Discussion 6
Understanding the Military’s Role
A. The War Power
Cicero said laws are inoperative in war (“silent enim leges inter arma”). In fact,
there is U.S. and international law that applies to the decision to use force and the manner
and method in which it is used. In U.S. practice there are three contextual legal questions
that should arise when the president or his advisors contemplate the use of military force.
First, does the president have the constitutional authority to use force? This is a “war
powers” question. Second, is the contemplated force lawful under international law? This
is a question involving that portion of the law of armed conflict1 known as the jus ad
bellum (thresholds for resorting to force). Third, are the means selected lawful? This is a
question involving the jus in bello (the law applying to the means and methods of war). It
may also be a question of U.S. criminal law to the extent that U.S. law implements
international law, or imposes obligations independent of international law.
Few areas of law are as important to national security as is this collective body of
domestic and international law. This is intuitive in situations in which force is actually
used; however, in other scenarios falling short of force the implied or express threat of
military intervention is itself the catalyst for policy influence. For the national security
lawyer, no area of law will have the same implications in defining a world of public order
and individual security. So too, in few areas, if at all, will the lawyer feel the same
measure of pressure from the policymaker. The importance of this law to the combatant
and civilian noncombatant is even more evident.
For the lawyer, policy consideration of resorting to force raises a threshold
question – is it lawful under U.S. law? The answer requires application of constitutional
text, theory, and practice. It may also require consideration of any applicable statutory
overlay, including the War Powers Resolution and case-specific legislation, like Public
Law 107–40, “Authorization for Use of Military Force,”2 or Public Law 107–243,
“Authorization for Use of Military Force Against Iraq Resolution of 2002.” The power to
resort to force is in some manner both separate and shared between the political branches.
This is clear from the outset of the Constitution. Of course, the preamble is not law, but it
does suggest that the authority to defend the country derives from the people and not a
singular branch of government or extraconstitutional theory of law. This is indicated as
well by the enumerated text.
The president’s enumerated war powers include those as commander in chief,
chief executive, and those authorities that collectively permit the president to conduct
foreign relations. These are the authorities cited by successive presidents in more than
100 “War Powers” [Resolution] reports as the legal basis for military action. In a time of
homeland security and terrorism it is also worth noting that the president is charged in
Article II “to take care that the laws be faithfully executed. If law were math, we might
add up the clauses and declare Congress the winner. However, the Constitution was
designed to avoid winners, an intentional system of overlap and friction. The question is
not whether the war power is shared. It is. The constitutional text demonstrates so. But
the powers are separate as well. The question is, how do these interlocking powers relate?
To what extent can the commander in chief exercise his authority absent Congress’s
exercise of its authority? If Congress acts, the question becomes one of Youngstown
analysis. Is the president acting at the zenith of his authority, or at its ebb? Or perhaps, as
suggested in Hamdan’s Footnote 23, is the president’s power extinguished?
Separate and shared powers were intended to serve as a check on the use of force
and as a method to ensure popular, or at least representational, support when force is used
by engaging both elected branches in the decision. That is black-letter principle, not
necessarily black-letter law. The legal issue is whether, when, and for how long the
president can resort to force solely pursuant to his authority, or in the face of
congressional opposition. But no matter how hard (or repetitively) advocates work to
extract the last ounce of authoritative weight from each word and phrase in the
Constitution and each page in the Federalist Papers, text alone does not resolve these
questions in a manner that lawyers and policymakers can agree is definitive and binding.
Because the Constitution provides for shared and separate powers between the
political branches, rather than exclusive responsibility, in the absence of controlling text,
questions involving the president’s authority to resort to military force often depend on
constitutional perspective. Text alone, for example, does not resolve how much unilateral
authority the president might properly derive from the commander-in-chief clause, or the
extent to which the “declare war” clause checks this authority. As a result, war power
questions are ultimately shaped by and depend on theory. War power theory runs a
continuum of view, from the narrow – Congress alone may authorize U.S. involvement in
combat – to the broad: the military instrument is an extension of the president’s foreign
affairs power and it is subject only to the positive and specific exercise of Congress’s
authority over spending, if that. Indeed, moving further along the spectrum, Justice
Sutherland’s extra-constitutional sovereignty argument in Curtiss-Wright might suggest
that the war power is an inherent executive authority independent of the Constitution. In
the opposite direction, the War Powers Resolution reflects a constitutional perspective,
that of the legislative branch in 1973, suggesting that prospective and inchoate exercise of
the congressional war power can extinguish the president’s later exercise of his authority
after sixty days.
Presidents have resorted to military force on numerous occasions in the absence
of implied or express congressional authorization. The congressional Research Service
has documented more than 225 such instances.12 Although these examples fall on a
factual continuum, it is possible to detect four general trends. In a majority of
circumstances, the president authorized immediate military operations of limited
duration, like the Mayaguez mission or an embassy evacuation (known in military
doctrine as a noncombatant evacuation operation, or NEO). Lawyers of all stripes
generally agree that the president may authorize the military to address an immediate
physical threat to U.S. lives or property, absent congressional authorization, consultation,
or, in some cases like the Mayaguez, in the face of congressional prohibitions.
A second category, which is more difficult to define and classify, or perhaps only
more difficult to agree upon, involves the use of the military for sustained, but limited
combat operations to accomplish limited (U.S.) policy objectives; for example, the use of
air power in the Balkans to halt Serbian involvement in Bosnia’s civil war (two weeks of
air strikes in 1996) or to stymie Serbia’s campaign of ethnic cleansing in Kosovo
(seventy-eight days of air strikes in 1999). In the latter case, one might note that while the
U.S. intent may have been narrow, the potential policy impact was not. In attempting to
protect two million Kosovars from ethnic cleansing, NATO was also influencing the
movement toward independence for Kosovo as well as the secession movement in
Montenegro. The 1989 invasion of Panama might be viewed in the same manner, due to
its limited duration and limited casualties, including forty U.S. dead; however, the U.S.
objective was more expansive aimed as it was at the removal and arrest of the de facto
head of government, Manuel Noriega.
Of course, in some cases it was only in implementation that operations proved
limited in nature, and less like “war” than anticipated, or for that matter, prolonged in
nature, and more like “war” than anticipated. For example, the 1994 U.S. intervention in
Haiti started with troops boarding aircraft prepared to engage a hostile force but landing
as peacekeepers. Clearly, the president believed he had the constitutional authority to go
in “hot” in the absence of affirmative congressional authorization. In other cases, narrow
“noncombat” operations have evolved into direct and sustained combat operations, such
as in Lebanon in 1982–1984, and in Somalia in 1991–1994. The threshold constitutional
determinations are made at the outset, before these ground truths are known; however,
the constitutional precedent is set with the factual result in mind; that is, seventy-eight
days of air strikes, not limited air strikes. Typically, the Congress has expressed support
for U.S. armed forces, but has not authorized, nor legislatively precluded, the president
from engaging in military interventions on the scale and duration of a Panama, Haiti,
Somalia, Bosnia, or Kosovo. Depending on one’s perspective, this might be viewed as a
product of political caution, a reflection of the political division between the branches at a
given moment, or a reflection of institutional perspective. Most likely there are 535
different explanations, which draw on all three perspectives.
Post-Vietnam, and thus following passage of the War Powers Resolution, the
United States has engaged in three sustained air-ground conflicts: the Persian Gulf War
(1990–1991),15 Afghanistan (2001–present), and Iraq (2003–present). In each case,
significant casualties were anticipated in advance of combat, including up to 20,000 in
the first Iraq conflict. In each case the Congress expressly authorized the use of force.
And, in each case, the president asserted an inherent authority to act alone, but
nonetheless sought and “welcomed” congressional authorization. Lawyers and
commentators were left to sort out whether such authorization was required or merely
prudential, signaling to an adversary that as a matter of law the president was acting at
the zenith of his power and as a matter of policy the nation was acting with uniform
political support (at least at the outset).
If not resolved by constitutional text, or theory, some argue the war powers
framework is defined by statute, in particular, the War Powers Resolution of 1973. As
recounted elsewhere,16 at the close of the Vietnam conflict, the Congress sought to
exercise its “war power” prospectively through creation of a statutory framework. The
War Powers Resolution was “necessary and proper,” proponents argued, in light of the
American experience in Vietnam and Cambodia. Proponents perceived that the president
had significantly expanded the presence of U.S. forces in Vietnam and secretly in
Cambodia without express statutory authorization, and perhaps, without congressional
knowledge. The resolution became law over President Nixon’s veto,17 and its
constitutionality has been disputed ever since.
Section 3 provides: “The president in every possible instance shall consult with
Congress before introducing United States Armed Forces into hostilities or into situations
where imminent hostilities are clearly indicated by the circumstances.” In practice,
presidents have authorized combat missions without consulting the Congress (e.g., the
1980 Iran hostage rescue mission); by notifying a limited number of members
immediately prior to an action (e.g., the 1986 Libya raid, or 1998 Afghanistan and Sudan
strikes); through extensive prior consultation (e.g., the 1999 Kosovo air war18); and
following a request for and receipt of express congressional authorization (e.g., Desert
Storm, Afghanistan, Iraq 2003). As noted in Chapter 4, this provision is generally viewed
as constitutional, in part, because it contains its own constitutional trap door – “in every
possible instance.”
The War Powers Resolution requires the president to report to the Congress
within forty-eight hours after introducing U.S. armed forces into hostilities or into
situations where hostilities are imminent; into the territory, airspace, or waters of a
foreign nation while equipped for combat; or in numbers that substantially enlarge U.S.
armed forces equipped for combat already located in a foreign nation.19 As a matter of
longstanding practice, the executive branch does not indicate under what section a report
is filed. This reflects the factual difficulty, and therefore legal difficulty, that executive
actors have in distinguishing among “imminent hostilities,” “ongoing hostilities,” and
situations where forces are “equipped for combat,” particularly where it is hoped that the
latter will deter the former. As importantly, a report involving “hostilities” would in
theory, trigger the sixty-day clock. Indeed, only one report, that pertaining to the
Mayaguez incident, has expressly cited to Section 4(a)(1) of the Resolution, and there the
predicate deployment was over by the time the report was filed.
The majority of such reports are inconsequential and ministerial, even pro forma.
In the case of a noncombatant evacuation operation (NEO) the mission is often over
before the forty-eight-hour report is filed. Nor should there be doubt that, if he felt it
necessary, the president could construe the reporting requirement as permitting
submission of a secret report if he determined that public disclosure might imperil U.S.
lives or ongoing operations. In the case of deployments whose contexts are inherently
short term, reporting elements like that requiring a statement as to the duration of a
deployment seem absurd as do reports reflecting small, but “significant,” changes to the
number of U.S. armed forces personnel deployed. For example, the doubling of a handful
of combat-equipped soldiers might be viewed by some as a “significant increase” in the
overall number of deployed troops requiring a supplementary report. To avoid such “hair
trigger” reports, the executive branch tends to submit reports of bland and flexible
generalization, decreasing the value of the report as a source of information.
Although the reports may not in fact serve to notify the Congress of deployments,
they can serve a useful bureaucratic purpose if executed in good faith. (Here it is helpful
not just to think of the Iraq war or the conflict in Afghanistan but also of a Bosnia or
hypothetical Darfur mission.) The reporting elements can serve as a vehicle within the
executive branch to identify internal policy fissures at a time when policymakers may be
focused on the predicate reasons for deployment. Further, because the report is sent under
presidential signature the report can serve as a useful test of purpose. Forty-eight hours
into the mission, do the national security agencies agree on how the mission is
characterized for the president, its goals, and its anticipated length? Does the president
agree with the bureaucracy’s characterization in the draft report or cover memorandum?
Have events played out as anticipated at the time the president approved the mission? For
these same reasons the reports are usually diluted to the lowest common denominator of
agreement so as to avoid placing the president on a policy limb. Nonetheless, in the
process of drafting even the generic report the executive may identify policy issues that,
while not ultimately reflected in a report, serve as touchstones for internal executive
consideration. Reports may also serve as a useful method of creating a paper trail of
congressional consultation for long-term deployments.
All administrations since the passage of the Resolution have objected to the sixty-
day clock on policy grounds. Among other things, policy critics argue, the resolution may
encourage opponents of U.S. intervention overseas to create “hostilities” so as to start the
clock. As President Nixon’s veto statement asserted, once running, U.S. enemies would
have incentive to hang on and wait it out, in anticipation of eventual U.S. withdrawal.
Whether this is a realistic argument is another matter. Since 1973, the Resolution has not
served to trigger the withdrawal of U.S. armed forces. However, as President Nixon also
asserted, “Until the Congress suspended the deadline, there would be at least a chance of
United States withdrawal and an adversary would be tempted therefore to postpone
serious negotiations until the 60 days were up.” This potential impact is harder to assess.
The clock may also turn legitimate policy debates over the merits of a deployment into
legal debates using arbitrary deadlines unrelated to military effect or world events.
As a matter of law, opponents of the provision argue that Congress cannot limit
through statute the president’s broad constitutional authority as commander in chief, to
conduct foreign relations, and as chief executive. Whatever the scope of presidential
authority, the Constitution does not define it in temporal terms. Therefore, if the president
may lawfully deploy U.S. forces in the first instance on his own authority then there is
nothing in the Constitution that limits that authority to sixty days. This position is
affirmed by longstanding executive branch practice, acquiesced to by Congress as a
voting institution. In addition, the Resolution identifies an inherent presidential authority
to protect the United States from attack, but does not recognize the president’s inherent
authority to protect U.S. nationals, such as those seized on board the Mayaguez. At
minimum, therefore, the Resolution would be unconstitutional as applied to such
scenarios. Thus, the Resolution may reflect the relative power of the political branches at
the height of Watergate, but it does not reflect constitutional law. In any event, the 93rd
Congress cannot bind a future Congress in the manner of its own constitutional
interpretation. The issue then is not whether Congress is free to act using its war power,
but whether Congress can exercise that war power in prospective manner to terminate the
president’s future use of his war power.
Legal proponents of the war powers clock argue that the Resolution is a legitimate
exercise of the “necessary and proper clause,” creating a framework that history has
shown is necessary if Congress is to meaningfully exercise its enumerated authorities
over “war.” The framework is necessary, because the president often presents the
Congress with a policy fait accompli and because the Congress itself lacks the political
will to play its constitutional role on a case-by-case basis. Much as the Congress cannot
extinguish the president’s constitutional authority through statute, presidential practice
and congressional inaction cannot have extinguished Congress’s own authority, later
asserted, in this case by the 93rd Congress. Further, proponents argue that the intent
behind a shared war power was to prevent a single person (a.k.a. the president) from
entering the United States into “war.” The sixty days, then, is a form of constitutional
measure to distinguish short-term uses of force from “war.”
As a matter of practice, the clock has generated considerable debate, but has not
in fact triggered the withdrawal of U.S. armed forces. Nor am I aware of evidence
(anecdotal or evidentiary) that the clock has dictated the pace or nature of military
strategy and tactics; for example, by changing the pace of a campaign against military
judgment. The clock has served as a reminder to those in the executive branch of the need
to meaningfully consult the Congress during ongoing military campaigns, in order to
deter questions of authority or potentially distracting lawsuits by members challenging
adherence to the Resolution in court. Executive branch actors also know that in the
absence of express authorization, regardless of their constitutional positions, an
administration will be required after sixty and ninety days to affirm and restate its legal
authority for proceeding with unilateral force in light of congressional, press, and public
inquiry. This “opportunity” to review the legal bidding is not such a bad thing in a
constitutional democracy. If past is prologue, it requires Republican administrations to do
more than assert blanket statements of constitutional authority, and Democratic
administrations to face squarely the constitutional issue presented, or risk the adjudication
of a court challenge on the basis of standing alone.
There have been two instances where the clock was arguably triggered. In the
case of U.S. combat in Somalia (1993), recounted in the book and movie Black Hawk
Down, the executive branch took the view that U.S. armed forces were not engaged in
continuous hostilities, but rather were subject to intermittent hostilities, each falling short
of sixty days in duration. Thus, the clock never tolled. Only in retrospect was it clear to
executive actors (and their lawyers) in Washington that during the summer and into the
fall of 1993, U.S. forces were in fact engaged in continuous combat in Somalia, most
notably with the warlord Mohammad Aideed. Nonetheless, it appears that the executive’s
position was taken in good faith, mimicking the lack of policy recognition in both
political branches that the United States was in fact engaged in daily combat and not
simply a string of snatch operations to seize Aideed and his compatriots. On October 3,
1993, ground truth came to Washington. Whether one accepts this constructive
“intermittent hostilities” argument or not, one detects with Somalia the beginning of an
operational code reflective of longstanding executive assertion. Whether constitutional or
not, the sixty-day clock is not in any event factually in play unless U.S. forces are
engaged in significant and sustained combat, or as some lawyers might argue in a world
“war.”
This operational code was affirmed during the seventy-eight-day Kosovo air
conflict where the president authorized, and executive lawyers ratified, more than sixty
days of continuous air combat in and over the Former Republic of Yugoslavia. A
subsequent lawsuit by thirty-one members of Congress challenging the conduct of the air
campaign beyond sixty days was dismissed for lack of standing (Campbell v. Clinton).
The executive branch argued that legislation passed expressly and solely to fund the
campaign constituted authorization for the purpose of the War Powers Resolution.
Acknowledging that the War Powers Resolution states that funding authorizations and
appropriations shall not constitute authorization for the purposes of the Resolution, the
executive branch argued that one Congress could not dictate the means by which a
succeeding Congress exercised its war power.26 In light of the Resolution’s language one
surmises that pivotal lawyers within the administration found this argument persuasive,
were disingenuous in their application of the law, or held the view that the sixtyto-ninety-
day clock was unconstitutional, at least as applied to the Kosovo conflict. In any event,
the president never affirmatively expressed a public view on constitutionality. Neither did
he certify to the Congress that an additional thirty days was necessary for a safe
withdrawal. The air campaign ended twelve days short of the ninety days arguably
permitted under the Resolution.
In review, five observations about the sixty-day clock emerge. First, most debates
about the war power start with the Constitution, pass through the statutory prism of the
War Powers Resolution, and eventually circle back to constitutional theory. The war
power debate is irresolvable. For each enumerated presidential authority there is as well a
congressional authority. Moreover, each political branch has an inherent incentive to
assert the broadest possible authority (in the case of the president to act unilaterally) and
to eschew legal concessions, out of concern that policy “precedent” will in some manner
serve as future concession in a different context. Further, presidents will do what they
believe necessary in the interest of national security. Therefore, presidents and their
lawyers unsure of what the future may bring will rarely, if ever, concede a limit to the
president’s authority to act in military defense of national security. This will surely
remain so in the context of a threat of WMD attack by terrorists or irresponsible nuclear
weapon states.
That being said, case law does instruct regarding the constitutional framework.
Youngstown in particular offers two timeless lessons. As the paradigm states, where the
president acts pursuant to congressional authority, he acts at the maximum of authority,
combining his authority with that the Congress can expressly or implicitly delegate.
Similarly, where the president acts in the face of congressional opposition he acts at the
nadir of his authority. This is an intuitive truism, but it is a constitutional principle that
can be lost in the heat of constitutional combat. Thus, even where the president may
(rightly) argue that he does not need legislative authorization (as executive lawyers are
invariably prepared to argue based on the theories presented above), he may nonetheless
benefit from its existence. A president that acts pursuant to Justice Jackson’s first
category will have flexibility to act in unintended ways and is more likely to sustain
public support in the face of setbacks and casualties. In the case of a sustained conflict, or
a failed conflict, the assent of both political branches adds legitimacy to military action.
These policy benefits, of course, must be weighed against the risk of seeking, but
not receiving, congressional authorization and thus confronting a situation where the
president is operating at the ebb of his authority rather than in the twilight of a
Youngstown category two where Congress is silent (or more likely votes to support the
troops, but not the president’s policy). Moreover, as a matter of policy, such a category
three circumstance may have the same practical effect as a funding cut-off by making the
use of military force politically untenable. Second, in practice the congressional war
power is not self-executing. In contrast, the president has an affirmative responsibility as
commander in chief to defend the country and to conduct foreign affairs. Thus events will
compel presidential response; Congress’s authority must be affirmatively exercised.
Again to Justice Jackson, “We may say that power to legislate for emergencies belongs in
the hands of Congress, but only Congress itself can prevent power from slipping through
its fingers.”28 That is what the 93rd Congress sought to accomplish with passage of the
War Powers Resolution, and its theoretically self-executing clock.
Third, a broad reading of presidential authority is not a modern response to world
war, cold war, the nuclear age, or potential WMD terrorism. Recent assertions of
authority represent a continuum of precedent with the past. What has changed is the
operational means of attack and defense, not the constitutional practice. In this sense, the
naval wars of the eighteenth and early nineteenth centuries might parallel the use of air
power in support of U.S. interests in later centuries. The WPR itself recognizes that the
president has independent authority to use force for certain purposes, including in cases
of armed attack. Moreover, whatever the president’s authority is, even in the view of the
93rd Congress, this authority must exist for at least sixty days, for the Resolution states
that it “shall not be construed as granting the president authority he would not have had in
the absence of this joint resolution.” Fourth, the War Powers Resolution seeks to delimit
what the Constitution has left unanswered, and do so in a quantitative manner, for which
there is no textual support. Thus, while the Congress clearly possesses “war power,”
found for example, in the appropriations power, as well as the textually exclusive
authority to “declare war,” nowhere is there a temporal delimitation to these terms.
Moreover, in the context of the founders’ time, a sixty-day limitation on the president’s
exercise of a shared, or unilateral executive authority, would have been absurd, for any
act of projecting U.S. military force beyond U.S. borders, that is, by sea, would
necessarily have taken more than sixty days to accomplish.
Finally, even if the Resolution’s clock is constitutional in some natural law sense,
the Resolution will never control the constitutional outcome, unless the president applies
it himself. But the executive branch has not felt bound by the sixty-day clock, either as a
matter of law or policy. That will surely not change in the face of a WMD terrorist threat
and the advent of new and unstable nuclear powers. As Justice Jackson observed,
presidential advisors may not be able (or willing) to define the scope of the president’s
powers as commander in chief, but they certainly “would not waive or narrow it by
nonassertion.” Presidents, who bear the burden and singular responsibility to protect the
United States, feel the same way.
Even were the clock to find its way substantively into court, it is doubtful a court
would reach too far in its substantive analysis. The clock is a generalized and inchoate
expression of one Congress’s constitutional view that is intended to bind the
constitutional view of a future Congress. Moreover, the clock would apply in a context
where the president will have acted in response to specific and articulated facts. In law,
the rules of statutory interpretation generally favor reading seemingly competing statutes
in a manner recognizing that specific language controls general language.31 Similarly, in
constitutional context, where competing constitutional claims are at issue, specific
applications of particularized powers have controlled generalized expressions of
legislative or executive power. This was true, for example, in the case of the spending
riders in place during the Mayaguez incident.
In those few cases where the Supreme Court has in fact balanced competing
constitutional claims it has applied the same principles. Thus, in Nixon the Court held
that the Watergate grand jury’s particularized investigative need outweighed the
president’s generalized concern about maintaining a deliberative process privilege.
Likewise, the Resolution does not address the specific military and foreign affairs context
in which the clock may actually toll. Thus, in constitutional balance, a Congress from the
past will have spoken, but with a soft inchoate voice. In contrast, the president will have
spoken with a current voice in a specific national security context, with specific foreign
policy and military consequences.32 The result is two expressions of constitutional view,
the one inchoate and the other specific to circumstance and national security need.
B. International Law
The law of armed conflict is based on textual instruments, most notably the Hague
Conventions, the Geneva Conventions, and Protocol I, as well as customary law.33
(Customary international law “consists of rules of law derived from the consistent
conduct of States acting out of the belief that the law requires them to act that way.”34)
Protocol I, for example, reflects both treaty text and customary international law.
Whereas the Parties view the Protocol as textually binding, the United States, which is
not a Party, views many of the Protocol’s statements as indicative of customary
international law. In substance, the law of armed conflict addresses, among other topics,
the resort to force, the use of force, the treatment of prisoners of war (POWs), the rights
of belligerency, war crimes, the protection of civilians, and neutrality. The law does so by
combining general principles (e.g., minimization of suffering) with specific tenets (e.g., a
prohibition on “perfidy”) and absolute prohibitions, such as that on the use of poison gas.
On a given basis, lawyers and policymakers must know and master specific provisions of
the law. For example, the “war on terrorism” has generated numerous issues and debate
regarding “detainees,” “unlawful combatants,” and “prisoners of war,” and more
generally, the applicability of the law of armed conflict to persons captured in a conflict
involving nonstate actors. The descriptive nomenclature chosen to describe these issues
may itself convey conclusions of law. These are essential questions, which have an
impact directly on the success of U.S. offensive and defensive operations, as well as on
perceptions about the values the United States projects overseas, and thus the success or
failure of U.S. public diplomacy.
Generally speaking, there are four widely recognized bases on which a state might
lawfully use military force: (1) self-defense and collective self-defense; (2) anticipatory
self-defense; (3) protection of nationals; and, (4) Security Council authorization.
Qualification is necessary because in practice, governments, practitioners, and academics
differ on their method of description and thus on the number of bases in law for using
force. Protection of nationals, for example, might be subsumed within the concept of
selfdefense or represented as a distinct legal basis. This is true of regional authorization
as well, which for reasons explained below, is included under the rubric of Security
Council authorization. Further, the United States has asserted a right of preemption.
However, there is insufficient clarity or continuity in U.S. practice to determine whether
preemption warrants treatment distinct from anticipatory self-defense, as the 2003
invasion of Iraq would suggest, or whether preemption is anticipatory self-defense
adapted to the WMD threat, as Al-Shifa might suggest. In any event, to the extent
preemption differs from anticipatory self-defense, for example, in the predicate
requirement of imminence, the doctrine does not (perhaps yet) reflect international law.
Scholars, practitioners, and governments have long recognized a right of self-
defense and of collective self-defense, in the case of requests for assistance where the
requesting state has a lawful basis to act in self-defense. Classic U.S. illustrations of the
exercise of the right after the UN Charter’s adoption in 1947 include the invasion of
Afghanistan in response to 9/11 and the 1986 U.S. air strikes on Libya following the La
Belle bombing in Berlin, which targeted U.S. service members. However, scholars and
governments have also long debated the threshold for resorting to force in self-defense in
light of the UN Charter’s limitation on the threat or use of force.36 By extension the
same is true of the right of collective self-defense where one state requests the assistance
of another in defending itself. The scope of the right takes on heightened importance in
light of the threat that jihadist terrorists may obtain and use weapons of mass destruction
as well as the advent of additional and dangerous governments obtaining nuclear
weapons.
The critical terms are “inherent” and “armed attack.” For lawyers embedded in
textual interpretation, this is critical text, for if there was an inherent right of self-defense
before the Charter, the Charter arguably could not have extinguished that right even as
the Charter seeks to limit that right to instances of armed attack. Two related issues arise.
First, must an actual armed attack occur before a state may act in lawful self-defense, and
if so, what constitutes “armed attack?”37 Second, must a state wait for an attack to occur
before defending itself, or does the inherent right of self-defense found in customary
international law include a right to defend in anticipation of an armed attack? For
lawyers, debate over the meaning of “armed attack” centers on the International Court of
Justice’s decision in Nicaragua v. United States (1986). In the case, the government of
Nicaragua sued the United States on the grounds that the United States had violated
international law – including the territorial integrity of Nicaragua – by providing arms
and training to the Contras and by mining Nicaraguan harbors in 1983.38 The United
States defended on the grounds that its actions were taken at the request of El Salvador
and in the collective self-defense of El Salvador. Prior to any U.S. activities, El Salvador
was the subject of cross-border incursions by Sandinista forces. Indeed, Nicaragua was
supporting forces within El Salvador seeking to overthrow the elected government in San
Salvador.
For those who seek clarity in law and find comfort in text, “armed attack” is a
seemingly attractive threshold. “Armed attack” has evidentiary grain. It is, in theory,
apparent to the world, and does not depend on subjective judgments about potential risk.
For lawyers, it is also the nomenclature used in the Charter and in existing international
“case law”; in other words, in those limited manifestations of international law found in
text. But as Nicaragua illustrates, the threshold is not as clear as one might presume in
practice. Moreover, it is not reflective of operational law, because it fails to account for
customary law and state practice. In particular, the ICJ’s 1986 characterization of the
factual predicate for armed attack is inconsistent with state practice in responding to
terrorism before Nicaragua and certainly afterward. States, including the United States,
have asserted a right to respond in self-defense to singular acts of “terrorist” violence.
Although total in reach and final for their victims, many of these incidents are clearly less
significant threats to the territorial integrity and political independence of the attacked
states than was an armed insurgency intended to overthrow an elected government.
Second, the court’s approach did not squarely address the evolving doctrine of
anticipatory self-defense.
Long before the Charter, let alone September 11, states recognized in military
doctrine and law a need to preempt imminent attack, and in some cases the possibility of
attack, rather than await the confirmation of armed attack. This is conceptually
illustrated, for example, in the war plans of the European alliances prior to World War I.
The German Schlieflen Plan, and those of other nations, was triggered not by actual
attack, but by indications of the mobilization of national armies that might attack. The
nature and necessity of reserve mobilization and the dependence on train transport to
reach tactical and strategic positions meant that states felt compelled to respond to
mobilization with countermobilization. The cascading effect resulted in armies
anticipating the need to defend not necessarily based on concrete intelligence of hostile
actions or intent, but based on mobilization necessities.39 For without
countermobilization there might be no opportunity to defend. Of course, the mobilization
itself might in turn confirm hostile intent, leading to a circular march toward war. For
American lawyers the study of anticipatory self-defense usually starts with Secretary of
State Daniel Webster’s response to the Caroline incident of 1837. The Caroline was a
private U.S. merchant ship used by U.S. sympathizers to run arms and supplies to
Canadian rebels.40 The supplies were shuttled to Navy Island located in the middle of the
Niagara River where the rebels had retreated and were regrouping. During a lull while the
ship was moored in New York, a British raiding party crossed the Niagara, set the
Caroline on fire, and sent the vessel over Niagara Falls. Two Americans were killed in
the process. The raiding party then withdrew to Canadian soil.
Here Webster identified the essential and related elements of anticipatory self-
defense: imminence, necessity, and proportionality. Indeed, regardless of the predicate
justification for resorting to force, under international law, the use of force must be
necessary and proportional in relation to the conduct addressed. These terms are not
authoritatively defined, and scholars and practitioners continue to debate their meaning as
applied. Indeed, lawyers generally agree that Webster’s formulation is too restrictive,
placing too much emphasis on the immediate, near instantaneous, nature of the threat.
This is certainly true with the advent of modern weapons like ICBMs and secret weapons
like WMD, where lack of knowledge of the need to defend may well prevent any
prospect of effective defense.
Illustrated in the context of the Caroline, the British response was arguably
necessary and proportional. On the one hand, because peaceful remonstration to U.S.
authorities regarding violations of Canadian territory went unheeded, military action was
necessary. The response was also arguably proportional, because the use of force was
limited to the destruction of the offending vessel; an invasion of New York, on the other
hand, would have been disproportionate to the predicate offense. On the other hand, the
U.S. actors had not directly attacked Canada nor manifested intent to do so. Forecasting
the ICJ’s later Nicaragua opinion, the crew of the Caroline had not crossed the threshold
of “armed attack.” They had supplied those who would do so in Canada; in doing so they
may have had commercial as well as ideological reasons. As importantly, the British
arguably could have accomplished their goal through lesser means by increasing the
pressure on Washington to stop its citizens from interfering in Canadian affairs or by
disabling the vessel and not by killing the Americans on board. Today, the concept of
anticipatory self-defense is generally accepted as black-letter law by most governments
and scholars, notwithstanding Nicaragua. Moreover, the elements are generally agreed
upon: an imminent threat of attack, a necessity of responding with military force to
prevent the attack, and a resort to force that is proportional to the anticipated threat or to
effectively deter the attack. The “classic” post-Charter example of anticipatory self-
defense remains the 1967 Arab-Israeli Six-Day War. The government of Israel correctly
assessed that the combined armies of Egypt, Syria, and Jordan were preparing to invade.
Israel struck first, destroying much of the Egyptian Air Force on the ground as well as
securing the Golan Heights and the Sinai.
Legal judgments depend on factual predicates. Determinations regarding the
necessity and proportionality involving resort to force are contextual. They also entail
judgments regarding the expected behavior of the recipient state. The United States has
long held that such judgments must be made and evaluated in the context of historical
practice. Where an equivalent measure of force may deter one actor, another actor may
demonstrate over time that only a magnification of responsive force will terminate the
unlawful action, a point demonstrated repeatedly by the actions of dictators like Hussein
and Milosevic. Therefore, lawyers evaluating policy options resorting to force must
understand and apply the policy and intelligence judgments influencing policy options
and not just abstract law. If policymakers believe a symbolic show of force (for instance,
a fly-by) will accomplish the permitted goal, a lawyer will find it difficult, applying the
principle of necessity, to concur in a significant use of force, such as the bombing of
national-level military targets in a capital city. These judgments may be particularly hard
to make in the context of anticipatory acts of self-defense, where the threat may be ill
defined, inchoate, or unstated, but nonetheless instant and sudden if realized, as in the
case of a WMD threat. These judgments are also difficult in an asymmetric terrorist
context, where terrorists do not resort to ordinary military methods of command,
mobilization, and attack, making it harder to discern the moment at which an attack is
imminent and to discriminate between responsible actors and civilians in response. Thus,
for lawyers, judgments about proportionality and necessity are hard to reach in the
abstract without an appreciation for the policy context, policy views, and factual context.
Sound national security process should therefore include a meaningful opportunity for the
national security lawyer to engage policymakers and intelligence officials on the facts to
inform judgments about the law.
The ICJ’s threshold for armed attack, to the extent it ever accurately reflected
customary international law, is hopelessly outdated when a single vector might carry the
smallpox contagion, or a suitcase-sized nuclear device could kill hundreds of thousands
of people. The time to react and defend is not clear. There are no mobilization train
schedules to watch and to warn. In this context as well, Secretary Webster’s
characterization of the predicate for exercising the right of anticipatory self-defense, “no
moment for deliberation,” seems firmly planted in the nineteenth rather than the twenty-
first century. Neither the ICJ nor the Charter, and surely not Daniel Webster, anticipated
the possession of weapons of mass destruction by nonstate actors. This dynamic compels
states to respond to indicators of intent and possibilities, as opposed to deeds of action.
The risks of mass casualties preclude waiting for confirmation of armed attack. Where
Webster had years to formulate his positions before transmitting them by letter across the
Atlantic, lawyers and policymakers today may literally have minutes to do the same as
they react to inchoate intelligence indicators.
The United States has sought to address this new threat in legal practice and
doctrine. This evolution began in the mid-1990s when the U.S. government determined
that it would apply not just the tools of law enforcement against the Al Qaeda threat but
also the law of armed conflict, including the right of anticipatory self-defense. As noted
earlier, this legal determination did not become public until after the 1998 Embassy
bombings and the subsequent U.S. response. In August 1998, the United States conducted
missile strikes against targets in Afghanistan intended to disrupt Al Qaeda by killing its
command, including Osama Bin Laden. The strikes were described, and defended using
the nomenclature of self-defense and anticipatory selfdefense, not law enforcement.43 As
important to the development of the law as this paradigm shift was the change in actual
U.S. practice. Concurrent with the U.S. strikes against Al Qaeda in Afghanistan, the
United States attacked and destroyed the AlShifa pharmaceutical plant in Khartoum,
Sudan.
Like Israel at Osirik, the United States could not hope to pinpoint the moment at
which the plant would produce viable chemical weapons. Neither could the United States
be confident it would detect the time and place where weapons or precursors might be
transferred to third parties. Once in third hands, the United States could not track the
weapons to determine in what manner they might be used. Thus, while the intelligence
picture was incomplete, depending in part on information and in part on intelligence
judgment, the security syllogism was complete. Al Qaeda had attacked the United States
before and vowed to do so again. The United States had information that Al Qaeda was
seeking chemical weapons. The United States possessed intelligence indicating, but not
confirming, that Al-Shifa was the site of chemical weapons activity. The United States
had information linking Osama Bin Laden to the Sudanese regime and which the DCI
judged linked Bin Laden to the Al-Shifa plant. From the standpoint of national security
decision-making the president’s choice was evident, and more so today, than at the time;
the intelligence judgment less so. However, the U.S. legal message was lost in part
because of variances in U.S. statements explaining the strikes as well as the
corresponding skepticism regarding the quality of the intelligence linking the Al-Shifa
plant to chemical weapons and to Bin Laden. As a result, it is hard to tell whether the
absence of legal objection reflected a degree of state and scholarly acceptance of the U.S.
legal argument, or whether it merely reflected that the focus of criticism was on the
intelligence underpinnings behind the strike and lingering doubts that the United States
had struck a civilian target.
What do Al-Shifa and Iraq tell us about U.S. legal policy, if anything, at this
time? First, there is continuity between Al-Shifa and Iraq. Both uses of force were
directed (at least in part in the case of Iraq) at preventing terrorists from obtaining
weapons of mass destruction. In both cases the U.S. action was predicated on intelligence
judgments rather than factual certainties, and in both cases the intelligence predicates
were subsequently put into question. However, there are differences as well in
nomenclature and perhaps in the application of imminence. In the case of Al-Shifa, for
example, the U.S. government held the view that the potential transfer of chemical
weapons could be imminent in the traditional sense of the word. In the case of
“preemption” the role of imminence is uncertain. The president’s 2003 State of the Union
Address seemed to suggest that imminence had been dropped from the legal equation
altogether.
Not surprisingly, after Iraq the preemption doctrine as a legal and policy prescript
for force has been pronounced both dead and alive. Some argue, with hindsight, that the
absence of WMD weapons in Iraq undermines the validity of a preemption doctrine.
Certainly, the Iraq war has undermined public and international confidence in the U.S.
capacity to accurately apply the doctrine, or perhaps alternatively, the capacity of the
policy decisionmakers to effectively use intelligence in doing so. That is a matter of
perspective. Doctrine or not, legal concepts embedded in the concept of preemption are
here to stay. First, as a synonym for anticipatory self-defense, preemption has always
been part of the fabric of international law and U.S. legal policy. Second, whatever one
calls the legal principle, after 9/11, no president will knowingly risk a WMD strike
against America or an ally because they failed to act on incomplete intelligence that such
an attack might occur. This trend was set in 1998, and it was repeated in 2003. It will
continue. As Dean Acheson reminded, “The survival of states is not a matter of law.”
Moreover, because sources of intelligence information are usually sensitive the
United States and Israel may be loath to make their full case in public, with specific data.
As an illustration, contrast the reaction to the U.S. strikes on Libya in 1986 with the U.S.
strike on Al-Shifa in 1998. In the former case, the United States identified a specific
source of signals intelligence plainly demonstrating Libyan culpability in the predicate
attack in Berlin. In the case of Al-Shifa, and U.S. concern regarding an ongoing WMD
threat from Al Qaeda, the United States was not prepared to put its full case on the table
of public opinion. The United States was left to express conclusions of fact and
judgments, but without the underlying sources of information. After Iraq it remains
uncertain whether states will assert a right of “preemption” or return to the vernacular of
anticipatory self-defense, while leaning forward in doing so. What is certain is that the
real and potentially catastrophic WMD threat will continue to put new stress on old and
theoretically settled constructs involving the right and scope of self-defense. The United
States will continue to wrestle with the concept of imminence, with each president
adopting and applying his view of the term in light of the intelligence presented and his
perception of the threat.
Indeed, such a right has been recognized since the time of Hugo Grotius, the
Dutch scholar whose seventeenth-century treatise is considered the baseline of modern
international law.57 Additional historical examples of the protection principle applied
include Grenada (1983), the siege of the International Legation in Peking (1900), and the
numerous instances recorded in War Powers reports involving noncombatant evacuations
from U.S. Embassies overseas. In addition, the Israeli operations to rescue imperiled Jews
in Ethiopia, Sudan, and Yemen during Operations Moses (1984), Joshua (1985),
Solomon (1991), and Magic Carpet (1950) arguably fall within this rationale. In theory,
such interventions do not threaten the territorial integrity or political independence of the
state in question because the goal is solely protective. In practice they need not do so.
Significantly, the right is predicated, like other uses of force, on application of the
principles of proportionality and necessity. Thus, the host state must be unwilling or
unable to provide the necessary protection itself. The concept is also subject to false
claim, as in the case of Nazi Germany’s seizure of the Sudeten Land to “protect” the
German population living there. The Serb interventions in Bosnia (1992) and Croatia
(1991) further illustrate the capacity for false assertion. Thus, an observer might test the
credibility of the protective claim by asking whether a state asserting such a right would
accept a reciprocal application on its own territory.
In the Iraq context in 1998 and 2003, the issue was one of competence to revive
“the all necessary means” language of UNSCR 678 (1990) following the 1991 Gulf War
cease-fire (UNSCR 687). In the case of Operation Desert Fox (three days of air strikes
against Iraq in December 1998), the United States argued that Iraq’s material breach of
the Gulf War cease-fire – the expulsion of UN weapons inspectors – reactivated the “all
necessary means” language. Under the law of armed conflict a material breach of a cease-
fire permits the party offended to resume the use of force. Thus, the critical questions
were (1) whether Iraq’s actions amounted to a material breach; and (2) whether such a
judgment was subject to determination by an individual state (or a subgroup of those
specifically offended, e.g., coalition members), or required the judgment of the Security
Council, which authorized the use of force in the first instance. In the latter case, then the
“reauthorization” of force would be subject to veto by a permanent member.
This same material breach argument was cited by the United States prior to the
2003 invasion of Iraq. The U.S. government also argued express authorization in the form
of UNSCR 1441, which states that “serious consequences” would result from Iraq’s
continued breach of UNSCR 687. In context, the U.S. government argued, “serious
consequences” had supplanted the normal nomenclature of “all necessary means” to
reauthorize the use of force. Other states, notably France, took the view that “serious
consequences” had been used in 1441 precisely because it did not trigger the use of force
and in any event the United States undertook to return for express consideration of
Security Council authorization to use force.59 This was a technical argument among
lawyers. On the world stage, as discussed earlier, the United States asserted the right to
preempt based on the potential threat posed by Iraq’s potential production of weapons of
mass destruction and the possibility it would pass those weapons to terrorists. There is
insufficient state practice, and certainly insufficient recognition of that practice, to
suggest that an operational code of tolerance presently exists to assert UN authority to use
force outside the “necessary means” language, absent a Security Council understanding
that in context other words carry the same meaning. However, Iraq resolutions
subsequent to 9/11, in particular UNSCR 1441, have implicitly opened the door to an
expanded vernacular of UN authorization to use force to include “serious consequences.”
On the other hand, there are advantages to a confirmed vocabulary of force.
Changes in vernacular from circumstance to circumstance will leave open the possibility
for misunderstandings regarding Security Council intent and the intent of the relevant
parties. At the extreme, if all words mean all things to all parties, then the value of
Security Council authorization and responsibility will diminish. For the United States,
such diplomatic sophistry cuts both ways. It may allow the United States to argue
Security Council authorization in gray contexts, but it may also dilute the importance of
the U.S. veto and result in dangerous reciprocal claims of authority. Moreover, members
of the Security Council may be hesitant to ratchet up the pressure on recalcitrant states
with increasingly robust resolutions if member states are concerned that states will use
such language as authorization to use force. The Council might choose alternative
language to authorize force. However, at this time, absent clear Council intent, it is hard
to argue that authorization can be assumed or implied from alternative language. As a
distinct matter, states may assert UN “sanction” when they act consistent with Security
Council resolutions calling for state parties to respond to a crisis or calling on states to
take action in response to a threat to international peace and security. But this is not
necessarily equivalent to language authorizing “all necessary means.” In such cases, the
United States has cited language in Security Council resolutions to justify use of military
force as “consistent” with Security Council resolutions, and thus “the Purposes of the
United Nations,” while not necessarily asserting Council authority in doing so.
States, including those most dedicated to the rule of law, have also asserted a right
to use force outside traditional constructs, when in their view a totality of contextual
circumstances justifies resort to military force. In the case of Panama, Grenada, and
Kosovo, for example, the United States pointed to a series of factors, or totality of
circumstances, justifying the use of military force. In each case the United States
referenced principles of traditional doctrine, such as collective self-defense in the case of
Kosovo and the protection of nationals in the case of Panama and Grenada. However, in
arguing a totality of other circumstances, the United States was effectively
acknowledging that the degree of force used exceeded that which was necessary and
proportional to protect U.S. nationals alone. The U.S. military intervention in Somalia, of
course, was also initially based on Council authorization on humanitarian grounds, but as
in the case of Lebanon, the mission evolved into hostile combat operations as the initial
legal basis drifted astern.
In the case of Lebanon, the invitation of the parties to the civil war to evacuate the
PLO provided a lawful basis for the United States to enter Beirut, albeit as peacekeepers.
However, after the initial successful withdrawal of the PLO, the American forces found
themselves drawn into the conflict for a variety of reasons including the necessity of
force protection. In Somalia, the United States intervened, consistent with Security
Council resolutions authorizing member states to address the ongoing humanitarian
catastrophe. Here too the mission evolved into hostile combat operations, and away from
the original international legal basis. From these interventions emerged the concept of
“mission creep”: the transformation of a military mission (gradually and generally below
the radar screen in Washington, hence “creep”) from one of limited scale and duration to
one of combat or “nation building” of indefinite duration. For lawyers the concept is
important, because the mission may move away not only from the original military
concept of operations but also from its legal underpinnings. This may have constitutional
implications if the president has provided only limited authorization, or the Congress has
delimited its funding of the operation in a relevant manner. It may also have international
consequences if allied support is predicated on the application of particular legal
doctrines. It may also undermine U.S. credibility if the U.S. basis for action under
international law does not appear to comport with events on the ground.
After the Kosovo conflict, the British government, humanitarian NGOs, and some
academic observers asserted that the conflict established as a matter of customary
practice a lawful right of humanitarian intervention. Moreover, in reviewing the conduct
of NATO’s air strikes, the International Criminal Tribunal for the former Yugoslavia
(ICTY) prosecutor concluded that NATO’s resort to force was based on humanitarian
intervention, but did not ultimately render an opinion on the resort to force, the
prosecutor judging that her mandate extended only to the methods and means used.63
This sentiment was also reflected by UN Secretary-General Annan, who espoused a
“developing international norm in favour of intervention to protect civilians from
wholesale slaughter.”64 With respect to Kosovo, the Secretary General said: “It is indeed
tragic that diplomacy failed, but there are times when the use of force may be legitimate
in the pursuit of peace.” This statement can be viewed as supportive not just of
humanitarian intervention, but also of alternative theories of legal authority outside the
Charter context. A parallel debate occurred within the U.S. government behind not very
closed doors as to whether the United States should, after the fact, recognize
humanitarian intervention as operational customary international law, not just as a policy
prescript.
If the president may lawfully resort to force under U.S. law,68 policymakers and
lawyers must also ensure the conduct of military operations is lawful because the law of
armed conflict is disjunctive in application.69 Lawful resort to force does not inherently
make lawful the means and methods used in applying that force. For example, the
principles of proportionality and necessity apply to the resort to force and as a distinct
matter to the methods and means of force. Moreover, in contrast to some areas of
international law that are soft in application (arguably including the law regarding the
resort to force) the law regarding the methods and means of warfare is “hard,”
operational law. It is reflected in international treaty text, customary international law,
and in U.S. domestic criminal statutes. It is also subject to U.S. punitive sanction, foreign
state punitive sanction, and, on a more episodic basis, international punitive sanction.70
As a result, the president as commander in chief not only has a duty to use force
effectively in the interest of U.S. national security, but to do so in a manner that “take[s]
Care that the Laws be faithfully executed.”71 Adherence to the law of armed conflict
(LOAC) is also longstanding U.S. policy, regardless of military context. Indeed, the law
is good national security policy and good military policy. However, U.S. perspectives on
the scope of the law and its application in legal policy are evolving in the face of the
enduring terrorist threat.
One function of the lawyer is to identify the critical legal policies at issue and to
ensure that their objectives are likely to be realized in a new context by an appropriate
and lawful application of force. This requires a process of meaningful and timely legal
input to decision-making. It also requires not just an understanding of the law but an
understanding of how the law often reflects good military policy. Adherence to the law
improves the prospects of, but of course does not guarantee, reciprocal application of the
same principles by one’s opponent. More broadly, as discussed in Chapter 3, the moral
authority of the United States to espouse the rule of law is founded in part on its
consistent and faithful adherence to the law. This is true in military operations as well.
Adherence to the law helps to garner and then maintain international support
(governmental, elite, and public) for US military operations. In the case of the 1999
NATO air campaign against Serbia application of the law was a sine qua non for the
NATO political consensus necessary to authorize NATO military operations.
Ground operations also tend to be more fluid than aerial operations against pre-
planned or fixed targets, at least where the United States maintains air supremacy or
superiority. Even when aerial targets are emergent, there is often some time for command
consideration as aircraft or weapons platforms move into position. In ground combat
there are fewer fixed targets, and emerging targets usually require immediate response.
Thus, there is rarely opportunity for senior commanders, including in context the
president, to apply policy and law to targets other than through the provision of
generalized guidance.
C. Constitutional Chain of Command
As a matter of U.S. law, military decisions must not only be made in accordance
with the proper substantive law but they must also be made by the appropriate person or
persons. Moreover, the appropriate person or persons must authorize not just the use of
force but also the manner and method in which force is used. These decisions are made
by, and then implemented through, the military chain of command. As a result, the
national security lawyer must understand the military chain of command to understand
where to provide meaningful advice and how to better appraise the operation of law and
consider if the process or practice warrants adjustment. That means the military authority
to conduct operations originates with the commander in chief. This authority is
transmitted to the war fighter through the chain of command. This can be accomplished
by the president delegating authority and responsibility to the secretary of defense or to
subordinate commanders. The president might also provide authorization in contingency
form, or through case-specific direction.
As a matter of constitutional principle, presidents have reserved the right to
exercise their responsibilities as commander in chief as they see fit, rather than as
legislated by the Congress. Where reasonable persons might disagree on how much “war
power” the president might derive from the commanderin-chief clause, there is no
reasonable debate that the president is the commander in chief, and the essence of
military command is the conveyance of lawful orders from commander to combatant
soldier. In theory, and sometimes in practice, the president can exercise command by
directing individual pilots,93 or destroyer commanders,94 or in a less tactical manner by
providing his commander’s intent to the secretary of defense and delegating the
responsibility for implementation to the secretary, who may himself delegate further, so
long as the president has not directed otherwise.
At the national level, military input is generated in a myriad of formal and
informal ways. Through the inter-agency process the Joint Staff will represent the
chairman and JCS at inter-agency working groups or through the endless process of
clearing cables, talking points, and speeches in the context of operational commitments,
crisis management, and policy development. The Joint Staff also prepare and update
contingency plans covering a full spectrum of potential scenarios. As a practical matter,
much of the chairman’s advice comes informally at Principals Committee meetings, in
NSC meetings, and in parallel phone calls on secure lines that mimic the more formal
process of principals and deputies meetings. In addition, the chairman’s advice is often
rendered through the secretary of defense to the president or the NSC principals.
Proponents also argued that a panel of equal chiefs sometimes resulted in equal
assignment of mission responsibility to each service, as opposed to task-organized forces
based on the specific needs of the mission and specific unit capabilities. The failed Iran
hostage rescue mission of 1980 served as Exhibit One in this regard. Among other things,
the operational force was cobbled together from off-the-shelf units from each service
eager to participate in the mission, rather than from a single unified command integrated
by training, equipment, and personnel. Moreover, a combination of necessary secrecy and
command relationships apparently kept critical operational dissent from reaching the
attention of the president. As a result, the Act also sought to address the perception and
reality that, on the Joint Staff service, loyalty colored staff work and that officers who
lost sight of their service identity were penalized when they returned to the parent
service’s arms. Thus, the Act requires military officers to serve joint tours to remain
competitive for promotion.
Critics of the Act and its implementation argue that it places too much authority in
the hands of the secretary of defense. The secretary is free to overrule the military chain
of command on operational as well as policy questions and to do so in a manner that may
not be evident to the president, or for that matter outside observers. By extension, critics
argue Goldwater-Nichols leaves too little authority in the hands of the professional
military, represented by the chairman and the joint staff. This observation has been made
of both strong and weak secretaries of defense and in time of war as well as relative
peace. In particular critics cite perceptions and realities related to two events. The first
involves Secretary Les Aspin’s 1993 decision delaying the transfer of armor to U.S.
forces in Somalia prior to the incidents portrayed in Black Hawk Down. The second
involves Secretary Donald Rumsfeld’s direction of the second Iraq war, and particularly,
decisions taken regarding the number of troops necessary to seize Baghdad and then
restore and hold order in Iraq. In both cases, as a matter of perception and perhaps fact,
the secretaries in question overrode professional military advice, or as appears more
likely in the second scenario, chose to select between competing advice. The competing
duties of the chiefs under the Act has also prompted criticism that the chiefs spend too
much time on administrative command, and not enough on advising with respect to the
operational matters.
How the chairman (and each service chief) executes his role depends on his sense
of his role as a statutory military advisor to the president and his view of military
tradition in policy and political context. As a matter of law, the chairman is subordinate
to and subject to the direction of the secretary of defense. This is a matter of statute, but
more importantly, of constitutional law, founded in the principle of civilian control of the
military instrument. At the same time, decisionmakers should not overlook that the
chairman serves as the senior military advisor to the president, and thus has a
responsibility to provide military advice independent of the secretary of defense’s views.
This responsibility is specified in the National Security Act and in Goldwater-Nichols. It
is also grounded in common sense. Presidents who do not probe for gaps between the
civilian leadership of the Department of Defense and the military leadership may learn
after that fact that critical points of disagreement subsequently play out on the battlefield
and in the pressroom and not just in the JCS tank. As with intelligence, too much focus
on the law rather than on leadership can detract from the real issues in command
relationship. Whether the command relationship between the secretary and the chairman
and the chiefs is a successful one has more to do with personality, leadership, and style
than the law.
From the combatant commander the chain of command generally flows to
subordinate operational commanders, task organized for each particular mission. Such
task forces typically have a ground element commander, an air element commander, and
a logistics element commander operating under the overall command of the task force
commander. A typical infantry chain of command for a unit deployed in combat might
run like this: commander in chief, secretary of defense, combatant commander, theater
commander, ground element commander, division commander, regimental commander,
battalion commander, company commander, platoon commander, squad leader, fire team
leader, and rifleman. However, it is important to emphasize, once the chain of command
proceeds from the president to the secretary of defense to the combatant commander,
there is no legally required or normative operational chain. Commanders are free to task
organize for particular missions and seek to employ lines of command that optimize
speed, efficiency, and accountability, while avoiding redundancy and delay. Thus, a
combatant commander may truncate the chain of command in the case of a hostage
rescue mission by personally communicating with the ground force element under the
command of a colonel, for example.
Within an operational unit, command and control can be shifted from one
headquarters to another on a tactical (taccon), operational (opcon), or administrative basis
(adcon). For example, if two divisions are advancing on parallel tracks, but one unit is
anticipating armored resistance, the Corps commander with command over both units
might shift a regiment of tanks usually assigned to one division to the other for the
purpose of the pending maneuver. This is a shift in tactical control. If the unit is going to
generally fall under the direction of the new division commander, then operational
control will be transferred. However, if the shift is temporary in nature, then the senior
commander will make clear to both units that administrative control has not shifted, thus
alerting the parent division that it is still responsible for resupplying the tank regiment in
question.
In contrast to other national security processes, the military chain of command is
clear and often crisp. It runs distinctly from the president, as commander in chief, to the
pilot in the cockpit or the soldier in the field. There is but one stop at each level of
command. The chain is transparent, with each member (generally) able to identify the
lawful links. Nonetheless, like other national security processes, the application of the
military chain of command warrants consistent contextual appraisal to ensure that in
operation it is effective and that it is accountable as a matter of policy and law. National
security lawyers should consider five factors in particular when assessing the operation of
the chain of command. There are also risks to vertical command, or better said, too much
vertical command, either because of layering or micromanagement. Vertical civilian
command is less important, indeed potentially disruptive, where the military objective is
set and the concept of operations calls for traditional and rapid maneuver warfare. First,
as the Long Commission demonstrated in the context of the Beirut bombing, vertical
command – in that case involving eleven layers between the president and the Battalion
Landing Team commander – can diffuse responsibility and accountability in dangerous
ways.110 Vertical command can take time and delay critical decision.
Second, where combat operations are fluid, vertical target decisionmaking is
inherently dysfunctional unless it is exercised through a commander’s intent or ROE.
This might be illustrated with reference to weaponized unmanned aerial vehicle (UAV)
platforms that can be deployed both as pointto-point weapons (that is, launched with a
specific coordinate in mind) or used to patrol for targets of opportunity. In the initial
mode, vertical commanders can appropriately participate in a target decision where the
target is pre-planned or fixed. In the latter case, the tactical setting will dictate that
command discretion and the LOAC be applied through rules of engagement or target-
class approval, rather than an assessment of specific target circumstances at the time of
attack. In summary, the constitutional chain of command should be exercised in a
contextual manner that accounts for a range of legal, policy, and military factors in
deciding when and how presidents, secretaries, and military commanders exercise
command and, in doing so, apply the law.
