BRIAN OREND

JUST AND LAWFUL CONDUCT IN WAR: REFLECTIONS ON

MICHAEL WALZER

(Accepted 2 February 2000)

“(T)he structure of rights stands independently of political alle-

giance; it establishes obligations that are owed, so to speak, to
humanity itself. . . ”
“There is no right to commit crimes in order to shorten a war.”

Michael Walzer1

“Jus in bello” is the Latin term just war theorists use to refer to
justice in war, to right conduct in the midst of battle. American polit-
ical theorist Michael Walzer has developed an account of jus in bello
which has been influential and subject to much debate, though not
much resolution. This paper will offer an updated, critical and thor-
ough examination of Walzer’s account, seeking thereby to further
work on the ethics of war and peace. The subject of wartime justice
has enjoyed something of a renaissance in philosophical interest,
no doubt owing to controversial events in such recent wars as the
Bosnian and Rwandan civil wars, Russia’s campaigns in Chechnya,
and NATO’s armed intervention in Serbia over Kosovo.
Walzer insists that jus in bello is a category separate from jus
ad bellum, which concerns the justice of resorting to war. A war
begun for the right reasons is, for Walzer, a war fought in response
to aggression, defined as “(a)ny use of force or imminent threat
of force by one state against the political sovereignty or territorial
integrity of another. . . ”2 Such state rights are themselves founded,
ultimately, upon individual human rights to life and liberty. The
most obvious example of an act of aggression would be an armed
1 Michael Walzer, Just and Unjust Wars (New York: Basic Books, 2nd ed.,
1991), 158, 210.
2 Walzer, Wars, 62.

Law and Philosophy 20: 1–30, 2001.

© 2001 Kluwer Academic Publishers. Printed in the Netherlands.
2 BRIAN OREND

invasion by one state, bent on taking over another, much as Iraq did
to Kuwait in August 1990. But this requirement of just cause, in
terms of resisting aggression, is not the only rule just war theorists
like Walzer insist on being followed prior to the resort to war. They
also stipulate that the war in question must be begun as a last resort,
be publicly declared by a proper authority, have some probability of
success, be animated by the right intention of resisting aggression
and also be expected to produce at least a proportionality of benefits
to costs.3 Walzer reasons that we have not finished our normative
labour once we have determined whether a state has resorted to war
justly, using these principles. For even if a state has resorted to war
justly, it may be prosecuting that war in an unjustified manner. It
may be deploying decrepit means in pursuit of its otherwise justi-
fied end. Just war theory insists on a fundamental moral consistency
between means and ends with regard to wartime behaviour. Since
Walzer views just war theory as the best interpretation of our shared
discourse on the ethics of war and peace, it follows for him that we
must likewise be committed to this core consistency: justified ends
may only be pursued through justified means.4
Concern with consistency, however, is not the only, or even the
main, reason behind our endorsement of separate rules regulating
wartime conduct. Such rules are also required to limit warfare, to
prevent it from spilling over into an ever-escalating, and increas-
ingly destructive, experiment in total warfare. If just wars are limited
wars, designed to secure their just causes with only proportionate
force, the need for rules on wartime restraint is clear. Even though
modern warfare has displayed a disturbing tendency towards totality
– particularly during the two World Wars – it does not follow that the
death of old-time military chivalry marks the end of moral judgment.
As Walzer proclaims, “(w)e still hold soldiers to certain standards.”5
Walzer claims that our shared war convention commits us, at its
deepest level, to three standards of jus in bello. The first is one of
3 For more on jus ad bellum, see: Walzer, Wars, 3–33 and 51–1125; R. Regan,

Just War: Principles and Cases (Washington, DC: Catholic University of America
Press, 1996), 3–86; J. B. Elshtain, ed. Just War Theory (Oxford: Blackwell, 1992);
and D. Luban, “Just War and Human Rights”, Philosophy and Public Affairs
(1980), 160–81.
4 Walzer, Wars, 3–20.
5 Walzer, Wars, 34–35.
 JUST AND LAWFUL CONDUCT IN WAR: REFLECTIONS ON MICHAEL WALZER 3

discrimination: armies are to discriminate or distinguish between

military and civilian targets, and aim their lethal force only at legit-
imate military, and military supply, targets. The second standard
commits armies to launch only proportionate force at these legit-
imate targets. Finally, armies are not to employ methods which are
intrinsically heinous; they may not commit actions which “shock
the moral conscience of mankind.” Walzer is emphatic that we hold
all soldiers to these three standards. Unlike our jus ad bellum judg-
ments, which tend to be binary, condemning one side and justifying
the other for resorting to war, our judgments about right conduct
apply across the board during wartime. Since jus in bello and jus
ad bellum are separate, it is an error to link the justice of conduct
to the justice of cause: soldiers fighting for a just cause can just as
readily run afoul of jus in bello as those prosecuting an unjust war.
We expect no more, no less, from soldiers of all sides than that they
adhere to the three standards of right conduct.6
Before examining Walzer’s understanding of the content of these
three standards, it is worth stressing how for him responsibility
for fulfilling jus in bello differs from the responsibility inhering
in jus ad bellum. Responsibility for the justice of resorting to war
rests on those key members of the governing party most centrally
involved in the decision to go to war, particularly the head of state.
Responsibility for the conduct of war, by contrast, rests on the
state’s armed forces. In particular, responsibility for right conduct
rests with those commanders, officers and soldiers who command
and control the lethal force set in motion by the political hierarchy.
In general, anyone involved in formulating and executing military
strategy during wartime bears responsibility for any violation of the
jus in bello standards, whose content will be specified below.

I. DISCRIMINATION AND NON-COMBATANT IMMUNITY

Walzer insists repeatedly that the requirement of discrimination

is the most important jus in bello rule. Soldiers charged with
the deployment of armed force may not do so indiscriminately;
rather, they must exert effort to discriminate between legitimate
6 Walzer, Wars, 34–49, 127–224.
4 BRIAN OREND

and illegitimate targets. How are soldiers to know which is which?

Walzer answers: a legitimate target in wartime is anyone or anything
“engaged in harming”. All non-harming persons or institutions are
thus immune from direct and intentional attack by soldiers. Since
the soldiers of the enemy nation, for instance, are clearly “engaged
in harming”, they may be directly targeted, as may their equipment,
their supply routes and even some of their civilian suppliers. Civil-
ians not engaged in the military effort of their nation may not be
targeted with lethal force. In general, Walzer asserts that “(a) legit-
imate act of war is one that does not violate the rights of the people
against whom it is directed.”7 In response, one might ask: how is
it that armed force directed against soldiers does not violate their
rights, whereas that directed against civilians violates theirs? In the
chaos of wartime, what exactly marks the difference?
One of the murkiest areas of Walzer’s just war theory concerns
the moral status of ordinary soldiers. His references to them exhibit,
on the one hand, a humane sympathy for their “shared servitude” as
“the pawns of war.” On the other, his references occasionally display
a glib callousness, as when he concurs with Napoleon’s (in)famous
remark that “soldiers are made to be killed.” 8 How can soldiers be
made to be killed when, as human beings, they enjoy human rights
to life and liberty? If it is our shared ideas about human person-
ality which (somehow) justify human rights, how can Walzer claim
that these possessors of personality be directly and intentionally
targeted with lethal force? 9 The answer must be that soldiers do
something which causes them to forfeit their rights, much as an
outlaw country forfeits its state rights to non-interference when it
commits aggression. This is indeed the case for Walzer: “(N)o one
can be forced to fight or to risk his life, no one can be threatened
with war or warred against unless through some act of his own he
has surrendered or lost his rights.” One could be forgiven for infer-
ring from this principle that only soldiers of an aggressor nation
7Walzer, Wars, 42–43, 135.
8Walzer, Wars, 37, 40, 136. James Dubik seems the first to note and explain
this murkiness, in his “Human Rights, Command Responsibility and Walzer’s Just
War Theory”, Philosophy and Public Affairs (1982), 354–371. See also Lackey,
“Theory”, 540–542.
9 Walzer, Wars, xxx; M. Walzer, “The Moral Standing of States”, Philosophy

and Public Affairs (1979/80), 209–229.

 JUST AND LAWFUL CONDUCT IN WAR: REFLECTIONS ON MICHAEL WALZER 5

forfeit their rights, since they are the only ones engaged in the kind
of rights-violative harm which grounds a violent, punitive response.
Interestingly, and perhaps problematically, Walzer denies this. He
believes that all soldiers forfeit their right not to be targeted with
lethal force, whether they be of just or unjust nations, whether they
be tools of aggression or instruments of defence.10
Walzer’s concept here is of “the moral equality of soldiers.” The
first “war right” of soldiers is to kill enemy soldiers. We do not, and
should not, make soldiers pay the price for the injustice of the wars
they may be ordered – perhaps even conscripted – to fight. That,
Walzer emphasizes, is the logically and morally separate issue of jus
ad bellum, which focuses on the responsibilities of political leaders.
But lawyers like the chief British prosecutor during the Nuremberg
trials, and philosophers like Thomas Pogge, ask: why shouldn’t we
hold soldiers responsible for the justice of the wars they fight? If
we held soldiers responsible in this regard, wouldn’t that constitute
an additional bar against aggressive war? Wouldn’t that account for
the fact that, even though the war was set in motion by others,
soldiers remain its essential executors? Wouldn’t that impose and
highlight an important responsibility for soldiers, namely, to refuse
to participate in the prosecution of aggressive war?11
Walzer experiences difficulty in answering this argument fully.
As an opening gambit, he contends that soldiers “are most likely to
believe that their wars are just.” But this alone cannot justify their
actions, since their beliefs may not be well-grounded, especially
considering the incentive they have to believe such justification in
the first place. Walzer also says that soldiers rarely fail to fight,
owing to “(t)heir routine habits of law-abidingess, their fear, their
patriotism [and] their moral investment in the state.”12 But the fact
that soldiers rarely fail to fight does not demonstrate that they are
always justified in fighting, especially if the cause is unjust. Walzer
next suggests that knowledge about the justice of the wars soldiers
fight is “hard to come by.” This is a surprising claim from a just

10 Walzer, Wars, 135.

11 Walzer himself notes the British prosecutor’s arguments in his Wars, 38.
This point has also been emphasized to me in conversation with Thomas Pogge,
philosophy professor, Columbia University in New York City.
12 Walzer, Wars, 127, 39.
6 BRIAN OREND

war theorist devoted to making that knowledge more accessible and

comprehensible. Perhaps, then, this is a reference to the soldier’s
general lack of higher education, as well as to government tenden-
cies towards secrecy. Fair enough, but ignorance at best constitutes
an excuse, and not a justification, for wilfully fighting in an unjust
war: it seems a stretch to assert that such ignorance can morally
ground a “war right” to kill enemy soldiers. Walzer’s subsequent
move appeals to the authority of Vitoria, who suggested that if
soldiers were allowed to pick and choose the wars they were willing
to fight in, the result would be “grave peril” for their country. But
this empirical generalization is speculative: why wouldn’t the result
actually be the preferred one, namely, that states would be seriously
hampered only in their efforts to prosecute an aggressive war which
they couldn’t justify to their soldiery?13
Walzer turns to his conventionalist methodology for assistance
in this regard: as a matter of fact, he suggests, we do not blame
soldiers for killing other soldiers. We blame soldiers only when they
deliberately kill either civilians or enemy soldiers kept by them as
disarmed prisoners of war. We extend to all soldiers caught in the
midst of battle the right to deploy armed force on behalf of their
own country.14 This is a true legal contention, and not an implaus-
ible moral one, but the latter is not so obvious as Walzer suggests.
Do we really believe that those soldiers who fought for Hitler, for
example, were utterly blameless for their bit part in the execution
of his mad aggression? No doubt, we tend to exonerate conscripts
like The Hitler Youth in the closing days of the war, presuming they
were far too young, gullible and propagandized to have made a free
choice. But what about those mature German soldiers who invaded
Poland, or France, at the war’s outset? It is not so clear to me that,
as a matter of fact, we do not blame them for fighting on behalf of
their country.
Walzer stresses more generally the pervasive socialization of
soldiers of any nation, their relative youth, their frequent conscrip-
tion, and their usual background as members of underprivileged
classes as grounds for not holding soldiers responsible for the wars
they fight. While soldiers “are not . . . entirely without volition”, he
13 13. Walzer, Wars, 39.
14 Walzer, Wars, 128.
 JUST AND LAWFUL CONDUCT IN WAR: REFLECTIONS ON MICHAEL WALZER 7

says, “(t)heir will is independent and effective only within a limited

sphere.” This sphere contains only those tactics and manoeuvres
soldiers are engaged in. It would thus constitute unfair “class legis-
lation” for us to hold soldiers like these responsible for the justice
of the wars they fight. We should focus on those most to blame, the
leaders who set the war in motion.15 But from the fact that polit-
ical leaders are mostly to blame for the crime of aggression, does
it follow that they are solely to blame, as Walzer insists? Perhaps
a compelling alternative would be to suggest that, for reasons
Walzer mentions, there should be a presumption against holding
soldiers responsible for the crime of violating jus ad bellum. But
this presumption does not preclude us from concluding, in particular
cases based on public evidence, that some soldiers of a particular
aggressor state either did know, or really should have known, about
the injustice of the war they were fighting, that they could have
refused to participate in it, and thus that they may be held respons-
ible, albeit with lesser penalties than the head of state. Such soldiers
would be like minor accomplices to a major crime. Walzer’s belief
about the absolute separateness of jus ad bellum from jus in bello
cannot, I suggest, be sustained in light of these considerations.
Walzer, in a tantalizing footnote, appears to flirt with a similar
linkage between the two traditional just war categories. With
specific reference to the soldiers of a democratic country engaged in
aggressive war (such as, he says, America during Vietnam), Walzer
stipulates that, as citizens, the soldiers should vote against the war
but, as soldiers, “they fight as members of the political community,
the collective decision having already been made, subject to all
the moral and material pressures. . . ” referred to in the preceding
paragraphs, like their mediocre education, etc. Walzer says that any
soldier with courage enough to refuse to fight such a war does
“act very well” and should be not only tolerated but honoured by
a just society. “That doesn’t mean, however, that the others [i.e. the
soldiers who still fight] can be called criminals.” Why not? Walzer’s
limp reply is to reiterate the socialization of most soldiers and to
comment that “(p)atriotism . . . is the ordinary refuge of ordinary
men and women, and it requires of us another sort of toleration.”
What precise sort Walzer fails to specify, yet it would seem a
15 Walzer, Wars, 40, 138.
8 BRIAN OREND

dangerous kind, as it weakens our condemnation and deterrence of

the very kind of aggressive war Walzer so strongly rejects.16
Even if we agree with the anti-Walzer proposal that some soldiers
may be held responsible for jus ad bellum violations, can we still
concur with Walzer’s idea that all soldiers generally remain legit-
imate targets during wartime? After several false starts, Walzer
offers us a compelling reason to do so: soldiers, whether just or
unjust, are “engaged in harm.”17 They bear arms effectively, are
trained to kill for political reasons and are “dangerous men”: they
pose serious threats to the lives and interests of those they are
deployed against, whether for a just cause or no. Walzer suggests
that an armed man trying to kill me “alienates himself from me
. . . and from our common humanity” and in so doing he forfeits
his right to life. This establishes, I believe, a strong prima facie
case that soldiers targeting other soldiers with lethal force is justi-
fied. Soldiers, whether for just or unjust reasons, remain among the
most serious and standard external threats to life and vital interests
during wartime. Only public, compelling and accessible knowledge
about the injustice of the cause of his own country can undermine
a soldier’s entitlement, in the face of such a threat, to respond in
kind.18
One implication of this general principle – namely, that a target
is legitimate only if it is engaged in harm – is that those who are
not “engaged in harm” cannot be legitimate targets during wartime.
Walzer suggests this is the most plausible, and publicly accessible,
sense of “innocence” in wartime. Its first application has to do with
soldiers themselves: when soldiers no longer pose serious external
threats – notably by laying down their weapons and surrendering
– they may no longer be targeted and should, in fact, be extended
benevolent quarantine for the duration of the war. He squares this
with his remarks about forfeiture thusly: “(T)he alienation [of the
right to life] is temporary, the humanity imminent.” Thus, ceasing to

16Walzer,Wars, 299–300 in the note.

17See Dubik, “Command”, 359–361, for an excellent analysis of Walzer’s
difficulties in this regard. See also Walzer, Wars, 138–160.
18 Walzer, Wars, 142.
 JUST AND LAWFUL CONDUCT IN WAR: REFLECTIONS ON MICHAEL WALZER 9

pose an external threat ceases the forfeiture, and the soldier’s human
rights spring forth intact.19
The second application of Walzer’s harm principle deals with
civilians. Even though some civilians may inwardly approve, or
even have voted in favour of, an unjust war effort, they nevertheless
remain externally non-threatening. They do not bear arms effect-
ively, nor have they been trained to kill, nor have they been deployed
against the lives and vital interests of the opposing side. Civilians
are not in any material sense “dangerous men.” Thus, “they have
done nothing, and are doing nothing, that entails the loss of their
rights.” So they may not be made the direct and intentional objects of
military attack.20 This is clearly controversial. Some thinkers argue
that the fact that civilian taxes fund the military renders null and
void any pretence of their being “innocent.” Civilians are causally
involved in financing the harm soldiers do. Others view nation-
ality as shared destiny, or suggest that modern warfare is totalizing
anyway and so wonder what the point of discrimination really is
in our age. These are not trivial arguments, especially the first
regarding taxation, but they fail to persuade. It is hard to see, for
example, how infants could be anything other than innocent during
wartime. Only the most dogmatic believer in collective responsib-
ility could deny this, and then at the cost of his credibility. There is,
moreover, little evidence that modern warfare is intrinsically total-
izing: the Persian Gulf War of 1991, for instance, did not escalate
into an indiscriminate slaughter. No doubt there are searching ques-
tions about the exact specification of “innocents” in wartime but I
follow Walzer in believing that it remains an important just war
category, needed not only to restrain violence but also to express
our strong moral commitment to punish only those who deserve it.
In the midst of what Clausewitz called “the fog of war”, one of the
most concrete and verifiable ways to cash out such desert is to define
it in terms of external engagement in serious harm.21

19 Walzer, Wars, 142, 46. Both the Hague and Geneva Conventions enshrine

these claims. See W. Reisman and C. Antoniou, eds. The Laws of War (New York:
Vintage, 1994), 149–231.
20 Walzer, Wars, 146–151.
21 D. Lackey, “A Modern Theory of Just War”, Ethics (1983), 540–542; T.

Nagel, “War and Massacre”, Philosophy and Public Affairs (1971/72), 123–144;
10 BRIAN OREND

Owing to these contentions, civilians should be thought of as

“innocent” of the war, and thus entitled not to be made the objects of
direct, intentional attack. That this norm of noncombatant immunity
is, as Walzer claims, the subject of very widespread, cross-cultural
concurrence is revealed by the fact that it is the most frequently and
stridently codified rule in the international laws of armed conflict,
especially the Hague and Geneva Conventions. “Noncombatants”,
Walzer emphasizes, “cannot be attacked at any time. They can
never be the objects or targets of military activity.”22 Difficulties
arise, though, when we consider those people who seem, simultan-
eously, to be both civilians and engaged in harming, such as civilian
suppliers of military hardware. What is the status of such people?
Walzer suggests that “the relevant distinction is . . . between those
who make what the soldiers need to fight and those who make what
they need to live like all the rest of us.” So targeting farms, schools
and hospitals is illegitimate, whereas targeting munitions factories
is legitimate. Walzer stresses, however, that civilians engaged in
the military supply effort are legitimate targets only when they are
engaged in that effort, so to target them while at home in residential
areas would be illegitimate: “Rights to life are forfeit only when
particular men and women are actually engaged in war-making or
national defence.” Walzer agrees with Thomas Nagel’s eloquent
explanation that “hostile treatment of any person must be justified in
terms of something about that person [his italics] which makes the
treatment appropriate.” We distinguish combatants from noncom-
batants “on the basis of their immediate threat or harmfulness.”
And our response to such threats and harms must be governed by
relations of directness and relevance.23
Walzer’s overall judgment on targeting is this: soldiers may target
other soldiers, their equipment, their barracks and training areas,
their supply and communications lines and the industrial sites which
produce their supply. Presumably, core political and bureaucratic
institutions are also legitimate objects of attack, in particular things
and R.K. Fullinwinder, “War and Innocence”, Philosophy and Public Affairs
(1976), 90–97.
22 Walzer, Wars, 151. For the Conventions, see Reisman and Antoniou, eds.

Laws, 47–132. See also G. Best, War and Law since 1945 (Oxford: Clarendon,
1994).
23 Walzer, Wars, 146, 219; Nagel, “Massacre”, 133–141.
 JUST AND LAWFUL CONDUCT IN WAR: REFLECTIONS ON MICHAEL WALZER 11

like the Defence Ministry. Illegitimate targets include residential

areas, schools, hospitals, farms, churches, cultural institutions and
non-military industrial sites. In general, anyone or anything not
demonstrably engaged in military supply is immune from direct
attack. Walzer is especially critical of targeting basic infrastructure,
particularly food, water, medical and power supplies. He criti-
cizes American conduct during the Persian Gulf War on this basis,
since very heavy damage was inflicted on Iraq’s water treatment
system, and presumably would also frown upon NATO’s targeting
the Serbian electric power grid during its 1999 armed intervention
on behalf of the ethnic Albanian Kosovars. While soldiers cannot
fight well without food, water, medicine and electricity, those are
things they – and everyone else in their society – require as human
beings and not more narrowly as externally threatening instruments
of war. Thus, the moral need for a direct and relevant response only
to the source of serious harm renders these things immune from
attack.24
Another serious perplexity about targeting concerns the close
real-world proximity of illegitimate civilian targets to legitimate
military and political ones: munitions factories, after all, are often
side-by-side with non-military factories, and at times just around
the corner from schools and residential areas. This leads us to the
complex issue of the Doctrine of Double Effect (DDE). The core
moral problem here is this: even if soldiers intentionally aim only at
legitimate targets, they can foresee that taking out some of these
targets will involve collateral civilian casualties. And if civilians
do nothing to lose their human rights to life and liberty, doesn’t it
follow that such acts will be unjust? Furthermore, since such acts are
constitutive of warfare – the very stuff and substance of the conduct
of war in our world – doesn’t it follow that war itself can never be
fought justly, and thus should never be resorted to, as the pacifist
concludes?
Though Walzer is initially suspicious of the DDE, in the end he
endorses one version of it as a plausible method for “reconciling
the absolute prohibition against attacking noncombatants with the
legitimate conduct of military activity.” The DDE stipulates that an
agent A may perform an action X, even though A foresees that X
24 Walzer, Wars, xx.
12 BRIAN OREND

will result in both good (G) and bad (B) effects, provided all of the
following criteria are met: 1) X is otherwise permissible; 2) A only
intends G and not B; 3) B is not a means to G; and 4) the goodness
of G is worth, or proportionate to, the badness of B. Assume now
that A is an army and X is an otherwise permissible act of war,
such as taking aim at a military target. The good effect G would be
destroying the target, the bad effect B the collateral civilian casu-
alties. The DDE stipulates that A may do X, provided that A only
intends to destroy the military target and not to kill civilians; that A
is not using the civilian casualties as means to the end of destroying
the military target; and that the importance of hitting the target is
“worth” the collateral dead.25
The first serious objection raised against the DDE concerns its
controversial distinction between intending Z’s death and “merely
foreseeing” that one’s actions will result in Z’s death. Many have
contended that the DDE is so elastic as to justify anything: all an
agent has to do, to employ its protective moral cloak, is to assert:
“Well, I didn’t intend that; my aim, rather, was this. . . ” On Walzer’s
behalf, it is clear that intentions are not infinitely redescribable,
nor irreducibly private, as this criticism seems to imply. Agents
are not free to claim whatever laudable intention they want in
order to justify their actions, however heinous. Intentions must meet
minimal criteria of coherence and, moreover, must be connected
to patterns of action which are publicly accessible. The criminal
justice system of most countries is predicated on these ideas: for
such serious crimes as murder, the case must be made by the prosec-
ution that the accused had mens rea, or the intent to kill. This is
done by offering third-person, publicly-accessible evidence tied to
the accused’s actions, behaviour and assertions leading up to the
time of the murder, as well as considering whether he had both
incentive and motive to commit the crime. Juries, as reasonable
and experienced persons, are then invited to infer the accused’s
state of mind. The plausibility of this procedure undermines the
popular academic claim that the DDE can be used to justify any
heinous action, whether in war or peace. Walzer agrees, suggesting
that we know the intentions of agents through their actions: “(T)he
surest sign of good intentions in war is restraint in its conduct.” In
25 Walzer, Wars, 153–154.
 JUST AND LAWFUL CONDUCT IN WAR: REFLECTIONS ON MICHAEL WALZER 13

other words, when armies fight in strict adherence to jus in bello –

taking aim only at legitimate targets, using only proportionate force,
not employing intrinsically heinous means – they cannot meaning-
fully be said to intend the deaths of civilians killed collaterally.
Their actions, focusing on military targets and taking due care that
civilians not be killed, reveals their intentions.26
What exactly constitutes “due care” by armies that civilians not
be killed during the prosecution of otherwise legitimate military
campaigns? For Walzer, it involves soldiers accepting more risks
to themselves to ensure that they hit only the proper targets: “We
draw a circle of rights around civilians, and soldiers are supposed to
accept (some) risks in order to save civilian lives.” This principle
might, for instance, entail that soldiers use only certain kinds of
weapons (eg. “smart” bombs, laser-guided cruise missiles), move
in more closely on the targets (eg. flying lower on a bombing raid),
gather and analyze intelligence on the precise nature of suspected
targets, perhaps provide some kind of advance warning to nearby
civilians, and certainly plan the tactic in advance with an eye
towards minimizing civilian casualties. Walzer suggests we locate
the limits of additional risk-taking that soldiers can and should
shoulder on behalf of those civilians they endanger at that point
where “any further risk-taking would almost certainly doom the
military venture or make it so costly that it could not be repeated.”27
Walzer maintains that civilians are not entitled to some implaus-
ible kind of fail-safe immunity from attack; rather, they are owed
neither more nor less than this “due care” from belligerent armies.
Providing due care is equivalent to “recognizing their rights as best
we can within the context of war.” Interestingly, Walzer concedes
that the requirements of due care – in particular, to formulate
strategies designed to minimize civilian casualties – reveal that, in
some respects, “utilitarian arguments and rights arguments . . . are
not wholly distinct.” We should note that a deontological paci-
fist would disagree, and insist that real respect for rights involves
rejecting this kind of aggregative thought about due care constituting

26 M. Walzer, “War and Peace in The Jewish Tradition” in T. Nardin, ed.

The Ethics of War and Peace: Religious and Secular Perspectives (Princeton:
Princeton University Press, 1996), 106.
27 Walzer, Wars, 151, 157.
14 BRIAN OREND

full respect for them. Civilians, the pacifist would conclude, are
entitled to fail-safe immunity and, since it cannot be provided to
them, war must forever remain unjust.28
What about the second criterion in Walzer’s DDE? Can it ever
be met to our satisfaction in the real world? On Walzer’s behalf,
it seems possible to discern whether a belligerent, such as country
C, is employing civilian casualties as a means both to its immediate
end of hitting the legitimate target and to its final end of victory over
rival country D. If there were greater civilian than military casualties
in D, for example, it would be clear which group of people was
bearing the brunt of C’s attack. Relatedly, if there were systemic
patterns – as opposed to unavoidable, isolated cases – of civilian
bombardment by C on the civilians of D, it would also be compelling
to conclude that C was directly targeting the civilian population of
D. Conversely, if the systemic pattern of C’s war-fighting indicates
its targeting of D’s military capabilities, with only incidental civilian
casualties resulting, then it would be reasonable to infer that C was
not trying to use civilian casualties as a pressure tactic to force D to
retreat and admit defeat.
The truly difficult aspect of Walzer’s DDE is the third criterion:
contending that the goodness of hitting the legitimate military target
is “worth”, or proportional to, the badness of the collateral civilian
casualties. A pacifist, for example, will always deny this. Is the
need to hit a source of harm sufficient to justify killing people
whom Walzer admits have done nothing to deserve death? Does
the source of harm have to pass some threshold of threat before
one can speak of the need for its destruction outweighing civilian
claims? If so, how to locate that threshold? More sharply, can one
refer to the ultimate “worth” of hitting the target to justify collat-
eral civilian casualties without referring to the substantive justice of
one’s involvement in the war to begin with?
Walzer does not confront these potent queries directly, though
one gains the impression from his work that the “worth” in question
refers simply to the target forwarding the war aims of the country
in question. Provided only that hitting the legitimate target will
contribute (how much?) to victory, the collateral civilian casualties
will be “worth” it. I do not believe that such an agnostic attitude
28 Walzer, Wars, 152 and 156, in the note.
 JUST AND LAWFUL CONDUCT IN WAR: REFLECTIONS ON MICHAEL WALZER 15

with regard to war aims will here suffice. I fail to grasp how it can
be morally justified to foreseeably kill innocent civilians in order
to hit a target that only serves the final end of an aggressive war.
The only justification sufficient, in my mind, to justify the collateral
civilian casualties would be that the target is materially connected
to victory in an otherwise just war. This suggests that aggressors not
only violate jus ad bellum, but in so doing face grave difficulties
meeting the requirements of jus in bello as well. To be as clear as
possible: to satisfy the jus in bello requirement of discrimination, a
country when fighting must satisfy all elements of the DDE. But it
seems that only a country fighting a just war can fulfil the propor-
tionality requirement in the DDE. Thus, an aggressor nation fighting
an unjust war may, for that very reason, also violate the rules of right
conduct. Here too we see that Walzer’s insistence on the separate-
ness of jus ad bellum and jus in bello may not be sustainable. Kant
may well have been more correct when he insisted on the need for
a consistent normative thread to be run though conduct during all
three phases of war: beginning, middle and end.29

II. PROPORTIONALITY

The jus in bello version of proportionality mandates that soldiers

deploy only proportionate force against legitimate targets. Walzer is
uncertain about the precise content of this requirement. He notes
that while the rule is designed to prohibit “excessive harm” and
“purposeless or wanton violence” during war, “there is no ready
way to establish an independent or stable view of the values”
against which we can definitively measure the costs and benefits
of a tactic. One case where he talks about, and endorses, a form
of proportionality involves the Persian Gulf War. During the War’s
final days, there was a headlong retreat of Iraqi troops from Kuwait
along a road, subsequently dubbed “The Highway of Death.” So
29 See my War and International Justice: A Kantian Perspective (Waterloo,
ONT: Wilfrid Laurier University Press, 2000), and my shorter article, “Kant’s
Just War Theory”, Journal of the History of Philosophy (April 1999), 323–353.
The most relevant primary source is I. Kant, The Metaphysics of Morals, trans, by
Mary Gregor (Cambridge: Cambridge University Press, 1995), especially 114–
124.
16 BRIAN OREND

congested did that highway become that, when American forces

descended upon it, it was a bloodbath whose aftermath was much
photographed and publicized. Although the Iraqi soldiers did not
surrender, and thus remained legitimate targets, Walzer suggests that
the killing was “too easy.” The battle degenerated into a “turkey
shoot”, and thus the force deployed was disproportionate. Perhaps
another example, from the same war, would be Saddam Hussein’s
very damaging use of oil spills and oil fires as putative means of
defense against an amphibious invasion of Kuwait by the Allies.30
Walzer insists that the “chief concern” in wartime is the question
of who may be targeted with lethal force. The question of what
means may be employed in the targeting is “circumstantial.” He
suggests that the elaborate legal rules defining what means may, and
what others may not, be employed during war is beside the point.
These rules – such as those prohibiting the use of chemical weapons
on the battlefield – may be desirable, he says, but are not morally
obligatory. After all, if solders may be killed, how much can it matter
by what means they are killed? While that is a persuasive way of
putting the matter, Walzer should not be flippant about setting these
rules aside, or assigning them second-place status in jus in bello,
behind discrimination. For the robust and elaborate set of legal rules
banning the use of certain weapons in wartime is, at the very least,
an important piece of evidence for any account of wartime ethics
which purports to be conventionalist. There is a vast number of
relevant conventions on this issue, aside from the canonical Hague
and Geneva Conventions, such as those banning the use of chemical
(1925 and second protocol 1996), biological (1972) and “excess-
ively injurious weapons” (1980). Also relevant are the conventions
against genocide (1948) and against methods of warfare which alter
the natural environment (1977).31
In addition to the thickly textured legal conventions, one might
suggest that there is a widely shared moral convention which stip-
ulates that even though soldiers may be targeted with lethal force,
some kinds of lethal force – such as burning them to death with
flame-throwers, or asphyxiating them with nerve gas – inflict so
much suffering and express such cruelty that they are properly
30 Walzer, Wars, 129, xxi.
31 Walzer, Wars, 42, 215; Reisman and Antoniou, eds. Laws, 35–132.
 JUST AND LAWFUL CONDUCT IN WAR: REFLECTIONS ON MICHAEL WALZER 17

condemned. Moreover, the reasoning which distinguishes between

legitimate and illegitimate weapons is very similar to the reas-
oning which generates the combatant/noncombatant distinction. For
example, there is a legal ban on using bullets which contain glass
shards. These shards are essentially impossible to detect. If the
soldier survives the shot, and the bullet is removed by surgery,
odds are that the glass shards will remain in his body. These shards
can produce massive internal injuries, long after the soldier has
ceased being “a dangerous man” to the other side. Parallel reasoning
was behind the 1999 passing into law of the International Treaty
Banning Land Mines: land mines, too frequently, remain weapons
of destruction long after the conflict is over. Finally, restrictions on
weapons can play a causal role in reducing destruction and suffering
in wartime, something which jus in bello as a whole is designed to
secure. Walzer doesn’t even explicitly object to particular weapons
on grounds that they are more likely than not to have serious
spillover effects on civilians, and thus run afoul of discrimination.
Biological weapons would fall under this category, as would many
land mines. Such a stance would be consistent with other judgments
one might expect, but does not hear, from him, such as criticizing
America’s extensive use of napalm and Agent Orange in Vietnam,
which inflicted long-term damage to Vietnamese agriculture. Walzer
is curiously unreflective about these considerations.32
Walzer recovers his reflectiveness about weaponry only when he
considers nuclear arms, which for a number of reasons have not been
declared illegal by ratified international treaty. “Nuclear weapons
explode the theory of just war”, Walzer famously declares.33 This
is a graphic but unfortunate formulation, for it seems to endorse
the popular academic view that just war theory is out of date in the
post-Hiroshima era. But Walzer cannot believe this, for he takes the
time to explain, using just war concepts, how the atomic bombing
of Japan was unjust.34 Thus, what his dramatic declaration must
32 Regan,Cases, 87–99, 136–150.
33 Walzer, Wars, 282. While there have been two UN General Assembly resol-
utions, in 1961 and 1972, banning the use of nuclear weapons (see Reisman and
Antoniou, eds. Laws, 66–67), these do not carry the binding force of a ratified
international treaty. Obviously, the fact that the world’s most powerful countries
are also nuclear powers inhibits the passing of such a treaty.
34 Walzer, Wars, 263–268.
18 BRIAN OREND

really mean is that nuclear weapons can never be employed justly.

Why not? First, and most crucially for Walzer, they are radically
indiscriminate weapons. Perhaps only a handful of the most volatile
biological weapons are more uncontrollable in their effects. Second,
nuclear weapons are unimaginably destructive, not just in terms of
short-term obliteration but also long-term radiation poisoning and
climate change, so that their use will always run afoul of propor-
tionality. Finally, there is the hint in Walzer that, owing to these
two factors combined, deliberate use of nuclear weapons – and
emphatically an all-out nuclear war – is an act evil in itself.

III. NO MEANS “MALA IN SE”

The most general rule of jus in bello which Walzer endorses is

that armies may never employ acts or weapons which “shock the
moral conscience of mankind.” This seems to be Walzer’s equiv-
alent of the traditional ban on “means mala in se”, or “methods
evil in themselves.” The imprecise yet interesting idea here is that
some weapons and means of war are forbidden not so much because
of the terribleness of the consequences they result in but, more
importantly, because they themselves are intrinsically awful. Is this
anything more than rhetorical heightening, an especially emphatic
banning of indiscriminate and/or disproportionate targeting? Walzer
believes so. Perhaps the most fruitful way to cash out his concept of
an intrinsically corrupt means is to define it as being rights-violative
in itself. Using rape as a tool of warfare is a plausible example. Rape
is ruled out here not so much because of all the pain it produces, or
because it is aimed at civilians, but because the act itself is rights-
violative, a disgusting disregard for the humanity of the woman
raped: a coercive violation of her bodily integrity and her entitle-
ment to choose her own sex partner(s).35 We might infer that, for
Walzer, methods like campaigns of genocide and ethnic cleansing
probably also fall under this category. We don’t have to do a cost-
benefit analysis to determine whether such acts are impermissible in
35 Walzer, Wars, 129–137; C. MacKinnon, “Crimes of War, Crimes of Peace”
in S. Shute and S. Hurley, eds. On Human Rights (New York: Basic Books, 1993),
83–110.
 JUST AND LAWFUL CONDUCT IN WAR: REFLECTIONS ON MICHAEL WALZER 19

warfare: we already judge such acts to be heinous crimes. The inten-

tional destruction, and/or forcible displacement, of whole peoples,
as Walzer suggests, is something we find “literally unbearable.”
Indeed, the international community passed a convention in 1948
banning genocide, and recently resorted to armed force over Kosovo
in Serbia to punish its practice. Nuclear weapons may also fall under
this category for Walzer because use of them implies deliberate
killing of the innocent, and on a wildly destructive scale. There
cannot be much doubt that nuclear weapons have indeed “shocked
mankind” and are the objects of continued fear and loathing.36

IV. REPRISALS

Walzer allows for reprisals in his just war theory. And this in spite of
his acknowledgment that “(n)o part of the war convention is so open
to abuse, is so openly abused, as the doctrine of reprisals.” Such a
doctrine permits a violation of jus in bello rules but only in response
to a prior violation by the opposing side. To his credit, Walzer
refuses to condone any violation of the principle of discrimination
as part of reprisal: “we must condemn all reprisals against inno-
cent people.” What of proportionality and no means “mala in se”?
While he does not explicitly say so, one supposes Walzer cannot
countenance a violation of the latter rule for mere reprisal purposes.
His single example of a justified reprisal focuses on proportion-
ality and prohibited weapons. He claims that Winston Churchill was
“entirely justified when he warned the German government early in
World War II that the use of [poison] gas by its army would bring
an immediate Allied reprisal.” Such threats by heads of state have
apparently become rather commonplace, since American President
George Bush warned Iraq in 1991 that, should it deploy chem-
ical weapons on the battlefield, America would reserve the right
to deploy other weapons of mass destruction, up to and including
nuclear armaments. It is important to note here that, presumably,
Walzer means that not only the threat but also the threatened action
are grounded by his doctrine of reprisal.37

36 Reisman and Antoniou, eds. Laws, 84–94; Walzer, Wars, 257, 323.
37 Walzer, Wars, 207–22; Regan, Cases, 172–178.
20 BRIAN OREND

Walzer justifies his permission for retaliations on the need to

enforce the rules of the war convention during battle: “(i)t is the
explicit purpose of reprisals . . . to stop the wrongdoing here [his
italics] with this final act” of jus in bello violation. Reprisals are
designed to make the enemy stop its own jus in bello transgres-
sions: state S violates proportionality, say, and state T responds in
kind so as to punish S and hopefully prevent future violations.38
Can we come up with relevant modern examples here? Walzer
might, on these grounds, commend America’s 1986 bombing of
Libya as retaliation for the latter’s involvement in terrorist strikes,
or America’s 1998 bombing of suspected terrorist sites in Sudan
and Afghanistan as reprisals for presumed involvement in American
embassy bombings throughout Africa. After all, with the important
exception of the Lockerbie jet bombing in 1988, Libya has seemed
to stop being a major state sponsor of terrorism since the Amer-
ican attack. And failure to respond to the embassy bombings in
Africa would have only invited further violence by anti-American
extremists.
Walzer’s reprisal doctrine is worrisome. It ignores the serious
likelihood that reprisals, far from chastening the state which origin-
ally violated jus in bello, will actually spur further violations. To put
it in just war terms, reprisals have dubious probability of success.
After all, what government is likely to just sit there and suffer a
violation of jus in bello? If it is the government that committed
the first (unbidden) violation, why would it hesitate to commit a
second one in response? If it is the government that received the first
violation, then it will, not implausibly, fear that a failure to respond
in kind will only whet its opponent’s appetite for more destruction.
Reprisal, in short, is a recipe for escalation, at its extreme risking
the onset of total war, a phenomenon just war theory utterly rejects.
Walzer might contend that a certain kind of reprisal may well
succeed in stopping escalation. Of course, it may, but what kind
of reprisal is that likely to be? Realistically, it seems, only a very
severe, disproportionate one. And while Walzer extends his reprisal
permissions solely in terms of relaxing proportionality (and not,
thankfully, in terms of relaxing discrimination), we can still ask:
how much is too much relaxation? Is there such a thing as “too much
38 Walzer, Wars, 207.
 JUST AND LAWFUL CONDUCT IN WAR: REFLECTIONS ON MICHAEL WALZER 21

relaxation” when it comes to reprisals? Might Walzer, for example,

condone the Gulf War “turkey shoot” incident on the Highway of
Death if it were in reprisal, say for Iraq setting Kuwait’s oil wells on
fire? Or if state T were to lose one brigade of its soldiers to nerve
gas unleashed unbidden by state S, does that mean for Walzer that, to
ensure an effective enforcement of the rules, T should now gas two,
or more, brigades of S’s soldiers in response? Or perhaps deploy
a tactical nuclear device against S’s battlefield positions? For me,
these are rhetorical questions. They also underline the precarious
position of the ordinary soldier in Walzer’s theory, subject as he is
to all of these measures.
Reprisal is a very tempting option in warfare, especially when
one notes that, given its nature, the aggressor nation will most likely
be the one which first violates jus in bello. And while relaxing
proportionality against legitimate targets may feel like a fitting
response to prior violations of a principle as important as discrim-
ination – for instance, gassing enemy soldiers who engaged in
civilian massacre – it is unlikely to achieve its more reasoned goal
of deterring future violations. As deterrence, reprisal is dubious. As
retribution, reprisal may seem elemental, yet it is unlikely to achieve
more than a modest, temporary satisfaction of popular outrage. It
would thus seem far better to adhere to the following policy on
reprisals, adopted from a familiar phrase: winning well is the best
revenge. But what if, Walzer would ask, one cannot put the first
two together: what if winning cannot be had, in the real world, by
fighting well and by resisting the sinful pleasures of revenge? What
if, ultimately, violating jus in bello seems the only way to stave off
devastating loss?

V. SUPREME EMERGENCIES

Walzer notes that, when it comes to war, “we want to have it both
ways: moral decency in battle and victory in war; constitutionalism
in hell and ourselves outside.”39 We are therefore confronted with a
grave dilemma when it looks as though we can win the war only by
setting aside the rules of right conduct. Walzer, to his credit, refuses
39 Walzer, Wars, 47.
22 BRIAN OREND

to indulge the fantasy that such situations cannot actually happen.

His way out of this dark dilemma, however, is one of the most
difficult and controversial aspects of his just war theory. It is his
doctrine of supreme emergency and it permits not merely violation
of proportionality against enemy soldiers but even of discrimination
against enemy civilians. It is something like the ultimate, no-holds-
barred reprisal against the ultimate threat. How does Walzer pose,
and then respond to, the problem?
Walzer stipulates that soldiers cannot appeal to military necessity
to set aside the three rules of right conduct, as elaborated above,
because these laws have already been structured with military neces-
sity in mind: “Belligerent armies are entitled to try to win their wars,
but they are not entitled to do anything that is or seems to them
necessary to win.”40 Furthermore, the threat of suffering run-of-the-
mill military defeat in wartime is not sufficient for a country to set
aside jus in bello: “(T)he rules of war may at some point become a
hindrance to the victory of one side or another. If they could then
be set aside . . . they would have no value at all. It is precisely then
that the restraints they impose are most important.” But sometimes
a country at war is faced with something much more dangerous
than run-of-the-mill military defeat. A country can sometimes suffer
what Churchill called a “supreme emergency.” Walzer suggests that
when “the very existence of a community may be at stake”, “the
restraint on utilitarian calculation must be lifted. Even if we are
inclined to lift it, however, we cannot forget that the rights violated
for the sake of victory are genuine rights, deeply founded and in
principle inviolable.”41
There is more than a mere taste of powerful paradox in this
doctrine of supreme emergency. Its full flavour is captured by
Churchill himself: “(W)e have a right, indeed are bound in duty,
to abrogate for a space some of the conventions of the very laws
we seek to consolidate and reaffirm.” Why did Churchill believe
that the British enjoyed such a remarkably permissive right during
WWII? He suggested that the British were “fighting to re-establish

40 Walzer, Wars, 131. It should be noted that, in most of the laws themselves,
the appeal to military necessity is explicitly ruled out as grounds for violation.
See Reisman and Antoniou, eds. Laws, xvii–xxxii.
41 Walzer, Wars, 195, 228.
 JUST AND LAWFUL CONDUCT IN WAR: REFLECTIONS ON MICHAEL WALZER 23

the reign of law and to protect the liberties of small countries. Our
defeat would mean an age of barbaric violence and would be fatal,
not only to ourselves, but to the independent life of every small
country in Europe.” “It would not be right”, Churchill averred, “that
the aggressive power should gain one set of advantages by tearing
up all laws, and another set by sheltering behind the innate respect
for law of its opponent. Humanity, rather than legality, must be
our guide.”42 Walzer’s supreme emergency doctrine has affinities
with Churchill’s. Walzer stresses that, since appeals to emergency
and crisis are inherent in the overheated wartime atmosphere, any
appeal to supreme emergency must be subjected to the most rigorous
public scrutiny. A supreme emergency exists only when there is
proof of a serious threat which is not only close and imminent but
also “unusual and horrifying.” There is an important ambiguity with
regard to what Walzer means by “unusual and horrifying.” On the
one hand, he suggests that the “ultimate horror” in question is a
serious threat to a people’s “survival as an independent nation”, to
“the survival and freedom of political communities.” The emphasis,
in other words, is on the grievous threat to the sovereignty of the
community in question. On the other hand, the “ultimate crisis of
collective survival” denoted by supreme emergency refers to “entire
peoples being enslaved and massacred”, i.e. the emphasis is not
merely on losing sovereignty but more on being subject to a further
set of appalling measures, like widespread murder, by an unjust
regime.43
Walzer’s only example of a real world supreme emergency
conflates both senses. It is Churchill’s Britain between mid-
1940 and late 1941. Walzer argues that Britain was justified
during this period in engaging in deliberate saturation bombing
of residential areas in Germany, even though such acts viol-
ated noncombatant immunity. For Germany’s blitzkrieg had left
it triumphant throughout Western Europe and the controversial
Molotov-Ribbentrop pact for now secured peace with the Soviet
Union in Eastern Europe. America, of course, had not yet entered
the European war during this time. Ignoring the massive assistance
Britain was receiving from its Commonwealth, Walzer concludes
42 Churchill quoted in Walzer, Wars, 245.
43 Walzer, Wars, 241, 251–254, 257.
24 BRIAN OREND

that the UK stood alone against the Nazi menace. And it was a
menace in both senses: to the sovereignty of the British people
collectively and to the individual human rights of many Britishers,
should the Nazis conquer their island nation. He asks: “(C)an one do
anything [his italics], violating the rights of the innocent, in order
to defeat Nazism?” He answers yes, and justifies himself thusly:
“Nazism was an ultimate threat to everything decent in our lives, an
ideology and a practice of domination so murderous, so degrading
even to those who might survive, that the consequences of its final
victory were literally beyond calculation, immeasurably awful. We
see it – and I don’t use the phrase lightly – as evil objectified in the
world.”44
Is Walzer implying that both senses must be met for the imminent
threat to be truly “unusual and horrifying”? He never says so
explicitly but is not inclined to make sharp distinctions between
a people’s collective right to political sovereignty and individual
rights to personal security, believing that the two bear close causal
connections in the real world. This indeed seems the most compre-
hensive reading of Walzer: a supreme emergency exists only when
there is proof of a close, potent and imminent threat of losing
sovereignty and then being subjected to massacre or enslavement.
One crucial consequence of this conception for Walzer is that only
just states may avail themselves of the supreme emergency escape
clause; only those who have met jus ad bellum may invoke supreme
emergency. And this for at least two reasons. The first is explicit
textual references to “a nation fighting a just war” invoking supreme
emergency, and as a result committing murder, “though for a just
cause.”45 The second, unstated by Walzer, is that both by fact
and definition a just state would not put another in a condition
of supreme emergency: only a brutal, aggressive, rights-violative
regime would do that. Though he resists it, and leaves it unstated,
it seems that Walzer has no choice but to acknowledge some kind of
deep connection between jus ad bellum and jus in bello.
Since Britain in 1940–1 was confronted by the Nazis with both
the loss of its sovereignty and the serious risk of being subjected

44 Walzer, Wars, 253; M. Walzer, “World War II: Why Was This War
Different?” Philosophy and Public Affairs (1971/72), 3–21.
45 Walzer, Wars, 323, 325.
 JUST AND LAWFUL CONDUCT IN WAR: REFLECTIONS ON MICHAEL WALZER 25

to massacre or enslavement, it was appropriate for it to use its

RAF bombers to lash out against the civilian population of Nazi
Germany, in the hopes that such might have some quelling effect on
the Nazi war machine. Walzer stresses, however, that these condi-
tions of supreme emergency evaporated by 1942, since by then both
the USA and USSR had joined the battle, turning the tide against
Hitler. This may be a bit strict, on his own terms, with regard
to the dating: the war’s outcome was arguably still in contention
throughout 1942 and even early 1943. It is clear, though, that by
mid-1944 the Allies could have expected eventual victory. In any
event, Walzer concludes that continued Allied bombing of German
residential centres – most graphically the razing of Dresden in 1945
– was unjust, since it targeted civilians and the supreme emergency
exemption no longer applied.46
Consider the quandaries regarding Walzer’s doctrine of supreme
emergency. What exactly is the status of this supreme emergency
“loop-hole” Walzer discerns in our shared war convention? Is it a
moral loop-hole, suggesting that it is just for a nation to set aside
the rules of justice for the sake of “a greater good”, like collective
sovereignty or a world without genocide? If so, then how can Walzer
appeal at the final moment to an aggregative, consequentialist
concept like “a greater good”, when throughout his just war theory
he ferociously denounces aggregative, consequentialist conceptions
of wartime morality, in particular utilitarianism? Just and Unjust
Wars, after all, is a systematic contrast between a rights-based, and
a utility-based, conception of wartime morality, and the bulk of it is
devoted to arguing in favour of the former. Rights, Walzer suggests,
“cannot simply be set aside; nor can they be balanced, in utilitarian
fashion, against this or that desirable outcome.” Indeed, when we are
“(c)onfronted by those rights, we are not to calculate consequences,
or figure relative risks, or compute probable casualties, but simply
to stop short and turn aside.” Most clearly: “rights and rules are
intended to bar utilitarian calculation.”47
46 Walzer, Wars, 258–263.
47 Walzer, Wars, 228, 268 and 230. See also M. Walzer, “Standing”, 222, n. 24:
“Nozick goes on to argue, on Kantian grounds, that rights must be understood as
constraints on action rather than as goals of a maximizing politics. Though I don’t
share his views as to the substance of a rights theory, the same conception of its
structure underlies my own position in Just and Unjust Wars.”
26 BRIAN OREND

How can Walzer, at the last moment, suggest that the need for a
fundamental consistency between the means and ends of warfare be
put aside? And how can he permit, at the international level, some-
thing which we would not commonly endorse in the interpersonal
case? Walzer himself, after all, admits that we do not think it morally
justified when person P, to prevent his own death at the hands of
murderous attacker R, reaches out and drags innocent person S
into the fray, using S as a shield between him and R. Walzer has
trouble squaring our condemnation of the interpersonal case with
our purported condoning of the interstate case: “(C)ommunities, in
emergencies, seem to have different and larger prerogatives. I am not
sure I can account for the difference, without ascribing to communal
life a kind of transcendence that I don’t believe it to have. Perhaps
it is only a matter of arithmetic . . . [perhaps rather] it is possible
to live in a world where individuals are sometimes murdered, but
a world where entire peoples are enslaved or massacred is literally
unbearable.”48
The reference to arithmetic is startling, and makes us mindful
of the utilitarianism Walzer is otherwise at pains to criticize. His
supreme emergency doctrine, in fact, bears a striking similarity to
one form of rule-utilitarianism: during ordinary conditions of war,
we are to adhere absolutely to the rules of jus in bello. However,
when confronted with the hardest case, we are to set aside these rules
and do what we must to prevail. The fact that human rights protec-
tions can ultimately be set aside may reveal what, for Walzer, is
really the primary political commitment: “the survival and freedom
of political communities – whose members share a way of life,
developed by their ancestors, to be passed on to their children – are
the highest values of international society.” Perhaps what supreme
emergency reveals about Walzer’s theory, like rule-utilitarianism
confronted with a hard case, is that individual human rights are
not as fundamental for him as shared ways of life. Alternatively,
one might ask Walzer: if the rules can be put aside here, why not
elsewhere? If in the end the triumph of the just state is what matters
most, then why have separate rules of jus in bello at all? Why not let

48 Walzer, Wars, 254.

 JUST AND LAWFUL CONDUCT IN WAR: REFLECTIONS ON MICHAEL WALZER 27

the just state avail itself of any means to crush the unjust aggressor
right from the start? 49
Walzer, in some sense, feels the sharpness of these questions.
He admits they force him into a paradoxical position: it is moral
to set aside the rules of morality during a supreme emergency.
“(I)n supreme emergencies”, he says, “our judgments are doubled,
reflecting the dualist character of the theory of war and the deeper
complexities of our moral realism; we say yes and no, right and
wrong [his italics]. That dualism makes us uneasy; the world is
not a fully comprehensible, let alone a morally satisfactory place.”
The deliberate killing of innocents, though murder, can nevertheless
be justified in a supreme emergency: it is simultaneously right and
wrong. At the same time, with respect to the same action, we say
yes and no. Bomb the residential areas deliberately – murder those
civilians – but do so only because you are “a nation fighting a just
war [which] is desperate and survival itself is at risk.” Walzer even
suggests that soldiers and statesmen really have no other choice but
to opt for the collective survival of their own communities. They
have no choice but to get their hands dirty, availing themselves
of brutal, rights-violative measures for the sake of the long-term
survival and rights-satisfaction of their own people.50
Calling this paradoxical position a “utilitarianism of extremity”,
though, is most unhelpful on Walzer’s part. For utilitarianism
implies a commitment to an overall greatest good, i.e. to an all-
things-considered justification, to a conviction that it’s better to
ensure long-term freedom and survival than to avoid the serious
stain of committing murder. Walzer talks like this when he stresses
that rights are “overridden” in a supreme emergency, presumably
by the “highest values” of shared and just communal existence.
Moreover, the utilitarian reference denies what Walzer previously
insisted on: the total separation and equal importance of jus ad
bellum and jus in bello. When push comes to shove, it is better
to “wager this determinate crime (the killing of innocent people)
against that immeasurable evil (a Nazi triumph).”51 Jus in bello, in

49 Walzer, Wars, 254.

50 Walzer, Wars, 325–328, 251–268; M. Walzer, “Political Action: The Problem
of Dirty Hands”, Philosophy and Public Affairs (1973), 160–180.
51 Walzer, Wars, 231, 259.
28 BRIAN OREND

a supreme emergency, is of lesser import than jus ad bellum. But

if this is true, then how is it wrong to violate jus in bello during a
supreme emergency? In other words, his reference to utilitarianism
does not fit together with his reference to paradox: utilitarianism
is designed to avoid paradox by offering up a coherent ranking
of the alternatives based on the goal of maximizing best overall
consequences.
It seems to me, on the basis of these considerations, that Walzer’s
doctrine of supreme emergency is muddled in conception and
dangerous in consequence. Sorting out the fragmented remarks,
we come to the conclusion that Walzer’s doctrine here is one of
two forms: it is either a form of consequentialism, not unlike rule-
utilitarianism, or it is a form of moral paradox. If it is a form of
consequentialism, mandating the ultimate survival of just political
communities, then it is at odds with either: 1) his own emphasis on
the inviolability of rights; 2) our conventional condemnation of the
same kind of action in the interpersonal case; or 3) his insistence on
jus in bello and jus ad bellum being totally separate. If on the other
hand it is a form of paradox, then it runs afoul of Walzer’s own
insistence that the best interpretation of our political commitments
during wartime must at least be logically coherent.52
Supreme emergencies, I believe, are not instances of genuine
paradox. They are, rather, cases of moral tragedy. A moral tragedy
occurs when, all things considered, every viable option one is
confronted with involves a serious moral violation. In a supreme
emergency, this is clear: if one violates jus in bello, one commits
murder and perhaps other crimes. On the other hand, if one does not
violate jus in bello, one’s omissions may contribute causally to the
death and devastation of one’s people at the hands of a brutal, rights-
violative aggressor. While the aggressor, of course, would bear the
biggest burden of blame for this, I think we would still criticize a
regime that stood by and did nothing for the sake of principle while
such destruction was meted out on its people.53 Thus, in a rare case
52 Walzer, Wars, 3–20; M. Walzer, Thick and Thin: Moral Argument at Home
and Abroad (Notre Dame: University of Notre Dame Press, 1994); M. Walzer,
Interpretation and Social Criticism (Cambridge: Harvard University Press, 1987).
53 I concur with Walzer’s notion that “let justice be done though the heaven’s

fall” is a rather hard doctrine for people to accept during a supreme emergency,
since here it really means what it says. Walzer, Wars, 230.
 JUST AND LAWFUL CONDUCT IN WAR: REFLECTIONS ON MICHAEL WALZER 29

of genuine supreme emergency we are not truly confronted with

options that are both right and wrong, rather, we are confronted
with options all of which are wrong. It is a moral blind alley: there
is no where to turn and still be morally justified.54 In other words,
I suggest we understand supreme emergency as a case where we
exit the moral realm and enter the harsh Hobbesian realm of pure
survival, where brutal, “do-or-die” measures will be taken and may
accordingly be excused but never morally justified. It is the only
realm of true necessity in wartime. The suggested alternative, then,
is this: in a supreme emergency, a just state will commit actions
which are morally wrong in order to save itself. While wrong,
such actions may nevertheless be excused on grounds of extreme
duress.55

VI. SUMMARY

Walzer defends three rules of right conduct during wartime: discrim-

ination and noncombatant immunity; proportionality; and a ban
on means mala in se. His defense of noncombatant immunity is
eloquent and formidable, but his insistence on it leads him to under-
estimate the question of prohibited weaponry. His allowances for
reprisals and supreme emergency strikes which violate right conduct
are very interesting yet deeply problematic. They are the darkest and
most dangerous aspects of his just war work. In my mind, these
allowances are far too permissive of actions which, at best, can
be excused on grounds of severe duress and, at worst, constitute
54 I believe this is the deepest meaning of the article which Walzer mistakenly
cites in favour of his own paradoxical reading of supreme emergency: Nagel’s
“Massacre”, 123–144. The DDE, it should be noted, cannot justify actions which
violate jus in bello during a supreme emergency. And this for at least two reasons.
First, the DDE stipulates that the action which produces both good and bad
effects must be otherwise permissible. But actions which violate jus in bello, like
murder, are not otherwise permissible. Furthermore, the DDE stipulates that the
bad effects cannot be the means whereby the good effects are to be achieved.
But in a supreme emergency, the bad effects (like massive civilian casualties) are
intended to bring about the good effects (to make the aggressor cease and desist).
55 For other views on Walzer’s supreme emergency escape clause, see: Lackey,

“Modern”, 540–542; and T. Nardin, Law, Morality and The Relations of States
(Princeton: Princeton University Press, 1983).
30 BRIAN OREND

some of the most appalling crimes which can ever be committed.

An underlying theoretical point throughout this piece has been that
Walzer cannot sustain his sharp distinction between jus ad bellum
and jus in bello. Kant’s older insistence that there be one just
war theory, woven into a consistent whole, thus appears to be an
important improvement on the most influential just war theory of
our time.56
University of Waterloo
Department of Philosophy
N2L 3G1
Waterloo, Ont.
Canada
E-mail: [email protected]

56 I would like to thank the anonymous reviewers, and all those at this journal.
This article discusses material that will appear in my forthcoming book, Michael
Walzer on War and Justice (Cardiff: University of Wales Press).