A US perspective on special operations and the law of armed conflict

Abstract

This article outlines and evaluates several unique challenges associated with legal support to special operations. It describes the essential differences between special tactics, techniques and procedures and conventional military operations. It offers a US perspective on how treaty and customary international laws apply to various special operations, including precision attacks against “high-value targets”, hostage rescue missions, and support to partner and proxy forces. It examines the benefits and drawbacks of heavy reliance on special operations during armed conflict, including an assessment of the legal challenges that arise when special operations forces become a primary feature of a military campaign. Finally, it highlights how the assumption of substantial military and political risk impacts the application of international law to special operations at the tactical level.

Introduction

Special operations forces (SOF) perform a variety of missions across the spectrum of operating environments. For instance, amid the armed conflict between Ukraine and Russia, SOF on both sides are collecting intelligence, engaging in sabotage and attacking critical targets.¹ In the Gaza Strip, the West Bank and Lebanon, Israeli special operations units are attacking Hamas and Hezbollah leaders and attempting to rescue hostages, some of whom are held underground.² And in the 2023 conflict over the Nagorno-Karabakh region, Azerbaijani forces deployed four ground-based SOF units and a special naval warfare unit “deep into the enemy's rear areas to target and destroy [Armenian] command and control networks, critical infrastructure, and air defense systems”.³

Other militaries, particularly those from Western countries, rely heavily on SOF to achieve tactical, operational and strategic effects during armed conflict.⁴ Inter alia, SOF conduct intelligence, reconnaissance and surveillance (ISR) activities,⁵ direct precision air strikes from both manned and unmanned aerial platforms,⁶ enforce United Nations (UN) Security Council resolutions,⁷ coordinate internet-based influence campaigns,⁸ and enable partners and proxies.⁹

SOF occupy a particular place of prominence in the US military. Today, US SOF are reportedly active in more than 140 countries and spend nearly $13 billion each fiscal year.¹⁰ They are the United States’ “primary tool” for combating transregional terror groups like al-Qaeda and the Islamic State of Iraq and Syria (ISIS) in the Middle East, Africa and Central Asia.¹¹ Consequently, the number of personnel assigned to US SOF units increased from 38,000 in 2001 to 73,000 in 2020,¹² a force larger than the German Army.¹³ Special operators endured over half of American military casualties between 2001 and 2022 despite comprising just 3% of US armed forces, a testament to the United States’ heavy reliance on SOF and the intensity of their combat operations.¹⁴

Many scholars and practitioners have assessed SOF-peculiar operational capabilities. Some have analyzed and criticized certain SOF drone strikes, raids and proxy warfare campaigns,¹⁵ while others have written about the ethics of special tactics and programmes.¹⁶ However, in open-source material, only a handful of scholars have examined how the law of armed conflict (LOAC) governs special operations.¹⁷ This article seeks to fill that void by exposing readers to various practical considerations and challenges associated with legal support to SOF.

The first part of this article defines special operations and describes why they are, in fact, special. The second part analyzes key legal issues associated with select special operations. First, it discusses how the law of armed conflict regulates high-value targeting (HVT). Second, it describes how the law applies to certain hostage rescue tactics, techniques and procedures. Third, it explains various legal challenges associated with SOF support to partner and proxy forces. Finally, in the third part of the article, we offer our conclusion, including a reminder that applying international law to military operations requires an appreciation of context and a balance between humanity and military necessity.

What are special operations and what makes them “special”?

American SOF trace their roots to the Office of Strategic Services (OSS). During World War II, the OSS's mission was to “‘effect physical subversion of the enemy,’ in three distinct phases: infiltration and preparation, sabotage and subversion, and direct support to guerrilla, resistance, or commando units”.¹⁸ Modern US domestic law reflects these activities and more. In accordance with the US Code, special operations tasks include “direct action, strategic reconnaissance, unconventional warfare, foreign internal defense, civil affairs, military information support operations, counterterrorism, humanitarian assistance, theater search and rescue, and such other activities as may be specified by the President or the Secretary of Defense”.¹⁹ The “other activities” phrase leverages the flexibility and adaptability of SOF; for example, the secretary of defence designated USSOCOM as the coordinating authority for countering the proliferation of weapons of mass destruction in 2016.²⁰

The North Atlantic Treaty Organization (NATO) and its member States also consider tasks like direct action, strategic reconnaissance, unconventional warfare and counterterrorism to fall within the ambit of special operations.²¹ So do the United States’ strategic competitors, Russia²² and China.²³ Additionally, some States employ specialized military units to conduct covert actions (e.g., illicit financing and political subversion) that are more regularly associated with civilian intelligence services.²⁴ However, it is important to note that euphemistic descriptions of conventional military operations as “special”, such as Russia's so-called “special military operation” in Ukraine, do not place such operations within the commonly accepted definition of special operations.²⁵

Special operations differ from conventional military operations in three significant ways. First, they are conducted by carefully chosen, highly trained and well-equipped military personnel.²⁶ In the US military, special operators are specially selected from their services, are older on average than their conventional counterparts, possess far greater operational experience, and complete advanced training in languages, cultures and special operations techniques.²⁷ American SOF also benefit from specific congressional funding for brain health care, cognitive performance assessments and family support programmes.²⁸ In other words, SOF are not “general purpose” (i.e., conventional) forces and, in our view, should not be systematically employed to conduct conventional military operations.²⁹

Second, unlike general purpose forces, “special operations require unique modes of employment, tactics, techniques, procedures, and equipment”.³⁰ These tactics and techniques may include high-altitude parachuting, special boat insertions, the wear of non-standard uniforms or indigenous clothing to assure secrecy or concealment, the use of non-standard weapons and ammunition, and other tactics that emphasize speed, shock, surprise and overwhelming violence.³¹ Unlike conventional warfare, which involves “large-scale, sustained combat operations”,³² special operations typically involve lightly armed, agile teams that “strike enemies where and when they are most vulnerable” and withdraw rapidly.³³ As one author has explained, SOF can “carry out tactical actions producing strategic effect out of proportion to the amounts of personnel and equipment committed”.³⁴ For example, only twenty-three US Navy SEALs participated in the 2011 raid against Osama bin Laden, which lasted “a few short minutes” but marked a significant milestone in the Global War on Terror.³⁵

Third, special operations “are characterized by … a high degree of risk”.³⁶ The employment of SOF, particularly in “hostile, denied, or politically and diplomatically sensitive areas”,³⁷ reflects an acceptance of “political or military risk not associated with operations by conventional forces”.³⁸ In the British Army, for example, SOF “can be authorised to operate at higher risk beyond the remit of conventional forces”.³⁹ As a result, national command authorities often select special operators to conduct the most high-risk or strategically important missions because of their talent, resources and experience.

Applying the law of armed conflict to certain categories of special operations

The United States often uses SOF for various types of operational missions. We have selected three categories of operational activity for close legal examination: HVT, hostage rescue operations and proxy force operations.

High-value targeting

SOF conduct “surgical strikes against sensitive targets” and the “capture of enemy leaders”.⁴⁰ These operations are commonly referred to as HVT operations and occur in both international and non-international armed conflicts. HVT is not a legal term of art; instead, it describes “lethal strikes, raids, and/or offensive cyber operations to capture or kill key enemy combatants and materiel”.⁴¹ During the conflict in Vietnam, for example, a special operations unit called OPS-35 identified, raided and attacked critical targets along the Ho Chi Minh Trail, which “compelled Hanoi to divert resources that would otherwise have been directed to military operations against South Vietnam”.⁴² In the decades that followed, SOF from numerous States sought to “decapitate” enemy forces by capturing or killing high-value individuals during armed conflicts in the Balkans, Libya, Panama, Sierra Leone, Somalia and Iraq.⁴³ These missions were not always successful, partly because of the high risks and difficulties of targeting senior leaders, and the utility of decapitation campaigns remains questionable.⁴⁴ Nevertheless, the prospect of degrading an enemy force's command structure remains an attractive military strategy that warrants high-risk efforts to remove senior leaders from the battlefield.⁴⁵

Following 11 September 2001, US SOF assumed primary responsibility for identifying and attacking senior terror group leaders like Al-Qaeda's Osama bin Laden and Abu Musab al-Zarqawi and the Taliban's Mullah Akhtar Mohammed Mansour.⁴⁶ Likewise, members of the Canadian Special Operations Forces Command, British 22 Special Air Service (SAS) and Australian Special Air Service Regiment targeted and interdicted key enemy leaders in Iraq, Afghanistan and Syria.⁴⁷ French SOF assumed responsibility for targeting the most senior members of Al-Qaeda in the Islamic Maghreb.⁴⁸ More recently, US SOF led coalition efforts to identify and kill “particularly charismatic or effective members of [the so-called Islamic State group]”, including its most senior leaders, Abu Bakr al-Baghdadi and his successor Abu Ibrahim al-Hashimi al-Qurayshi.⁴⁹

Targeting senior leaders like these requires talent, operational experience, integration with other government agencies, moral intuition,⁵⁰ unique assets and authorities,⁵¹ decentralized decision-making, and speed. Special operations personnel are selected, trained and equipped to identify and either capture or attack high-value enemy targets. In addition, they often enjoy relatively permissive rules of engagement (ROE) and regular access to embedded legal advisers who specialize in the LOAC.

Proportionality

The legal rules governing the conduct of hostilities in armed conflict always apply to special operations, but HVT presents unique compliance challenges, especially in the realm of proportionality. The proportionality rule, which is customary international law in both international and non-international armed conflicts,⁵² requires combatants to “refrain from … launch[ing] any attack which may be expected to cause incidental loss of civilian life, injury to civilians, [or] damage to civilian objects … which would be excessive in relation to the concrete and direct military advantage anticipated”.⁵³ A commander's assessment of “military advantage” is not limited to the “immediate tactical gains but may be assessed in the full context of the war strategy”.⁵⁴ Still, anticipated military advantages must always be “concrete and perceptible … rather than … hypothetical and speculative”.⁵⁵ In theory, and regardless of a conflict's classification, effective HVT operations deliver tangible effects.

Because the LOAC permits attacks against military objectives “when it is known that some civilian deaths or injuries will occur”,⁵⁶ attacks against genuinely “high-value” targets, which might include people or objects, might lawfully cause extensive collateral damage, so long as it is not excessive in relation to the military advantages gained by the attack.⁵⁷ In other words, because an attack against a high-value target can produce particularly significant concrete and direct military advantages, extensive collateral damage may not be unlawfully excessive.⁵⁸ This is a blunt reality of war and one of the most daunting challenges associated with legal support to lethal special operations.

Take, for instance, a situation in which SOF have positively identified an elusive and particularly dangerous chemical weapons developer who is located in an urban centre inside a conflict zone. Suppose further that intelligence indicates that the chemical weapons developer is building a mustard gas weapon and planning to use it to attack military forces in the region. In that case, the law allows the SOF commander to attack the chemical weapons developer in the urban centre even though some civilian harm may occur as a direct result. However, if credible intelligence indicates that the chemical weapons developer is planning a large-scale chemical attack against civilians in a densely populated city outside of the conflict zone (i.e., a terrorist attack), the tolerance for collateral damage may be much higher.

The challenge lies in determining how much harm the LOAC and domestic policy tolerate.⁵⁹ To answer this question, a commander and their legal adviser must understand and consider the law, contextual factors, the applicable ROE, and the staff's recommendations. Additionally, because “there is no agreed-upon metric for determining how much civilian harm would be excessive in light of the expected military advantage”, the commander must rely on his or her experience and intuition.⁶⁰ In a potential future large-scale armed conflict, highly proscriptive ROE and policy-driven targeting restrictions commonly used during counterterrorism operations may be replaced by broader, intent-driven ROE. This will be particularly true for SOF, which generally operate with relative autonomy and independent initiative as opportunities arise.⁶¹ As a result, these challenges will only increase in future conflicts.

Extreme targeting dilemmas are not hypothetical for commanders within SOF or their supporting legal advisers.⁶² In Iraq and Syria, US SOF identified and targeted ISIS chemical weapons developers like Abu Malik, a chemical engineer who worked in Saddam Hussein's “Muthana”, where he developed sarin gas weapons before joining Al-Qaeda in 2005.⁶³ US SOF killed Abu Malik in 2015 near Mosul after discovering intelligence indicating that he was coordinating mustard gas attacks against civilians in Europe. According to Gregory Koblentz, an expert on chemical and biological weapons, “[i]f Abu Malik had survived, his experience working for Saddam's program would have made the threat of the Islamic State's chemical weapons much higher”.⁶⁴ In our view, the removal of Abu Malik from the battlefield in Iraq delivered a direct and concrete blow to ISIS and spared many innocent civilians from the horrors of a chemical weapons attack.⁶⁵ Consequently, the operation to kill him may have also justified substantial harm to the civilian population in Mosul.

Throughout Operation Inherent Resolve in Iraq and Syria, US SOF attacked other high-value targets, including senior leaders and members with unique technical, scientific and financial skills. They did so under ROE that sometimes explicitly permitted civilian casualties and damage to civilian objects.⁶⁶ In particular, US SOF targeted ISIS's external operators (i.e., those seeking to conduct mass casualty attacks in cities outside of the conflict zone) before they could coordinate large-scale suicide attacks against civilians.⁶⁷ As recently as April 2023, US SOF conducted an air strike in Syria, killing Khalid ‘Aydd Ahmad al-Jabouri, a senior ISIS leader responsible for planning terror attacks in Europe.⁶⁸ While US Central Command quickly pointed out that the strike caused no collateral damage, the LOAC would have tolerated collateral damage provided that the responsible commander determined that it was not excessive in relation to the concrete and direct military advantages anticipated from the attack.

Nevertheless, “society is intolerant of civilian casualties”.⁶⁹ Retrospective proportionality assessments conducted by external organizations, media outlets and governments are often unforgiving because they are based on the actual effects of an attack rather than the harm anticipated by the commander at the time of the attack.⁷⁰ Consequently, many individuals have criticized US SOF for harming civilians, labelling some military operations as “unnecessary”,⁷¹ “shameful”⁷² and “reckless”.⁷³ Even after US SOF killed ISIS leader Abu Bakr al-Baghdadi in Syria in 2019, mainstream media reports focused on allegations that US forces had killed two civilians near the targeted compound but not on the necessity of, or the military advantages gained by, killing ISIS's most senior leader.

Under the LOAC, one cannot judge whether collateral damage is excessive without grasping the totality of the circumstances and a military objective's value.⁷⁴ Therefore, critics should never disregard the apparent military benefits of an attack just because it caused civilian harm. Admittedly, most outside observers will not fully understand the military advantages gained by attacking high-value targets because the intelligence used to characterize targets as “high-value” is typically derived from protected human sources or classified electronic surveillance programmes. Thus, armed forces cannot adequately explain the lawfulness of some attacks that cause collateral damage. Still, “[c]ombat … is an inherently messy, imprecise, bloody business”, and as Professor Michael Schmitt has argued, critics must “be cautious about wrapping lex ferenda in the cloak of lex lata”.⁷⁵

On the other hand, it is crucial to recognize how counterproductive civilian harm can be. Even legally justified collateral damage can “erode international support and affect foreign policy”.⁷⁶ As retired general Stanley McChrystal observed in Afghanistan, “[w]e're going to lose this … war if we don't stop killing civilians”.⁷⁷ The pursuit of high-value targets can impede an accurate view of the strategic landscape. In addition, some special operations personnel may view every potential target as “high-value”, which is a subjective and imprecise term. Theoretically, SOF should not bear primary responsibility for targets of “low” or “moderate” value. However, when SOF assume primary responsibility for an entire military campaign, some targets may be described as “high-value” even if they are not.

The relatively low-intensity campaigns of Iraq and Afghanistan revealed this operational design flaw. For example, as the inspector-general of the Australian Defence Force noted in the Afghanistan Inquiry Report,

[w]hile … Special Forces provide an attractive option for an initial deployment, it is a misuse of their capability to employ them on a long-term basis to conduct what are essentially conventional military operations. Doing this on a protracted basis in Afghanistan detracted from their intended role in the conduct of irregular and unconventional operations and contributed to a wavering moral compass.⁷⁸

In our view, commanders and staff officers within SOF must refrain from labelling routine targets (e.g., “rank and file” members of a hostile force or basic enemy fighting positions) as “high-value” because doing so may unintentionally result in causing unnecessary collateral damage. Effective SOF legal advisers must understand this dynamic and help commanders assign appropriate values to targets when conducting proportionality assessments.

Feasible precautions

Like all armed forces engaged in combat, SOF must also take feasible precautions when planning and conducting attacks. Article 57 of Additional Protocol I (AP I) requires combatants to “take all feasible precautions” to minimize the “incidental loss of civilian life, injury to civilians and damage to civilian objects”.⁷⁹ The International Committee of the Red Cross (ICRC) considers this principle to be customary international law applicable in international and non-international armed conflicts.⁸⁰ The United States has not ratified AP I and is therefore not bound by it, but US officials acknowledge that the obligation to “take feasible precautions” reflects customary international law.⁸¹ The Department of Defense (DoD) Law of War Manual recognizes the duty to “take feasible precautions to reduce the risk of harm to the civilian population” and considers the word “feasible” to mean “practicable or practically possible, taking into account all circumstances ruling at the time, including humanitarian and military considerations”.⁸²

Within SOF, feasible precautions include actions like increasing the number of airborne ISR assets monitoring a target, including commercial, off-the-shelf drones; using human intelligence sources to verify the presence of civilians within a target building; providing advanced warnings to civilians through innovative, internet-based messaging; and using specialized munitions to minimize collateral damage. For example, prior to conducting attacks against chemical weapons developers in Iraq and Syria, US SOF “selected special incendiary bombs designed to generate a small blast radius but intense heat, to vaporize weapons, supplies and any residual gases that might otherwise escape”.⁸³ Similarly, US SOF have reportedly relied on the AGM-114R9X variant of the Hellfire missile, which uses six blades and kinetic energy to destroy targets.⁸⁴ These inert missiles do not use any explosives and therefore cause “little damage beyond the target”.⁸⁵

Next, when conducting raids,⁸⁶ SOF might employ a “tactical callout” to compel enemy fighters to surrender, placing them hors de combat,⁸⁷ and allow civilians to escape before a raiding force enters the structure.⁸⁸ For example, on 3 February 2022, a small contingent of US SOF conducted a raid in Syria's Idlib province that led to the death of Islamic State leader Abu Ibrahim al-Hashimi al-Qurayshi. While the national command authority considered an air strike to kill al-Qurayshi, the president ultimately chose to conduct a raid in order to mitigate the potential for harm to civilians in the compound.⁸⁹ Before dawn, special operations aviation personnel transported some two dozen assaulters to the town of Atmeh, Syria. During a “tense standoff”, the raiding force blasted warnings in Arabic over loudspeakers while one soldier conducted a “tactical callout” with a bullhorn, telling everyone in the house to surrender.⁹⁰ As a result, ten civilians were spared from harm before Qurayshi detonated a suicide vest on the third floor, killing himself, his wife and two of his children.⁹¹ These are just a few examples of feasible precautions that SOF may employ to mitigate civilian harm during high-risk special operations.

In addition, SOF can enable “feasible precautions” for general purpose forces by conducting special reconnaissance⁹² behind enemy lines, and sometimes in civilian clothes, to verify targets before a larger operation. For example, during the Persian Gulf War, “SOF placed navigation beacons at key F-111 ingress points allowing for critical radar updates, thus improving bombing accuracy and reducing the possibility of collateral damage”.⁹³ In this regard, strategic reconnaissance may require a trade-off between legal principles (i.e., failing to distinguish oneself but for the purpose of taking feasible precautions), thus increasing the legal risk to the special operators in order to decrease the risk of physical harm to civilians and civilian objects.

The tactics of hostage rescue operations⁹⁴

The non-international armed conflict between the US-led coalition and ISIS in Syria, Iraq and elsewhere was marked by a series of hostage situations. As Human Rights Watch noted in 2020, ISIS's expansion in Iraq and Syria “featured horrendous public abuses”, including detentions and kidnappings.⁹⁵ Hostage rescue operations are a core special operations activity.⁹⁶ The following section analyzes specific tactics related to hostage rescue operations. While it includes examples from domestic hostage rescue operations to illustrate the use of specific tactics, the analysis focuses on the legal aspects of hostage rescue during armed conflict.

The specialized SOF units that train for and execute hostage rescue missions must grapple with two fundamentally competing interests. On the one hand, the primary purpose of any hostage rescue mission is to save lives. However, to succeed in achieving the principal goal, special operators may need to employ lethal force against the hostage-takers or their sentinels. When hostage-taking occurs during armed conflict, the hostage-takers may be subject to attack under the LOAC's rules regarding combatants,⁹⁷ organized armed groups,⁹⁸ or civilians directly participating in hostilities.⁹⁹ However, hostage-takers are often co-located with their hostages, which significantly complicates the tactical operation and demands unique tactics, techniques and procedures. As a result, rescuers seek to separate hostage-takers from their hostages and protect the hostages from the effects of violence.

Riot control agents

One technique that US SOF may employ to separate the hostage-takers from their hostages is the use of riot control agents (RCAs). Article II(7) of the Chemical Weapons Convention (CWC) defines RCAs as “[a]ny chemical … which can produce rapidly in humans sensory irritation or disabling physical effects which disappear within a short time”.¹⁰⁰ Classic examples of RCAs include tear gas and pepper spray, which cause “irritation to the eyes, mouth, throat, lungs, and skin”.¹⁰¹ Other RCAs include the chemical agent chloropicrin, which induces “severe vomiting and excessive tear formation”.¹⁰² In the past, militaries and specialized police units have used RCAs to separate hostage-takers from hostages (i.e., to enable the distinction between lawful targets and civilians), to incapacitate hostage-takers temporarily, or to flush hostage-takers out from enclosed areas and tunnel systems. While many of these operations have occurred during peacetime (e.g., a law enforcement hostage rescue operation occurring outside of an armed conflict), the United States’ unique position on the CWC allows US SOF to use RCAs in an armed conflict in certain, limited circumstances.

Due to their temporary effects, the United States does not consider RCAs to be “chemical weapons” banned by the Protocol for the Prohibition of the Use in War of Asphyxiating, Poisonous or Other Gases, and of Bacteriological Methods of Warfare.¹⁰³ However, Article I(5) of the CWC, which applies in both international and non-international armed conflicts, prohibits the “use [of] riot control agents as a method of warfare”.¹⁰⁴ It is not entirely clear how the CWC's phrase “as a method of warfare” operates regarding the use of RCAs in armed conflict because the treaty does not define the term.¹⁰⁵

In the absence of a treaty definition and as a condition for ratifying the CWC, the United States interprets “method of warfare” consistent with Executive Order 11850, which predates the CWC.¹⁰⁶ This domestic presidential policy specifically authorizes the use of RCAs in war “in defensive military modes to save lives”.¹⁰⁷ As a result, when approved by a proper authority, US military personnel may use RCAs in situations where “civilians are used to mask or screen attacks and civilian casualties can be reduced or avoided” and during “rescue missions in remotely isolated areas, of downed aircrews and passengers, and escaping prisoners”.¹⁰⁸

While the United States looks to how RCAs are being employed to determine whether their use is lawful (i.e., are the RCAs being used as a method of warfare?), other States may look to the classification of the conflict between the hostage-rescuers and hostage-takers. The CWC does not prohibit the use of RCAs for “law enforcement including domestic riot control purposes”.¹⁰⁹ States, particularly those in Europe, generally rely on law enforcement organizations or paramilitary forces (e.g., Germany's GSG-9, France's GIGN, Belgium's SIE, and Ireland's Garda ERU)¹¹⁰ to conduct hostage rescue operations domestically and abroad.¹¹¹ Incidentally, some scholars have expressed a view that the use of RCAs during a hostage rescue operation conducted by specialized law enforcement units may not be prohibited by the CWC, even during armed conflict.¹¹²

Hostage rescue operations are fundamentally about saving lives. Thus, US SOF may lawfully use RCAs to separate hostages from hostage-takers within a building that has been transformed into a defensive strongpoint, to disable hostage-takers physically, or to flush hostage-takers from a building in which hostages are isolated, restrained or unable to escape. While the use of tear gas or pepper spray in an underground tunnel, like those beneath Gaza,¹¹³ might incapacitate everyone within the tunnel, including hostages, it would not kill or permanently injure them. In other words, in our view, US SOF may use RCAs during hostage rescue operations to increase the chances that hostages will survive, not as a method of warfare.

Use of expanding bullets

Another potential tactic unique to hostage rescue operations in armed conflict is the use of expanding ammunition. Hostage rescue operations typically involve close quarters and short ranges, as they often occur within a confined space. Therefore, hostage-rescuers may prefer to use expanding bullets to mitigate the risk that a fired round may strike and then pass through a hostage-taker, travel through a wall, or ricochet, ultimately hitting or killing a hostage or fellow rescuer. Unlike standard, jacketed bullets, such as the United States’ M855A1 round, which are used to penetrate hard targets, expanding bullets flatten upon impact.¹¹⁴ This flattening effect increases the bullet's stopping power by maximizing its surface area upon initial impact but decreases its ability to continue penetrating the object. As the late Professor Yoram Dinstein noted, in the context of hostage rescue operations, “[t]he use of [expanding] bullets maximizes the chances that the hostage-taker will be instantly put out of action (preventing him from injuring the hostages); and it also minimizes the risk that a bullet would pass through his body and strike a hostage”.¹¹⁵

The ICRC takes the position that customary international law prohibits the use of “expanding bullets” in both international and non-international armed conflicts.¹¹⁶ The ICRC derives its position from a variety of sources,¹¹⁷ including the 1899 Hague Declaration concerning Expanding Bullets, which states that “[t]he Contracting Parties agree to abstain from the use of bullets which expand or flatten easily in the human body, such as bullets with a hard envelope which does not entirely cover the core or is pierced with incisions”.¹¹⁸ The United States is not a party to this instrument, and also, by its own terms, the 1899 Hague Declaration only applies to “war” between the contracting parties.¹¹⁹

The United States does not view the 1899 Hague Declaration as reflective of customary international law in armed conflict.¹²⁰ Instead, since 1899, the United States has maintained the position that expanding bullets, like all other munitions, are “only prohibited if they are calculated to cause superfluous injury”.¹²¹ This position is consistent with the customary international law prohibition regarding superfluous injury and unnecessary suffering¹²² and the United States’ corresponding treaty obligations under Article 23(e) of the 1907 Hague Regulations, which prohibits parties from “employ[ing] arms, projectiles, or material calculated to cause unnecessary suffering”.¹²³

In addition, the Rome Statute of the International Criminal Court makes the use of expanding bullets a war crime in both international and non-international armed conflicts.¹²⁴ However, a required element of those offences is that “[t]he perpetrator was aware that the nature of the bullets was such that their employment would uselessly aggravate suffering or the wounding effect”.¹²⁵ Again, many States have concluded that Article 8 of the Rome Statute only criminalizes “the use of expanding bullets that are also calculated to cause superfluous injury”.¹²⁶

Importantly, as the late Hays Parks noted in a 1990 legal opinion for US Special Operations Command, “the fact that a weapon causes suffering does not lead to the conclusion that the weapon causes unnecessary suffering, or is illegal per se”.¹²⁷ Instead, the “calculated to cause” standard requires an analysis of the design intent of the munition or weapon.¹²⁸ In the case of expanding bullets, they were specifically designed to increase stopping power and decrease penetration in order to limit collateral damage. This design characteristic makes expanding bullets particularly effective during hostage rescue missions, during which forces must quickly neutralize a threat while minimizing the chances that the bullet will over-penetrate the target or structure. As a result, expanding bullets are not “calculated to cause unnecessary suffering”, and nor would their use for these purposes be unlawful.¹²⁹

In conclusion, based on the United States’ view, US SOF may lawfully use expanding ammunition during hostage rescue missions in armed conflict, so long as it is not used for the purpose of causing unnecessary suffering. As with RCAs, specialized law enforcement units that conduct hostage rescue operations may lawfully use expanding bullets for law enforcement purposes, even during armed conflict.¹³⁰ During hostage rescue operations, there are significant “humanitarian concerns for protection of the innocent civilians taken hostage or otherwise placed at risk, the members of the rescue force, or the civilian population in the surrounding area”.¹³¹ Therefore, hostage-rescuers may use expanding ammunition “in such very close life-threatening situations” for the purpose of “employ[ing] a projectile that deposits all of its energy in the target. This provides for high target selectivity by maximizing the disabling effect on the target while minimizing the aforementioned risk to hostages or dangerous material.”¹³²

The wear of non-standard uniforms

Finally, surprise is an essential element of hostage rescue. According to Tomer Israeli, who served in both Shin Bet (Israeli Secret Service) and the Israeli Army Special Forces unit Sayeret Matkal, “surprise is the key”.¹³³ Similarly, US Army doctrine recognizes surprise as a critical element of a successful raid, explaining that in such a raid, “the unit overcomes any enemy forces at or near the objective in a violently executed surprise attack using all available firepower for shock effect”.¹³⁴

History reveals that to preserve the element of surprise, hostage-rescuers have worn both non-standard uniforms and civilian clothes when conducting hostage rescue operations. For example, in 1977, German GSG-9 and British SAS soldiers wore civilian clothing during the successful hostage rescue of Lufthansa Flight 181 in Mogadishu, Somalia.¹³⁵ Members of the SAS wore non-standard uniforms again in 1980 during a hostage rescue operation in the Iranian Embassy in London.¹³⁶ The American commandos involved in Operation Eagle Claw (the failed attempt to rescue American hostages in Tehran in 1980) wore non-standard uniforms approved by the Joint Chiefs of Staff and the president.¹³⁷ While most of these examples are of domestic hostage rescue operations, during an armed conflict, military members must distinguish themselves from the civilian population if they wish to remain entitled to prisoner of war (PoW) status and retain combatant immunity.¹³⁸

As a result, the tactic of wearing non-standard uniforms or civilian attire during a hostage rescue operation in an armed conflict requires additional analysis. Article 4(A)(2) of Geneva Convention III is often misconstrued as requiring combatants to wear uniforms during armed conflict,¹³⁹ but as Hays Parks explained in his seminal article, Article 4(A)(2) contains “no language requiring military personnel to wear a uniform, nor prohibiting them from fighting in something other than a full, standard uniform. Nor does it make it a war crime not to wear a uniform.”¹⁴⁰

During armed conflict, military members may lawfully wear non-standard uniforms.¹⁴¹ This attire includes items such as camouflage-pattern garments, distinctive insignia, or military equipment (e.g., body armour) which otherwise distinguishes the wearer from the civilian population.¹⁴² A non-standard uniform usually complies with Article 4(A)(2)'s conditions, allowing the wearer to maintain their entitlement to PoW status and their combatant immunity.¹⁴³ While international law does not prohibit military members from wearing civilian attire to collect intelligence or conduct acts of sabotage,¹⁴⁴ they would not be entitled to PoW status or combatant immunity if captured during an international armed conflict.¹⁴⁵ In sum, hostage-rescuers conducting operations in an armed conflict may lawfully wear standard uniforms, non-standard uniforms, or even civilian attire in certain circumstances.

But suppose the hostage-rescuers wore civilian clothes and concealed their weapons until the last possible moment before launching a surprise attack against the hostage-takers. Article 37(1) of AP I prohibits “kill[ing], injur[ing] or captur[ing] an adversary by resort to perfidy”. AP I defines perfidy as

[a]cts inviting the confidence of an adversary to lead him to believe that he is entitled to, or is obliged to accord, protection under the rules of international law applicable in armed conflict, with intent to betray that confidence …. [This includes] the feigning of civilian, non-combatant status.¹⁴⁶

As explained by Yoram Dinstein, “[t]he act of feigning civilian status becomes unlawful perfidy – and a war crime – if (and only if) there is an intentional betrayal of confidence by inviting the enemy to believe that a person is a civilian and then, e.g., pulling a hidden weapon and opening fire”.¹⁴⁷

While the United States is not a party to AP I, it acknowledges that it is customary international law that “combatants may not kill or wound by resort to perfidy” during armed conflict.¹⁴⁸ Further, as the ICRC has noted, the “rule [prohibiting perfidy] is supported by official statements and other practice pertaining to non-international armed conflicts”.¹⁴⁹ Thus, while SOF can certainly conduct reconnaissance of an objective area in civilian clothing, they must distinguish themselves from the civilian population during the actual hostage rescue despite the vital importance of surprise. Practically, special operators may comply with this obligation during armed conflict by simply wearing a distinctive insignia, load-bearing equipment (e.g., a vest or belt used to carry ammunition, firearms or first aid) or body armour and carrying arms openly. However, it remains imperative that SOF legal advisers be cognizant of these obligations and the tactical nuances of the operation (i.e., the context) when advising commanders during hostage rescue operations.

Proxy force operations

SOF do not only attack high-value targets or respond to crises; theatre-level commanders often incorporate SOF into broader campaign plans. One way SOF can achieve military objectives abroad, particularly in hostile, denied and politically sensitive environments, is by working “by, with, and through” foreign partners or proxy forces.¹⁵⁰ The US military understands “by, with, and through” to mean an approach in which “operations are led by … partners, state or nonstate, with enabling support from the United States or U.S.-led coalitions, and through U.S. authorities and partner agreements”.¹⁵¹ Special operations can provide partners and proxies¹⁵² with training, equipment, intelligence and planning advice, and in armed conflict, they may direct or conduct combat operations alongside those forces.¹⁵³

In the US military, SOF are primarily responsible for “assess[ing], train[ing], advis[ing], and assist[ing] host-nation military and paramilitary forces” to enable “these forces to maintain the host nation's internal stability, to counter subversion and violence in their country, and to address the causes of instability”.¹⁵⁴ SOF can also enable resistance movements or insurgencies to “coerce, disrupt, or overthrow a government or occupying power by operating through or with an underground, auxiliary, and guerrilla force in a denied area”.¹⁵⁵ Within the US military, responsibility for this type of “unconventional warfare” belongs to US Army Special Forces,¹⁵⁶ also known as the Green Berets.¹⁵⁷

In 2014, US SOF began providing training, equipment and close air support to “vetted Syrians”, now known as the Syrian Democratic Forces (SDF), to destroy ISIS's so-called caliphate in Syria and prevent ISIS from conducting attacks outside of the conflict zone.¹⁵⁸ In Libya and Somalia, US SOF provided “direct, on-the-ground support” to local forces battling ISIS, Al-Shabaab and Al-Qaeda.¹⁵⁹ SOF from other nations, including the United Kingdom, France and Canada, have also reportedly supported partner operations to defeat terror networks in the Middle East and Africa.¹⁶⁰

A significant challenge arises, however, when an effective partner or proxy force engages in unlawful acts. In some circumstances, the only combat-proven partners may be local indigenous or guerrilla forces that are inexperienced or undisciplined. While “by, with, and through” operations may initially mitigate political risk, if a partner or proxy force violates the LOAC, the relationship could cause significant domestic and international repercussions for the supporting nation or jeopardize an entire operation. Indeed, the US military recently cancelled a series of military engagements in Africa, including “special operations and cyber courses”, due to human rights abuses among proposed partner forces.¹⁶¹

The potential dilemma for SOF is that building relationships with a partner force requires substantial time and effort. Special operators may spend years or decades operating in a particular region of the world so they can commit to a deep cultural education, develop language fluency and build long-term relationships with State and non-State groups within that region.¹⁶² US SOF endeavour to build a global network of “true partners who can confront their own threats” and who share intelligence about challenges to US interests.¹⁶³ When a particularly effective or habitual partner force violates the LOAC, withdrawal can be difficult. Often, the partner force may be the only source of reliable human intelligence in an area, a guarantor of safe access to an otherwise hostile region, or the only effective means for combating a threat. Embedded legal advisers can help ensure that special operators fulfil their duty to identify, document, report and assist with correcting LOAC violations, even though reported abuses may derail hard-earned relationships and strategically important programmes.

The United States’ support of El Salvador during its non-international armed conflict with the Farabundo Martí National Liberation Front (Frente Farabundo Martí para la Liberación Nacional, FMLN) highlights this challenge. In the late 1970s, El Salvador, a strategic US partner in Central America, faced collapse due to civil unrest and insurgent activity.¹⁶⁴ By 1981, the armed conflict had intensified, and as a result, the US Congress authorized the deployment of fifty-five Army Green Berets and Navy SEALs to help the Salvadoran military fight the FMLN and restore peace to the country.¹⁶⁵

Immediately, respect for human rights and the LOAC became one of the biggest challenges for SOF in El Salvador.¹⁶⁶ Special Forces personnel had to build rapport with their Salvadoran military counterparts, many of whom were “notorious human rights violators”.¹⁶⁷ In addition, many of the Salvadoran military rank and file were untrained conscripts who had been “led to believe [by their officers] that villagers who sympathized or aided and abetted the insurgents were legitimate military targets under the rules of engagement”.¹⁶⁸ Consequently, the Salvadoran military remained “in garrisons abusing civilians” rather than conducting effective operations against the FMLN.¹⁶⁹

However, US SOF could not overtly provide LOAC training because even mentioning the term “was a real turn-off” to the Salvadoran military.¹⁷⁰ Therefore, the special operators had to rely on creative means to encourage compliance. For example, US SOF “emphasized the distinction between combatants and civilians by teaching ambush techniques [that incorporated a civilian into] the scenario. If the soldiers killed the civilian … they failed the exercise.”¹⁷¹

While El Salvador's compliance with the LOAC throughout the conflict was mixed, SOF's subtle LOAC training helped professionalize the Salvadoran military, and increased US pressure caused the Salvadoran government to implement LOAC reforms.¹⁷² As a result, the Salvadorans became less abusive, which led to the FMLN's loss of a “valuable source of propaganda and recruiting justification”.¹⁷³ Ultimately, the Salvadoran military successfully repelled the FMLN's attempts to overthrow the government, and in 1992, the two sides agreed to a formal ceasefire.¹⁷⁴

International law requires States to respect the LOAC and the treaties to which they are bound.¹⁷⁵ However, whether States are obligated to ensure that others respect international law is subject to debate.¹⁷⁶ While the ICRC broadly interprets this obligation, some States, including the United States, have taken a narrower view.¹⁷⁷ Despite this view, the United States, as a matter of policy, seeks to promote adherence to the LOAC by others¹⁷⁸ and looks to its partners’ compliance with the LOAC in “assessing the lawfulness of [the United States’] assistance to, and joint operations with, those military partners”.¹⁷⁹

Specifically, US federal law prohibits the DoD from providing “assistance to units of foreign security forces that have committed a gross violation of human rights”.¹⁸⁰ While the so-called Leahy Laws do not define the term “gross violation of human rights”, the Department of State (DoS) has interpreted it to include “torture or cruel, inhuman, or degrading treatment or punishment, prolonged detention without charges, causing the disappearance of persons …, or other flagrant denial of the right to life, liberty, and the security of person”, as well as “rape under color of law”.¹⁸¹ Other acts, such as LOAC violations, can also fall within this definition.¹⁸² Thus, before US SOF provide any training, equipment or assistance to a foreign force, the DoS must make sure that the unit and certain key members in the unit have not committed any Leahy Law violations.¹⁸³ Once embedded with a foreign force, if members of that force violate the LOAC, US SOF must immediately report, and potentially investigate, the allegation even if no US personnel were involved.¹⁸⁴ If the allegation is substantiated, the US government must then help the foreign government take “all necessary corrective steps”,¹⁸⁵ to include “bringing the responsible members of the unit to justice”,¹⁸⁶ in order for all US assistance to that unit to continue. Practically, embedded US SOF typically help foreign partner forces take the necessary corrective actions because of their relationship and co-location.

Additionally, States “are obliged by international law to refrain from actively encouraging violations of either treaties to which they are Party or ‘cardinal’ rules of the law of war”.¹⁸⁷ In its 1986 Nicaragua judgment, the International Court of Justice (ICJ) found that the US Central Intelligence Agency (CIA) had provided its Nicaraguan Contra proxy force with significant training, equipment, financing and direction.¹⁸⁸ Despite this, the Court determined that the Contras were not subject to the CIA's control and did not attribute their LOAC violations to the United States.¹⁸⁹ This case illustrates the challenges involved in attributing a surrogate force's LOAC violations to the supporting State.¹⁹⁰

In the Nicaragua case, however, the ICJ did find evidence that the CIA had actively encouraged the Contras to violate the LOAC via the dissemination of a poorly worded psychological warfare manual, which was reportedly written by an unsupervised, low-level employee.¹⁹¹ Consequently, the Court held “[t]he United States is thus under an obligation not to encourage persons or groups engaged in the conflict in Nicaragua to act in violation of the provisions of Article 3 common to the four 1949 Geneva Conventions”.¹⁹² The Court derived this obligation from “the general principles of humanitarian law” and stated that common Article 3 “constitutes a minimum yardstick” applicable in both international and non-international armed conflicts.¹⁹³ As a result, special operators who are deployed to remote regions of the world, and their supporting legal advisers, must ensure that all surrogate force training, planning and advice promotes compliance with the LOAC.

Conclusion

SOF from around the world conduct high-risk and politically sensitive missions in dangerous environments. Through the employment of special tactics and unique equipment, SOF can achieve strategically important effects during armed conflict. There is, of course, no special operations exemption to international law, but legal advisers, commanders and commentators must always appreciate how contextual factors affect the application of the LOAC to special operations. The tactical nuances of an operation can drastically affect the application of a legal rule, prohibiting conduct in one setting or permitting action in another. The law represents a challenging balance of interests that special operations personnel and their legal advisers must seek to master.