch order. 239

the Secretary of Defense, who may add comments), 153(c) (Chairman shall submit directly to
Congress a report on the needs of the combatant commands, including the Chairman’s views
on whether the President’s budget request is deficient); DANIEL MAURER, CRISIS, AGENCY,
AND LAW IN U.S. CIVIL-MILITARY RELATIONS 43-45 (2017) (independence of the Chairman
under Goldwater-Nichols). See also Intelligence Reform and Terrorism Prevention Act
(IRTPA), Pub. L. 108-458, § 102A, 118 Stat. 3644 (2004), (codified as amended at 50 U.S.C.
§ 3024(a)(2) (2020)) (Director of National Intelligence as head of the intelligence community
shall ensure that intelligence provided to Executive and Legislative Branches is “objective
[and] independent of political considerations”).
237 See Woolf, supra note 29; Blair, Strengthening Checks on Presidential Nuclear
Launch Authority, supra note 6; Lin, A Two-Person Rule for Ordering the Use of Nuclear
Weapons, Even for POTUS?, supra note 30.
238 See 10 U.S.C. § 162(b) (military chain of command runs from President to Secretary of
Defense to commanders of joint combatant commands); id. § 163(a) (orders may be passed
through the Chairman of the Joint Chiefs of Staff). The Director of National Intelligence is
not in the military chain of command but is a logical reporter to Congress on threats as head
of the intelligence community and director of the National Intelligence Program. See 50 U.S.C.
§ 3023(b).
239 There is no claim in the public record that the President has the ability to contact
military subordinates so low on the chain of command (such as a submarine crew) that they
could not be expected to have access to information about an adversary nuclear threat. Instead,
the kind of lower level officers the President inferentially is most likely to be able to contact
include the acting deputy to an unavailable – or the successor to a relieved – Secretary of
Defense, Chairman of the Joint Chiefs of Staff, or Commander of the U.S. Strategic
Command, or else officers (perhaps junior generals or admirals, or field-grade officers) at the
Pentagon’s National Military Command Center (NMCC). If the usual threat conference
officials where incapacitated or out of communication due to adversary nuclear attack, the all
but certainly obvious fact of an attack to anyone in the nuclear chain of command would be

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 67

Next, with modifications the NFCA extends the covert action statute’s
reporting provisions to the nuclear context. As discussed above, Title 50
of the U.S. Code bars covert action to influence conditions in the United
States, and bars them abroad (including use of force) absent a presidential
written finding and report to Congress – ideally in advance, and in extremis
as soon as possible thereafter.240 The work being done here is normative
(underscoring that the President exercises Article II authority within a
statutory framework and is accountable) and informative (notifying
Congress that the Executive Branch is up to extremely sensitive things).
In the draft NFCA, the covert action mechanism is reworked and
extended to multiple actors, for several purposes.
Section (b)(3)(B) calls for a special role for the Vice President, a
standing NSC member made coordinator for nuclear matters by paragraph
(b)(2)(B) of the NFCA: to inform the congressional leadership when the
NSC has begun to deliberate about nuclear use. 241 Congress has a strong
interest in knowing of such deliberations, especially where there is no
imminent threat already driving reports under (b)(3)(A) or public concern.
Communication through a single person would facilitate protection of
sensitive information (subject to declassification procedures addressed
later in the NFCA). Importantly, the special nuclear responsibilities of the
Vice President where there is time to deliberate would foster an
expectation of involvement in an historic decision – investing the Vice
President in the NFCA’s process norm-building.242 Additionally, the
President would have to know that if the Vice President’s role under the
NFCA were bypassed when time was available for NSC deliberation that
the President in the process would be giving the Vice President
information the latter (depending on the circumstances) might use to argue
for the President’s removal – and the Vice President’s succession to the

enough for the officer to be confident that the legal authority (plainly the President’s repel-
attacks authority would be operative), “extreme circumstances,” and imminent threat prongs
of the NFCA had been satisfied. The official could then report the threat to Congress “as soon
as possible,” assuming the official and the Congress survived.
240 See 50 U.S.C. § 3093. The statute includes a series of options for the form of the report
and how widely it is shared within Congress, in view of the sensitivity of covert operations.
Regarding the legislative framework, see BAKER, supra note 40, at 148-58; W. MICHAEL
REISMAN & JAMES E. BAKER, REGULATING COVERT ACTION 116-22 (1992); See also Robert
S. Chesney, Military-Intelligence Convergence and the Law of the Title 10/Title 50 Debate, 5
J. NAT’L SEC. L. & POL’Y 539 (2012) (convergence between operations and authorities for
military and intelligence activities, with particular reference to modern covert action statute).
241 I thank Ned Foley for the suggestion to think about a special role for the Vice
President.
242 The Vice President is a two branch actor, serving also as President of the Senate. See
U.S. CONST., art. I, § 3, cl. 4.

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

presidency. As Madison advised, ambition can be made to counter

ambition.243
The NFCA’s next provision, in paragraph (b)(3)(C) is the most similar
to the covert action statute. This provision requires a presidential finding
that the NFCA’s stipulations have been observed, and reporting of that
finding to Congress. Briefing on this part of the NFCA during every usual
nuclear command and control orientation would serve to remind the
Commander in Chief that Congress has regulated use of “the bomb.”
The three layers of reporting just outlined may well be enough to serve
the NFCA’s expectation creation aims. Also, as the number of people
communicating classified information goes up so too does the risk of a
leak. If Congress still wanted more expectation-building around NSC
“good process” where adversary attack is not imminent, however,
Congress could enact additional reporting requirements for cabinet
officers.244 Section (b)(3)(D) of the NFCA calls for the Secretaries of State
and Defense and the Attorney General to report to Congress
contemporaneously with the President’s report – meaning, separately and
roughly the same time. These cabinet officers would be on the hook to
explain how the nuclear strike makes sense in view of their equities
(foreign policy interests, military necessity, the rule of law) and complies
with the NFCA. As with the other notifications and finding, this reporting
obligation would not formally impede presidential action. But it would
provide an additional reminder that nuclear use is regulated by statute and
another process prompt for presidential subordinates to reflect on the
President’s compliance with the law.245

4. Protecting the Statute

Finally, Section (e) of the NFCA is written to complicate interpretive

tactics sometimes used by the Executive to avoid statutory limitations.246

243 See THE FEDERALIST NO. 51 (1788) (James Madison) (“Ambition must be made to
counteract ambition” and government must be structured “to control itself.”). Of course, the
Vice President may be a pliant personality who always defers to the President. Process rules,
expectations, and norms can only do so much.
244 These cabinet members in a multitude of contexts are already subject to an array of
reporting requirements.
245 The NFCA in Section (b)(4) additionally calls for ongoing consultation with and
reporting to Congress, in language informed by § 3 of the War Powers Resolution. See War
Powers Resolution, Pub. L. No. 93-148, 87 Stat. 555.
246 Such provisions do not always generate compliance. But the statute is stronger with
them. See Jonathan F. Mitchell, Legislating Clear Statement Regimes in National Security,
43 GA. L. REV. 1059, 1098–1100 (2009) (national security statutes requiring a clear

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 69

The NFCA in Section (e) borrows FISA’s exclusivity provision to make

clear that it controls regarding the nuclear decision process. 247 It requires
that the statute can only be amended or repealed by explicit reference. 248
It includes a severability provision.249 The draft NFCA also provides
interpretive guidance meant to frustrate creative reconstruction of the
statute in secret or via use of minority legal interpretive theories. 250

statement do not always receive Executive compliance, and must be buttressed by

procedural checks); Trevor Morrison, Constitutional Avoidance in the Executive Branch,
106 COLUM. L. REV. 1189, 1213-14 (2006) (courts will apply clear statement rules to protect
under-enforced constitutional values).
The NFCA § (b)(7) also prevents the President from appointing whomever she chooses
(for example, a pliant, unqualified, or corrupted person) to perform functions under this
statute or in the nuclear chain of command, in the event that the specified official is
unavailable, has resigned, or been relieved. Under the NFCA, their role would be assumed
by their organization’s next-in-command.
247 See 50 U.S.C. § 1812(a). (Precedence is perhaps a better term than exclusivity
because statutes operate in the context of the Constitution and administrative directives, but
for legal consistency the NFCA uses FISA’s term of art). President George W. Bush ordered
surveillance outside of FISA after 9/11 in violation of a provision in Title 18 of the U.S.
Code that made FISA the exclusive authority for foreign intelligence surveillance. See 18
U.S.C. § 2511(2)(f); SAVAGE, supra note 77, at 183–85. The move was heavily criticized.
See, e.g., Letter from Constitutional Law Scholars and Former Government Officials 2 (Jan.
9, 2006), https://perma.cc/LL5A-HW9R. The Executive and Legislative Branches then
agreed to bring the extra-FISA surveillance activities under statute. See Protect America Act
of 2007, Pub. L. No. 110-55, 151 Stat. 552 (2007); FISA Amendments Act of 2008, Pub. L.
No. 110-261, 122 Stat. 2436 (2008). In the process, Congress re-enacted the exclusivity
provision within FISA at 50 U.S.C. § 1812(a), reinforcing its legal and normative pull. See
id. at § 102(a), Pub. L. No. 110-261, 122 Stat. 2459.
248 Lack of an explicit reference or other clear statement provision in the Non-
Detention Act, 18 U.S.C. § 4001a, made it easier for the Supreme Court in Hamdi to agree
with Executive Branch arguments that the post-9/11 AUMF satisfied the statute or was a
constructive amendment. See Hamdi v. Rumsfeld, 542 U.S. 507, 508 (2004) (O’Connor, J.,
plurality op.). The Executive similarly argued that the AUMF implicitly authorized
surveillance outside FISA. See GONZALES, supra note 203. When Congress brought the
extra-FISA surveillance activities under statute, see Protect America Act of 2007, Pub. L.
No. 110-55, 151 Stat. 552 (2007); FISA Amendments Act of 2008, Pub. L. No. 110-261,
122 Stat. 2436 (2008), it also enacted a clear statement rule. See 50 U.S.C. § 1812(b).
249 This provision is modeled on the War Powers Resolution provision. See War Powers
Resolution, Pub. L. No. 93-148, 87 Stat. 555, § 9. See also I.N.S. v. Chadha, 462 U.S. 919,
932 (1983) (such “unambiguous language” creates presumption of severability).
250 This “secret law”-combating provision is informed by Executive Branch and FISA
court interpretation of Sec. 215 of the USA PATRTIOT Act in secret regarding telephony
metadata in a way that, when leaked, appeared nothing like the statute on its face. See Orin S.
Kerr, A Rule of Lenity for National Security Surveillance Law, 100 VA. L. REV. 1513, 1525
(2014) (critiquing secret legal interpretation). The NFCA provision enacts a Public Law
Supremacy Rule. See Rudesill, Coming to Terms with Secret Law, supra note 49, at 301-05,
338-42 (describing rule). See also Appendix, NFCA § (b)(5) (other transparency measures).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

B. Potential Objections and Alternatives

The draft NFCA is one approach to a process-generating statute. With

the intent of stimulating further discussion and research, this section
addresses several potential objections and sketches alternatives.

1. Potential Objections

Is the NFCA too strong? There are certain to be claims that statute
would undermine nuclear deterrence by limiting launch authority,
statutorily giving roles to other Executive Branch officials, or expanding
Congress’s involvement. But the NFCA legally proscribes and
normatively seeks to frustrate launch where there is no imminent threat, in
a context in which national security officials are certain to be deferential
to a sober president’s judgments and eager to protect the country. Its
provisions mainly provide new bases for subordinates to report to
Congress and – as a matter of conscience rather than formal process –
decide that the President is dangerously impulsive. The NFCA would do
nothing to change the nuclear command and control system’s ability to
generate decision where adversary missiles are in the air. It would have no
impact where a senior official could reasonably defer to the President’s
judgment that adversary nuclear attack is imminent in the next three days,
or where presidential subordinates believed that the President had in good
faith carefully evaluated alternatives. Those are low bars that operate only
regarding the conscience of senior subordinates, and only against a Rogue
or Precipitous President. Recalcitrant subordinates (reasonable or
unreasonable) can still be relieved. Congress’s role is one of receiving
reports, and potentially passing legislation (probably requiring a super-
majority for veto-override). None of these are bases on which a remotely
rational adversary might start a nuclear war that would trigger the
President’s repel-attacks authority and result in their destruction.
Critics might claim that Congress has never denied the Commander in
Chief the use of existing forces, or that once raised Congress cannot restrict
the President’s use of a particular weapon. 251 Neither is true. Congress has
since the Founding variously conditioned or banned uses of extant military
forces, and often legislated with great specificity. Under statute, U.S.
forces could seize only ships sailing to (and therefore not from) French
ports during the Founding Era naval war with France, generally may not

251 See Emerson, supra note 175, at 210 (similar argument); The Prize Cases, 67 U.S.
635, 670 (1863) (in case lacking conflicting statute, dicta stating that the President is able to
“determine what degree of force the crisis demands”).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 71

be used for domestic law enforcement, may not be used to suppress

insurrections or other domestic disorder absent statutory requirements,
may not operate at polls or interfere in U.S. elections, may not interrogate
in a manner at odds with the Army Field Manual or torture, may not detain
or punish the enemy contrary to legislated rules and procedures, and may
not be used to conduct covert action or national security surveillance
absent certain certifications and showings.252 It makes no obvious sense,
and does not accord with the constitutional history, to think that statutory
limits of this kind are valid if applied to all weapons and forces but invalid
if applied only to one type of weapon (especially a uniquely powerful one).
Military forces may not be ordered to violate properly enacted statute
whether general or specific, violate the law of armed conflict or otherwise
commit war crimes, or violate treaties ratified by the Senate that prohibit
a variety of military activities (including targeting wounded or
shipwrecked enemy troops, and testing nuclear weapons anywhere except
underground)253 and specific weapons (including chemical weapons, a
particular weapon of mass destruction that like nuclear weapons was
created by Congress in abundance). 254

252 See Little v. Barreme, 6 U.S. (2 Cranch) 170, 177-78 (1804) (President’s order to seize
U.S. ships going to and from French ports illegal under statute); Posse Comitatus Act, ch. 263,
§ 15, 20 Stat. 145, 152 (1878) (repealed, re-enacted, and codified as amended at 18 U.S.C. §
1385) (generally prohibiting federal forces from engaging in law enforcement); 10 U.S.C. §§
251-55 (2016) (Insurrection Act); 18 U.S.C. §§ 592-93 (troops generally may not be deployed
at polls, or interfere in elections); 18 U.S.C. §§ 2340, 2340A & 42 U.S.C. § 2000dd
(interrogation restrictions); Hamdan v. Rumsfeld, 548 U.S. 557, 624 (2006) (President’s order
regarding detainees invalid under UCMJ statute); 50 U.S.C. § 3093 (covert action); 50 U.S.C.
§ 1801 et seq. (FISA).
253 See Little, 6 U.S. (2 Cranch) at 170 (seizing Danish ship contrary to statute); War Crimes
Act, 18 U.S.C. § 2441 (2006) (criminal penalty); Uniform Code of Military Justice arts. 2, 18,
21, (codified at 10 U.S.C. §§ 802, 818, 821 (2019)) (jurisdiction to try violations of law of
war); Convention (I) for the Amelioration of the Condition of the Wounded and Sick in Armed
Forces in the Field, opened for signature Aug. 12, 1949, 75 U.N.T.S. 31 (wounded and sick
soldiers out of the fight may not be attacked and must be provided food, medicine, freedom
from abuse, and other basic protections); Convention (II) for the Amelioration of the Condition
of Wounded, Sick, and Shipwrecked Members of the Armed Forces at Sea, opened for
signature Aug. 12, 1949, 75 U.N.T.S. 85 (applying similar standards of protection for sailors);
Convention (III) Relative to the Treatment of Prisoners of War, opened for signature Aug.
12, 1949, 75 U.N.T.S. 135 (protecting prisoners of war); Convention (IV) Relative to the
Protection of Civilian Persons in Time of War, opened for signature Aug. 12, 1949, 75
U.N.T.S. 287, (protecting non-combatants); Treaty Banning Nuclear Weapons Tests in the
Atmosphere, in Outer Space, and Under Water, opened for signature Aug. 5, 1963, 480
U.N.T.S. 43.
254 See Convention on the Prohibition of the Development, Production, Stockpiling and
Use of Chemical Weapons and on their Destruction, opened for signature Jan. 13, 1993, 1974
U.N.T.S. 45.

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

In view of Congress’s powers and the nuclear stakes, another objection

might be that the draft NFCA is too modest. For example, the President
might try to wiggle around the NFCA through arguments about its
particular definitions and other terminology, together with demands for
deference to the President and Executive Branch in national security. The
statute would also not stop the President from attempting a nuclear
Saturday Night (firing) Massacre to find subordinates willing to execute
problematic orders. In the end, doing more than this statute provides via
legislated proscriptions, funding limits, and reporting requirements would
require Congress (or the Congress, states, or people through a
constitutional amendment) to attempt a larger adjustment to the
constitutional equilibrium.

2. Potential Alternatives

This article endeavors to catalyze further discussion by scholars and

practitioners and encourage new thinking about nuclear launch authority.
Following are several alternative statutory directions.
A more permissive NFCA could allow nuclear use in the face of attack
by non-nuclear weapons. The United States deployed nuclear forces to
Europe to deter Warsaw Pact conventional attack, and U.S. nuclear
doctrine has long reserved the possibility of nuclear use in response to
threats from chemical, biological, and other weapons of mass
destruction.255 This article and its draft statute have reserved without
judgment this controversial basket of issues, which would need to be
addressed.
Alternatively, a more expansive statute would embrace use of force
generally. That would amount to a rewrite of the WPR. Or, a less intrusive
statute – one regarding force generally or limited to nuclear weapons –
could be enacted as legislative advice. Of course, because Presidents have
in recent decades pushed against the strictures of the WPR and other hard-
law statutes, their regard for a “Sense of the Congress” statute would likely
be lower still.256

255 The Trump Administration controversially has suggested that the United States
might use nuclear weapons in response to catastrophic cyber threats. See 2018 NPR, supra
note 3, at 38 (U.S. nuclear capabilities hedge against nuclear and non-nuclear threats including
cyber); David E. Sanger & William J. Broad, Pentagon Suggests Countering Devastating
Cyberattacks with Nuclear Arms, N.Y. TIMES (Jan. 16, 2018), https://perma.cc/BVT6-5HFY.
256 The salience of such hortatory provisions and reporting requirements are under-
studied parts of the administrative state and in the national security context in particular.

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 73

If Congress instead sought to write a stronger statute than outlined

here, one can postulate several options.
For example, Congress could integrate into the NFCA a “no first use”
provision, or subsequently pass the Markey-Lieu or even Fulbright
proposals that would require ex ante congressional legislation absent an
initiated adversary attack. Alternatively, Congress could require that the
reports in Section (b)(3) of the NFCA be ex ante – delivered to Congress
before the President could act. Any President is sure to object, as would
separation of powers formalists and presidentialists.257 Even so, the
question warrants new analysis. While it is considering deviations from
formalist separation of powers, the renewed conversation might revisit and
perhaps update Cold War-era proposals for permission from a
congressional leadership committee for first use of nuclear weapons. 258
Another limited but potentially justified departure from formal
separation of powers doctrine – one that could see Congress asserting its
authorities against those of a defiant President in Youngstown Category
Three – is borrowing a limited Executive Branch process step from FISA’s
Sec. 702. As discussed in Part IV.C, Sec. 702 requires the Attorney
General and Director of National Intelligence (DNI) to make certifications
(and the FISA court to approve them) before the government may conduct
certain intelligence collection inside the United States.259 A revised NFCA
might require certification that there is an imminent threat in the judgment
of either of the top two civilian intelligence officials, the Secretary of
Defense and DNI. The Secretary is probably the official most likely to
inform the President of an imminent threat of adversary nuclear attack. The
DNI is the senior U.S. intelligence officer. 260 The Secretary and the DNI

257 For discussion of inter-branch controversy over ex ante reporting in the covert
action context, see REISMAN & BAKER, supra note 240, at 121. In the face of a veto Congress
there permitted the President temporal latitude.
258 A variation of the Federation of American Scientists proposal discussed in Part IV.A
is providing a role for the Vice President as President of the Senate (a two-branch officer), and
the Speaker of the House.
259 See 50 U.S.C. § 1881a (2018). The legislative and statutory history is worth note. Sec.
702 was written and amended in the wake of the George W. Bush Administration’s violation
of FISA with warrantless wiretapping after 9/11, and with congressional knowledge of an
infamous hospital room confrontation in which the Bush White House sought to pressure the
Justice Department’s leadership to reauthorize part of the warrantless collection program
under internal Executive Branch rules and oversight. See SAVAGE, supra note 77, at 190–94.
Congress in Sec. 702 decided against reliance on such Executive self-regulation and gave the
Attorney General and DNI statutory certification responsibilities. Regarding FISA’s
constitutionality, see supra note 217; Rudesill, The Land and Naval Forces Clause, supra note
154, at 457-61.
260 See 50 U.S.C. § 3023(b) (2020).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

would have the benefit of information flows from the North American
Aerospace Defense Command (NORAD), the U.S. Strategic Command
and other combatant commands, and the 17-agency U.S. Intelligence
Community. Requiring one of these top officials to certify in their
independent judgment that there is a temporally imminent threat guards
against a President who concocts or wildly exaggerates a threat. Allowing
either official to certify (in contrast Sec. 702 requires two certifications)
reduces the extent of the intrusion on the Commander in Chief and
provides operational flexibility in the event of a fast-moving threat. If there
truly is good reason to think there is an imminent threat of adversary
nuclear attack, their recognition of the threat would happen in the course
of doing their jobs. If the Secretary or DNI were eliminated or otherwise
could not be contacted due to an attack that is already underway, the statute
could waive the requirement.261
Plausible argument could be offered that the departure from formalism
would be minimal and appropriate in view of the Rogue President risk. 262
Our constitutional system has permitted other deviations from formalism
to serve important congressional and national equities. 263 But this section
702-inspired certification would certainly be contested by presidentialists
and also by thinkers who share the concurrent powers (majority) view of
separation of powers, because the certification would – however
minimally and valuably – give a presidential subordinate a potential veto

261 If instead neither of these officials and none of their successors are available to make
the certification because the President is intent on a nuclear strike and has relieved the entire
civilian leadership of the Defense Department and Office of the Director of National
Intelligence, a good assumption is that the President is mentally unfit. The NFCA at that point
could have the nuclear command and control system fail-safe (fail into safe mode) until the
President is removed, resigns, transfers power to an Acting President, or comes to their senses.
262 Scholars have in recent years presented compelling evidence that undermines the
originalist case for the Unitary Executive Theory – the idea that the President possesses all,
and indivisible, executive authority available to the federal government, including a raft of
national security powers that inhere to the state. Professors Peter Shane and Julian Davis
Mortensen have, respectively, presented powerful originalist evidence that the Founders were
comfortable with division of executive power and that its original meaning was simply the
power to implement (execute) law created by some other authority. See Peter Shane, The
Originalist Myth of the Unitary Executive, 19 U. PA. J. CONST. L. 323 (2016); Julian Davis
Mortenson, Article II Vests Executive Power, Not the Royal Prerogative, 119 COLUM. L. REV.
1169 (2019); Julian Davis Mortenson, The Executive Power Clause, 168 U. PA. L. REV. 1269
(2020). Cf., Seila Law LLC v. Consumer Financial Protection Bureau, 591 U.S. __ (2020)
(statute that restricted President’s removal authority of an Executive Branch official
unconstitutional).
263 See Raven-Hansen, supra note 15, at 794 (deviations from formalist separation of
powers justified to advance goals within Congress’s authority).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 75

on use of force, disrupting the chain of command. 264 Alternatively, critics

may point out that FISA draws authority in part from the Fourth
Amendment (whereas the NFCA would not), or they may question
Congress’s intent in Sec. 702, or they could distinguish use of force and
intelligence collection. One can imagine thoughtful responses to these
points, auguring toward an expanded conversation in this direction. 265

CONCLUSION

Whatever statute or executive order requires, a committed Rogue or

Precipitous President may still succeed in pushing to execution a nuclear
strike order the nation comes to regret deeply. There is too much
uncertainty, and perhaps too much presidential power. 266 But a
constitutional equilibrium that includes some sensible efforts at statutory
regulation of the President’s access to “the bomb” is preferable to the status
quo. Accordingly, this article recommends a statute that proscribes use
where “extreme circumstances” do not pertain, and where adversary
nuclear attack is not imminent and the executive has not reviewed a full
slate of options. This article looks to Title 50 U.S. Code for long-used
reporting requirements, reworking them for nuclear decisions in ways that
would help build expectations and norms of careful appraisal and good
process as time permits. A committed Commander in Chief would be able
to relieve a lot of subordinates before being relieved in turn (via the
Twenty-Fifth Amendment or impeachment), but a nuclear Saturday Night
Massacre would be the more difficult the firmer and broader the ground
on the basis of which subordinates could resist. The suggested Nuclear
Forces Control Act provides such new ground, and a reporting-driven
prompt for the statute’s invocation potentially before the question of
legality under the law of armed conflict is reached. More bases on which
to push back inferentially buys more time for resistance to spread and
grow, and more possibility that a Rogue or Precipitous President will be
stopped or come to their senses. In turn, greater clarity about the rules of
nuclear use and greater confidence that any nuclear use would be necessary

264 See Barron & Lederman, Lowest Ebb, Part II, supra note 164, at 945-46 (statute may
not assign ultimate decision authority to subordinate instead of Commander in Chief).
265 For example, along with other constitutional provisions supporting both Sec. 702 and
a 702-like NFCA provision, the Declare War Clause would provide authority for the latter
where it does not for the former.
266 “Perhaps the strictures of the Constitution are such that the Congress cannot select a
constitutionally valid scheme . . . But if our options are so limited, then we are already far too
late.” Carter, supra note 140, at 123.

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

and legal would buttress public confidence in government and the nation’s
nuclear forces.
This article has argued that it is time to replace inherited notions that
nuclear weapons are constitutionally special with the understanding that
they ought to be made statutorily special. Because of enduring nuclear
nightmares, nuclear-conventional convergence, the value of “good
process” and statutes that facilitate it, change in military technology and
the international security environment, doctrinal developments, and
entrenchment of the covert action and FISA statutory frameworks – thanks
to all these post-Cold War developments, Congress can and should act.
Executive abuse of authority prompted Congress to legislate limits
and process rules for some of the nation’s most sensitive national security
activities, including covert action and national security surveillance.
Tragically, impulsive employment – or attempted illegal employment – of
the nation’s nuclear loaded weapons may be necessary to prompt Congress
to craft a new regime balancing presidential nuclear command and
statutory control. But the nation need not wait for another moment of
presidential impairment, nor for an atomic atrocity. Participants in the
revived national conversation about nuclear command and control should
think anew about legislative solutions using Congress’s under-utilized
Article I powers. The constitutional history of congressional acquiescence
regarding nuclear weapons need not be our constitutional fate. 267

267 For the concept, see PHILIP BOBBITT, CONSTITUTIONAL FATE: THEORY OF THE
CONSTITUTION (1982).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 77

APPENDIX: DRAFT STATUTE

SECTION 1. SHORT TITLE.

This Act may be cited as the “Nuclear Forces Control Act.”
SECTION 2. CONTROLS ON USE OF NUCLEAR WEAPONS.
Title 50 U.S. Code, Chapter __, is amended by adding Sec. ___:
(a). Purposes and Congressional Authority.
(1) Purposes. The purposes of this Section are to exercise
Congress’s constitutional authority to control use of the
forces it creates; to inform congressional oversight of
the international security environment and of the
Executive Branch; to prevent unnecessary or otherwise
illegal use of nuclear forces in any case and particularly
where the United States, its forces, or allies do not face
temporally imminent nuclear attack; to facilitate the
most careful and thorough decision-making process
within the Executive Branch that is practicable; and, to
provide definitions and interpretive guidance to
facilitate compliance.
(2) Constitutional Authority. This Section reflects
exercise of the powers of Congress pursuant to Article,
I, sec. 8, cl. 1, 11, 12, 13, 14, 18, and Article I, sec. 9,
cl. 7 of the Constitution to “provide for the common
Defence and general Welfare,” “declare War,” “raise
and support Armies,” “provide and maintain a Navy,”
“make Rules for the Government and Regulation of the
land and naval Forces,” “provide for calling forth the
Militia” to “repel Invasions,” discipline and govern the
militia,268 “make all Laws which shall be necessary and
proper for carrying into Execution” these powers,
control appropriations, and conduct oversight.
(b). Rules for Nuclear Forces, Funding Limitation, etc.
The following rules and conditions shall govern operational
employment of nuclear forces.

268 Although nuclear-capable aircraft, missiles, and submarines are operated by federal
regular military forces, the broader military establishment that supports their operations and
the command and control system includes extensive participation by the National Guard. The
Guard is the “organized militia” under statute. See 10 U.S.C. § 311(b)(1) (2016).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

(1) Rules for Nuclear Forces. Nuclear weapons are not

provided for operational employment, and shall not be
operationally employed, unless:269
(A) law other than this Section provides legal authority
for use of force, including nuclear weapons;
(B) use of nuclear weapons is militarily necessary in
extreme circumstances to defend the vital interests
of the United States and its allies, and is otherwise
legal; and,
(C) one of two circumstances pertain:
(i) the United States, its forces, or allies face
imminent nuclear attack; or,
(ii) nuclear attack on the United States, its forces,
or allies is not imminent, and the military,
diplomatic, legal, intelligence,
environmental, and other implications of
both nuclear and non-nuclear options have
been carefully evaluated, and non-nuclear
weapons will not succeed in defending the
vital interests of the United States and its
allies.
(2) National Security Council and the Nuclear
Command, Control, and Communications System
at the National Leadership Level.
(A) National Security Council. Consistent with the
direction of the President, the National Security
Council established by 50 U.S.C. 3021 shall advise
the President in the evaluation of nuclear and non-
nuclear options, particularly as provided in
paragraph (b)(1)(C)(ii).
(B) The Vice President. The Vice President shall be
responsible for the coordination of National
Security Council interagency review of nuclear
threats, capabilities, and decisions.
(C) Legal Advice. The nuclear command, control, and
communications system shall be configured to
facilitate the involvement of the Attorney General,

269 This statute, like the covert action law and FISA, is centered around a bar on activity
contrary to its definitions and processes.

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 79

and other legal advisors as appropriate, to the extent

practicable.
(3) Notifications, Finding and Reporting
Requirements.
(A) Threat Notification to Congress.270 Whenever in
their independent judgment nuclear attack on the
United States, its forces, or allies is imminent, the
following officials shall notify the congressional
leadership as soon as possible, and thereafter keep
the congressional leadership fully and currently
informed until the threat is no longer imminent: the
Secretary of Defense, the Director of National
Intelligence, the Chairman of the Joint Chiefs of
Staff, the Commander of the U.S. Strategic
Command, and any other person in the military
chain of command contacted by the President with
an order for operational employment of nuclear
weapons.271 Notifications and updates may be
provided in oral or written form, as extensive as
circumstances warrant, transmitted separately, or
transmitted together if in agreement about the
imminence of the threat.
(B) National Security Council Deliberation
Notification to Congress. The Vice President shall
notify the congressional leadership within 48 hours
of initiation of National Security Council review of
potential operational employment of nuclear
weapons. The notification shall be provided in
written form but may initially be provided in oral
form.
(C) Presidential Finding and Report. Nuclear weapons
are not provided for operational employment, and
shall not be operationally employed, unless the
President determines and explains in a written

270 The covert action statute requires that the DNI and other presidential subordinates
keep Congress “fully and currently informed” regarding covert actions. The implication is that
these are already-authorized covert actions. In view of the much higher potential stakes
associated with nuclear threats and use, this model statute temporally moves up the reporting
requirement to track the emergence of the threat.
271 Effective implementation of the statute would be facilitated by training for the nuclear
chain of command in the statute and on how to report to Congress.

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

finding that the criteria in paragraph (b)(1-2) are

met, and reports that finding to the congressional
leadership.
(i) Timing. The finding shall be reported as soon
as possible after decision and before
operational employment of nuclear weapons.
(ii) Exception. Where time does not permit the
preparation and transmission of a written
finding before operational employment of
nuclear weapons, as soon as possible the
President shall make an oral report to the
congressional leadership, and thereafter a
written finding shall be prepared and
transmitted to the congressional leadership.
(D) Reports Contemporaneous to Presidential Finding.
Where nuclear attack on the United States, its
forces, or allies is not temporally imminent,
contemporaneous with the presidential finding in
(b)(3)(C) and with a similar timing requirement and
exception, the following officials shall report to the
congressional leadership as follows:
(i) The Secretary of State shall report regarding
how employment of nuclear weapons serves
the vital interests of the United States and
otherwise complies with this Section;
(ii) The Secretary of Defense shall report
regarding how the employment of nuclear
weapons is militarily necessary and
otherwise complies with this Section;
(iii)The Attorney General shall report regarding
how the employment of nuclear weapons is
legal and otherwise complies with this
Section; and,
(iv) Any other person in the military chain of
command contacted by the President with an
order for the operational employment of
nuclear weapons shall report regarding how
such employment complies with this Section.

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 81

(4) Congress.
(A) Consultation. The President in every possible
instance shall consult with Congress before and
after employment of nuclear weapons.
(B) Reporting and Ongoing Oversight. With due regard
for the protection from unauthorized disclosure of
classified information related to intelligence and
nuclear weapons, the President, the Vice President,
and the heads of all departments, agencies, and
military commands involved in the contemplated or
executed operational employment of nuclear
weapons:
(i) shall cooperate with the Congress in the
creation and maintenance of secure methods
for transmission of notifications, findings,
and reports under this Section;
(ii) shall keep the congressional leadership
and relevant committees fully and currently
informed of (a) contemplated, planned, and
executed operational employment of
nuclear weapons, (b) relevant threat and
weapon effects assessments, and (c) the
functioning of the nuclear command,
control, and communications system; and,
(iiI) shall furnish to the relevant committees
any information or material related to
operational employment of nuclear
weapons which is in the possession,
custody, or control of any department,
agency, or entity of the United States
Government and which is requested by a
relevant committee.272
(5) Public Transparency. If classified the written
notifications, finding, and reports stipulated in (b)(3)
shall be declassified not later than one year after
transmission to Congress, an unclassified summary
shall be published, or an unclassified statement shall be

272 This provision is borrowed in modified form from the covert action statute, 50 U.S.C.
§ 3093(b) (2019), with addition of the President and Vice President and removal of explicit
mention of the DNI. See also War Powers Resolution, Pub. L. No. 93-148, 87 Stat. 555, § 4(b)
(1973) (presidential reporting).

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

published indicating that one or more communications

were transmitted to Congress and stating when more
information will be published.273
(6) Funding Limitation. No appropriations shall be
available for activities not in compliance with this
Section.
(7) Vacancies. Should a Senate-confirmed officer
identified in this Section be unavailable, the
responsibilities identified in this Section and their role
in the nuclear chain of command shall be vested in the
person duly acting in their role, provided that such
official shall be the Senate-confirmed officer of the
unavailable official’s department, who has duly
ascended to the acting position by virtue of their
department’s specific order of succession. In any such
case and without regard to the Federal Vacancy Reform
Act, no person other than the official exercising an
acting role by virtue of their department’s specific order
of succession may perform their functions under this
Section or in the nuclear chain of command.
(c). Criminal Sanctions. Any order to employ nuclear weapons
where the conditions in paragraph (b)(1) have not been met is an
illegal order. Any person participating in the execution of the order
who knows that the conditions in paragraph (b)(1) have not been
met, or who could reasonably ascertain whether the conditions in
paragraph (b)(1) have been met, shall be punishable by a fine of
not more than $10,000 or imprisonment for not more than five
years. This Section does not apply to persons who transmit,
execute, or otherwise act on an authenticated order to employ
nuclear weapons, who have no knowledge or reasonable ability to
know whether the conditions in paragraph (b)(1) have been met.
(d). Definitions.
(1) “Operational employment” means intentional launch
against an adversary and detonation of one or more
nuclear warheads.

273 This transparency provision is modeled on secret law-combating stipulations of the

USA FREEDOM Act, Pub. L. No. 114-23, § 402, 129 Stat. 268, 279–82 (codified as amended
at 50 U.S.C. 1872) (2015) (declassification of Foreign Intelligence Surveillance Court
decisions with significant legal interpretations, or publication of a redacted version or a
summary).

Electronic copy available at: https://ssrn.com/abstract=3689074

 NUCLEAR COMMAND AND STATUTORY CONTROL 83

(2) “Independent judgment” means an assessment not

directed by any other person or entity.
(3) “Imminent” means reasonably possible within the next
72 hours, based on assessment of adversary capabilities
and intentions. Adversary possession of the
technological capability of launching a nuclear attack,
without other significant indications of adversary
intent, does not create imminence under this Section.
(4) “Contemporaneously” means at approximately the
same time as the President’s finding is reported, and at
the least as soon as possible thereafter.
(5) “Unpublished” means not available to the public.
(6) “Congressional leadership” means the Speaker of the
House, the Majority Leader of the Senate, and the
Minority Leaders of the House and Senate. These
officials shall share notifications, findings, and reports
received pursuant to this Section, written or oral, in full,
with the chair and ranking members of the Armed
Services Committees of the House and Senate, the
House Foreign Affairs Committee and the Senate
Foreign Relations Committee, and the House
Permanent Select Committee on Intelligence and the
Senate Select Committee on Intelligence.
(e). Rules of Construction.
(1) Exlusivity. This Section shall be the exclusive means
under which decisions about operational employment
of nuclear weapons may be conducted.
(2) No Implied Amendment or Repeal. This Section shall
not be interpreted to be amended, qualified, superseded,
or repealed except by explicit reference to this Section
in subsequent Public Law.
(3) Severability. If any provision or part of any provision
in this Section or application thereof is held invalid, the
remainder of the Section shall not be regarded as
invalid.
(4) Supremacy of Public Law. This Section, and any
classified or otherwise unpublished legal authority,
order, directive, rule, memorandum, or other guidance
or interpretation construed to relate to this Section, shall
be interpreted in a manner deferential to the public
meaning of this Section: the understanding a

Electronic copy available at: https://ssrn.com/abstract=3689074

11 JOURNAL OF NATIONAL SECURITY LAW AND POLICY __ (2020)

knowledgeable and reasonable person would have of

this Section, employing majority approaches to
interpreting law, and considering only information
actually available to the public.

Electronic copy available at: https://ssrn.com/abstract=3689074