RESTRICTED

DEFENCE SERVICES COMMAND AND STAFF COLLEGE

MIRPUR, BANGLADESH

AIR WING
2024

AP-1D: RULES OF ENGAGEMENT

RESTRICTED
 RESTRICTED

DEFENCE SERVICES COMMAND AND STAFF COLLEGE

AIR WING

AIR POWER MODULE

RULES OF ENGAGEMENT AP - 1D
OBJECTIVES

Training Objective

1. Understanding the basic principles of International Law, with particular ref to the
laws of armed conflict related to air ops.

Enabling Objectives

2. Be familiar with and understand the laws of international armed conflict related to
air op.

3. Understand the applications and the implications of Rules of Engagement with

particular ref to air ops in peace, during pre-hostility pd and in war.

i
RESTRICTED
 RESTRICTED

DEFENCE SERVICES COMMAND AND STAFF COLLEGE

AIR WING

AIR POWER MODULE

RULES OF ENGAGEMENT AP - 1D
CONTENTS

Page

Objectives i
Contents ii
Introduction vi

CHAPTER – 1 - LAW OF ARMED CONFLICT

Legal Background

Air Law

Definitions 1-2
History 1-3
The Chicago Convention 1-4
State Aircraft 1-5
Air Navigation Issues 1-5
Rules of Aerial Warfare 1-6

Rules of Engagement

Factors to be Considered 1-7

Further Action 1-9
Purposes of ROE 1-9
Prerequisites of ROE 1-9
Content 1-9
Examples 1-10
Promulgation 1-10
Summary 1-11

Annexes:

A. Rules of Engagement - Definition of Terms. 1A-1 to 1A-3

B. Examples of Rules of Engagement. 1B-1
C. Extracts from the Army Field Manual, Vol 1. 1C-1 to 1C-4
D. Self Defence – Sanremo Handbook on ROE 1D-1 to 1D-5

ii
RESTRICTED
 RESTRICTED

CHAPTER – 2–RULES OF ENGAGEMENT IN THE AIR

Introduction 2-1
The Role of the Air Defence Commander 2-1
Identification During Peace 2-2
Identification During Periods of Tension 2-3
Identification in General War 2-4
IFF 2-5
General Air Defence Considerations 2-6

Annexes:

A. International Law, Rules of Engagement and

Air Warfare by Christopher Greenwood, M A. 2A-1 to 2A-8
B. Chariots of Fire: Rules of Engagement in
Operation Deliberate Force by Lt Col Ronald M Reed. 2B-1 to 2B-35
C. Questionnaire 2C-1

iii
RESTRICTED
 RESTRICTED

DEFENCE SERVICES COMMAND AND STAFF COLLEGE

AIR WING

AIR POWER MODULE

RULES OF ENGAGEMENT AP - 1D

Introduction

1. During periods of peace or tension, actions taken by one nation against another
can be misinterpreted; in the worst case, such actions can lead to an outbreak of
hostilities between the nations. At the same time, nations must protect their own forces
during the difficult period before a war breaks out. Rules of Engagement (ROE) are
provided to help commanders make sensible and rapid decisions in situations short of
war. Once war breaks out, the situation is more clearcut and decision-making on
engaging the enemy is much easier, although, of course, military actions are still
subject to International Law and Convention.

2. ROE apply equally to the land, sea and air scenarios and it is essential that
officers at all levels that are likely to be involved, however tenuously with ROE, have a
clear understanding of their nature, and the political and legal implications underlying
their formulation.

Aim

3. This precis aims to introduce students to the subject of ROE with particular
emphasis on rules of engagement in the air.

Definition

4. Rules of Engagement are directives that a government may establish to

delineate the circumstances and limitations under which its own ground, naval and air
forces will initiate and continue combat engagement with enemy forces.

Background

5. The conduct of military operations is circumscribed by the provisions of

international and national law. The former includes the principles set out in the United
Nations Charter and the Hague and Geneva Conventions, the provisions of treaties and
agreements to which nations are a party, besides the tenets of customary international
relations. National law includes both the law of the country concerned, to which its
forces are always subjected wherever they may be serving, and the law of the country
to which a force may be deployed. Guided by this legal framework, it is for ministers to
decide on the deployment and commitment of the country's armed forces and to set the
parameters within which those forces can operate. Rules of Engagement are the
means whereby ministers provide political direction and guidance to commanders at all

iv
RESTRICTED
 RESTRICTED

levels governing the application of force. They are approved by ministers and may only
be changed with ministerial authority.
6. ROEs define the degree and way force may be applied and are designed to
ensure that such application of force is carefully controlled; ROEs are not intended to
be used to assign specific tasks or as a means of issuing tactical instructions. In
passing orders to a subordinate, a commander at any level must always act within the
ROE received but is not bound to use the full extent of the permissions granted.

7. ROEs are usually written in the form of prohibitions or permissions. When they
are issued as prohibitions; they will be orders to commanders not to take certain
designated actions: when they are issued as permissions, they will be guidance to
commanders that certain designated actions may be taken if the commanders judge
them necessary or desirable to carry out their assigned tasks. The ROEs are thus
issued as a set of parameters to inform commanders of the limits/constraints imposed
or freedom permitted when carrying out their assigned tasks. The conformity of any
action with any set of ROE in force does not guarantee its lawfulness, and it remains
the commander's responsibility to use only that degree of force, which is necessary,
reasonable and lawful in the circumstances.

v
RESTRICTED
 RESTRICTED

CHAPTER - 1

LAW OF ARMED CONFLICT

Legal Background

1. History. ROE originated from the Law of Armed Conflict, codified in 1899
and 1907 Hague Peace Conferences. This, in turn, is based on:

a. Basic principles, such as humanity and military necessity.

b. Custom or conduct of individual nations in conflicts (Customary Law).

c. Formal international agreements, such as treaties and resolutions.

2. Principles and Requirements. Principles of the law of armed conflict are the
same in land, sea or air warfare, with differences in application only. The laws are
international rules which govern the conduct of all armed conflicts; they are binding
upon belligerents, neutral states, and individuals - especially upon members of the
armed forces and apply whether or not war or conflict has been declared.

3. International Agreements. The following major International Agreements

are in existence:

a. Hague Conventions (1907). Outline the laws and customs of wars

and conflicts on land; deal basically with fighting an enemy who is still fighting.

b. Geneva Conventions (1949). Deal with the treatment of an enemy

who has stopped fighting: wounded, sick, POWs (whether war has been
declared) and civilians.

c. Hague Draft Rules of Air Warfare (1923). These draft rules are not
binding as they were never formally ratified, but they are nonetheless regarded
as pervasive authority.

4. Minor International Agreements. The following minor International

Agreements also apply:

a. Geneva Gas Protocol (1925) prohibiting the use of gases and

bacteriological methods of warfare.

b. Biological Weapons Convention (1972) adding to the above more

importantly.

c. Two Additional Protocols to the 1949 Geneva Conventions concluded at

the Diplomatic Conference between 1972 and 1977, signed by, amongst others,
the USA and UK but not yet ratified. These Protocols expand the Geneva
Conventions, reiterate the need to give general protection to civilians, detail the

1-1
RESTRICTED
 RESTRICTED

duties of military commanders in this and related respects, and provide for the
establishment of legal advisers to commanders.

5 The Future. Recent wars have focussed attention on the need for better training
for staff and commanders on the Laws of Armed Conflict.

AIR LAW

6. The mission of air forces is to fly and fight. To carry out this mission air force
personnel need to know and understand the legal regime which affects air operations.
After the Gulf War, General Colin Powell reinforced this point when he stated that
decisions were impacted at every level, (the law of war) and proved invaluable in the
decision-making process.

7. The legal regime which affects air operations is closely related to both the
international law of armed conflict (which regulates the full spectrum of armed conflict)
and the important body of law, known as, the law of the sea.
Some basic legal concepts which impact air operations are:

a. Definitions of air law terms.

b. History of air law.

c. The Chicago Convention.

d. State Aircraft.

e. Air navigation issues.

f. Rules of aerial warfare.

Definitions

8. First, let us define some of the terminologies integral to an understanding of air

law.

9. Air Law. That body of law governing the use of airspace and its benefits for
aviation, the public and the nations of the world. The major sources of air law are:

a. Multi-lateral and bilateral treaties.

b. National domestic law.

c. Contracts between international agencies and companies.

d. Customary international law.

10. Airspace. There is no authoritative definition as to the boundaries of airspace.

The international practice is to recognise the uppermost boundary as being somewhere
between the lowest altitude at which a satellite remains in orbit and the highest altitude

1-2
RESTRICTED
 RESTRICTED

at which an aircraft can fly. Consequently, the uppermost boundary is generally

recognised as being somewhere between 50 and 80 kilometres above the earth's
surface.

11. Aircraft. In the 1944 Chicago Convention, an aircraft is defined as any machine
which can derive support in the atmosphere from the reactions of the air other than the
reactions of air against the earth's surface. This definition includes:

a. Fixed wing and rotary wing aircraft.

b. Airships.

c. Gliders.

d. Balloons.

It does not include hovercraft, cruise missiles or rockets.

History

12. Since man first started exploiting the third dimension, the air above the surface
of the earth, regulations have emerged to control its use. In 1794, the balloons of the
Montgolfier brothers disturbed the local populace. As a result, the local authorities
passed laws restricting the times during which the balloons could be used. Additionally,
after balloons were used in the US Civil War for military purposes, the potential of air
power was recognised. As a result, the French in the 1890s called for restrictions on
the military use of aircraft.

13. Despite the provisions of the Hague Convention of 1897, which forbade the
dropping of bombs from balloons, and the draft Hague Code of 1923 which placed
restrictions on aerial warfare, no firm set of rules emerged to provide practical and
authoritative guidance to military airmen. Consequently, the following issues arose
during wars in which air power played a role:

a. Limits on aerial bombardment.

b. Definitions of legitimate military targets.

c. The status of civil aircraft.

d. Which laws of armed conflict (LOAC) were applicable to airmen.

e. Determining when airmen were Horse De Combat (out of combat).

These issues gave rise to great controversy concerning the lawful and moral
boundaries on the use of air power. While some customary international law
developed, it was only after Vietnam that the international community addressed that
part of the LOAC relating to air power. This resulted in additional protocols to the 1949
Geneva Convention being established in 1977.

1-3
RESTRICTED
 RESTRICTED

The Chicago Convention

14. Because the international community recognised the benefits of civil aviation and
the need to set some international controls, negotiations were entered into and, in
1944, a comprehensive code entitled the Convention on International Civil Aviation
(Chicago) was agreed upon. This multi-lateral treaty provided a set of fundamental
guidance for the control and development of international civil aviation. Signatories of
this convention have grown from the original 50 to include almost every nation in the
world.

15. The Chicago Convention developed some fundamental principles and concepts.
The following is a list of some of these:

a. Every nation has complete sovereignty over its own airspace.

b. A basic distinction exists between civil and state aircraft.

c. The convention applies only to civil aircraft.

d. There are established freedoms of air navigation over areas not subject to
territorial sovereignty.

e. Civil aircraft can transit through national airspace without specific

clearance, though landing rights for scheduled services may be subject to
bilateral agreements.

f. The International Civil Aviation Organisation (ICAO) is the single

international controlling authority for international civil aviation.

16. The Chicago Convention also set out some basic rules regarding national
sovereignty. These included:

a. No aircraft may enter national airspace without permission. Permission

may be specific, for example, diplomatic clearances for a flight; or general, for
example, permission granted under a treaty.

b. Every aircraft must identify itself in another nation's airspace.

c. Any aircraft in another nation's airspace must obey all reasonable orders.

d. Control by a territorial state cannot expose aircraft and occupants to

unreasonable danger; and

e. Subject to state laws, nations should allow an aircraft to enter its airspace
in an emergency.

1-4
RESTRICTED
 RESTRICTED

State Aircraft

17. State aircraft are not subject to the restriction of the Chicago Convention and
cannot take advantage of its provisions. Consequently, they are not subject to ICAO
rules and procedures. Examples of state aircraft are military aircraft, customs and
police aircraft, and VIP and mail-carrying aircraft. Because state aircraft are not
covered by the Chicago Convention, they must have diplomatic clearance to enter
another nation's airspace. The basic rule for state aircraft is that they must conduct
themselves with due regard to the safety and welfare of other air users. However, as a
matter of policy and practice, most military aircraft obey ICAO procedures for safety
reasons, unless mission requirements dictate otherwise.

18. Military aircraft must be distinguished from civil aircraft. They are required to
have visible external military and national markings. The additional protocols define
military aircraft as those aircraft crewed by military personnel registered on a military
aviation register and destined to form part of the armed forces in conflict. There are
specific provisions relating to the status and protection of medical aircraft in the
protocols. Only military aircraft can engage in combat operations.

Air Navigation Issues

19. As discussed previously, state aircraft need diplomatic clearance to transit

through another nation's airspace. This means that aircraft have no right of innocent
passage over territorial waters, like the rights all ships enjoy. Other than this, the
freedom of the air can be exercised by all aircraft over the high seas, exclusive
economic zones (EEZ), air defence identification zones (ADIZ) and flight information
regions (FIR). FIRs are areas established by ICAO and are administered by national
air traffic control authorities. They establish no sovereign rights, and state aircraft are
not subject to control within them.

20. Nations can and do intrude into other nations‟ airspace. Some can be justified,
others cannot. Some examples of justifiable intrusions are as follows:

a. Intrusion based on duress or emergency.

b. For national self-defence e.g., overflight of Cuba in 1962.

c. Israel's flight into Entebbe.

It is important to note that the Israel attack against the Osirak nuclear reactor was
justified based on pre-emptive self-defence, but this claim was not generally accepted
by the international community.

1-5
RESTRICTED
 RESTRICTED

Rules of Aerial Warfare

21. Rules of aerial warfare have been slow to develop. However, those now in
existence are detailed and wide-ranging. Unfortunately, the area is plagued by
controversy. It is beyond the scope of this precis to give a detailed analysis of these
rules; however, it is important to introduce the three basic principles of aerial targeting.
These three principles are as follows:

a. The principle of humanity.

b. The principle of military necessity.

c. The principle of proportionality.

These are the building blocks of LOAC.

22. The principle of humanity recognises that there are limits to the methods and
means of warfare. On humanitarian grounds, certain targets are afforded special
protection. This means they cannot be attacked unless certain conditions apply.
Civilians and civilian objects are not subject to attack, nor are religious and cultural
sites, hospitals, medical personnel, and military forces who are out of combat (for
example, POWS, units who have surrendered, the injured and shipwrecked). The
Geneva Convention and its additional protocols establish a body of law which sets out
the protection given to these individuals and objects.

23. LOAC recognises that, in the pursuit of military objectives, the legitimate use of
force may cause incidental injuries and collateral damage. Accordingly, the following
guidance is crucial:

a. Combatants can use the amount of force needed to achieve a military

objective.

b. Such use of force cannot be prohibited by LOAC.

c. There must be the least possible expenditure of life feasible in the

circumstances.

d. The force used must be able to be regulated by the user.

Accordingly, military necessity recognises that lives will be lost, and damage inflicted to
private property, but such loss of life and damage must be incidental to the military
mission. These LOAC principles accord with the three principles of war: maintenance
of the aim, concentration of force, and economy of effort.

24. Proportionality is the requirement to balance the principles of humanity and

military necessity. Military necessity cannot justify all use of force, for example,
indiscriminate attacks are never justified. There are absolute boundaries. An air
commander, while planning an attack, must weigh up the consequences of the attack
with the value of the objective to be achieved. That is, the amount of death and
destruction caused must be proportional to the military advantage anticipated. For

1-6
RESTRICTED
 RESTRICTED

example, it would be wrong to level an entire city simply because it had a few soldiers
garrisoned there. However, it would be acceptable to specifically attack the barracks or
any factories which manufacture weapons in the city. Additionally, the Geneva Protocol
provides specific guidance on all these principles. Commanders have a heavy
responsibility under LOAC, one which has been almost universally accepted by the
nations of the world.

RULES OF ENGAGEMENT

25. Rules of engagement are required for certain operational tasks and control of air
defence resources like interceptors, SAMs and AAA in peacetime, and in periods of
tension or limited conflict when it is necessary to supplement or modify the normal
guidance with detailed rules to ensure that the conduct of the Military Commander is
accurately related to the needs of the specific political situation, or any changes in this
situation. Some definitions of terms of ROE are in Appendix 1.

Factors to be Considered

26. In considering a decision to use force or threatening conduct to achieve a given

aim, various factors must be considered. They include:

a. The National Political Policy.

b. The Legal Aspect.

c. The Enemy's Capabilities.

d. The Enemy's Intentions.

e. Own Capability.

f. Deterrence.

These factors are discussed below.

27. The National Political Policy. National Policy is overriding and must be
concisely stated so that military commanders have a clear background from which to
frame their military policy. There are 3 possible courses:

a. To Escalate. Escalation involves taking such reprisal action to an enemy

attack that the level of tension is raised to a new threshold to force the enemy to
withdraw or to apply extra force himself, i.e., taking up a more threatening
posture.

b. To Maintain the Status Quo. Maintaining the status quo involves

keeping the situation at its present level by matching act for act, i.e., dealing with
trouble as it occurs.
c. To Reduce Tension. Reducing tension involves avoiding, as far as
possible, being provoked into using force like avoiding trouble as far as possible.

1-7
RESTRICTED
 RESTRICTED

28. The Legal Aspect. While the use of force in peacetime is generally in
opposition to the Rule of Law, there are occasions when a particular aspect of
International Law may be overriding. This might include a prohibition of entry into
another country's territorial water, or airspace, or over her borders; perhaps to avoid
dragging another party into the conflict. Under this heading, guidance may also be
given as to whether International Law may be broken, and in what circumstances.
Extracts from the Army Field Manual, Vol 1, 'Legal Constraints on the Use of Force' are
given in Appendix 2.

29. Enemy Capabilities. Is the enemy's capability such that we can afford to allow
him to commit a Hostile Act before retaliating? In other words, is our survivability at
stake if he strikes first? If this is the case, then his intentions become of overriding
importance, while if his capability does not give him the initiative, then the problem is
eased.

30. Enemy Intentions. As stated above, in some circumstances, an enemy's

intentions must be gauged when considering Rules of Engagement and Hostile Intent
must be clearly defined. Between his extreme options of overt aggression, or "self-
defence", there are 3 relevant tests that would be prudent to apply:

a. Mistake. Has he misidentified our units, or have we misidentified his?

On many occasions, errors in identity and navigation have resulted in Mistaken
Hostile Intent and force being used.

b. Surveillance. Is his interest limited to surveillance ? If so; does this

constitute sufficient reason for retaliation ?

c. Provocation. Are his actions intended to harass and possibly provoke a

response ? If so, is this for a military reason (e.g. baiting a trap) or for a political
end ?

31. Our Capability. Beyond the obvious preliminary assessment of our force's
ability to retaliate effectively to an enemy attack, or to pre-empt such an attack, it is
important to decide whether the enemy can also cope with the resulting (possibly
escalated) situation. Related to this is the degree of control and communication
between threatened units and higher command. Ideally, the authority to use force
should be held at a level where political/military consultation can quickly take place. If
he is closely supervised, a sentry can be given simple rules to follow because he would
be able to report back quickly if complications develop. On the other hand, ships and
aircraft may well have their communications disrupted by jamming. Unless special
measures are taken to safeguard these communications, their instructions must contain
an element of blanket approval for specific action without recourse to higher authority.

32. Deterrence. The military aim normally includes an element of deterrence, and
the Rules of Engagement should be balanced against the importance of the deterrence
being effective. The enemy is more likely to be deterred from hostile action by rules
which permit retaliation, than, for example, by a posture of self-defence.

1-8
RESTRICTED
 RESTRICTED

Further Action

33. Conflict. In applying the factors discussed above to a threatening situation,

there will inevitably be some conflict between them; the most obvious being a situation
where a pre-emptive attack is indicated for military reasons but is prohibited by national
policy. In such an instance, the penalty for accepting such an overriding political
directive must be made clear to higher command levels.

34. Review. The Rules of Engagement applicable during a state of tension will no
longer apply once hostile acts have occurred on both sides. Furthermore, as the
tension is increased by incidents over a period, it will be necessary to review the rules
frequently considering the changing situation.

Purposes of ROE

35. Means by which politicians control the military's application of force.

36. Designed to ensure that the conduct of military commanders - at all levels - is
accurately related to the requirements of a specific political situation.

37. Guide commanders on freedom of action and how force may be used.

Prerequisites of ROE

38. Be clear, precise, and tactically realistic.

39. Contain and explain enough of the philosophy behind the political policy, and the
policy itself, to enable commanders to deal correctly with unforeseen circumstances.

Content

40. Three tiers of ROE:

a. Political Policy. Determined at the ministerial level and issued to provide

background for commanders. Three standard policies, expressed in shorthand
terms are:

(1) Alpha - De-escalation.

(2) Bravo - Maintenance of the status quo.

(3) Charlie - Escalation: to take the initiative even if this involves

increasing the level of action or use of force.

b. Military Policy. States the policy to be followed by subordinate

commanders and the priorities required if circumstances change.

1-9
RESTRICTED
 RESTRICTED

c. Detailed Rules. These amplifying instructions control the actions of

subordinates within the limits and spirit of the military policy. They are of 3
types:

(1) Instructions, for example, `is to‟.

(2) Authorizations, for example, `is permitted'.

(3) Prohibitions, for example, `is forbidden‟.

Such detailed rules usually indicate the most appropriate political policy to aid
local interpretation and initiative.

Examples

41. General Rules

Desig Military Policy Political Policy

Aircraft Captains are to respond to any Bravo: The present
aggression with tactful firmness and are tension should not be
nnn to exhibit a determination to meet any allowed to deteriorate
escalation but are not to exceed that further when reacting to
carried out by the enemy. harassment.
Remarks: Enemy action may be that of a junior commander and not an
accurate reflection of national intent.

42. Specific Rules.

Desig Detailed Rules Instructions Political Policy

Warning shots from aircraft Charlie: The risk of escalation is
nnn
are permitted. acceptable within declared constraints.
Remarks: Military Policy: Minimum force which may include the firing of
weapons is To be used to attain the aim.

Promulgation

43. Shorthand political policies, specific rules and military policies are detailed in
operational and tactical instructions issued in peacetime by controlling authorities.
Implementation is ordered by a brief signal referring to the document and numbered
Rules (nnn above). Additions and exceptions may also be signalled, together with an
appreciation of the political situation in amplification of the political policy.

44. ROE, based on International Law, are an indispensable part of the national
defence, deterrent posture, and policy:

a. They must be realistic, easily workable, and credible to all sides.

b. No ROE can lawfully prevent the sovereign right to act in self-defence.

1-10
RESTRICTED
 RESTRICTED

Summary

45. In sum, the Rules of Engagement may be defined at 3 levels:

a. The political policy, and legal aspects.

b. The military policy derived from the political policy and legal aspects.

c. Amplifying instructions to units in the form of specific DOs and DON'Ts to

cover all likely situations.

46. All rules must be kept under constant review by commanders, to ensure that
instructions to units reflect the changing situation.

Annexes:

A. Rules of Engagement - Definition of Terms.

B. Examples of Rules of Engagement.

C. Extracts from the Army Field Manual, Vol 1.

1-11
RESTRICTED
 RESTRICTED

ANNEX A TO
CHAPTER-1 TO
AP 1D

RULES OF ENGAGEMENT - DEFINITION OF TERMS

1. In the context of Rules of Engagement, certain terms have specific meanings.
The more important of these are listed below.

2. Sovereignty of Airspace. International airspace is the airspace over the high

seas or any unclaimed territory which is not the subject of the sovereignty of any state.
The Chicago Convention (1944) provides that the sovereignty of a state extends to the
airspace above its territory (land mass and territorial water adjacent thereto).
Sovereignty of a coastal state thus extends to the airspace over its territorial waters
which most states now claim to be 12 nautical miles as laid down by the UN Law of the
Sea Convention (UNCLOS) 1982.

3. Air Defence Identification Zone (ADIZ). An airspace of defined dimensions

within which the ready identification and control of aircraft is required.

4. Intruder. A potentially hostile unit may be operating in international waters and

violating no law but may be engaged in surveillance or intelligence-gathering
operations.

5. Scramble. This is the order given to launch an aircraft as quickly as possible,

usually followed by mission instructions.

6. Air Defence Alert (ADA). Air defence resources, interceptors, SAMs and AAA
are maintained in different states of readiness, termed alert states, to deal with
intruders. Each alert state (Alert 5 etc) is identified by the maximum number of minutes
allowed for the weapon system to get into action - an aircraft to get airborne after the
order to 'scramble', a SAM system to perform missile launch, or an AAA system to
engage the target.

7. Interception. Interception is an action taken by a manned aircraft to establish

contact to identify and/or engage an airborne object. Depending on the situation this
could involve diverting an intruder from its track, flight path or mission, forcing it to land
on friendly territory, or shooting it down if considered hostile.

8. Harassment. This is defined as a calculated act of a provocative nature, falling

short of Hostile Intent, designed to inconvenience or alarm.

9. Shadow. This means to observe and maintain contact (not necessarily

continuously) with an object. Shadowing may be overt or covert.

10. Trail. This means to detect and classify a target, and having done so, to
successfully track the target until contact is considered lost.

11. Lure. This is to take active deceptive measures to attract the attention of an
intruder and move the intruder away from a sensitive area.

1A-1
RESTRICTED
 RESTRICTED

12. Reconnaissance. This is obtaining, through visual observation or the use of

various sensors, information on the activities and resources of an enemy or potential
enemy; additionally, to secure data concerning the meteorological, hydrographic, or
geographic characteristics of a particular area.

13. Surveillance. Surveillance is an operation conducted during peacetime and

periods of tension aimed at building up and maintaining a comprehensive picture of the
enemy's deployment, movement, or exercise activity at sea.

14. Mark. This is to maintain contact with a target from such a position that the
marking unit is at short notice to take offensive action (i.e., weapon systems may be
employed effectively with minimum delay).

15. Counter Mark. This is to provide a direct means of countering the threat
presented by a marker.

16. Intervention. Intervention is an action taken to divert a ship or aircraft from its
track, flight path or mission.

17. Buzzing. Action by an aircraft to emphasize its presence by flying low over a
specific area or target.

18. Covert Action. Action is carried out in such a way as to disguise or hide one's
presence or purpose.

19. Hostile Act. Aggressive action against national or designated friendly forces,
property, or territory.

20. Hostile Intent. Action by a potentially hostile force, which appears to be

preparatory to a hostile act. It is not always possible to determine a given set of criteria
which clearly define hostile intent. Examples of Hostile Intent by enemy units are given
below:

a. A missile-armed submarine which is on the surface, within weapon range

and headed in the direction of a major unit, with its missile house open and its
fire control radar energized.

b. A surface ship which is within weapon range of own major unit with its
missile house open, its fire control radar energized and both radar and missile
launcher pointing at the major unit.

c. An aircraft which approaches a major unit in a known missile launch

profile with its missile fire control radar energized and operating in either the
acquisition or track mode.

d. A submarine closing or being detected within 5000 yards of a major unit.

e. An aircraft which carries out minelaying in international waters.

1A-2
RESTRICTED
 RESTRICTED

21. Hostile Unit. A unit which has committed a hostile act or is declared hostile by
an authorised commander.

22. Identification. Determination of the identity of a unit by means which include

visual recognition, electronic interrogation, electronic support measures, acoustic
information, track behaviour, flight plan correlation or hostile action.

23. Interrogation. Action taken to determine identity.

24. Physical Harassment. Harassment has the element of direct contact or risk of
collision.

25. Positive Identification. Assured determination of identity by specified means.

26. Potentially Hostile Unit or Force. Unit or force of a State or Organization

which has clearly stated or demonstrated its will to commit a hostile act against national
interests.

27. Riding Off. Manoeuvring own unit between escorted unit (s) and opposing force
to cause that force to turn away.

28. Self Defence. Action to protect oneself or one's unit when faced with "an
instant, overwhelming need, leaving no choice of means and no moment for
deliberation."

29. Simulated Attack. Action is carried out when a unit manoeuvres into a firing or
bombing position with necessary sensors energised and with weapons ready to fire or
drop.

30. Visual Illumination. Activation of an artificial light source to illuminate the target
within the visual light spectrum.

31. Warning Off. Informing intruding units that their actions are interfering with
operations may lead to the taking of countermeasures.

1A-3
RESTRICTED
 RESTRICTED

ANNEX B TO
CHAPTER-1 TO
AP 1D

EXAMPLES OF RULES OF ENGAGEMENT

1. Promulgation. An Officer in Technical Command (OTC) at sea will probably

receive a national Rules of Engagement signal from the national commander in chief.
The signal will first state the National Political Policy and then go on to list the Rules
that are in force. These Rules are, of course, carefully aligned to the National Political
Policy.

2. Examples. Examples of rules which might be used with any one political policy
are given below:

Political Policy Rules

Reduce Tension 1. Every effort is to be made to avoid conflict.

2. Warning shots are forbidden.

3. Movement of armament is forbidden.

Maintain Status Quo 1. When the enemy is about to engage with weapons,
minimum force is to be used to prevent the successful
employment of those weapons.

2. Hold the submarine down.

3. Movement of armament is permitted.

Escalate 1. Carry out unrestricted offensive operations using
conventional weapons.

2. Jamming is permitted.

3. Following a successful attack on one or more

friendly units, authority is granted to destroy all enemy
warships in the area.

1B-1
RESTRICTED
 RESTRICTED

ANNEX C TO
CHAPTER-1 TO
AP 1D

EXTRACTS FROM THE ARMY FIELD MANUAL (VOL 1)

SECTION 7 - LEGAL CONSTRAINTS ON THE USE OF FORCE

Knowledge and Observance of the Law of Armed Conflict

1. To ensure compliance with the law of armed conflict, the Geneva Conventions of
1949 and the Additional Protocols of 1977 require states to disseminate their provisions
as widely as possible and to include their study in military training programmes.

Development of the Law of Armed Conflict

2. The law of armed conflict is derived from practice, custom and treaty. As a
matter of humanity, the law contains both the limits which have been placed on the
degree of force which may be used to overcome an enemy who is still fighting and the
principles which govern the treatment of individuals who are the victims of armed
conflicts.

General Principles

3. Limitations on the Means and Methods of Warfare. The principle that the
choice of means and methods of warfare is not unlimited has become firmly established
in the 20th Century. The principle is not related to any weapon system.

4. Unnecessary Suffering. The principle prohibiting unnecessary suffering was

first expressed formally in the preamble to the Declaration of St Petersburg of 1868. It
was confirmed by the Hague Rules of 1907 and the Geneva Additional Protocol I of
1977.

5. The Principle of Humanity. The Martens clause in the preamble to the Hague
Rules of 1899 and 1907 provides that in cases not covered by the rules the inhabitants
and the belligerents remain under the protection and governance of the principles of the
law of nations derived from the usages established among civilised peoples, from the
laws of humanity and the dictates of public conscience.

6. Distinction between Combatants and Non-Combatants. According to the

preamble to the Declaration of St Petersburg, `the only legitimate object which states
should endeavour to accomplish during war is to weaken the military forces.
Combatant and non-combatant distinction has become blurred. It is now clear that all
members of the armed forces (including resistance groups), other than medical
personnel and chaplains, are treated as combatants.

7. Treatment of the Civilian Population. Neither the civilian population as a

whole nor individual civilians may be attacked. This principle only excludes direct

1C-1
RESTRICTED
 RESTRICTED

attacks on civilians. It does not exclude proportionate, incidental damage caused by

attacking military objectives.

8. Undefended Towns. Attacks or bombardments by any means whatsoever of

undefended towns, villages, dwellings, or buildings are prohibited. However, the rule
does not prevent attacks on military objectives in undefended places.

9. Limitation of Attacks to Military Objectives. This traditional rule was

confirmed by the preamble to the Declaration of St Petersburg. The difficulty has
always been experienced in trying to define what amounted to a military objective. In
World War II, it was thought legitimate to attack civilian morale. This, however, resulted
in such hostile criticism that it cannot be regarded as an internationally accepted
practice. Geneva Additional Protocol I of 1977 has attempted to define military
objectives.

10. The Rule of Proportionality. This rule, which strives to maintain a balance
between military necessity and humanity, is the very nub of the law of armed conflict
and has long been regarded as a principle of customary law. It has been fundamental in
shaping conventions on the law of armed conflict, in arms control and the doctrine of
reprisals. It has also been used in the doctrine of self-defence. The rule forbids attacks
which may be expected to cause incidental loss of civilian life, injury to civilians,
damage to civilian objects, or a combination thereof which would be excessive with the
concrete and direct military advantage anticipated.

11. Prohibition of Indiscriminate Attacks. Only a thin line divides incidental

damage from negligent, indiscriminate attacks. Blatantly indiscriminate, in the sense of
blind, attacks have been condemned in the past, offending the rule referred to in
paragraph 49 above. There has been much controversy as to whether certain weapons
are in themselves so imprecise that the likelihood of hitting an identified military target is
too low. However, it must be emphasised that indiscriminate effects are caused by the
use rather than the nature of the weapon. It may be concluded that no weapon is per se
indiscriminate. It is always the use that must be considered.

12. Reprisals. Reprisals are otherwise unlawful acts justified by customary law to
redress a violation of the law by the enemy. Reprisals must be preceded by efforts to
redress the violence by other means or, at the very least, an ultimatum. They must be
aimed at rectifying a wrong. They must be proportionate and must cease as soon as
the violation complained of ceases. Reprisals cannot be taken against persons and
objects specifically protected by treaty against reprisals, for example, prisoners of war,
the wounded and sick, medical personnel and facilities and protected persons.

Treaties Dealing with Weapons

13. Apart from the general provisions, already referred to, dealing with unnecessary
suffering, several treaties deal specifically with the use of weapons. Drawn up in
peacetime, they are a compromise between military and humanitarian interests:

a. The Declaration of St Petersburg prohibits the use of projectiles below

400 grammes in weight which are either explosive or charged with fulminating or
inflammable substances. State practice indicates that the Declaration does not

1C-2
RESTRICTED
 RESTRICTED

apply to tracers, used for target indication or ranging, nor to anti-aircraft and anti-
armour projectiles.

b. The Hague Declaration II of 1899 prohibits projectiles the sole object of

which is the diffusion of asphyxiating or deleterious gases.

c. The Hague Declaration III of 1899 prohibits the use of bullets which
expand or flatten easily in the human body.

d. The Hague Rules 1907, Rule 23(a) prohibits the use of poison or
poisoned weapons.

e. The Geneva Gas Protocol of 1925 prohibits the use of bacteriological

methods of warfare and the use of asphyxiating, poisonous or other gases and
all analogous liquids, materials, or devices. Although the terms of the Protocol
contain an absolute prohibition on the use, some nations, including the United
Kingdom, the United States, and the Soviet Union, have entered reservations to
the effect that it will cease to be binding if enemy armed forces fail to respect its
prohibitions.

f. The Geneva Weaponry Convention of 1981, signed by the United

Kingdom but not yet ratified, prohibits, or restricts the use of certain conventional
weapons which may be excessively injurious, for example, fragmentation
weapons whose fragments cannot be detected in the human body by X-ray or
which have indiscriminate effects on the civilian population.

The Legality of Nuclear Weapons

14. There is no general prohibition on the possession of nuclear weapons by states,

nor is there any rule of international law dealing specifically with their use. However,
the law of armed conflict would apply, as in the case of other weapons.

Effect of Geneva Protocol I of 1977

15. The Protocol has confirmed some of the traditional rules of the law of armed
conflict, modified others and introduced some new rules, the most important of which
are:

a. The prohibition of methods or means of warfare which are intended, or

which may be expected, to cause widespread, long term and severe damage to
the natural environment.

b. The obligation to verify the legality of new means or methods of warfare.

c. The prohibition of starvation and attacks on objects indispensable for the

survival of the civilian population.

d. The prohibition of attacks on works containing dangerous forces, i.e.,

dams, dykes, and nuclear power stations.

1C-3
RESTRICTED
 RESTRICTED

e. Protection for medical aircraft.

16. The United Kingdom has signed but not yet ratified the Additional Protocols of
1977. Further information may be found in the Manual of Military Law Part III (and the
Tri-Service Manual on the Law of Armed Conflict JSP 381).

1C-4
RESTRICTED
 RESTRICTED

ANNEX D TO
CHAPTER-1 TO
AP 1D

EXTRACTS FROM HANDBOOK OF IIHL, SANREMO

SELF-DEFENCE

1. Categories of Self-Defence. International law and the domestic laws of all

nations recognize a right of self-defence, which is the use of force to defend against
attack or imminent attack. Self-defence is available in all situations, including armed
conflict. National laws differ on the definition and content of the right of self-defence.
Consequently, individuals and units will exercise this right by their respective national
laws. For some nations, self-defence is not governed by ROE. For some nations, the
terms “hostile act” and “hostile intent” are related to mission accomplishment, rather
than self-defence. The Handbook has adopted what the drafters view as the more
generally accepted view while recognizing that nations may have different approaches.
In any event, the Handbook is intended for use by all nations. For this Handbook, the
right of self-defence is considered on four levels.

2. Individual self-defence. This refers to the right of an individual to

defend himself or herself (and in some cases other individuals) from an attack or
imminent attack. Some nations permit commanders to limit individual self-defence in
the same way as for unit self-defence.

Annex B Use of Force in Individual Self-Defence

Series 10
Purpose: To regulate the right to use force in individual self-defence.

Rule
10 A Use of force, up to and including deadly force, in individual
self-defence is permitted, except as follows: (SPECIFY).
10 B Use of non-deadly force in individual self-defence is permitted.
10 C Use of force, up to and including deadly force, in individual
self-defence is permitted.
10 D Use of non-deadly force in defence of property where there is
likelihood that destruction of, or damage to, that property will
lead to the injury of (SPECIFY persons) is permitted.

10 E Use of force, up to and including deadly force, in defence of

property where there is a likelihood that destruction of, or
damage to, that property will lead to an imminent threat to the
life of (SPECIFY persons) is permitted.
Note: Defence of property in such situations is an exercise of
the right of individual self-defence, unit self-defence, or the
right to protect the specified persons. Specified persons might
include members of the Force, own state nationals, all
civilians, etc. See Series 40 for measures regarding the use of

1D-1
RESTRICTED
 RESTRICTED

force to protect property where no imminent threat to life

exists.
10 F-Z Spare.

3. Unit self-defence. Unit commanders have the right to defend their unit and
other units from their nation in the face of an attack or imminent attack. For some
nations, the concept of unit self-defence is both a right and an obligation, whereas for
others the concept is only a right. Some nations permit the right of unit self-defence to
be limited by orders from higher authority. Unit self-defence may be extended to units
and individuals from other nations when authorised by the applicable ROE.

Annex B
Use of Force in Unit Self-Defence
Series 11
Purpose: To regulate the right to use force in unit self-defence.
Rule
11 A Use of force, up to and including deadly force, is permitted in
unit self-defence, except as follows: (SPECIFY).
11 B Use of non-deadly force in unit self-defence is permitted
11 C Use of force, up to and including deadly force, in unit self-
defence of (SPECIFY groups or units, e.g. the Force, civilians)
is permitted.
11 D Use of non-deadly force in unit self-defence of (SPECIFY
groups or units) is permitted.
11 E Use of non-deadly force where there is the likelihood of
property damage that will lead to injury of persons in own unit
or other units of own nation is permitted.
11 F Use of non-deadly force where there is likelihood of property
damage that will lead to injury of persons in (SPECIFY groups
or units, e.g. the Force, civilians) is permitted.
11 G Use of force, up to and including deadly force, where there is
a likelihood of property damage that will lead to an imminent
threat to life in own unit or other units of own nation is
permitted.
Note: Defence of property in such situations is an exercise of
the right of unit self-defence in situations where the unit or
persons in the unit are in peril. See Series 40 for measures
regarding the use of force to protect property.
11 H-Z Spare

4. Protection of Others. This refers to the right to defend specified persons

(who are not part of the Force) against an attack or imminent attack. For some nations,
the right of individual self-defence or unit self-defence may not include the right to use
force to defend another nation‟s citizens.

Annex B Use of Force for the Protection of Others

1D-2
RESTRICTED
 RESTRICTED

Series 12
Purpose: To regulate the use of force for the protection of persons who are not
members of the Force.
Rule
12 A Use of force for the protection of others is prohibited.
12 B Use of non-deadly force for the protection of others in
(SPECIFY group) is permitted.
12 C Use of force, up to and including deadly force, for the
protection of others of (SPECIFY group) is permitted.
12 D Use of force, up to and including deadly force, to prevent the
commission of a serious crime against (SPECIFY) persons is
permitted.
12 E-Z Spare.

5. National self-defence As recognised in Article 51 of the United

Nations Charter, refers to the right of a nation to defend itself against armed attack, and
for most nations, the threat of imminent armed attack.

Annex B Use of Force in National Self-Defence

Series 13
Purpose: To regulate the use of force in national self-defence.
Rule
13 A Use of force in national self-defence is prohibited.
13 B Use of non-deadly force in national self-defence of (SPECIFY
nation) is permitted.
13 C Use of non-deadly force in national self-defence of (SPECIFY
nation) when authorised by (SPECIFY authority) is permitted.
13 D Use of force, up to and including deadly force, in national
self-defence of (SPECIFY nation) is permitted.
13 E Use of force, up to and including deadly force, in national
self-defence of (SPECIFY nation) when authorised by
(SPECIFY authority) is permitted.
13 F-Z Spare.

Decisions on whether national self-defence will be invoked are retained at the highest
levels of governmental or executive authority.

6. Hostile Act and Hostile Intent. For this Handbook, the right to use
force in self-defence arises in response to a hostile act (attack) and/or demonstrated
hostile intent (threat of imminent attack).

1D-3
RESTRICTED
 RESTRICTED

GUIDANCE ON HOSTILE INTENT (Appendix 4 to Annex A)

7. Defining Hostile Intent

a. Hostile intent is the threat of the imminent use of force. A determination of

hostile intent is based on the existence of an identifiable threat recognizable
based on both capability and intention.

b. The right to use force in self-defence arises either when forces have been
attacked and/or when there is demonstrated hostile intent. Demonstrated hostile
intent exists when there is a reasonable belief that an attack or use of force is
imminent, based on an assessment of all the facts and circumstances known at
the time. Some nations permit the right of individual self-defence and unit self-
defence to be limited by a force commander (see Series 10 and 11 in Annex B).

8. Ascertaining Hostile Intent. In determining whether an entity is

demonstrating hostile intent, forces will use their best judgment and consider available
intelligence, political and military factors, indications and warnings, and all other
relevant information concerning the capabilities of possible threats in the area of
operations.

9. Indicators of Hostile Intent. No checklist of indicators will conclusively

determine hostile intent. The following are examples of actions that may, depending on
the circumstances, demonstrate hostile intent:

a. Aiming or directing weapons.

b. Adopting an attack profile.
c. Closing within weapon release range.
d. Illuminating with radar or laser designators.
e. Passing targeting information.
f. Laying or preparing to lay naval mines.
g. Failing to respond to the proactive measures listed below.

10. Proactive Measures that May Assist in Ascertaining Hostile Intent.

In addition to the above indicators of hostile intent, time and circumstances
permitting, forces should take proactive measures to assist in determining the intent of
an opposing entity or force, including, but not limited to:

a. Verbal query (see Appendix 7 to Annex C).

b. Verbal warning (see Appendix 7 to Annex C).
c. Visual signals.
d. Noise signals.
e. Physical barriers.
f. Changing course and speed to determine if continuing to maintain an attack
profile.
g. Illuminating with fire control radar.
h. Firing warning shots.

1D-4
RESTRICTED
 RESTRICTED

ESCALATION OF FORCE (EOF) IN SELF-DEFENCE (Appendix 5 to Annex A)

11. Introduction. In all self-defence situations, when confronted with

circumstances in which the use of force may become necessary, it is appropriate to
exercise control over the application of force to ensure that the use of force is
justifiable. Consequently, when time and circumstances permit, the use of force must
always be graduated (also referred to as escalation). Escalation of force (EOF) may
involve sequential actions that begin with non-deadly force measures and may
graduate to deadly force measures. EOF procedures are designed to avoid unjustifiable
use of force. EOF procedures also may assist in the determination of hostile intent (see
Appendix 4 to Annex A).

12. Graduated use of force. Graduated use of force requires that

individuals confronted with a need to use force should always aim to use the least
harmful option available in those circumstances. Indeed, one principal purpose of the
graduated use of force is to create operational time and space in the hope that there
will be no need to escalate to the use of deadly force in self-defence.

13. Use of force options. A variety of options for the use of force may be
available in any given situation. The options available will often include:

a. Presence.
b. Verbal and visual warnings, including the display of weapons.
c. Soft physical pressure.
d. Hard physical pressure.
e. Non-lethal weapons (such as batons).
f. Lethal weapons (such as firearms).

14. General Considerations. Several general considerations should be

taken into account with EOF policy, options, and training:

a. EOF is concerned with employing the necessary option. Use of force

options must be read within their context on every occasion – that is, the
assessment as to what the minimum first response shall be should be made on a
case-by-case basis. Use of an excessive option, where a less harmful option
could reasonably have achieved the aim of neutralizing or removing the threat in
the circumstances encountered, may have legal consequences for individual
users of force.

b. Where time and circumstances permit it is expected that less harmful

options (for example, warnings or warning shots) will be exercised before more
harmful options are used.

c. On some occasions, for operational reasons, ROE may limit access to

certain less harmful EOF options. For example, ROE may prohibit the use of
warning shots. However, it must be remembered on all occasions that ROE and
EOF procedures do not limit the right of self-defence. Subject to any limitations
promulgated in ROE (either series 10 or 11), all necessary and proportional
means and actions may be used in self-defence.

1D-5
RESTRICTED
 RESTRICTED

d. Force preparation should include scenario-based training in EOF situations

that members of the Force are likely to encounter during the operation, such as
checkpoint or access control operations.

e. Use of proactive measures to determine hostile intent (see paragraph 4.4 of

Appendix 4 to Annex A) and EOF measures are similar and may serve the same
purpose.

15. Use of Force in Self-Defence. Subject to any limitations promulgated in ROE

(either series 10 or 11), all necessary and proportional means and actions may be used
in self-defence. Where time and operational circumstances permit, military forces
should warn the threatening entity to allow it to withdraw or cease its threatening 4 ©
actions. Military forces are permitted to use force in self-defence only if non-forceful
alternatives to prevent or deter the attack or imminent attack:

a. Have been exhausted,

b. Are unavailable, or
c. Are deemed insufficient to defend forces in those circumstances.

The use of force is generally authorised so long as the hostile act or hostile intent
continues. The use of force must be proportional, which means that the nature,
duration, and scope of force used should not exceed what is required. (Note: The
concept of “proportionality” in self-defence should not be confused with the concept of
“proportionality” in international armed conflict, which is related to attempts to minimize
collateral damage.)

16. Pursuit. Self-defence, unit self-defence, protection of others, and national

self-defence include the authority to pursue and engage forces that continue to
demonstrate hostile intent. ROE may limit the extent to which pursuit is authorised,
depending upon the military and political situation. Pursuit in self-defence should be
distinguished from “hot pursuit,” which, for this Handbook, is a measure that applies
only in a maritime law enforcement context and is defined in customary international
law and Article 111 of the 1982 Law of the Sea Convention.

17. Relationship between Self-Defence and Mission Accomplishment ROE.

Individuals and units have the right to defend against attack and imminent attack.
Generally, ROE issued for a mission do not limit this right. Because national laws and
policies differ, there will not always be consistency in a multinational force as to when
the right to use force in self-defence ends and the use of force for mission
accomplishment begins. Inconsistencies should be clarified in the planning process.

1D-6
RESTRICTED
 RESTRICTED

CHAPTER - 2

RULES OF ENGAGEMENT IN THE AIR

Introduction

1. Since the Air Force operates in a wide range of roles, the Rules of Engagement
under which it operates will vary greatly. How complicated the rules become will
depend on the aircraft's role and task, but even the pilot of a transport aircraft flying in
an area where there is international tension must operate with a clear set of instructions
in case of interception. This is a simple case and for the more aggressive roles the
problems of laying down rules are far more complicated and the rules even more
necessary.

2. Any aircraft has the potential, either by infringement of foreign airspace or by an

overreaction to incursions by foreign aircraft, to rapidly increase international tension.
Since it will normally be our intention to reduce tension or at the most to maintain the
status quo, the Rules of Engagement will be directed at the use of minimum force.
They must also allow for the fact that there are a variety of reasons for infringement by
the other side, other than the wish for escalation, for instance, a navigational error. The
rules must cater for this.

3. In the air, it is difficult for a pilot to establish the intent of a possible hostile
aircraft without allowing himself to become too vulnerable. Allowing that he can act in
self-defence, the pilot can hardly wait for the other side to open fire since in a missile
engagement this will probably be fatal. The ROE applicable to a particular situation is,
therefore, likely to be related to the scale of activity. For instance, one, possibly hostile
aircraft approaching our national airspace at a high level may be considered an
acceptable risk meriting an identification sortie, whereas multiple low-level intrusions
would be automatically declared hostile and destroyed. Where possible the rules
should define such circumstances and indicate who has the authority to interpret them
and to declare an intruder hostile and order attack.

4. The ability of a pilot to refer decisions and the final interpretation of the current
rules to higher authority may not always be possible since communications may suffer
from blank spots, equipment failure or jamming. The rules must, therefore, be
comprehensive enough for the aircraft captain to be in no doubt of the correct course of
action when he is out of touch.

The Role of the Air Defence Commander

5. The rules of engagement largely determine the steps an Air Defence

Commander (ADC) must take to identify a possible hostile element before engaging. It
is, therefore, necessary to elicit the form of rules likely to be imposed during peace,
periods of tension and general war:

2-1
RESTRICTED
 RESTRICTED

a. Peace. In peace, an ADC is unlikely to be authorised to engage an

aircraft from a potential enemy country even if it is over the ADC's homeland
unless it commits a hostile act, such as:

(1) Opens its bomb doors or drops bombs.

(2) Lands or drops troops.

b. Period of Tension. In a period of tension, the main concern of the

government would be not to do anything that might exacerbate the situation. An
ADC would, therefore, only be cleared to engage potential enemy aircraft if they
behaved in a decidedly hostile manner, for example, they were armed and were
heading directly towards a vital target at a low level and high speed. However,
even under these circumstances, the air defence commander would, almost
certainly, hold fire until the aircraft violated his borders or territorial waters.

c. General War. In general war, the ADC would have to be `reasonably

sure' that the aircraft was not friendly before engaging.

Identification During Peace

6. In peacetime, the following considerations apply:

a. The ADC's task is to ensure that there are no unauthorized flights over the
country by foreign aircraft. He must, therefore, be reasonably sure of the identity
of all aircraft approaching the frontiers.

b. The air traffic flow into a modern industrialised country is likely to be

heavy.

c. Commercial pressures would permit only safety restrictions to be imposed

on traffic flow and routing.

7. To intercept and visually identify every aircraft approaching the country would
place an intolerable burden on fighter resources. However:

a. All flights into a country must submit flight plans.

b. Flights by foreign aircraft into a country must be given prior approval.

8. The ADC is also helped by the fact that:

a. All allied military aircraft which may be required to cross international

borders are fitted with IFF.

b. Civil commercial aircraft which fly international routes are fitted with a
transponder which is equivalent to military mode 3.

2-2
RESTRICTED
 RESTRICTED

c. Military IFF identifies the service, the present controlling authority and the
aircraft type and tail number.

d. Civil aircraft transponders identify the present controlling authority.

9. Categories of Unknown. The main categories of unknown are:

a. Commercial aircraft on whom the flight plan has been delayed.

b. Lost private pilots and defectors.

c. Enemy aircraft on training/reconnaissance/probe missions.

Identification During Periods of Tension

10. The ADC would have to positively identify all aircraft approaching the country.
This raises several critical questions:

a. Is IFF good enough to ensure positive identification? Should the ADC

accept that an aircraft giving the correct responses is friendly? Could the aircraft
have copied the codes being used by other aircraft? IFF may be acceptable if
the aircraft is following an approved flight plan and in contact with either an air
defence or an air traffic radar unit.

b. Should the ADC engage an aircraft not showing the correct codes or no
IFF at all? The answer is surely "No"! Unless the aircraft conforms with sub-para-
a, all the ADC could do as a first step would be to scramble a fighter to identify
the aircraft.

11. Other measures can be taken to reduce the load on the fighter force:

a. Establish an ADIZ.

b. Set up specific air corridors for air traffic within the ADIZ.

12. An ADIZ is an airspace in which special controls pertain to facilitate identification:

a. The following rules would apply:

(1) Aircraft should only enter an ADIZ if they have received prior
clearance.

(2) Aircraft entering an ADIZ without clearance are investigated by an

armed fighter.

2-3
RESTRICTED
 RESTRICTED

b. To be able to carry out the threat of intercepting all aircraft violating an

ADIZ, the boundaries should:

(1) Be inside high-level radar cover at least.

(2) Be within the fighter radii of action.

c. The value of the ADIZ is that:

(1) It should ensure that all friendly movements within the ADIZ are
known.

(2) Because of (1), all unknown movements within the ADIZ are likely
to be by aircraft belonging to the potentially hostile power.

d. An ADIZ is of most value and easy to enforce if over its own territorial
waters and/or territory and reasonably deep. They are of less value or difficult to
enforce if narrow or over international waters. In these instances, the enemy
could wear down the fighter force by frequent flight towards or into the ADIZ.

13. In addition to assisting with the identification problem and reducing the load on
the fighter force, the establishment of an ADIZ might influence the conduct of a
campaign in the following ways during a period of tension.

a. Established too early it could heighten tension,

b. It adds another important factor for the potential enemy to consider. He

knows that he has complete freedom of action outside the ADIZ. He also knows
that if he enters, he could escalate the situation.

Identification in General War

14. In general war, there is likely to be frequent passage through our own lines by
friendly aircraft outbound for or inbound from counter-air, interdiction, or close support
missions.

15. As has been concluded earlier, the only way to positively identify an aircraft is by
visual inspection. If an ADC was required to do this:

a. He would be unable to use his long/medium range SAM. (Short-range

SAMs could be used provided the ADC was happy with the crew's standard of
recognition training).

b. Each fighter interception would take very much longer, target penetration
would increase and the number of interceptions a fighter might be able to make
on a particular raid could be reduced.

c. Fighter aircraft closing to identify fighter-bombers might well find

themselves in a manoeuvring engagement. Such manoeuvres would certainly
prolong the interception. They might even result in the loss of fighter aircraft.

2-4
RESTRICTED
 RESTRICTED

d. Aircraft with long-range head-on snap-down capability would be seriously

affected. These aircraft have, or should have, the capability to engage several
targets simultaneously but not if they must close in to identify their targets first.

IFF

16. IFF is regrettably not dependable yet. This raises further questions concerning
air defence operations:

a. In a situation in which an aircraft is giving either a nil or an incorrect

response should the ADC engage? It could be that:

(1) The aircraft is an enemy.

(2) The aircraft is a friendly one with either a u/s or an incorrectly set
IFF.

The key to the situation is seen in track behaviour. If the aircraft is following laid
down procedures on routing, height, and speed for example, it should be
investigated by a fighter; if it is not, the aircraft should be engaged without further
ado.

b. In a situation in which an aircraft is giving the correct responses, can it be

automatically assumed to be friendly? In this event:

(1) Could it be an enemy copying our codes ?

(2) Can we assume that the system of changing mode 3 codes

regularly and frequently would overcome code copying ?

The key once again is track behaviour. If the aircraft is following the laid down
procedure, the ADC would be justified in assuming that it was friendly; if not he
should scramble a fighter to investigate.

c. Would it be possible to positively control movements into and out of the

battle area ? If this were instituted:

(1) Could the system cope with the likely volume of paperwork?

(2) Are adequate communications likely to be available?

(3) Is such a system likely to be too cumbersome and impose

unacceptable delays ?

2-5
RESTRICTED
 RESTRICTED

d. Would it be possible to use either horizontal or vertical corridors ? In such

a case:

(1) Might they impose unacceptable restrictions on routing ?

(2) Could the enemy identify and use them?

17. As should have been concluded earlier, an ADC could not rely completely on IFF
and strike/attack aircraft will have to accept some restrictions on their freedom,
particularly on the return leg, if the defences are not to inflict heavy losses on our own
aircraft.

General Air Defence Considerations

18. Due to the difficulties surrounding the identification problem in the application of
ROE the following important points need to be considered in the scenario:

a. We should always try to defend in depth and have a mix of weapons; just
as fighters are not the complete answer, nor are missiles.

b. An effective AEW capability is essential if we are to be able to deal with

the low-level attacker successfully. It should be noted that the latest generation
of attack aircraft are all optimised for low-level operation.

c. Whether we like it or not we are likely to have to work under very

restrictive rules of engagement at least at the start of any future conflict. These
rules will further tend to sway the balance in favour of the aggressor.

d. Modern air defence equipment and weapons must have effective ECCM
fits. Also, we should not forget that EW can be made to work for both sides.

e. Unless we can find some answer to the identification problem, we will not
be able to utilize fully the capabilities of the next generation of interceptors. For
example, the F4 has a head-on, snap-down capability against low-level targets;
the advanced fighters will have the same capability but against several targets
simultaneously. These aircraft will only be able to exploit these capabilities if
they are released from the requirement to identify their target positively before
engaging.

Annexes:

A. International Law, Rules of Engagement and Air Warfare by Christopher

Greenwood, M A, LIB - The Hawk 84.

B. Chariots of Fire: Rules of Engagement in Operation Deliberate Force by LtCol

Ronald M Reed.

C. Questionnaire.

2-6
RESTRICTED
 RESTRICTED

ANNEX A TO
CHAPTER 2 TO
AP-1D

INTERNATIONAL LAW, RULES OF ENGAGEMENT AND AIR

WARFARE
By Christopher Greenwood, MA, LLB

Christopher Greenwood is a Fellow of Magdalene College, Cambridge,

and a lecturer in the Faculty of Law of the University of Cambridge, where
he teaches public international law. His undergraduate career was also
spent at Magdalene, where he read law from 1973-77 and was President
of the Cambridge Union Society. He was called to the Bar in 1978.
Publications include articles on various aspects of internal law and
contributions to the international law reports of which he is the assistant
editor.

There is something vaguely unreal, about the whole idea of 'laws of war'. War - or
armed conflict as it has come to be called now that declarations of war are virtually
unknown - is/generally regarded as the antithesis of law. Most observers of
international relations would agree with Cicero's maxim that amidst tire clash of 'arms, laws
are silent. When nations are reduced to fighting each other international law, so its critics
say, has failed and international lawyers should leave the services to get on with the
business of fighting and winning, unhindered by legal arguments which are at best
irrelevant and at worst a positive hindrance.

Yet the concept of rules of war is as old as international law and nearly as old as
war itself. Moreover, the rules of engagement to which military and particularly air
operations are subject generally endeavour to reflect legal, as well as military and
political, considerations. The purpose of this article is to give a brief review of the rules
by which international law seeks to control the use of force and then attempt an answer
to the critics' question: how far do their rules matter when you are at the 'sharp end' in
an international conflict?

The rules of international law on the use of force fall into two distinct groups.
First, there is a body of rules designed to deter States from resorting to force at all by
restricting the circumstances in which it is legitimate to employ armed force of any kind.
Although this branch of the law can trace its origins back to the just war theories first
formulated by writers like St Thomas Aquinas, it is only in this century that it has been
of much practical importance. Secondly, there are the laws of war or armed conflict,
properly so-called, which seek to ensure that if a conflict does break out military
operations are conducted as humanely 'as possible. These rules have a more
substantial history and have long been an important part of military codes. They will
therefore be considered first.

2A-1
RESTRICTED
 RESTRICTED

The Laws of Armed Conflict

Derived from humanitarian principles, military ethics and notions of chivalry, the
laws of armed conflict were for a long time largely unwritten. Today most of these rules
have been codified (though often at a very high level of generality) in international
agreements which have also made substantial amendments and additions to the
traditional law. The principal agreements are:

The Hague Conventions of 1899 and 1907.1

The Geneva Conventions of 1949.2

The Geneva Protocols of 1977.3

There are also several treaties dealing with specific issues, such as the Geneva
Gas Protocol of 19254. All these agreements are binding on the States which are parties
to them and, insofar as they reflect contemporary customary international law, on all
States. In addition, there are various texts which are not in the form of international
agreements, and which have no binding force, but frequently provide evidence of
customary international law. For present purposes, the most important of these texts is
the 1923 Hague Draft Rules on Aerial Warfare.

The laws of armed conflict apply in an international conflict regardless of whether

any of the participants consider themselves to be at war5. However, there is no
satisfactory definition of armed conflict. The rules of armed conflict do not apply to
isolated and minor incidents such as the interception of an intruding aircraft. Since no
clear test exists for determining when such incidents have reached the level of an
armed conflict and since the commander on the spot may not know whether a particular
incident is isolated, it may be very difficult to determine whether a given case is
governed by the law of armed conflict or normal peacetime rules.

What impact do the laws of armed conflict have on operational decisions,

especially in air operations? The 1949 Genera Conventions (the most important
and probably the most effective of the provisions currently in force) concentrate on
the protection of non-combatants and former combatants who have been put out of
the fighting by being wounded or taken prisoner. Many of the rules which they
contain are not, therefore, directly relevant to most front-line operations, Even the
Civilians Convention, which might have been expected to have a greater impact in
this respect, is chiefly concerned with the protection of civilians who are the power
of an enemy State (such as internees and the inhabitants of occupied territory)
rather than with the general civilian population which might be affected by aerial
bombardment. From the standpoint of a commander making operational decisions,
therefore, the most significant provisions of the Geneva Conventions will usually be
1
Texts reproduced in the Manual of Air Force Law.
2
Four conventions dealing with the wounded and sick on land, the wounded and sick at sea, prisoners of
war and civilians; the texts are reproduced in the Manual of Air Force Law.
3
Protocol I deals with international conflicts, Protocol Il with internal conflicts. The Protocols are not yet in
force for the majority of States, including the United Kingdom and the United States. The texts are printed
in Cmnd. 6927.
4
Text reproduced in the Manual of Air Force Law.
5
See Article 2 of the four Geneva Conventions.

2A-2
RESTRICTED
 RESTRICTED

those relating to the protection and marking of hospitals, hospital ships, etc.

The rules which deal with methods and means of warfare and with what is
and what is not a legitimate target have a greater potential impact. However, it is in
precisely this area that the law is least suited to application by Air Force
commanders. Most of the relevant rules are based on the Hague Conventions of
1899 and 1907, which were drafted before the potential of air power had been
appreciated. Many of the provisions of the Hague Conventions are too general to
be of much assistance at· the operational level. Where they are more specific, they
are often ill-suited to air operations. For example, Article 25 of the Hague Rules on
Land Warfare, 1907, provides:

The attack or bombardment, by whatever means of towns, villages,

dwellings, or buildings which are undefended is prohibited. (Emphasis
added).

This provision looks to be of the greatest importance for air operations, but it
was, of course, drafted with land warfare alone in mind. It therefore provides no
answer to the difficult question of how far a town situated in territory which is
occupied and defended by the enemy can be regarded as undefended - a common
problem for an air force attacking targets behind the front lines,

That is not to say that there are no rules of international law governing air
warfare. Customary rules have evolved because of State practice during the
numerous conflicts since the invention of air power. However, customary rules, of
their nature, tend to be general and it is often difficult to discover what they are, let
alone apply them properly. As Professor de Saussure, a former US Air Force
officer, wrote in 1970, in an article entitled 'The Laws of Air Warfare: Are There
Any?':

This paucity of conventional rules has left airmen stranded for authoritative
and practical guidance. The airmail is indeed subject to the general laws of
war to the same general extent as the sailor and the soldier, bill where
does he look for special rules governing his air activity?6

The Draft Rules of Aerial Warfare drawn up in 1923 were an attempt to meet this
need but they were not adopted by States or embodied in a binding agreement.

This picture, however, will change radically when the United Kingdom and the
other NATO States ratify the First Geneva Protocol of 1977. Part IV of the Protocol is
designed to lay down far more precise rules for the conduct of combat operations and is
likely to be of particular significance for air-to-ground attacks. Article 48 begins by
restating the basic rule that civilians and civilian objects are not legitimate targets and
requiring that operations be directed only against military objectives. Indiscriminate
attacks are prohibited7. While this provision is not particularly new - such a principle
was already part of customary international law by 1977 - the Protocol contains three

6
USAF JAG Law Review Vol. XII, NO.4, page 244.
7
Since the Protocol applies to all kinds of military operation, the prohibition in Article 48 will also apply to
guerrilla attacks by 'liberation movements'. The Protocol does not, however, apply to nuclear warfare; see
the UK reservation on signing the Protocol.

2A-3
RESTRICTED
 RESTRICTED

important innovations.

First, the list of civilian and other protected objects which are not to be regarded
as legitimate targets is very extensive. Not only are the lives of the civilian
population not a target, nor are its morale and attacks which, in the words of the
1943 Casablanca Directive, are aimed at 'undermining the morale of the ...
people to the point where their capacity for armed resistance is fatally weakened'
will thus be unlawful8. Objects indispensable to the survival of the civilian
population, particularly food and agricultural resources, are protected 9. Article 56
also protects 'works or installations containing dangerous forces, namely dams,
dykes and nuclear electrical generating stations', even where these objects are
military objectives or are situated near military objectives if the risk to the civilian
population from an attack would be great. This protection ceases only if the work
or installation concerned is used in 'regular, significant and direct support of
military operations' and if the attack is the only way of stopping this support.

Secondly, the Protocol contains more detailed rules for the protection of the
civilian population from the incidental effects of attacks on military targets. The
prohibition of indiscriminate attacks is given greater substance by rules which
prohibit target area bombing10 and attacks.

... which may be expected to cause incidental loss of civilian life, injury to
civilians, damage to civilian objects or a combination thereof, which would
be excessive about the concrete and direct military advantage anticipated11.

Finally, Article 57 of the Protocol lays down a series of precautions which are to
be taken by those who plan or decide upon an attack. These include a duty to do
everything feasible to verify that the target is a legitimate one to minimise civilian
casualties and to refrain from any attack if the harm to civilians is likely to
outweigh the concrete and direct military advantage anticipated. While decisions
on these matters must be taken based on such information as is reasonably
available at the time, if subsequent information reveals that the target is not, in
fact, a lawful one or that far greater civilian losses are likely, the attack must be
suspended or cancelled where this can be done. If there is any doubt about
whether a building (such as a church or a school) which is normally in civilian
use is being used for military purposes, it must be presumed to remain a civilian
object12.

The rules outlined above are intended to strike a balance between military
necessity and the reasonable protection of the civilian population. While far from
perfect, they go some way to meeting de Saussure's criticism of the lack of precise,
concrete rules for the conduct (If air warfare. Many of the provisions codify existing
customary law and should make it easier to draft rules of engagement which can be
applied by those involved in planning and executing air operations at whatever level of
command.

8
Article 51(2).
9
Article 54; see also Article 55.
10
Article 51(5)(a).
11
Article 51(S)(b).
12
Article 52(3).

2A-4
RESTRICTED
 RESTRICTED

The Impact of the United Nations Charter

While the law of armed conflict seeks to regulate the conduct of hostilities, this
century has also seen the rapid - if not always effective development of a body of rules
which aim to restrain States from resorting to force in the first place. The most important
provision in this sphere is Article 2(4) of the United Nations Charter:

All members shall refrain ill their international relations from the threat or
use of force against the territorial integrity or political independence of any
State, or in any other manner inconsistent with the purposes of the United
Nations.

This prohibition must be read in conjunction with Article 51 of the Charter:

Nothing in the present Charter shall impair the inherent right of individual or
collective self-defence if an armed attack occurs against a Member of the
United Nations until the Security Council has taken measures necessary to
maintain international peace and security ....

There is some doubt about the scope of Article 2(4) and, about whether it
prohibits any unilateral use of force which cannot be justified as self-defence. Doubts
also exist about the use of force by regional organisations - a problem highlighted by
the Grenada intervention in which the United States attached great legal significance to
the actions of the Organisation of East Caribbean States. Nevertheless, it is self-
defence which is the principal justification for the use of force by States and, for most
purposes, any unilateral resort to force will be unlawful unless it can be justified by
reference to the principle of self-defence.

It therefore becomes essential to understand the scope of the 'inherent right' of

self-defence. Unfortunately, Article 51 has given rise to more disputes than almost any
other provision of the Charter. Two problems must be noted here. First, there is a
controversy about the concept of anticipatory or pre-emptive self-defence. Defensive
measures in anticipation of an imminent attack were a feature of the law before the
Charter but would appear to have been excluded by the requirement in Article 51 that
there be an armed attack. Yet Article 51 expressly preserves the 'inherent right' and
thus, presumably, the pre-1945 law. Secondly, it is not clear how far the reference to
'collective self-defence' in Article 51 permits a State to go in assisting an ally when it is
under no direct threat itself.

These questions must be answered at the highest political and military levels. On
the face of it, they concern only the initial decision to commence using force and it is
therefore easy to assume that they have no relevance to the planning or conduct of
operations once hostilities have started. Such an assumption would be wrong, for these
rules of law apply not only to the initiation but also to the conduct of hostilities. It is here
that the doctrine of self-defence is more than merely a refinement of the 'just war'
theory. The concept of a just war was that a State should go to war only if it had a just
cause. If, however, it had such" cause, the just war concept imposed no restraint on the
scope or conduct of the war. That is not the case with the doctrine of self-defence. It is
an innate part of any concept of self-defence'- whether in international or national law -

2A-5
RESTRICTED
 RESTRICTED

that it entitles the victim of an attack to use such force as is reasonably necessary to
defend himself. If he goes beyond that limit, he ceases to be acting in self-defence and
he becomes an aggressor.

If a State which resorts to force is to remain within international law, therefore, it

must ensure not only that its initial resort to force is a reasonable measure of self-
defence but also that its subsequent conduct or hostilities do not exceed what is
reasonably necessary for self-defence. It is this continuing application of the Charter
throughout an international conflict which makes this branch of international law
important to those responsible for the day-to-day planning and conduct of hostilities as
well as those responsible for the initial decision to use force.

In case this requirement seems to be asking the impossible of operations officers

two qualifications need to be borne in mind. First, it would be unreasonable to expect
an understanding of the finer points of interpretation of the United Nations Charter at
will levels of command. Rather, it is the function of rules of engagement to incorporate,
along with the military and political restraints, an appreciation of the legal constraints on
operational freedom. Secondly, any notion of confining operations to the minimum of
force necessary in the circumstances should be tempered by the common-sense
maxim stated by an English judge in a self-defence case: 'One cannot expect detached
reflection in, the face of an upraised knife.' The limitations of self-defence must
themselves be applied reasonably.

The Sinking of the General Belgrano

The application of the two bodies of rules described above to operational

decisions taken during a conflict may be illustrated by reference to one of the incidents
of the Falklands conflict. The sinking of the Argentine cruiser General Belgrano was one
of the most controversial British actions during the conflict. It would be unprofitable to
speculate about the facts of that incident, for at the time of going to press those facts
were the subject of such dispute, but it is possible to undertake a brief analysis of the
legal questions raised by the incident because of those facts which are established.

In one sense, the legality of HMS Conqueror's action in sinking the Belgrano is
clear. There was no violation of the laws of armed conflict. By 2 May 1982 when the
Belgrano was attacked, there could be no doubt that Britain and Argentina were in a
state of armed conflict. Although neither State regarded itself as being at war, the week
before the Belgrano sinking had seen the fighting which led to the recapture of South
Georgia, Argentine air attacks on the Task Force and British air attacks on Port Stanley
airfield. The laws of armed conflict were, therefore, clearly applicable. In these
circumstances, a group of Argentine warships was manifestly a legitimate target for a
British submarine.

It is more difficult to give a clear-cut answer to the question of whether the

sinking was a reasonable measure of self-defence and hence lawful under the Charter.
The British position was that the original Argentine invasion of the Falkland Islands and
the Dependencies was an armed attack in violation of Article 2(4) of the Charter a view
which was largely accepted by the Security Council when it adopted Resolution 502
calling upon Argentine to withdraw. Under Article 51 of the Charter, therefore, the
United Kingdom was entitled to use such force as was reasonably necessary to

2A-6
RESTRICTED
 RESTRICTED

repossess the islands. It is in this light that the legality of the Belgrano sinking has to be
considered.

The question, therefore, is not whether the Belgrano was about to attack
elements of the Task Force (thus entitling Britain to take defensive measures against
that attack) but whether the Belgrano and the other Argentine warships that were at sea
posed such a threat to the operation to retake the islands that attacking the Belgrano
became a reasonable and necessary step towards the achievement of that objective,
That question can only be answered by someone in possession of all the information
that was available to the war cabinet and the command centre at the time when the
order to attack the Belgrano was given. If on that information, it was reasonable to
conclude that the Belgrano and its escorts were a serious threat to the operation to
retake the islands and if nothing had happened to deprive the United Kingdom of the
right to repossess the islands in self-defence, the sinking was lawful.

Once the question is seen in this way, it becomes apparent that the exact course
of the Belgrano at the time that it was attacked, while still of some importance is far less
significant than some commentators have made but, for it was the Belgrano's threat to
the overall operation rather than the danger which it might have posed to ships in the
immediate vicinity which was critical. Similarly, the fact that the vessel was sunk outside
the Total Exclusion Zone is far from decisive. A glance at the Exclusion Zone
pronouncement shows that the United Kingdom was making clear that Argentine ships
and aircraft which came within 200 miles of the Falklands would be presumed to be
hostile, but it expressly preserved Britain's right to take measures of self-defence
outside the zone, a fact which had been stressed in a further communication to the
Argentine Government a few days before the attack on the Belgrano. While the fact that
the ship was outside the exclusion zone at all material times vas a fan to be considered
in determining what force was reasonably necessary, it was certainly not conclusive.

Conclusion

There are, therefore, substantial bodies of international law which fall to be

applied by those responsible for the planning and conduct of military operations. Even if
operations officers will usually be required to apply rules of engagement based in part
upon the rules of international law rather than the rules of international law themselves,
some awareness of the relevant provision of international law may give valuable insight
into the reasoning behind those rules of engagement.

Yet the question remains do these rules of international matter in the real world?
It is, of course, rare that a legal tribunal will solemnly adjudicate on the legality of an act
like the sinking of the General Belgrano, although a glance at the volumes of war
crimes trials after the Second World War will show that such inquiries are not as rare
i1S one might imagine. The possibility of a trial is not the reason that the laws of armed
conflict and the use of force matter. They are important for quite different reasons.

2A-7
RESTRICTED
 RESTRICTED

In the first place, these laws are not the arbitrary products of lawyers' minds.
They have evolved over a long period because of State practice, humanitarian
principles, military ethics and much else. The consequence is that they reflect, albeit
imperfectly, vividly held concepts of international morality and international political
expectations, an authoritarian regime engaged in total war may well feel that it can
ignore such considerations but most conflicts since 1945 have been limited wars. When
a State engages in limited hostilities for objectives short of the total subjugation of an
opponent, it will want to prevent the conflict from becoming more extensive than is
necessary. It will also wish to minimise the harm which the conflict may do to its
relations with neutral States; -In these circumstances, a flagrant disregard for
international law can do immense harm.

That international opinion is sensitive to such a disregard for the law is plain, one
has only to look at the number of those States normally hostile to the whole idea of
colonial possessions who were sufficiently affronted by Argentina's invasion of the
Falklands to support, or at least acquiesce in the British action. Similarly, the skilful
presentation of Britain's case, including the legality of its actions to recover the islands,
did much to win over public opinion in the United States to favour United States support
for Britain. Elsewhere, it has been allegations about the use of poisonous gas and
illegal attacks on neutral shipping which have provided virtually the only news about the
Gulf War to arouse international interest in 198. Of course, legal considerations are not
the only factors in these cases, but they are nevertheless of considerable importance,

In a democracy, there is a further factor to be considered - the effect of a breach

of international law on domestic opinion. Many modern conflicts - the Falklands,
Lebanon, Vietnam are very public affairs, fought under the watchful eve of the television
cameras. Violations of the laws of war will often reach a large audience and can have a
very adverse effect on public support for a war, the My Lai massacre example, did much
to undermine public support in America for United States improvement in Vietnam.

International law cannot prevent wars any more than having a law of murder will
automatically put a stop to illegal killing. The law of armed conflict cannot ensure that
wars will be fought cleanly and with as much humanity as possible. However, the fact
that the law cannot be 100% - even 50% - effective is no reason to disregard it
altogether. A law of war may strain the imagination but war without law is too horrible to
contemplate.

2A-8
RESTRICTED
 RESTRICTED

ANNEX B TO
CHAPTER 2
AP-1D

CHARIOTS OF FIRE : RULES OF ENGAGEMENT IN

OPERATION DELIBERATE FORCE
Lt Col Ronald M Reed

The title of this chapter doesn't refer to an Academy-Award-winning Hollywood

movie dealing with a race to victory. Rather, it refers to the effective, albeit highly
centralized and restricted, rules of engagement (J3.0E) that applied to military air
operations during the North Atlantic Treaty Organization's (NATO) Operation Deny
Flight from 1992 to 1995 and its sub-operation, Deliberate Force, in August and
September 1995. Maj Gen Hal Homburg, code name Chariot, director of the Combined
Air Operations Centre (CAOC) at Vicenza, Italy was the lowest-level commander who
had the authority to approve weapons release during these two operations. Unless a
superior commander authorized such a release, NATO aircraft struck nothing in the
tactical area of operations without Chariot's approval. This fact, combined with some
unique provisions for close control and coordination with United Nations (UN) forces
(the dual key" phenomenon, discussed later) meant that ROE had to be restrictive
enough to satisfy the political and operational sensitivities of both NATO and UN
political and military authorities. At the same time, ROE needed to be flexible enough to
provide for force protection as well as the accomplishment of assigned military
objectives. The Deny Flight/Deliberate Force ROE which successfully walked this
tightrope of competing influences offers some insight into appropriate ROE for coalition
peace operations of the future.

Before discussing the implementation of ROE in Deliberate Force, one must

provide a framework for analysing these rules. This chapter does so by examining the
general nature of ROE, particularly its function and importance in military operations: by
presenting several historical examples that illustrate certain issues which have affected
military operations: by addressing the actual ROE for air operations in Deliberate Force
and following tile evolution of ROE from tile beginning of NATO air operations in Bosnia
in 1992 until the conclusion of Deliberate Force in September 1995; by identifying
several key issues that highlight the more problematic situations faced by military
planners and operators in the Balkans: and by discussing implications for tile future
development of ROE.

Nature and Definition of ROE

Abraham Lincoln said that force is all-conquering, but its victories are short-lived."2
This assertion illustrates the concept that unbridled force may help achieve short-term.
tactical-level success, but without some strategic direction or guidance, the chance of
realizing overall objectives is diminished. Carl von Clausewitz recognized this fact in his
discussion of the relationship between policy and war: "At the highest level, the art of
war turns into policy-but a policy conducted by fighting battles rather than by sending
diplomatic notes. The assertion that a major military development, or the plan for one.
should be a matter for purely military opinion is unacceptable and can be damaging"

2B-1
RESTRICTED
 RESTRICTED

(emphasis in original).3 The contemporary mechanisms that interconnect the political-

policy sphere of influence with the military-strategic sphere of influence and that ensure
compliance with national and international legal constraints are rules of engagement.

By delineating the circumstances and limitations under which one may use military
force, ROE provides the method to ensure relevancy and congruency between the
means (military force) and the ends (political/diplomatic objective) one seeks, while
always ensuring compliance with international law arid the law of armed conflict. These
political/diplomatic military and legal influences converge to form the basis for ROE.
Capt J. Ashley Roach, a judge advocate in the US Navy, notes that ROE results from a
composite of these three factors (fig. 14.1).4 The relative influences of each of these
factors will vary according to the circumstances; therefore the diameters of the circles
as well as the degree of convergence differ according to the situation.5 The
political/diplomatic circle represents the assurance that military operations are
conducted by national policy According to Roach. ROE should be flexible enough to
accommodate changing circumstances and should be designed to allow military
courses of action that advance political intentions with little chance for undesired
escalation or reaction.6

Figure 14.1. ROE Influences (Adapted from Capt J. Ashley Roach, "Rules
of Engagement," Naval War College Review 36, no. 1 [January-February
1983]: 46-48)

The military circle takes into consideration the practical operational

considerations for a particular military situation. ROE represents the upper
bounds on the freedom of the commander to use military force toward
successful mission accomplishment ROE guides the commander in
balancing the enemy threat (and the concurrent needs for self-defence) with
the need to avoid conflict escalation. According to ROE expert D P
O'Connell the conduct of operations in tense situations always involves a
nice balance of threat and counter-threat on the part of both sides and the
main purpose of rules of engagement is to prevent that balance (from)
being disturbed by thrusting the apparent necessity of self-defence too
obviously upon one player rather than upon the other8 O'Connell would
consider a tension situation to be a confrontation between two or mort'
military forces in which both the threat of the use of force and the desire to
avoid escalation are present. The military circle, therefore, represents this
balance of threat capability and intent that one must consider when

2B-2
RESTRICTED
 RESTRICTED

developing appropriate ROE.

The legal circle represents the combined domestic and international

legal considerations that one must adhere to in any military operation. The
laws of war provide the absolute limit on the use of force in any conflict
scenario. Dome-sue laws and regulations also provide restrictions on the
employment of force (such as restrictions on the use of certain weapons,
including approval requirements of the National Command Authorities
(NCA) for use of nuclear weapons or chemical and riot-control agents).
Although obligations of US and international law always influence and limit
ROE the rules normally operate well within the boundaries of the law.9

Commanders often limit the use of force by making decisions that are
more restrictive than legal or ROE constraints. For example, W. Hays Parks
cites the freedom-of-navigation exercise conducted against Libya's claim
over the Gulf of Sidra in 1981.10 Even though both international law and
ROE in effect at the time allowed the use of force in self-defence against
demonstrated hostile intent, the commander issued orders that US forces
not fire unless first fired upon. In this case, two Libyan Su-22 Fitters
engaged two Navy F-14s on a head-to-head intercept, clearly
demonstrating hostile intent. The F-14s withheld fire until the Su-22s fired
air-to-air missiles. In response, the Navy fighters quickly evaded the
oncoming missiles and then downed both Fitters. This example shows the
distinction between a commander's rights under the law, his or her authority
under ROE, and the exercise of his or her discretion.

The legal circle also represents ROE's raison d'etre: the inherent right
to self-defence, which is the foundation for the US Standing Rules of
Engagement (SROE). When considering ROE, one normally first thinks of
the need to restrict the use of force and provide guidance to commanders
on the constraints within which they must operate. In reality, however,
written SROE does not exist today because of a need to restrict military
operations politically; rather, SROE resulted from incidents involving the
USS Stork and the US Marine Corps barracks in Beirut, Lebanon, as well as
the need to ensure that commanders understand the right and obligation to
protect themselves and their forces.12 The commander's principal task in
any military operation is to take all necessary and reasonable actions to
protect his or her forces from attack or threat of imminent attack. The legal
standard for the use of armed force in self-defence remains the same
whether protecting the individual, the unit, the aircraft or the nation: a
situation must require the use of force (necessity) and the amount of force
must correspond to the situation giving rise to the necessity
(proportionality).13
Necessity arises not only when an armed attack occurs (hostile act)
but also when one confronts the threat of imminent attack (hostile intent). In
other words, a commander need not absorb the first shot before returning
fire. The often-perceived requirement to fire only if first fired upon is a
legally false important concept in today's world of high-speed and high-
lethality weaponry. Waiting to be fired upon can have disastrous results.
Nevertheless, several nations view the issue of hostile intent in a different

2B-3
RESTRICTED
 RESTRICTED

light, especially in military peace operations. At a recent legal conference in

the Netherlands, Canadian and British representatives indicated that they
do not follow ROE based on hostile intent in any peacekeeping operation,
even to the point of suffering casualties before using military force.14 In their
opinion the need to maintain impartiality and to avoid becoming a party to
the conflict carries a higher priority than the right of self-defence. They view
casualties incurred because of this inhibition of the use of force as the cost
of doing business. As discussed later, international law does not require this
restriction and it is inconsistent with US SROE.

Joint Publication (Pub) 1-02. Department of Defence Dictionary of Military

and Associated Terms defines ROE as directives issued by a competent
military authority which delineate the circumstances and limitations under
which United States forces will initiate and/or continue engagement with
other forces encountered.15 Although this definition adequately defines ROE
at the strategic operational level one can more practically define ROE at the
operational and tactical levels.16

Several issues relating to the US definition of ROE become apparent

immediately. Under the current US SROE competent military authority refers
to the NCA.17 Although combatant commanders may augment SROE to
reflect unique political and military policies, threats, and missions specific to
their areas of responsibility, 'one must submit any changes resulting in
different rules governing the use of force to the: „CA for approval Obviously,
political policy makers use SROE as a mechanism to ensure that military
commanders completely understand when and how to use force to support
policy objectives. Therefore, policymakers have the final authority on ROE
for the use of force in any given situation.

Next, delineating the circumstances under which one may use military
force relates to the fact that situations often drive ROE. The amount of force
one can use in a given situation depends upon a variety of political, military
and legal factors that meld to create the contextual environment for that
engagement. The situation may have geographic implications. For example,
during the Rolling Thunder campaign in Vietnam, air commanders could not
attack targets within a 30-mile radius from the centre of Hanoi, a 10-mile
radius from the centre of Haiphong, and within 30 miles of China. 19 Another
type of geographical implication is the exclusion zone. In the Balkans
conflict, the entire area of Bosnia-Herzegovina represented a no-fly zone
(NFZ); ROE provided for the engagement of any unauthorized aircraft
operating within that zone.20 Similarly, ROE also applied to other geographic
zones in the Balkans.21
Another aspect of the situational nature of ROE relates to the type of
conflict. ROE for peacekeeping differs from ROE for peace enforcement or
limited war. Peacekeeping generally involves the most restrictive ROE due
to the need to maintain strict impartiality on the part of the peacekeeping
force. Because of the need to restrict the use of force, ROE for
peacekeeping operations typically limits it to self-defence only. Peace
enforcement and limited war, however, presuppose the use of force as a
coercive mechanism to change the behaviour of a particular party. ROE for

2B-4
RESTRICTED
 RESTRICTED

these operations often is less restrictive and authorizes the use of force in a
wider range of situations.

Adam Roberts notes that peacekeeping is, notoriously, a very different

type of activity from more belligerent or coercive use of force, and the
differences cause serious problems. The three principles on which
peacekeeping operations have traditionally been based (impartiality,
consent of host states, and avoidance of use of force) are different from the
principles on which other uses of force have been based. Further, the
dispersion of forces and their lightly armed character, mean that they are
intensely vulnerable to reprisals in the event force is used on their
behalf.22This problem, discussed later, played a significant role in the
development of ROE for NATO military operations in the Balkans. The
inherent incompatibility between the UN's role in providing peacekeepers on
the ground and NATO's role in providing peace enforcers in the air became
obvious during Deny Flight, which occurred before the execution of
Deliberate Force in August 1995.

Additionally delineating the circumstances refers to the basic purpose

of any ROE: the inherent right to self-defence. According to the introduction
to the US SROE, the purpose of these rules is to implement the inherent
right of self-defence and provide guidance for the application of force for
mission accomplishment.23 Further, these rules do not limit a commander's
inherent authority and obligation to use all necessary means available and
to take all appropriate action in self-defence of the commander's unit and
other U.S. forces in the vicinity.24 The concept of self-defence in SROE
covers situations involving both hostile acts and hostile intent. As
mentioned above, SROE defines the elements of self-defence as (1)
necessity (a hostile act occurs or a force or terrorist unit exhibits hostile
intent) and (2) proportionality (the force used must be reasonable in
intensity, duration, and magnitude, based on all facts known to the
commander at the time, to decisively counter the hostile act or hostile intent
and to ensure the continued safety of US forces).25

SROE‟s provisions apply not only to individual self-defence but also to

national collective, and unit self-defence. National self-defence is defined as
the act of defending the United States, U.S. forces and in certain
circumstances, U.S. citizens and their property, U.S. commercial assets and
other designated non-U.S. forces to reign nationals and their property from
a hostile act or hostile intent.26 Collective self-defence a subset of national
self-defence authorizes the NCA to provide for the defence of designated
non-US forces personnel and their property. Unit self-defence entails the act
of defending a particular unit of US forces, including elements or personnel
thereof and other U.S. forces in the vicinity against a hostile act or hostile
intent.27 In essence U.S. forces have the authority and obligation under
SROE to use all necessary means available, and to take all appropriate
action to defend themselves, their unit other US forces in the vicinity and
(with NCA approval) non-US forces against hostile acts or hostile intent.

The precise meaning of all necessary means depends upon the

2B-5
RESTRICTED
 RESTRICTED

unique circumstances of a particular situation. SROE provides guidelines for

self-defence: (1) Attempt to Control without the Use of Force: (2) Use
Proportional Force to Control the Situation: and (3) Attack to Disable or
Destroy,28 each of which focuses on military necessity and proportionality.
Normally, the use of force in a self-defence scenario is a measure of last
resort. When time and circumstances permit, one should warn a hostile force
and give it a chance to withdraw or cease threatening actions. When one
must use force, the nature, duration, and scope of the engagement should
not exceed whatever is needed to "decisively counter the hostile act or
hostile intent and to ensure the continued safety of U.S. forces or other
protected personnel or property. Any attacks in the name of self-defence to
destroy or disable a hostile force are allowed only when they must be used
to prevent or terminate a hostile act or hostile intent. When a force ceases to
pose an imminent threat, one must terminate any engagement of that force.
Thus, the provisions for self-defence permit no retaliatory strikes. If a hostile
force remains an imminent threat, however, US forces may pursue and
engage it under the concept of immediate pursuit of hostile foreign
forces.30The discussion of self-defence in SROE assumes the commitment
of a hostile act or the existence of the threat of force by a foreign or terrorist
unit.

Although it is generally easy to identify and define a hostile act on the

ground and in the air, the elements that constitute hostile intent vis-a-vis
aerial ROE are not easily delineated. For example, at what point does an
enemy aircraft or surface-to-air weapon provide sufficient indications of
hostile intent to warrant the use of force in self-defence? Normally,
manoeuvring into a weapons-release position or illuminating friendly aircraft
with fire-control radar indicates hostile intent. The presence of other factors,
however, such as defensive manoeuvring capabilities of the friendly aircraft
or stealthy characteristics, might negate the threat.

To give adequate guidance on what is and is not allowed one might

be tempted to define and legislate every conceivable type of threat that a
pilot could encounter. However, not only is it impossible to predict every
eventuality, but the more detailed the ROE becomes, the more such rules
restrict the flexibility and judgment of the person best able to correctly
assess the threat pilot in the air. This difficulty illustrates one of the most
troublesome dilemmas in drafting effective ROE-the desire for clear
unequivocal guidance as to what pilots may or may not do and for
maximum latitude in exercising their judgment and discretion.31

During the stages leading up to Deliberate Force the US Navy on

several occasions requested more specific ROE regarding air-to-air and air-
to-ground threats of hostile intent. General Homburg, sensitive to the need
for flexibility and judgment, responded by stating. You guys are professional
pilots; I'm not going to give you a cookbook, and I'm not going to tie your
hands.32 This guidance is consistent with the conclusions of Parks: In
preparing ROE for a particular situation threat, or operation less always is
better than more, to allow the individual in the 'hot seat' maximum latitude in
making decisions when being confronted with a threat.33 As mentioned later

2B-6
RESTRICTED
 RESTRICTED

in this chapter even though ROE for Deliberate Force extensively restricted
the employment of military force the rules never infringed upon the
judgment of the pilot to respond to an immediate and unavoidable threat.

Ultimately ROE bridges the gap between the policy maker and the
military commander. The primary consideration for the policy maker is
ensuring control over the use of military force so it best serves national or
alliance objectives. The primary consideration for military commanders is
ensuring that they can defend their forces and employ them to accomplish
the assigned mission most effectively. According to a RAND study by Bradd
C. Hayes, these two considerations provide both the foundation and
dilemma for people who draft the ROE dilemma of action versus reaction. 34
To maintain close political control and avoid the escalation of hostilities.
ROE in peacetime (including peacekeeping operations) tends to be more
restrictive. However, as Hayes correctly points out the political price for
reaction and hesitancy can prove unacceptably high-witness the incidents
involving the USS Stark and the Marine barracks in Beirut.35 Conversely
pre-emptive: self-defence can also have high political costs-witness tile
shoot-down of the Iranian commercial airliner by the USS Vincennes.36

Because one must often sacrifice military objectives to achieve

political objectives.37 Writers of ROE must develop and evaluate these
rules within the context of both the political and military considerations of a
particular operation. According to Clausewitz, war is not merely an act of
policy, but a true political instrument, a continuation of political intercourse
carried on with other means .... The political object is the goal war is the
means of reaching it, and means can never be considered in isolation from
their purpose.38Consistent with the description of the nature and purpose of
rules of engagement, several historical examples illustrate their influence
and impact on past military operations in ways that build on the framework
used here to analyse ROE in Deliberate Force.

Background/Historical Examples

The bombing of the US Marine barracks in Beirut, the attack on the

USS Stark, and the shoot-down of the civilian Iranian airliner by the USS
Vincennes illustrate several important matters about the impact of ROE on
military operations. Each of these incidents provides insight into how ROE
can affect military operations and suggests issues one should consider
when drafting ROE.

The bombing of the Marine barracks in 1983, which cost 241 soldiers,
sailors, and marines their lives, shows how interpretation of ROE can have
disastrous results. Because the commander of the 24th Marine Amphibious
Unit considered his mission to maintain a presence in a peace operation he
did not want his troops to look or act like an occupation force.39Although
the Marines initially faced a permissive environment in Beirut, it became
increasingly hostile the longer US forces remained. 40 Neither the immediate
commander nor the chain of command noticed this shift in the threat
environment, and this lack of sensitivity led to ROE, in effect at the time of

2B-7
RESTRICTED
 RESTRICTED

the incident, that did not require the sentries on duty to have a round in the
chamber of their weapons or a magazine inserted into their weapons. 41

At about 0621 on 23 October 1983, a heavy truck loaded with

explosives entered the parking lot at Beirut International Airport, made a
right-hand turn, crashed through a concertina-wire barrier, drove past two
sentries and through an open gate, travelled about 450 feet, and passed
three large drainage pipes before ramming into the building that housed the
battalion landing team. The truck immediately detonated, destroying the
building, and killing 220 marines, 18 sailors, and three soldiers. The
Department of Defence (DOD) Commission Report indicated that during the
few seconds, it took the truck to ram into the team's building, the sentries
could take no action to stop the truck because their weapons were
unIoaded:42 Even though written ROE specifically stated that the marines
could defend themselves against both hostile acts and demonstrated hostile
intent the commander's interpretation (or misinterpretation) of the mission
and the threat environment resulted in an overly restrictive application of the
rules. Having the sentries put their bullets in their pockets violated the first
rule of any inherent right of self-defence.

The crew of the USS Stark found itself operating in the Persian Gulf in
support of reflagging operations for Kuwaiti oil tankers. On 17 May 1987, an
Iraqi F-1 Mirage launched two Exocet missiles at the Stark. The United
States accepted Iraq's apology, which claimed that the attack responsible for
the deaths of 37 sailors and the wounding of 21 others was unintentional.43 A
Navy study chaired by Rear Adm Grant Sharp concluded that the rules of
engagement that were in existence on May 17 1987 were sufficient to
enable Stark to properly warn the Iraqi aircraft, promptly of the presence of a
US warship and if the warning was not heeded the rules of engagement
were sufficient to enable Stark to defend herself against hostile intent and
imminent danger without absorbing the first hit. 44 The study determined that
the commander of the Stark failed to appreciate the obvious change in the
threat environment in the central Persian Gulf. It also concluded that the
commander/watch team improperly understood the use of fire-control radar
to illuminate a threatening aircraft as a measure short of the deadly force-
an act that could have secured the ship's safety. 45 Like the Beirut bombing
this incident illustrates the potential pitfalls of commanding officers failing to
ensure the inherent right to self-defence under ROE in a –non-combat
environment will a dynamic threat.

In both previous examples, the problem did not arise so much from
overly restrictive ROE as from a lack of sensitivity on the part of the
commanders to a changing threat environment. Both examples illustrate
the tendency in non-combat situations and peace operations to be overly
controlling and to curtail reaction to the threat. An effective ROE, therefore,
should be highly flexible so that one can meet changes in the threat
environment will an effective response. At the same time both examples
illustrate that no matter how robustly one writes ROE, nothing can
compensate for the poor judgment of a commanding officer, who must
ultimately decide how to react and employ forces in any given situation.

2B-8
RESTRICTED
 RESTRICTED

The underuse or overuse of force in any scenario may negatively affect

mission accomplishment. The genius of the commander, as Clausewitz
would say, determines the effectiveness of the use of military force in any
military operation.

In contrast to the Beirut and the Stark incidents, the one involving the
USS Vincennes shows how aggressive, pre-emptive self-defence can be
militarily and politically costly. In this case, misinterpretation of the threat
caused a US Navy Aegis cruiser to shoot down an Iranian civilian airliner.
Regardless of whether the blame lies with an overaggressive, trigger-
happy commander or with human and/or mechanical error during a
confused naval battle this tragic mistake resulted in the death of 290
civilians aboard the airliner.

Although the official investigation of the Vincennes incident absolved

the captain and crew from fault, several questions arose after the
completion of the investigation.46 Nearly four years after the incident,
Newsweek ran an article that described the story of a naval fiasco, of an
overeager captain, panicked crewmen and the (Over-up that followed.47 In
response the director of the Oceans Law and Privacy Department at the
Naval War College drafted a memorandum refuting many of Newsweek's
claims.48 Despite differences in interpretation, everyone agrees upon certain
matters of fact.

On the morning of 3 July 1988, the USS Vincennes was on duty in the
Persian Gulf, assisting in escort operations for reflagged Kuwaiti oil tankers.
Responding to notification by the USS Montgomery that approximately 13
Iranian gunboats might be preparing to attack a merchant ship, the
Vincennes proceeded north to investigate. After one of the Iranian vessels
fired upon a helicopter performing routine morning patrol from the
Vincennes, the latter became involved in a surface engagement with the
gunboats. During this time an Iranian civilian airliner took off from Bandar
Abbas (a joint military/civilian airfield in Iran) on a flight path that would take
the airliner over the area of the naval engagement. Misidentifying the
airliner as an Iranian F-14, the Vincennes issued 11 warnings that it should
remain clear of the area and then fired two SM-2 missiles, shooting it down.
Approximately three minutes and 45 seconds elapsed from the initial
identification of the aircraft as possibly hostile until the shootdown.49

The previously mentioned RAND study of naval ROE noted that the
rules in the Persian Gulf applicable during the Vincennes incident had been
changed after incidents involving the USS Stark and USS Samuel B. Roberts
(the latter had struck a mine in the Gulf approximately one month before the
Vincennes arrived) to encourage anticipatory self-defence.50 This study
however did not blame hair-trigger ROE for the shoot-down of the Iranian
airliner. Rather it concluded that the hostile environment and ongoing sea
battle played a prominent role in the decision and that ROE probably did not
significantly affect the decision to use force to deal with the presumed
threat.51

2B-9
RESTRICTED
 RESTRICTED

The official DOD report concluded that under ROE, the primary
responsibility of the commanding officer is the defence of the ship from
attack or threat of imminent attack.52 Based on information he believed to
be true at the time, the captain of the Vincennes shot down the Iranian
airliner in self-defence-a clear case of the-damned if you do, damned if you
don‟t dilemma faced by military commanders operating in non-combat
environments euphemistically called operations other than war. The best a
commander can hope for is ROE flexible enough to be relevant to a
changing threat environment information and communications to appreciate
when those changes take place, and experience and judgment to make the
correct decision when faced with a threat that might require forcible
response.

The Beirut Stark and Vincennes incidents all illustrate the importance not
only of having effective and flexible ROE but also of the judgment of the
commander in every conflict. Of course, the commander's, for that matter,
judgment depends on the real-time information available at the point of
decision, The fog and friction of war often inhibit or distort the flow of
accurate and timely information. ROE, therefore, must provide guidance
consistent with the political and military realities of the conflict to bridge the
gap between military means and political ends in an uncertain environment.
The rest of this essay uses this test of consistency and relevance to explore
and evaluate the evolution and execution of ROE in Operation Deliberate
Force.

Evolution of ROE in Operation Deliberate Force

The rules of engagement for Deliberate Forces arose from various

changes and additions in NATO military operational tasking in the Balkans
brought about by numerous United Nations Security Council resolutions
(UNSCR) relating to Bosnia Herzegovina. Consequently, any full analysis of
ROE for Deliberate Force must begin with an examination of the genesis
and evolution of ROE from the inception of NATO air operations in
Operation Sky Monitor, through Deny Flight, and culminating in the air strike
operations conducted in August and September of 1995. To truly
comprehend ROE requires an understanding of the complex political and
military environment that faced the military planners and operators during
the entire period of operations. The changing character of this political and
military environment, in turn, caused ROE to evolve from a policy of not
using force in the simple monitoring of the UN-declared NFZ to a robust use
of force in the air strike campaign executed in August and September of
1995. The changing environment shaped the intent and purpose of formal
ROE and reflected the impact of multinational/multi-alliance influences on its
formulation.

A flow of contributions and guidance provided the framework for ROE

development in Deny Flight/Deliberate Force (fig. 14.2). The UNSCRs served
as the legal basis under the UN Charter and international law for the use of
military force by nations and/or regional organizations or arrangements in
support of specific provisions outlined in the resolutions. In response to a

2B-10
RESTRICTED
 RESTRICTED

UNSCR a regional organization (in this case NATO's North Atlantic Council
NAC) issues to forces under its control a mandate authorizing the use of
military force the latter limited by the UNSCR and further restricted by any
NATO concerns. The mandate is translated into a military operations plan
(OPLAN) which includes an annexe detailing ROE to be used in the NATO
military operation. For Deliberate Force, the commander in chief of Allied
Forces Southern Europe (CINCSOUTH) promulgated OPLAN 40101,
Operation Deny Flight and the supporting ROE annex. 53 An operations
order (OPORD) contains the next level of ROE in this case. OPORD
45101.5, Deny Flight M issued by the commander of the 5th Allied Tactical Air
Force (COM5ATAF).54 One should note that each subordinate level of ROE
can be no less restrictive than the combined ROE for each of the higher
levels. Subordinate ROE, however, can be more restrictive as long as the
rules remain congruent with the intent of superior commanders and overall
political objectives. Special instructions (SPINS) provide additional ROE
guidance. For Deliberate Force COM5ATAF and the CAOC director issued
periodic SFINS that included a section on ROE and Commander's Guidance
(SPINS 028 was in effect during Deliberate Force). Lastly, each air tasking
message (ATM), which provides daily specific information on targets selected
for attack may contain information regarding special ROE for that attack. To
fully examine ROE for Deliberate Force, one must therefore follow the rules
of evolution through each of these steps. Beginning with NATO's first
involvement in October 1992.

2B-11
RESTRICTED
 RESTRICTED

UNSCR

NAC Decision

OPLAN 40101

OPORD 45101.5

SPINS

ATMs

Figure 14.2. Development of ROE

On 16 October 1992, NATO forces began Operation Sky Monitor in

response to UNSCR 781, which requested member states to assist the
United Nations Protection Force (UNPROFOR) in monitoring the ban on
military flights in the airspace of Bosnia-Herzegovina.56 Although the
resolution created an NFZ, it did not authorize the engagement of
unauthorized flights:-Rather it called upon member states to nationally or
through regional agencies or arrangements (read NATO) provide technical
monitoring and other capabilities to monitor Compliance with the NFZ, 57
Accordingly. ROE for Sky Monitor limited the use of force in self-defence,
with no provisions for the engagement of aircraft based solely upon their
violation of the NFZ.58 The rules that applied during this period of operations
derived from Supreme Allied Commander Europe (SACEUR) Support Plan
10001D, NATO Europe Integrated Air Defence.59 Which attempts to balance
the desire to limit the possibility of provocation and escalation with the
sovereign right to take action in self-defence.60 Under this ROE, one could
engage aircraft only for self-protection. For all other situations, including the
protection for all other friendly forces, pilots had to request specific
approval.61 The restriction on the defence of friendly forces is contrary to the
previously discussed US SROE notion of collective and unit self-defence.62
However, it illustrates the type of ROE expected in an
observation/peacekeeping military operation in which any use of force is
extremely circumspect.

NATO airborne early warning (NAEW) aircraft already involved in

Operation Sharp Guard (the naval monitoring and subsequent embargo
operations in the Adriatic) carried out the monitoring of the NFZ. The addition
of a NAEW orbit established over Hungary with the support of the Hungarian
and Austrian governments in late October 1992 enhanced the monitoring of
the NFZ. The UN noted that more than five hundred flights violated the NFZ
from 16 October 1992 to 12 April 1993.63

The numerous violations of the NFZ ban resulted in the enactment of

2B-12
RESTRICTED
 RESTRICTED

UNSCR 816 adopted on 31 March 1993, which provided an extension of the

NFZ ban to include all fixed-wing and rotary-wing aircraft in the airspace of
the Republic of Bosnia-Herzegovina. It also included provisions for member
states, subject to close coordination with Secretary-General Willy Claes and
UNPROFOR, to take all necessary measures to ensure compliance with the
NFZ ban on flights.64 The resolution led to an NAC decision on 8 April 1993
to enforce the NFZ with NATO military aircraft. This decision resulted in the
development and implementation of OPLAN 40101, which began at noon
Greenwich mean time (GMT) on 12 April 1993 with aircraft from France, the
Netherlands, and the United States.65

Annex E to OPLAN 40101 contained ROE applicable to the

enforcement of the NFZ. The planners who wrote the ROE were vigorously
supervised by the UN, NATO, and individual national authorities.66
Attempted to include the maximum amount of military flexibility within the
politically charged planning environment. Since the NAC approved all ROE
anyone of the 16 sovereign NATO member nations had veto power over the
rules. The military feared that this lowest common denominator approach
would produce ineffective watered-down ROE.67 Fortunately this fear was
never realized, and the resulting ROE provided for the most part robust rules
for self-defence and mission accomplishment. A five- or six-man team that
comprised a small Black Hole type of planning group led by Lt Gen Joseph
Ashy commander of Allied Air Forces Southern Europe (COMAIRSOUTH)
tightly controlled the whole planning process for ROE at Headquarters Allied
Forces Southern Europe (AFSOUTH).

Within the first two weeks of operations, a significant problem arose

with the enforcement of the NFL. ROE provided for the termination of all air
use by the parties within Bosnia making no distinction between fixed and
rotary-wing aircraft. The UNSCR specifically addressed the ban of all
unauthorized fixed-wing and rotary-wing aircraft within Bosnian airspace.
Unfortunately, the engagement of helicopters proved problematic. Ultimately
COMAlRSOUTH issued guidance (endorsed by the UN and NATO) that
defined away the problem by determining that helicopters had no military
significance.69 The thinking was that the risks of accidentally shooting down
a UN helicopter or some other helicopter transporting civilians or casualties
were so great that they outweighed the military significance of rotary wing
aircraft. The memory of the shoot-down of the US Army Blackhawk
helicopter in Iraq by US Air Force F-15s in April 1993 had some bearing on
this conclusion.70

Even though COMAIRSOUTH decided not to engage helicopters the

written ROE remained unchanged. During NATO's monitoring and reporting
of helicopter flights, the authority remained in place to engage helicopters
under certain circumstances.71 This situation illustrates the point that writing
interpreting and implementing ROE are not always coextensive highlighting
the importance of judgment on the part of on-scene commanders. Since
"legislating" the military significance of any item is impossible, a responsible
commander-such as COMAIRSOUTH-must ensure the interpretation and
execution of ROE by the overall concept of operations and political

2B-13
RESTRICTED
 RESTRICTED

sensitivity associated with a given situation.

The next significant change in air operations occurred as a result of

UNSCR 83672 which responded to threats against the previously created
~safe areas" of Sarajevo, Bihac, Srebrenica, Gorazde, Tuzla and Zepa. 73 It
did so by authorizing the use of air power to support the UNPROFOR
mandate of deterring attacks against the safe areas and responding in self-
defence to any attack, incursion, or deliberate obstruction in or around those
areas that affected the freedom of movement of UNPROFOR or protected
humaI1itarian convoys. At a meeting on 10 June 1993, NATO foreign
ministers agreed that NATO would provide protective airpower in case of
attacks against UNPROFOR in Bosnia-Herzegovina by the request for UN
member-state assistance.74 A report by the UN secretary-general on 14 June
1993 asked NATO to ~prepare plans for provisions of the necessary air
support capacity, in close coordination with me and my Special Representative
for the former Yugoslavia" (emphasis added).75 UNSCR 836 and the request
of the UN secretary general led to the deployment of close air support (CAS)
aircraft to the Southern region and NATO's air cover. UNPROFOR.76 The
specific requirement for close coordination with the UN resulted in the
infamous dual-key process.

In addition to the deployment of CAS aircraft, NATO decided to make

immediate preparations for stronger measures, including air strikes, against
the people responsible for the strangulation of Sarajevo and other areas in
Bosnia-Herzegovina, as well as those responsible for wide-scale
interference with humanitarian assistance. At an NAC meeting on 2-3 August
1993, NATO military authorities had the task of drawing up in close
coordination with UNPROFOR operational options for air strikes including
appropriate command and control (C2) and decision-making arrangements
for their implementation. On 9 August 1993, they produced a memorandum
listing operational options for air strikes approved by NAC.77 Out of this
memorandum flowed several crucial ideas and themes that ultimately
affected both ROE and air operations executed during Deliberate Force.
These included C2 coordination between the UN and NATO (the dual key),
the proportionality of force used in air strikes sensitivity to collateral damage
military necessity phased approach to expanded air strikes and the breakout
of target categories into option-one, -two, and -three targets. OPLAN 40101,
change two, annex E. "Rules of Engagement reflects these issues and
themes.78

The breakout of potential targets into three categories had its basis in
the concepts of proportionality military necessity and gradual application of
force.79 The NAC memorandum's discussion on the use of force articulated
these concepts.80 The discussion on air strike options noted that the selection
of targets needed to take into consideration proportionality as well as the
importance of showing resolve and capability: the selection should also
discourage retaliation.81 The concept of operations entailed a phased
approach in which the first phase commenced with an initial use of airpower,
limited in time and scope but robust enough to achieve the desired effect.
Thereafter, if required and when authorized by the appropriate political

2B-14
RESTRICTED
 RESTRICTED

authority, NATO would conduct air strikes in phases that focused first on
the immediate environs of Sarajevo or other areas. These strikes would
have the specific purpose of assisting in the relief of the siege, facilitating
the delivery of humanitarian assistance, and supporting UNPROFOR in the
performance of its mandate. NATO might have to employ subsequent
phases to expand operations to encompass targets that influenced the
sustainability of the siege forces.82 Planners therefore needed to group and
prioritize targets to show target type and geographic locality to assure
congruence with political and legal mandates.

As a result of these considerations. NAC decision MCM-KAD-084-93

established option-one two and three target sets to meet these
requirements. Option-one targets encompassed the first-strike phase and
included not only militarily significant targets but also those such as
specific artillery batteries participating in a siege that visibly impeded or
prevented the implementation of UNSCRs. 83 Option-two targets covered
the initial follow-on phase and included direct and essential support items
such as artillery heavy weapons supply points and munitions sites, C 2
facilities, and early warning radar and surface-to-air missile (SAM) sites.84
Option-three targets covered the expanded operations phase and included
targets of strategic value outside the immediate areas under siege. Many
of the same types of targets discussed in option two but not located in the
immediate area of the siege fell into option three, as well as items such as
military-related petroleum, oil, and lubricants and anything that would tend
to degrade overall military capability throughout Bosnia Herzegovina. By
breaking out the potential targets into separate option sets, military and
political authorities could control the escalation of violence by authorizing
attacks in a phased and gradual approach. For each of the target options,
one fact remained constant matter the option chosen, no attack could
occur without close coordination with the UN through UNPROFOR.

The issue regarding C2 coordination between the UN and NATO

resulted in what has come to be known as the dual-key process. From the
very beginning of Deny Flight, an ongoing dialogue had occurred regarding
what the term close coordination meant and who controlled the use of
military force within Bosnia-Herzegovina.85 As noted above the
requirement to coordinate closely with the UN before initiating CAS or air
strikes came from the language in the UNSCR and the report of the UN
secretary-general. In practice, the dual-key process required approval of
the appropriate level of authority in the NATO and UN chains of command
(fig. 14.3) before execution of any weapons release.

2B-15
RESTRICTED
 RESTRICTED

Figure 14.3 NATO and UN Chains of Command

The NATO chain of command proceeded directly from SACEUR

through CINCSOUTH to COMAIRSOUTH to COM5ATAF and finally to the
CAOC director. The approval authority for CAS support on the NATO side
was delegated all me way down to COMBATAF and the c.-\OC director.86 On
the UN side however the approval authority proved somewhat problematic.
The UN secretary-general himself held approval authority for the first use of
CAS. This UN UNPROFOR ground unit came under attack. a tactical air
control party (TACP) or forward air controller (FAC) assigned to the unit
would initiate a -Blue Sword" CAS request. The request would go through the
air operations control center to Sarajevo for evaluation by the UN PROFOR
commander and. if approved to the force commander of United Nations
Peace Forces (FC UNPF) in Zagreb where a crisis action team cell would
also evaluate the request. It then went to Yasushi Akashi the UN secretary-
general's special representative (SGSR) to the former Yugoslavia and.
finally, to New York for approval by the UN secretary-general himself.

The first attempted request for CAS by a UNPROFOR ground unit

provides an example of the difficulties associated with the dual-key
approach. On 12 March 1994, a French TACP in the Bihac area called for
CAS to stop a Serbian 40-millimeter antiaircraft artillery (AAA) attack on
UNPROFOR troops. Even though an AC-130 was in the area and had the
offending artillery piece in its sights, approval from the UN side of the dual
key was not forthcoming. Due to problems in locating Akashi, the request
was not approved until six hours after the TACP's request. In the meantime,
the AC-130 left the area and came back two or three times. Unfortunately,
when the approval finally came through, the TACP and the AC-130 could no
longer positively identify the target so the former called off the CAS
request.87

The dual-key process also caused some initial confusion regarding the
distinction between receiving approval to drop 'weapons and actual
clearance to do so. OPLAN 40101, change two provided ROE conditions
required for weapons release.88 Concerns for collateral damage and the
imperative to avoid fratricide resulted in ROE that not only required approval
from both UN and NATO channels but also required identification and
clearance from the FAC. Because there was no requirement to ask the FAC
2B-16
RESTRICTED
 RESTRICTED

whether he had received prior approval through the UN chain of command,

the pilot could assume that if the FAC gave clearance for the attack, the
latter had requisite approval from the UN to do so. After receiving Chariot's
approval, the pilot could attack a positively identified target if the FAC gave
clearance to do so.

For the entire period leading up to Deliberate Force, UNPROFOR

ground personnel received approval for CAS a total of three times. The first
use of CAS occurred on 10 April 1994 in response to a request made by
UNPROFOR military observers in Gorazde. After approval by the SGSR, two
US Air Force F-16Cs dropped bombs under the control of a UN FAC.
Approval from the UN chain of command took less than two hours.88 The
next day, UNPROFOR again requested air protection for UN personnel in
Gorazde. Two US Marine Corps F/A-18A aircraft also under the control of a
UN FAC bombed and strafed targets. The last CAS request came in July
1995 during the siege of Srebrenica by the Bosnian Serb army (BSA). 90
Often instead of an approved CAS mission NATO aircraft would engage in
air presence demonstrations conspicuously showing themselves to the
offending ground forces by flying over the area at high speed with
afterburners to scare the enemy into stopping the attack. Even though the
aircraft released no weapons in these passes the demonstrations often
proved effective in temporarily halting attacks on UNPROFOR positions. 91

The ineffectiveness of the dual-key process came under intense fire

after the fall of the Srebrenica and Zepa safe areas in July 1995. Even
though NATO aircraft were available and present during the Bosnian Serb
siege of Srebrenica the UN did not turn its key until nearly three days after
the attack had begun. By this time CAS could not save the safe area. A
report presented by the Dutch Ministry of Defense in December 1995
blamed the UN's misinterpretation of Bosnian Serb objectives and the UN
command structure for withholding NATO airpower to deter the BSA
attack.92 The report stated that despite repeated requests for CAS from the
Dutch peacekeepers on the ground in Srebrenica-with NATO CAS aircraft
on airborne alert over the Adriatic-the UN approved only one last-minute
CAS mission.93 which helped the peacekeepers regroup north of Srebrenica
but did not stop the BSA from taking the town. As a consequence of this
incident, Secretary-General Claes transferred the UN key from Akashi to
Gen Bemard Janvier the FC UNPF. At this point, decisions about how and
when to use military force fell to the UN and NATO military commanders. 94

The close coordination and consultation procedures also applied to

the use of air strikes in response to violations of the UNSCRs. Because
UNPROFOR a lightly armed UN peacekeeping force could protect neither
itself nor the safe areas from BSA attacks NATO airpower became both a
mechanism for providing force protection (ie, CAS) for UN personnel and a
threat to deter the BSA.95 In February 1994: NAC decided to establish a 20-
kilometre (km) exclusion zone around Sarajevo declaring that 10 days after
2400 GMT on 10 February 1994 heavy weapons not removed from this
zone or turned over to control would be subject to NATO air strikes. 96 NAC
authorized CINCSOUTH in close coordination With the UN to air strikes

2B-17
RESTRICTED
 RESTRICTED

against artillery or mortar positions in or around Sarajevo (including areas

outside the exclusion zone) that UNPROFOR considered responsible for
attacks against civilian targets in that city.97 Because the threat of airpower
proved successful in forcing effective compliance with the NATD ultimatums
the air strikes were not required. Ultimately NAC established exclusion
zones around each of the remaining safe areas. The threat of using air
strikes against a limited exclusion zone or a specific piece of military
hardware influenced military operations throughout Deny Flight and
Deliberate Force.

After Krajina Serbs fired on the Bihac safe area from Croatian territory
and in light of the history of problems experienced by UN forces in the
Krajina area dating back to September 1993 UNSCR 908 extended CAS
support to the territory of the Republic of Croatia on 31 March 1994. Later
that year UNSCR 958 extended the mandate under UNSCR 836 for CAS
and air strikes to the Republic of Croatia. 98 Change four to OPLAN 40101
reflected these UNSCRs both of which prompted subsequent NAC
mandates.99 The resolutions and mandates tended to react in a limited and
proportional manner to the nature of the situation in Bosnia and Croatia.
Each time the UN responded to an act of Serbian aggression it raised the
ante-but never in a proactive or preventive manner.

Throughout the remainder of Deny Flight, NATO conducted limited air

strikes by the provisions of UNSCR 836 and 958. One such strike occurred
in response to attacks by Bosnian Serb aircraft flying out of Udbina airfield
in Serb-held Croatia. Although NATO attacked Udbina under UNSCR 958
the UN-not wanting to kill anyone-prohibited strikes against aircraft on the
ground.100Although NATO complied with this restriction it refused to conduct
the air strike without hitting the enemy air defence system that protected the
airfield. Although the UN had the same reservations about killing people
who operated the integrated air defence system (lADS) equipment, NATO
prevailed in the discussion by pointing out that it would conduct suppression
of enemy air defences (SEAD) strikes strictly to defend the attacking
aircrews.101 This reluctance by the UN to cause any kind of casualty
(whether military or civilian) carried through to the execution of Deliberate
Force.

In addition to CAS and air strikes, air-to-air engagements also took

place before Deliberate Force. On 28 February 1994, the engagement and
shoot down of four fixed-wing Galebj/Jastreb aircraft in the airspace of
Bosnia-Herzegovina demonstrated ROE's air-to-air procedures. After
NAEW aircraft detected unknown aircraft south of Banja Luka on the
morning of the 28th, two NATO aircraft (GS Air Force F-16s) intercepted
and identified six Galebj/Jastreb aircraft in the area. By this phase of ROE,
the NAEW aircraft issued two warnings to land or exit the NFZ, ignored by
the Galebj/Jastreb pilots. Indeed, during the warnings the violating aircraft
dropped bombs in Bosnia-Herzegovina. Following the ROE and after
receiving Chariot's approval, the NATO fighters engaged the planes and
shot down three of them. A second pair of NATO fighters (also US Air

2B-18
RESTRICTED
 RESTRICTED

Force F-16s) arrived and shot down a fourth violator. The two remaining
enemy aircraft left the airspace of Bosnia-Herzegovina. 102

This incident illustrates the phased and stepped nature of ROE

developed for Deny Flight. The political and military sensitivity of operations
in the Balkans required flexible ROE able to control the use of force as the
threat warranted and able to avoid the escalation of hostilities. ROE drafted
for OPLAN 40101, after the enactment of UNSCR 836 and the resulting
NATO Military Committee Memorandum MCM-KAD-084-93, contained the
same provisions existing at the time NATO executed Deliberate Force.
Planners linked ROE to a phased approach to operations that depended
upon the situation at hand.

Each of the ROE phases reflects several of the elements noted

earlier about MCM-KAD-084-93. The concepts of proportionality military
necessity, and collateral damage dominated the thought process
surrounding the development and execution of ROE for air strike options.
The single most defining element of every planning and execution decision
was the overriding need to avoid collateral damage and escalatory force.
Thus by focusing on the concepts of proportionality military necessity and
Collateral damage, NATO planners hoped to keep military options congruent
with political objectives. The degree to which NATO adhered to these
principles is likely the ultimate reason for the success of the mission and the
initiation of the peace process in Dayton. Ohio.

After the fall of Srebrenica and Zepa. ROE remained the basic call
unchanged and ready for the upcoming execution of Deliberate Force. The
fall of these two safe areas precipitated the discussions and decisions that
led to the planning and ultimate execution of this operation.

Implementation of ROE in Operation Deliberate Force

After the fall of the Srebrenica and Zepa safe areas in July 1995. initial
discussions by NATO and UN officials dealt with an appropriate response to
the Bosnian Serbs should they make any moves toward taking Gorazde
(later extended to include Sarajevo Tuzla and Bihac). The NATO London
Conference on 21 July 1995 decided that Man's attack on Gorazde would be
met by substantial and decisive airpower.103 NAC decisions of 25 July and 1
August specified meeting further Bosnian Serb action with a firm and rapid
response aimed at deterring attacks on safe areas and employing if
necessary the timely and effective use of airpower until attacks on or threats
to these areas had ceased. 104 The conference may have been a way to
finesse the UU and avoid a confrontation in the UN Security Council where
the Russians probably would have vetoed an increased use of force over the
Bosnian Serbs.105 By creating an NAC decision that "interpreted" already
existing UNSCR mandates the conference avoided the possibility of a
problem with the UN Security Council.

The NAC decision established "trigger" events that would initiate

graduated air operations as determined by the common judgment of NATO

2B-19
RESTRICTED
 RESTRICTED

and U. T military commanders:(1) Any concentration of forces and/or heavy

weapons and the conduct of other military preparations which in the common
judgment of the NATO and UN Military Commanders presented a direct
threat to the: remaining UN Safe Areas or (2) Direct attacks (e.g ground
artillery or aircraft) on the designated safe-areas.106The trigger events
applied equally to each of the safe areas. Adm Leighton Smith,
CINCSOCTH, pointed out that confirming the first trigger would have been
difficult for each of the safe areas because the BSA already had a high
concentration of forces and heavy weapons located there. Ultimately it would
have come down to the difficult matter of gauging the intent of the BSA
(although NATO did have several indicators for doing SO).107

By the London agreement, once the UN and NATO military

commanders agreed that a trigger event had occurred, NATO authorized the
attack of targets associated with option two from MCM-KAD-084-93,
including concentrations of forces. The NATO/UN military commanders could
continue the air strikes as long as they thought they were needed to defend
a safe area and a wider geographic area (zone of action [ZOA]). The strikes
also could target any concentrations of troops deemed to pose a serious
threat to the UN safe area.l08 The NAC decision assigned execution authority
for air actions to Lt Gen Michael Ryan, COMAIRSOUTH.109

Because of the negative experience with the dual-key process and the
perceived inability of the UN political structure to make timely decisions, NAC
stressed the importance of the UN's transferring execution authority for air
actions to its military commanders. 110In apparent agreement with this
request, the UN secretary-general transferred the key from Akashi to
General Janvier as mentioned above. This action showed the UN's trust and
confidence in the ability of military leaders to ensure that any military use of
force would comply with UN mandates and policy guidance without the need
for direct political oversight. Although never executed the NAC decision also
included provisions to authorize expanded operations against option three or
elements thereof if any of the warring factions continued offensive operations
against the safe areas despite air strikes under option two.111

Lastly, the NAC decision addressed the use of SEAD aircraft against
the BSA's lADS.112 NATO would attack enemy IADS if and when air strikes
in support of a safe area commenced and would continue to attack only for
the duration of that operation. One can consider attacks on enemy IADS a
form of preemptive self-defence since lADS pose a threat to friendly air
forces operating in proximity. This policy proved troublesome for many UN
political leaders who viewed attacks on anything not directly located in or
around the safe area exclusion zone as verging on option three.

The NAC decision generated a memorandum of understanding (MOU)

between General Janvier and Admiral Smith, providing the UN interpretation
of the NAC decisions discussed above.113 This MOU covered the phasing of
operations, and operation considerations [including ZOAs, lADS. and air-
land coordination), targeting arrangements (including targeting boards and
approved target lists), and conditions for initiation (trigger events). The

2B-20
RESTRICTED
 RESTRICTED

memorandum led to discussions between General Janvier and Admiral

Smith regarding the types of targets that could be struck and the ZOAs that
would apply if a trigger event occurred.

Initially the UN 'wanted the ZOA limited to closely relate to the attacked
or threatened safe area; it also felt that NATO should direct air strikes only
toward those forces that committed the triggering offence. According to
Admiral Smith, General Janvier thought that the ZOA should extend the
current exclusion zone to 25 km rather than 20 km. Admiral Smith and
NATO, however, thought ZOAs should apply to a much wider geographic
area as long as one could establish a connection to the threatened safe area
and the target.114 After some debate, the UN proposed two ZOAs dividing
Bosnia Herzegovina in half with an overlap in the area around Tuzla (fig.
14.4).115 Accordingly should a trigger event occur in Sarajevo or Gorazde,
planners would activate the southeast ZOA, and aircraft would strike targets
associated with that zone. Alternatively, if Bihac were the location for the
trigger event, air strikes would focus on the northwest ZOA If Tuzla were
attacked or threatened either one or both ZOAs could be activated.

Regardless of the location of the trigger event, NATO could strike lADS
wherever it affected friendly air operations. Consequently, NATO developed
Deadeye Southeast and Deadeye Northwest as two lADS target sets, both
of which were tied to the self-defence needs of the aircrews operating in the
tactical area of operations rather than to ZOAs.116 For all other targets, the
ability to target was based on the location of the trigger event and the active
ZOA.

Figure 144 Zones of Action for Deliberate Force (From Corona briefing
slides, Lt Gen Michael Ryan, Headquarters AIRSOUTH, Naples, Italy, 5
December 1995, United States Air Force Historical Research Agency,
Maxwell AFB, Ala., H-3)

Targeting for Deliberate Force did not mount to an all-out strategic

bombing campaign designed to destroy the Serbs' industrial infrastructure and

2B-21
RESTRICTED
 RESTRICTED

capability to wage war. Rather. NATO limited the targets and used force in a
graduated manner to compel the behaviour of the Bosnian Serbs. General
types of targets for Deliberate Force included LADS (early warning
radar/acquisition radar/SAM sites/communications. etc.) fielded forces (heavy
weapons/troop concentrations). C2 (headquarters/command
facilities/communications) direct and essential support (ammo/supply depot
and storage/supporting garrison areas/logistics areas) and lines of
communications (transportation choke points and bridges). 117 In selecting
specific targets within these categories the Joint Targeting Board (UN and
NATO) worked hard to find easily hit targets with limited potential for collateral
damage.

In addition to the normal concerns about collateral damage, General

Janvier remained extremely sensitive to casualties (including BSA casualties ).
When the general reviewed one of the initial target lists he demanded that a
target labeled barracks be removed because he wanted to avoid killing
people.118 Admiral Smith had failed to explain to General Janvier the distinction
between targets and desired mean point of impact (DMPI). Targets could
involve several individual DMPIs within a target complex. The fact that
planners describe a target as a barracks does not mean that aircraft would
strike a barracks building. The same barracks complex might include a very
lucrative ammunition storage area with a different DMPI than that of the
barracks. Understanding General Janvier's concern, Admiral Smith merely
redefined the DMPls selected for attack, and General Janvier agreed to the
targets.119

The trigger event for Deliberate Force was the shelling of a Sarajevo
marketplace presumably by the BSA. When Admiral Smith saw the results of
the shelling on Cable News Network (CNN) he called the NATO liaison to the
FC UNPF. He told him to tell General Janvier that if the UN determined that
BSA had fired the shell, then NATO requested air strike options according to
the MOU.120 Once Lt Gen Rupert Smith, commander of UNPROFOR (who
was sitting in for the vacationing General Janvier confirmed BSA responsibility,
the UN and NATO turned their keys, and Operation Deliberate Force began.
General Janvier later concurred with this decision.

During Deliberate Force several issues relating to the interpretation and

execution of ROE arose, the most prominent being the degree to which
concerns over collateral damage drove the planning and execution of air
strikes. The most telling indication of the concern over collateral dam age was
the fact that General Ryan personally selected every DMI because he felt that
the political sensitivity of the operation demanded strict accountability on the
part of the air commander. He believed that every bomb dropped or missile
launched not only had a tactical-level effect but a possible strategic effect.
Accordingly, Ryan directed his staff to evaluate all proposed targets and
DMPIs for their military significance and their potential for high medium or low
assessments of collateral damage. For example, intelligence personnel
described the Pale Army supply depot as a key BSA supply but with a high
potential for collateral damage. 121 Similarly they assessed the Pale ammo
depot south as having medium potential for collateral damage and the

2B-22
RESTRICTED
 RESTRICTED

Jahorina radio communication (RADCOM) station as having low potential. 122

These examples illustrate the spectrum of collateral damage probabilities for
various targets. Based upon the importance of each target and its potential for
collateral damage, General Ryan selected the specific DMPls for the attack.

In addition to the actual selection of DMPls, the concern for collateral

damage often drove weapons selection and tactics for weapons release.
Precision-guided munitions (PGM) were the weapons of choice. Of the more
than one thousand munitions dropped during Deliberate Force, nearly 70- per
cent were precision munitions. 123 Of all munitions dropped by US aircraft more
than 98 percent were precision munitions. 124 Although ROE did not specifically
require PGMs it did state that target planning and weapons delivery will include
considerations to minimize collateral damage. 125

At one point during the operation, an aircraft bombed a bridge, using the
standard profile of attacking along the length of the span and released two
PGMs on the designated DMPI. Unfortunately, the second PGM went long and
destroyed a farmhouse located next to the end of the bridge. 126 As a result
ROE underwent modification via the ATM/SPINS to require a ' much more
restrictive approach for attacking bridges.127 Aircrews would now have to make
a dry pass over the targeted bridge, attack on an axis perpendicular to the
bridge rather than along its length and release only one bomb per pass.

When first notified of this change in ROE the commanders and aircrews
at Aviano Air Base (AB) became very concerned feeling that it placed pilots at
increased risk. Despite Aviano's initial refusal to fly the mission is directed
pilots eventually flew the mission successfully. 128 Upon hearing the concerns
of the aircrews over the dry-pass ROE. General Ryan and General Hornburg
decided to rescind the ROE restriction in the ATM of 12 September 1995 the
next day.129 This incident illustrates the difficulty commanders faced in
reconciling the competing demands of force protection and minimization of
collateral damage balance that shifted as operations progressed.

Every section of ROE included a statement concerning the need to

minimize collateral damage. These included sections dealing with the
engagement of air-to-air and surface-to-air systems (take into account the
need to minimize collateral damage): 130 With CAS (limit collateral damage to
the minimum that is militarily feasible) 131 and with SEAD (no unacceptable
collateral damage).132 Clearly concerns about collateral damage represented
the defining issue for Deliberate Force.

Thirty-five hundred sorties and more than one thousand dropped

munitions produced only two confirmed instances of any significant collateral
damage (the farmhouse incident mentioned above and an accidental strike on
a water-treatment plant).133 When questioned by a CNN journalist about
reports of significant amounts of collateral damage, Admiral Smith pointed out
that if such damage had indeed occurred, CNN would be filming the damage
instead of interviewing him. 134 The careful selection of DMPls by General
Ryan, combined with careful attacks by NATO aircrews, resulted in one of the
most precise operations ever conducted.

2B-23
RESTRICTED
 RESTRICTED

Concern for minimizing collateral damage also resulted in one of the

more controversial ROE issues in the Deliberate Force dual correlation
requirement for reactive SEAD strikes. SPINS required all SEAD mission
commanders to see that planning and positioning of SEAD aircraft included
consideration of methods to ensure engagement of the desired target,
minimize chances of engaging unintended targets, and mitigate the impact of
possible high-speed ant radiation missile CHARM) ambiguities. 135 Due to these
concerns, aircraft detecting a possible SAM engagement were directed to
depart the immediate threat area, using an appropriate defensive profile and
countermeasures as required. Only in extreme situations (e.g., experiencing a
threat without reasonable means of escape) were aircrews authorized to use
any measure for self-defence.136 Preemptive HARM strikes were not
authorized without Chariot's approval, and pilots could employ reactive HARM
strikes in Bosnia-Herzegovina or Croatia only if one of the following: conditions
existed.

A. Positive indication of a hostile act (confirmed missile or

projectiles fired). or

B. Dual correlation of positive indications of hostile intent

further defined as aircraft illuminated by surface-to-air (SAM)/anti-
air artillery (AAA) fire control radar and/or any air defence system
radar directly related to these SAM/AM systems.

C. Other instances of demonstrated hostile intent including

other instances by systems other than those described in para B of
this section when approved by Chariot. 137

Although the dual-correlation requirement infringed upon the inherent

right to self-defence, it did reduce the likelihood that reactive and preemptive
SEAD operations would cause otherwise avoidable collateral damage. Dual -
correlation ROE was restrictive but leadership preferred to err on the
conservative side.138 Even though NATO had a limited number of SEAD
aircraft the chance of HARM ambiguity represented too great a risk for
fratricide or collateral damage to allow more extensive use of lethal SEAD by
aircraft without dual-correlation capability. 139

Dual correlation potentially reduced collateral damage; however, it

imposed tactical handicaps on NATO fliers US Navy aircrews for example
were not pleased by the restriction the policy placed on their use of preemptive
HARMs.140 Most SEAD aircraft had no problems with dual correlation since
both the aircraft and the weapon they carried typically HARNs had separate
internal capabilities to achieve correlation. However, for other less capable
aircraft, ROE required that unless an off-board platform provided the second
correlation the aircraft would have to depart the immediate threat area.

The section of ROE dealing with combat search and rescue (CSAR) also
caused some concern. NATO ROE authorizing search and rescue (SAR)
aircraft and naval vessels to use "self-defence force as necessary (emphasis

2B-24
RESTRICTED
 RESTRICTED

added) 141 to ensure the recovery of survivors was inconsistent with US ROE
for CSAR which provided for a greater degree of force in recovering survivors
by allowing minimum force as necessary (emphasis added). 142 Since US ROE
applied to Operation Provide Promise (the humanitarian airlift operation into
Bosnia) missions and NATO ROE applied to Deny Flight and Deliberate Force
missions special operations forces designated to go into the same threat area
to recover survivors could theoretically be authorized a different level of force
depending upon what the aircraft was doing when it was shot down. For
example, if a special-ops helicopter gunship were approaching the location of
the survivor of a downed aircraft the helicopter under ATO ROE could not fire
upon an enemy truck with armed personnel unless it committed a hostile act or
demonstrated hostile intent toward the helicopter (reactive). Under US ROE,
however, the helicopter could use the minimum force necessary to prevent the
enemy truck from capturing the pilot or interfering with the pickup (preemptive).

The difference between the levels of force authorized in the two ROEs
resulted. in part, from the fact that NATO doctrine does not address the
"combat part of CSAR. Understandably, SAR-without the combat perspective-
relates to issues of self-defence. CSAR, a more robust form of SAR that is
unique to the US military, seeks to recover survivors within a hostile threat
environment.

The Navy's Sixth Fleet made several requests to use US rather than
NATO ROE for CSAR operations. In a NATO conference that addressed the
issue of US versus NATO ROE, NATO members agreed that US ROE would
apply to any US forces engaged in a CSAR mission. 143 However, no one made
changes to the written ROE in OPLAN 40101. During Deliberate Force the
actual ROE for a CSAR mission proved very sketchy, especially considering
the political sensitivity of the operation. 144 The Navy continued to press for
more specific ROE and urged the adoption of US SROE. According to Maj Dan
Bush, legal advisor to the CAOC during Deliberate Force, Generals Ryan and
Hornburg decided that because Deliberate Force was a NATO operation,
NATO ROE-in accordance with OPLAN 40101-would apply to everyone
involved. Anyone not using NATO ROE would fail to appear on the ATM
schedule.145 Fortunately, ROE for CSAR was not put to the test during
Deliberate Force.

Targeting and execution differed from ROE in the context of casualties.

Commanders for Deliberate Force repeatedly stressed that stuff (things) rather
than people were the targets for this operation. 146 Although the Bosnian Serb's
lack of personnel compared to the Bosnian Muslim and Croatian forces,
constituted a potential centre of gravity, planners decided not to target them.
In essence, NATO strategy sought to wage aggressive peace rather than
war. Since one of the key strengths of the BSA was its armour and heavy
weapons, NATO planned to take away as much of that strength as quickly
as possible.

NATO's desire to minimize the loss of civilian and military life

constrained tactics and target selection in other ways. In the parlance of the
CAOC staff, structures associated with soft pudgies (people) were not

2B-25
RESTRICTED
 RESTRICTED

intentionally targeted.147 If they were, aircraft often hit them at night, when
people were less likely to be around. For the same reason, General Ryan
decided to move attacks on bridges, originally scheduled for the daytime, to
night. 148

At the same time, ROE and military commanders restricted targeting

options and execution to minimize collateral damage, commanders
continually stressed the importance of force protection. Admiral Smith's
three priorities for Deliberate Force, in order of precedence, were (1) force
protection, (2) minimization of collateral damage, and (3) effective strikes on
targets.149 Specifically designed to ensure the safety of NATO forces, ROE
maximized force protection by limiting the exposure of NATO aircraft to
threatening situations and by allowing aircraft to use force in self-defence
against both hostile acts and demonstrated hostile intent. ROE limited the
exposure of NATO aircraft to threats by prohibiting operation over land in
Bosnia or Croatia without SEAD protection and by prohibiting non-SEAD
aircraft from operating within known SAM threat rings. 150 If an aircraft were
engaged or threatened with demonstrated hostile intent without a
reasonable means of escape (as determined by the pilot at the time)
aircrews could use any measure for self-defence.151the latter defined in
OPLAN 40101 as action taken in consonance with international law to
protect oneself or other Friendly Force in the vicinity. 152

This definition closely tracks the definition of self-defence found in US

SROE discussed earlier in this chapter. ROE therefore proved robust
enough to authorize the use of force to protect NATO forces from any attack
or demonstration of hostile intent. ROE that provided the authority to act and
the flexibility to use sound judgment in determining when and how to use
force in self-defence alleviated concerns about force protection, the lessons
of the USS Stark and the Marine barracks in Beirut were not forgotten.

The combined concerns of force protection collateral damage and the

overall success of the mission resulted in very restricted operations. Aircraft
could strike only assigned DMPIs-no targets of opportunity.153 SPINS
contained specific reference to these constraints: For fixed sites, aircrews
will attack ATM-specified DMPIs only, even if these DMPIs have previously
been hit.154In fact the ATM reinforced the ROE: the only valid target DMPIs
are those assigned via the ATM process or directly assigned real-time by
the CAOC battle staff director Target DMPIs assigned via the ATM are
only valid for the period of that ATM. 155

This tight control resulted in what officials at AvianoAB termed the

effective but inefficient use of air power. From the tactical perspective,
constant changes in targets as well as interference in weapons and tactics
proved frustrating. From the strategic perspective however, NATO air
planners felt they could accept a little inefficiency particularly if that was the
price of avoiding an incident such as the one that occurred in the AI Firdos
bunker during Operation Desert Storm. To ensure effectiveness, leadership
gave supreme importance to the minimization of collateral damage and
ordered the development and implementation of ROE to translate these

2B-26
RESTRICTED
 RESTRICTED

concerns into meaningful guidance for the effective prosecution of the

Deliberate Force.

Implications for the Future

Having explored ROE for Operation Deliberate Force one becomes

aware of several implications for the use of airpower in future peace
operations. First and perhaps the most obvious, is the degree to which the
military restricted its own operations in Deliberate Force. The restrictions
placed on targeting remind one of those used in Vietnam. Unlike the political
restrictions placed on military operations in Vietnam however in Deliberate
Force the military's own restraint limited operations. We may therefore be
witnessing a 180-degree shift in the relationship between political and
military influences in ROE. Rather than politicians tying the hands of the
military the military may now be tying its hands. 156 The Air Force Studies
and Analysis Division came to a similar conclusion in its report on Deliberate
Force: There was nothing in the upper levels of DELIBERATE FORCE's
command and control structure to compare with President Lyndon
Johnson's sometimes weekly sessions picking targets for the Rolling
Thunder campaign over North Vietnam in the 1960s. To an impressive
degree, guided bombs had permitted Air Force officers to internalize the
kind of restraint Johnson wished to impose upon them. In DELIBERATE
FORCE, it was General Ryan himself who exercised most of the restraint. 157

This internalization of restraint by the military leadership is due in

many respects to the years of inculcation in the law of armed
conflict.158NATO political authorities trusted the ability of their military
leadership to take general guidance and plan and execute military
operations consistent with that guidance. One can attribute part of this trust
to the perception that US and NATO military forces understand and comply
with the laws of war. Since Vietnam, US forces have consistently shown that
they conduct military operations in an extremely professional manner and
exercise the utmost restraint in using force to achieve an objective. The tight
control that General Ryan placed over military operations likely will reinforce
the perception that in many respects, the military is a self-regulating
instrument of power. In the future, this demonstration of restraint may allow
the NCA to feel confident that given mission-type orders the military will plan
and execute operations consistent with political objectives and in
compliance with the laws of armed conflict. The commander on the scene
cognizant of the political objectives and sensitive to the unique threat
conditions facing forces in the area of operations is in the best position to
draft effective ROE. The latitude given by political authorities and the degree
to which the military controls its operations are key factors in the
development of acceptable and effective ROE.

A related issue is the importance of congruence of context, objectives

and means in cont1icts such as Deliberate Force. A complete divergence of
objectives and means occurred between NATO and UN forces during Deny
Flight. On the one hand, UN forces who sought to keep peace that did not
exist operated under very restrictive self-defensive ROE and had limited

2B-27
RESTRICTED
 RESTRICTED

military capability. Despite operating as the UN Protection Force they could

not even protect themselves. On the other hand, NATD forces, present
under peace-enforcement provisions, sought to deter or compel behaviour
through the use of military force. NATO ROE, therefore, provided for a much
more robust use of force to achieve the desired objectives.

Differences in ROE on when and how the UN and NATO could use
military force doomed the dual-key process from the start. NATO always
viewed the use of force in terms of compelling the Bosnian Serbs to do or
not do something. However the UN viewed force in the much more limited
context of self-defence. Only after Srebrenica and Zepa fell and the UN
decided to pull back its forces to allow NATO to conduct air strikes, did
congruence occur among the situation in Bosnia, the political objectives and
the use of military force to obtain those objectives. ROE for a peace
operation should relate not to its title (peacekeeping, peace enforcement,
etc) but to the reality of the situation on the ground. If no peace agreement
exists and fighting continues then one should write ROE to provide for the
compellent use of military force until the factions agree to a truce. If a
UNSCR proves insufficient to warrant the more aggressive type of ROE,
then one should deploy no military forces until the adoption of such a
resolution.

The ROE concept of categories of target options, designed to keep

the level of force to the minimum amount necessary, was an excellent
method of ensuring maximum control over escalation. However, the use of
target-option categories can be harmful if one exhausts the target list
without having achieved the political objectives. At that point one either
escapes the level of force to the next target-option category or the use of
force loses credibility. Such was the case in Deliberate Force: negotiators
had great concerns that NATO would run out of option two targets before
convincing the Bosnian Serbs to accept a truce and join peace talks. An
interview with Ambassador Christopher Hill revealed that the US State
Department and Ambassador Richard Holbrooke harboured exactly these
concerns.159

The fact that officials at the State Department had concerns about the
execution of the air operation highlights the need for a State Department
liaison to the air operations centre in any operation in which one uses
airpower as a coercive tool to achieve diplomatic leverage. Future
operations similar to Deliberate Force probably will use airpower to provide
an environment conducive to diplomatic negotiations. To ensure complete
congruence between the desires and concerns of the negotiation team and
the execution of the air operation, the liaison should coordinate these
concerns with military personnel who plan and execute the air operations.
Such a person would not necessarily exercise political oversight but act as a
conduit to ensure connectivity among diplomatic negotiators, military
commanders and planners.

The last major implication of the implementation of ROE in Deliberate

Force is the high percentage of precision weapons used in the operation. To

2B-28
RESTRICTED
 RESTRICTED

reiterate, nearly 70 per cent of all munitions used and more than 98 per cent
of munitions dropped by the United States were precision weapons.
Concerns over the minimization of collateral damage led to this
unprecedented reliance on PGMs. The issue for ROE is whether the use of
PGMs has minimized the chance for collateral damage to the point that one
expects zero collateral damage and legally requires it for future operations.

The law of armed conflict does not require that one conduct military
operations to eliminate the possibility of collateral damage. Arguments that
the existence of PGMs has increased the standard of care required to
conduct military operations to this point or that one cannot use dumb bombs
because they are indiscriminate remain legally untenable. The law of armed
conflict requires that the application of force be by the tests of military
necessity humanity and proportionality.160 Military necessity involves the
right to use any degree or means of force not forbidden by international law
to achieve a military objective. Humanity related to necessity entails the
infliction of suffering injury or destruction not necessary for accomplishing
legitimate military purposes. Proportionality provides the link between the
concepts of military necessity and humanity by balancing the degree of
likely damage to noncombatants with the military value of the proposed
target 161 The general immunity of noncombatants from attack does not
prohibit operations that may cause collateral damage (including death.
injury. or destruction of property). International law requires only that the
military balance the value of the proposed target with the likelihood and
degree of collateral damage. Therefore although PGMs may increase the
number of targets that one could justifiably strike by lowering them to
tolerable levels the collateral damage coincident with striking them by
international law includes nothing that requires their use against any target
of military value. Thus judge advocates should assist in the drafting of ROE
to ensure that no one misinterprets the requirements of international law
and that no one places restrictions based upon erroneous applications of
the law on planning and executing air operations.

The expectation of zero collateral damage however is another matter.

The more the military uses PGMs and shows through the international
press media, guided bomb units hitting the crosshairs overlaying targets of
all kinds, the more the public will expect such precision in the future. As
discussed earlier, such expectations create political pressures that have
just as much impact on ROE and military operations as do legal obligations.
The reality of the situation is that guided bombs sometimes miss. US Air
Force analysis of PGMs after Deliberate Force revealed that nearly one-
third missed their individual aiming points on targets (because of human
error, weather problems, and weapon problems equally).162 The careful
DMPI selection process and the professional weapons-employment tactics
by aircrews kept the amount of collateral damage low. Unfortunately,
military leadership sometimes overstates its case to show the effectiveness
of operations, as did Secretary of Defense William Perry after the
conclusion of Deliberate Force: From Aviano and the decks of carriers in
the Adriatic, we launched one of the most effective air campaigns that
we've ever had. It was over one thousand sorties. Every target that had

2B-29
RESTRICTED
 RESTRICTED

been designated was destroyed, and there was zero collateral damage.
This was a rare instance whereby combination of exclusive use of
precision-guided ammunitions and very strict rules of engagement we
conducted this massive campaign with no damage, no damage to civilians,
no collateral damage of any kind.163

Although the air operations in Deliberate Force achieved a high

degree of success in minimizing collateral damage, one cannot say that no
collateral damage of any kind occurred. Such overstatement may create
unrealistic and potentially dangerous expectations for future air operations.
Political and military leaders need to be sensitive to the fact that the military
can minimize but never eliminate collateral damage. General Hornburg
stressed that this was the mindset of personnel at the CAOC.164 Regardless
of political pressures to minimize collateral damage in the future, one should
base ROE on reality rather than abstract ideals and therefore refrain from
writing rules of engagement under the constraints of zero collateral damage.
Minimizing collateral damage instead of achieving zero collateral damage
should become the political standard for ROE.

In conclusion, the ROE drafted and implemented in Deliberate Force

effectively balanced the competing interests of force protection,
minimization of collateral damage, and mission accomplishment. The CAOC
military leaders' close control over operations proved appropriate
considering the political and military realities of the situation in Bosnia-
Herzegovina. Although the circumstances in Bosnia are unlikely to be
repeated many lessons relating to the evolution and implementation of ROE
will remain applicable to future uses of airpower in support of peace
operations. As long as we evaluate these lessons from the perspective of
the context in which they arose, Operation Deliberate Force offers an
example of the value of well-conceived and masterfully implemented rules
of engagement.

Notes

1. Commander in chief. Allied Forces Southern Europe (CINCSOUTH) Operations Plan

(OPLAN) 40101, Operation Deny Flight change four, 3 May 1995 (NATO Confidential)
Information extracted is unclassified. (Operation Deliberate Force was a subset of Operation
Deny Flight. The OPLAN for the latter contained ROE that applied during Deliberate Force).
2. Correct Quotes, version 10:CD-ROM (NovatoCalif: WordStar International. Inc 1994).
3. Carl Von Clausewitz on War trans and ed Michael Howard and Peter Paret
(Princeton, NJ: Princeton University Press, 1989) 607.
4. Captain Roach uses four circles in his diagram, dividing the political and diplomatic
inf1uences into two circles Sinn these issues are usually very closely related. I have
combined them into one circle. Each of the circles comprises a multitude of individual factors
that fall under the umbrella of that particular generic factor. Capt J. Ashley Roach, "Rules of
Engagement." Naval H'ar College Review 36. no. I (January-February 1983): 46-48. The
three circle approach is consistent with current doctrine as expressed in the Joint
Commander's Handbook for Peace Operations (Fort Monroe, Va Joint War Fighting Center. 28
February 1995). 74.
5. Roach. 46.
6. Ibid. 47.
7. Ibid. 48.

2B-30
RESTRICTED
 RESTRICTED

8. D. P. O Connell, The influence of Law on Sea Power Annapolis: Naval Institute Press
1975), 180.
9. W. Hays Parks Righting the Rules of Engagement, US Naval Institute Proceedings
115. no. 5 (May 1989): 87.
10. Ibid. 88.
11. Ibid.
12. Col Philip Johnson, Chief International and Operations Law. Headquarters United
States Air Force (USAF)/JAI interviewed by author Pentagon. Washington. D.C 20
December 1995.
13. Roach. 50.
14. Johnson interview.
15. Joint Publication 1-02 Department of Defense Dictionary of Military and Associated
Terms. 23 March 1994.329.
16. Scott R. Morris ed Operational Law Handbook (Charlottesville. Va US Army Judge
Advocate General's School, 1998) 8-1.
17. Chairman of the Joint Chiefs of Staff instruction (CJCSI) 3121.01. Standing Rules of
Engagement for us Forces. 1 October 1994 par 6.
18. Ibid. par. 6a.
19. Mark Clodfelter. The Limits of Airpower: The American Bombing of North Vietnam(New
York: Free Press. 1989). 119.
20. Col Roberto Corsini the Balkan War: What Role for Airpower Airpower Joumal no. 4
(Winter 1995): 55.
21. These include the 20 km exclusion zone for heavy weapons around each UN-
designated safe area, the Croatian restricted-operating zont's and the wnes of action related
to trigger events for Deliberate Force discussed later in this chapter.
22. Adam Roberts, From San Francisco to Sarajevo: The UN and the Use of Force,
Survival: The IISS Quarterly 37 no. 4 (Winter 1995-1996): 14.
23. CJCSI 3121.01, par. la.
24. Ibid enclosure A, par. 2a. To emphasize this passage the winters placed It in all-
capital letters and boldface.
25. Ibld enclosure A. par. 5d.
26. Ibid enclosure A. par. 5b.
27. Ibid enclosure A. par. 5c.
28. Ibid enclosure A. par. 8.
29. Ibid.
30. Ibid enclosure A. par. 8b.
31. Parks. 84.
32. Maj Dan Bush. USAF interviewed by author Department of State. Washington. D.C 18
December 1995. Major Bush was the legal advisor to the CAOC. Vicenza. Italy. from June
1995 until October 1995.
33. Parks. 93.
34. Bradd C. Hayes. Naval Rules of Engagement: Management Tools for Crisis. RAND
Note N-2963-CC (Santa Monica, Calif.: RAJND. July 1989), 21.
35. Both the Stark and Beirut Marine barracks incidents as well as the Vincennes incident
receive more extensive analysis later in this chapter.
36. Hayes. 21.
37. Ibid., 24.
38. Causewitz.87.
39. Martha Lynn Craver and Rick Maze, "Reports Criticize Military for Beirut Security: Air
Force Times, 2 January 1984. 4.
40. Report of the DOD Commission on Beirut Inremarional.4.irport Terrorist Act, October 23.
1983 (Washington. D.C.: Commission on BeirutInternationalAirport Terrorist Act. 20
December 1983).4.
41. Ibid., 89. The commander explained that he decided not to permit insertion of
magazines in weapons on interior posts to preclude accidental discharge and possible injury
to innocent civilians.
42. Ibid 94.
43. Richard C. Gross. Lessons Learned the Hard Way: The USS Srark. Defense Science
&Electronics 6. no. 12 [December 1987): 9.
44. Ibid 10.

2B-31
RESTRICTED
 RESTRICTED

45. Ibid.
46. Rear Adm William M Fogarty Investigation Report: Formal Investigation into the
Circumstances Surrounding the Downing of Iran Air Flight 655 on 3 July 1985 (U)
(Washington. D.C. Department of Defense 28 July 1988).
47. John Barry and Richard Charles, Sea of Lies. Newsweek. 13 July 1°92. '29-39.
48. Richard J. Grunawalt director Oceans Law and Policy Nava1 War College
memorandum to dean, Center for Naval Warfare Studies subject: USS Vincennes (CG 49)
and the Shoot down of Iranian Airbus Flight 655. 18September 1995.
49. Fogarty. 7.
50. Hayes 54. The RAND study noted that the definitions of hostile act and hostile intent
were revised and that one should now consider all aircraft and ships potentially hostile. All
nature and friendly shipping became eligible for US assistance.
51. Ibid 5.
52. Fogarty.13.
53. CINCSOUTH OPLAN 40101 change four. (NATO Confidential) Information extracted is
unclassified.
54. COM5ATAF OPORD 45101.5. Deny Flight 30 September 1993 NATO Confidential
Information extracted is unclassified.
55. COMSATAF SPINS 02S. 241530Z August 1995. (NATO Confidential Information
extracted is unclassified.
56. UNSCR 781 9 October 1992. This resolution established the NFZ in the airspace of
Bosnia-Herzegovina in order to ensure the safety of the delivery of humanitarian assistance
and as a decisive step for the cessation of hostilities in Bosnia-Herzegovina. The ban of
military flights did not apply 10 UNPROFOR flights or others in support of UN operations
including humanitarian assistance. The UNSCR did not authorize enforcement of the ban but
requested that UNPROFOR and member states monitor compliance with the NFZ.
57. Ibid.
58. Operation Deny Flight fact sheet; on-line. Internet, 25 February 1999 available from
http/ chinfo.navy.mil/navpalib/intl/bosnia/denyflt,txt.
59. SACEUR Support Plan 10001D. -NATO Europe Integrated Air Defense. (NATO
Confidential) Information extracted is unclassified.
60. Ibid. (NATO Confidential) Information extracted is unclassified. COM5ATAF OPORD
45101.5. par. 4, incorporates the restrictions of the NATO Europe Integrated Air Defense ROE.
61. SACEUR Support Plan 10001D. (NATO Confidential) Information extracted is
unclassified.
62. CJCSI 3121.01, A-4. These provisions discuss the inherent right to use all necessary
means available and to take all appropriate action to defend one's self, one's unit, and other
friendly forces, including personnel and their property.
63. Deny Flight fact sheet.
64. UNSCR 816. 31 March 1993. This resolution which provides the iegal basis to engage
violators of the NFZ specifically indicates that it acts under chapter seven of the United Nations
Charter, which provides for peace-enforcement powers through a general system of collective
security.
65. Deny Flight fact sheet.
66. Lt Col Lowell R. Boyd Jr., tape-recorded interview by Lt Col Robert Owen.
Headquarters AIRSOUTH, Naples, Italy. 6 December 1995. United States Air force Historical
Research Agency (AFHRA], Maxwell Air Force Base (AFB). Ala.
67. Ibid.
68. Ibid. At that time, General Ashy was COMAJRSOUTH and Lieutenant Colonel Corzme
worked on the planning staff. The term Black Hole carne from the nick name of the location
where a small cadre of personnel planned the air operation again Iraq Personnel in the Black
Hole produced all of the daily tasking for air strikes.
69. Lt Gen Michael Ryan presentation at the School of Advanced air power Studies
Maxwell AFB. Ala8 February 1996.
70. Boyd interview.
71. COM5ATAF SPINS 028. par. 3D(2)(B) 1. (NATO Confidential) Information extracted is
unclassified.

72. UNSCR 836. 4 June 1993. This resolution provided the basis for NATO CAS support
of UN ground forces as well as the authority to respond to attacks incursi ons or obstruction of

2B-32
RESTRICTED
 RESTRICTED

freedom of movement inor around the safe areas. In essence the seeds for Operation
Deliberate Force were planted with the enactment of this UNSCR.
73. UNSCR 819. 16 April 1993; and UNSCR 824. 6 May 1993. These resolutions created
the listed safe areas within Bosnia-Herzegovina.
74. Deny Flight fact sheet.
75. S/25939. Report of the Secretary-General Pursuant w Security Council Resolution 836
(1993) (New York: United Nations. 14 June 1993), par. 4.
76. North Atlantic Military Committee memorandum to UN secretary general subject:
MCM-KAD-084-93. Operational Options for Air Strikes in Bosnia-Herzegovina. 8 August 1993.
(NATO Secret) information extracted is unclassified. This memorandum led to the deployment
of CAS aircraft to the Southern Region and served as the basis for providing CAS support.
77. Ibid. (NATO Secret) Information extracted is unclassified.
78. CINCSOUTH OPLAN 40101 change two. 13 August 1993. (NATO Confidential)
Information e extracted is unclassified.
79. MCM-K\D-084-93 enclosure par. 2. (NATO Secret) Information extracted is
unclassified.
80. Ibid par. 5. (NATO Secret) Information extracted is unclassified.
81. Ibid par. 9b. (NATO Secret) Information extracted is unclassified.
82. Ibid. (NATO Secret) Information extracted is unclassified.
83. Ibid par. 9b(l). (NATO Secret) Information extracted is unclassified.
84. Ibid par. 9b(2). (NATO Secret) Information extracted is unclassified.
85. Lt Col Chris Red Campbell interviewed by author Maxwell AFB. Ala22 March 1996.
Stationed in Zagreb during this time. Campbell indicated that a lack of understanding existed
between NATO and the UN as to who was really in charge with respect to the enforcement of
the UNSCRs specifically concerning the use of airpower in the tactical area of operations UN
personnel generally took the position that they controlled the trigger mechanism for use of
force and that every action had to be coordinated with them before extrication. The issue of
whether ROE allowed air-to-air engagement without the coordination and/or approval of the
UN was hotly debated.
86. CINSOUTH OPLAN 40101 change two E-1 NATO Confidential Information extracted
is unclassified.
87. Boyd interview .
88. CINCSOUTH OPLAN 40101 change two. phase three step four subpar (6). (NATO
Confidential) Information extracted is unclassified.
89. Boyd interview;
90. Deny Flight fact sheet.
91. Boyd interview.
92. Joris Janssen Lok 'The Netherlands: Learning the lessons of Srebrenica Jane's
Defence Weekly 25 no. 5 (3 I January 19961: 35.
93. Ibid.
94. Adm Leighton Smith CINCSOUTH. Operation Deliberate Force videotaped lecture
presented at the AirWarCollege. Maxwell AFB. A1a 9 November 1995. AFHRA.
95. Ibid.
96. Deny Flight fact sheet.
97. Ibid.
98. UNSCR 958. 19 November 1994. This resolution extended the provisions of UNSCR 836
to Croatia and was the basis for establishing restricted operating zones there.
99. CINCSOUTH OPLAN 40101 change four. (NATO Confidential) Information extracted is
unclassified.
100. Smith lecture.
101. Ibid.
102. Deny Flight fact sheet.
103. Operation Deliberate Force fact sheet. AFSOUTH/PI0 Naples Italy6 November 1995: on-
line. Internet, 4 March 1996 available from gopher /marvin.ste.nato.int:70 /00 /yugo/ df0611.
104. NATO NAC Notice C-N(95)65, 26 July 1995, AFHRA. CAOC-06 (NATO Secret)
Information extracted is unclassified; and MCM-KAD-057-95 NATO Air Operations to Stabilize
Bosnia-Herzegovina beyond Gorazde. 31 July 1995 AFHRA, CAOC-06-03 (K.J\TO Secret)
Information extracted is unclassified.
105. Lt Gen Michael Ryan lecture, School of Advanced Airpower Studies Maxwell AFB. Ala7
February 1996.

2B-33
RESTRICTED
 RESTRICTED

106. MCM-KAD-057-95, annex A-I. (NATO Secret) Information extracted is unclassified.

107. Smith lecture.
108. MCM-KAD-057-95, annex A-2. (NATO Secret) Information extracted IS unclassified.
109. Ibid annex A-3 (NATO Secret) Information extracted is unclassified.
110. Ibid annex A·4. (NATO Secret) Information extracted is unclassified.
111. Ibid. (NATO Secret) Information extracted is unclassified.
112. Ibid annex A-6. (NATO Secret) Information extracted is classified.
113. CINCSOUTH memorandum of understanding with FC UNPF subject UN Interpretations
of NAC Decisions, 10 August 1995. (NATO Secret) Information extracted is unclassified.
114. Smith lecture.
115. From Corona briefing slides Lt Gen Michael Ryan Headquarters AIRSOUTH Naples Italy
December 1995, AFHRA, H-3.
116. Ibid.
117. CINCSOUTH memorandum of understand par 14 NATO Secret Information extracted
is unclassified.
118. Smith lecture.
119. Ibid.
120. Ibid.
121. Target description Pale army supply depot target no COO06 Aviano AB Italy AFHRA
AVI-43. (NATO Secret) Information extracted is unclassified.
122. Target description Pale ammo depot south target no. COO1 Aviano AB Italy (NATO
Secret) Information extracted is unclassified; and target description Jahorina RADCOM station
target no D087 Aviano AB Italy (NATO Secret) Information extracted is unclassified.
123. Operation Deliberate Force fact sheet Of the 1,026 munitions dropped. 708 were
precision (including laser-guided bomb/guided bomb unit [GBD) -10. -12. -16 and -24; AS30L;
standoff land-attack missile: GBU-15; Maverick; and Tomahawk land-attack missile) and 318
were non precision (MK-82 MK-83 MK-84 and cluster-bomb unit [CBU]-87).
124. Ryan interview.
125. COM5ATAF OPORD 45101.5 change one toannex E par 6a(3) 6 July 1994. (NATO
Confidential) Information extracted is unclassified.
126. Maj Gen Hal M Homburg videotaped interview by Lt Col Robert Owen and author. 12
March 1996 Maxwell AFB Ala AFHRA.
127. CAOC air tasking message 11 September 1995 MSGID/AIRTASK/ HQ 5ATAF
/11/SEP/ /PERID/110300Z/TO:1202592// (NATO Secret) Information extracted is unclassified.
128. Lt Col Gary West commander 510th Fighter Squadron recorded interview With Lt Col
Bradley Davis and Maj John Orndorff Aviano AB Italy. 20 February 1996.
129. CAOC air tasking message 12 September 1995 MSGID/AIRTASK/ HQ 5ATAF/12/SEP/
/PERI0/120300Z/TO:1302592// (NATO Secret) Information extracted is unclassil1ed.
130. CINCSOUTH OPLAN 40101 change four annex E par 5a(6) (NATO Confidential
Information extracted is unclassified.
131. Ibid annex E par 5d(4) (NATO Confidential Information extracted is unclassified.
132. Ibid annex E par 6a(1)(d) NATO Confidential Information extracted is unclassified.
133. Homburg interview.
134. Smith lecture.
135. CM5ATAF/CAOC Deny Flight SPINS 02S 2430Z August 1995 par. 3D(2)(D) 1 NATO
Confidential Information extracted is unclassified and Capt Ken Calise US Navy C-5 and
battle-staff director CAOC Vicenza Italy interview by Maj Mark Conversino and author 16
January 1996.
136. COM5ATAF/CAOC Deny Flight SPINS 02S par 3D(3) (NATO Confidentia1) information
extracted unclassified.
137. Ibid par 3D(2)(D)4 (NATO Confidential) information extracted is unclassified.
138. Cahse interview.
139. Ibid.
140. Bush interview and Maj Keith Kiger Sixteenth Air Force Weapons and Tactics
interviewed by Maj Mark Conversino and author Aviano AB Italy 19 January 1996.
141. CINCSOUTH OPLAN 40101 change four annex E par 5a(8) (NATO Confidential)
Information extracted is unclassified.

142. Maj Steve Lrwin CAOC legal advisor tape-recorded interview with author AvianoAB,
Italy18 January 1996.

2B-34
RESTRICTED
 RESTRICTED

143. Ibid.
144. Bush interview.
145. Ibid.
146. Hornburg interview.
147. Kiger interview.
148. Calise interview.
149. Smith lecture.
150. COM5ATAF/CAOC SPINS 028 (NATO Confidential) Information extracted is
unclassified.
151. Ibid (NATO Confidential) Information extracted is unclassified.
152. CINCSOUTH OPLAN 40101 change four appendix one to annex E par 11 (NATO
Confidential) Information extracted is unclassified. Note the changes in the definition of
friendly forces that occurred from change two to change four of the OPLAN.
153. Ryan interview.
154. COM5ATAF/CAOC SPINS 028 par 3D(2)(C)2 (NATO Confidential) Information
extracted is unclassified.
155. CAOC air tasking message 8 September 1995 MSGID/AIRTASK/ HQ
5ATAF/8/SEP//PERID/80300Z/TO: 90259//. (NATO Secret) Information extracted is
unclassified.
156. Congress Senate Senator Thurmond of South Carolina Speaking in Reference to the
Rules of Engagement for the Vietnam War, 94th Cong 1st sess Congressional Record 6 June
1975, vol. 121, pt. 14. 17558. Senator Thurmond states that "the public would not approve of
it. They would not approve of sending their men into battle with their hands behind their
backs to fight that war to make gun fodder, so to speak out of them. The parents would not
approve of it the county did not approve of it.
157. Dr. Wayne Thompson and Maj Tim Reagan Deliberate Force: A Distinctive
Campaign. Washington. D.C., December 1995. (Secret) Information extracted is unclassified
158. Lt Col John G. Humphries. Operations Law and the Rules of Engagement in
OperationsDesert Shield and Desert Storm: Airpower Journal 6. no. 3 (Fall 1992): 38
Humphries discusses the importance of the trust between the NCA and the· mi litary that
developed partly due to the military‟s intensive efforts after Vietnam to train and inculcate the
ideas associated with the law of armed conflict.
159. US Ambassador Christopher Hill, interviewed by Lt Col Robert Owen, 27 February
1996.
160. Air Force Pamphlet (AFP) 110-31, International Law: The Conduct of Armed Conflict
and Air Operations, 1976, pars. 1-5. 1-6, 5-3. 6-2.
161. Ibid.
162. Ryan interview.
163. William J. Perry, secretary of defense, DOD news briefing, speech to the Adjutants
General Association of the United States, 7 February 1996.
164. Hornburg interview.

2B-35
RESTRICTED