Norm challenges and norm death: The inexplicable?

Case 1: nuclear non-proliferation norm

In the late 1940s, the USA vigorously promoted the idea of nuclear non-proliferation. With the development of nuclear capabilities in the former Soviet Union, France, the United Kingdom and China, the necessity of a multilateral approach became evident (Bunn, 2003). In 1968, negotiations at the United Nations (UN) were concluded with the Non-Proliferation Treaty (NPT). Subsequently, the non-proliferation norm became strongly institutionalized, comprising several treaties. Today, the NPT is almost universal in scope, with 188 member states in total, and indefinite duration (Rublee, 2009: 38). ³

Due to its institutional embedding, it can be expected that any changes of the regulatory scope require negotiations (Hypothesis 1). Indeed, negotiations appear to be the predominant mechanism to change the NPT as it subject of a review conference very five years. However, the regulatory scope of the norm has also been challenged frequently from states outside the regime. Several non-member states, that is, India, Israel, North Korea and Pakistan, have acquired nuclear weapons and thereby infringed upon the norm of non-proliferation (Shaikh, 2002). Moreover, NPT member Iran has violated the non-proliferation norm by developing a nuclear programme (Sagan, 2006).

However, normative change can only be expected if the norm challengers are stronger than the norm supporters (Hypothesis 2). Firstly, the group of norm challengers is weakened as many of them are not members of the NPT and therefore cannot engage in negotiations, and by their lack of coherent opposition. Some norm challengers do not reject the regime itself. Israel, for instance, supports the aims of the NPT, yet sees itself as unable to follow it due to security concerns. India and Pakistan, in contrast, not only infringe upon the norm but also reject the regime (Cohen, 1998: xviii, 470; Perkovich, 1999: xii, 597). However, the overwhelming majority of states remain committed to non-proliferation. In several regions of the world, nuclear weapon-free zones have been declared (Rublee, 2009: 39). Furthermore, states with the technical capabilities to produce nuclear weapons, such as Germany, Spain, Canada, Italy and Japan, refrain from developing nuclear weapons programmes. Moreover, countries such as Ukraine, South Africa and Belarus relinquished possession of nuclear weapons, or weapons programmes. Hypothesis 3 is only applicable if the norm challenger is stronger than the actors who prefer the status quo. The latter condition is not met in this case; as a result, Hypothesis 3 cannot be examined.

In summary, despite the fact that several countries have become de facto nuclear powers, the NPT norm was not abolished through norm violations. There is still overwhelming support for non-proliferation. Still, although the norm is embedded in an institutionalized negotiation system, the disarmament element was challenged to some extent by the major nuclear powers through their modernization programmes. Hypothesis 1 can therefore not be corroborated; yet for Hypothesis 2, the norm remained unchanged.

Case 2: norm against wartime plunder

The norm against wartime plunder prohibits the removal and appropriation of historical and cultural valuables and art by the victorious power. Already in the 17th century, the idea took hold that art requires particular protection during times of war. However, it was not until the 19th century that such a protection was effectively implemented in practice. After the victory over Napoleon in 1815, the Allied forces returned art plundered by the French to its home countries. In 1874, the norm became codified in a very precise manner consisting of two articles included in the Brussels Declaration. According to Article 38, plunder was prohibited without exception. In addition, Article 8 prohibited the confiscation and intended destruction of historical monuments, art and academic works. However, the norm was not embedded in an institutionalized negotiation system in which the actors regularly convened, but was integrated into the Hague Convention (Sandholtz, 2007: xi, 71–100). ⁴

As Hypothesis 1 suggests, the norm was challenged by norm violations rather than negotiations. During World War I (WWI), the parties mainly complied with the norm. Although some artefacts were destroyed, no party resorted to plunder. Only during WWII did Germany systematically challenge the norm. Hitler’s troops confiscated vast amounts of art in all occupied countries. The actions were rhetorically justified by the necessity to retrieve German art lost in previous wars, and the right to confiscate Germanic art. However, such actions directly contradicted the norm against wartime plunder (Sandholtz, 2008: 115). Germany’s challenge of the norm did not lead to a weakening or the abolition of the norm, as a strong alliance of states backed the norm. During the war, the Allies signed a declaration condemning the confiscation of historical artefacts and art in occupied territories. ⁵ Accordingly, the Allied powers returned stolen art systematically after the end of the war (Sandholtz, 2008: 117–120). Finally, the national socialist leaders were charged with the plundering of the occupied territories during the Nuremberg Tribunals (Sandholtz, 2008: chapters 6 and 7). These considerations appear not to be influenced by a cost–benefit analysis, as the Allies could have easily confiscated German art after their victory. The rejection of wartime plunder was instead rooted in the conviction that cultural treasures are an expression of the identity of a nation and should therefore not be taken as spoils of war. ⁶ Only the USSR initially requisitioned German art, yet ceased the practice after four transports (Sandholtz, 2008: chapter 6). In this case, the Allies were militarily stronger than the norm challenger Germany. After the Germany’s defeat, the Allies reinstated the status quo of the norm (Hypothesis 2). Hypothesis 3 assumes that the precision of the norm has an influence on the specific outcome of norm challenges, namely whether the norm is partially weakened or dies completely if the norm challenger is stronger than the actors who prefer the status quo. Since the latter condition is not met, Hypothesis 3 cannot be examined.

Case 3: norm against forcible intervention

Article 2 IV of the United Nations Charter (UNC) contains one of the core principles of the international community: the prohibition of threatening or using force against the territorial integrity of another state (Simma, 1999: 1–22). Although the norm against forcible intervention is widely accepted, it has been incrementally curtailed since the 1980s. In many cases of gross human rights violations, forcible interventions appeared to be the only feasible solution to protect innocent lives. The norm and the enforcement of human rights standards contradicted each other. Any attempt to comply with both norms generated a cognitive dissonance. The genocide in Rwanda in 1994 underscores this problem. Although the international community was well informed about the events in the country, no intervention was conducted. In hindsight, then-UN Secretary General Kofi Annan described the absence of intervention as a ‘sin of omission’ ⁷ and former US president Bill Clinton even apologized to survivors (Shattuck, 2003: 390).

A normative change took place within the UN negotiation system. A first attempt to change the norm was undertaken by the Security Council (SC). In 1992, the humanitarian situation in Somalia was depicted as a breach of international peace according to UNC Chapter VII. This enabled the SC to authorize the use of force to remedy the plight of the people. On the same basis, the interventions in Haiti (1994), Bosnia (1995) and East Timor (1999) have been legitimized (Malanczuk, 1997: 402–415). The advantage of authorizing forcible interventions through the SC was that the norm against forcible intervention was not violated. However, the disadvantage was the dependency on consensus in the SC. This became apparent during the Kosovo Crisis in 1999. Disagreements between SC members prevented a resolution sanctioning intervention. Eventually, NATO forces intervened without a UN resolution. In order to avoid the occurring cognitive dissonance of breaking the norm to protect human rights, rhetorically the Kosovo operation was deemed illegal according to the letter of international law, but legitimate in a moral sense (Independent International Commission on Kosovo, 2000: 372). This allowed an emphasis on both the need to protect human rights and the compliance with the norm against forcible intervention.

Against the backdrop of these developments, a Canadian-initiated Commission – the International Commission on Intervention and State Sovereignty (ICISS) – formulated the concept of contingent sovereignty (ICISS, 2001: IX). The new concept left the state sovereignty under normal circumstances untouched, yet allowed intervention in the domestic affairs of states in cases of gross human rights violations. The ICISS even considered a ‘responsibility to protect (R2P)’ people in cases of gross human rights violations to be in existence. R2P was not deemed to abolish the norm against forcible intervention, yet was considered to limit its application under particular circumstances.

In line with Hypotheses 1 and 2, the norm against forcible intervention was successfully altered through negotiations. A watershed event in this regard was the UN World Summit in 2005. Canada and the Group of 77, the main supporters of R2P, planned to use this forum to persuade other actors. The African Union also supported normative change, as it had already set a precedent by establishing a mechanism to implement R2P in Africa (Bellamy, 2006: 158). However, globally the terrain was difficult as most of the important permanent UN SC members were hesitant to support the norm (Bellamy, 2006: 152). During the summit, negotiations between the sceptics and the proponents took place and resulted in a tacit consensus. The final document of the UN World Summit in 2005, however, indicated a broad international support for alteration (Bellamy, 2005: 31–53). Over the course of the last decade this initially tacit support has grown significantly. In particular, the SC has incorporated R2P in its resolutions numerous times. ⁸ R2P is an important international norm curtailing the norm on forcible intervention (Bellamy and Williams, 2011).

As anticipated by Hypothesis 3, the norm environment together with the norm precision influenced the specific outcome of normative change. The norm against forcible intervention was characterized by little precision, while the environment was subject to slow change with the emergence of a competing norm. At first glance, the norm appears to be precise. Different norms with overlapping regulatory scopes clearly indicate the prohibition of forcible interventions. ⁹ The norm addresses all members of the international community and does not permit any deviations. However, the UNC contains three exceptions: UNC Article 51 permits self-defence, which includes the violation of the territory and sovereignty of the aggressor; UNC Chapter VII authorizes the SC to decide on forcible measures; and UNC Articles 53 and 107 specify the ‘enemy state clause’. Moreover, the exceptions are defined only vaguely and allow for great leeway in their interpretation. ¹⁰

Since the non-intervention norm was of little precision, the norm was not challenged directly and the vagueness of the norm allowed for incremental curtailment of its regulatory scope within the UN negotiation system in the form of R2P.

Case 4: the permission of nationalistic terror

The term ‘terrorism’ is difficult to define and concerns, in the broadest sense, any violence aimed at intimidating political opponents in order to achieve one’s own political objectives (Hoffman, 2001: 15). The term came into use during the French Revolution and referred to an instrument of the state to enforce order and prevent anarchy. However, with the excesses of the French Revolution and the emergence of democratic principles, state terror lost its legitimacy as an appropriate instrument of governance. This perception did not change during the 1930s, when authoritarian regimes used terror to intimidate and subdue their populations. In the 1980s and 1990s, terrorism had been outlawed domestically and the use of force by non-state actors had been generally delegitimized internationally (Thomson, 1994). Although initially a legitimate tool in the hands of the state, terrorism was no longer an appropriate means to achieve political goals.

In contrast to this general trend, however, one specific form of terrorism was still considered legitimate during the 20th century: nationalistic terror employed by non-state actors against an oppressor or occupying force (Sandholtz, 2008: 110). Three factors play into the existence of a norm permitting the use of terror for nationalistic purposes. Firstly, many states supported nationalist movements, which employed terroristic means. Secondly, in the debate on the international level on non-state actor violence, terrorism had a special status. Members of the UN repeatedly declared violent non-state actors as illegitimate. However, in the very detailed enumerations of violent groups, groups resorting to terrorism were never included (Sandholtz, 2008: 118). Thirdly, international initiatives, which were initially meant to prevent terrorism, supported the legitimacy of the political demands of terrorist groups. For instance, the ad hoc committee of the UN General Assembly on terrorism exempted groups employing terrorist means against imperialism, exploitation and apartheid from anti-terror measures (Franck and Lockwood, 1974: 46–91; Sandholtz, 2008: 114–117).

This norm was embedded into the UN negotiation system. Accordingly, as expected, it was challenged during negotiations within the framework of the system (Hypothesis 1). In the early 1980s, the legitimacy of terrorism as a means to fight oppression was decreasing. A new consensus emerged within the UN. General Assembly Resolution 40/61 (1985) declared terrorism unanimously as a criminal act. Subsequent resolutions in the 1990s aimed at eliminating terrorism (1994), preventing terrorist attacks (1997) and preventing the financing of terrorism (1999) (Sandholtz, 2008: 120). Likewise, the SC has condemned terrorist acts in several resolutions since 1985 (Bantekas, 2003: 315–333).

Hypothesis 2 anticipates a strong norm challenger to successfully push for normative change. Influential members within the UN system, such as Western states and permanent SC members, sought to abolish this norm. However, their initiatives were met with resistance – developing countries in particular further advocated the differentiation of ‘common’ and ‘political’ violence (Verwey, 1981: 69–92). This resistance, however, did not prevail. Indeed, some of the developing states are still opposed to the Western view, but nevertheless (although hesitantly) supported later resolutions of the General Assembly.

The norm was precise, as it did not legitimize terror in general, but only nationalistic terror. ¹¹ The norm environment in the 1980s was rather unstable. Firstly, terrorist attacks had continuously increased since the 1960s. While there were 132 attacks in 1968, the number had risen to 438 attacks in 1985. ¹² Furthermore, the threat had become internationalized. Hijacked airplanes or cruise liners, as well as the hostage situation at the Olympic Games in Munich, convinced many states of the necessity of counterterrorism measures. Secondly, many of the newly independent states considered their role as defenders of the status quo rather than as facilitators of revolutions. Many were opposed to an increased role of non-state actors on the international level. In particular, the spread of democratic procedures opened new avenues for participation and conflict resolution (Sandholtz, 2008: 121). The high precision of the norm led to its rapid degeneration, culminating in the death of the norm. While nationalistic terror was a legitimate tool during the 1960s and 1970s, it became delegitimized during the 1980s.

Case 5: anti-mercenary norm

The prohibition of mercenarism is widely accepted in the international community, as 100 resolutions critical of mercenaries were passed by the General Assembly. Even the SC condemned the use of mercenaries in the 1960s and 1970s (Percy, 2007a: 373–374). The mercenary ban has been included in many international treaties, for instance the Additional Protocol of the Geneva Convention (AP I) (1978) or the International Convention against the Recruitment, Use, Financing and Training of Mercenaries (1989). However, the strength of the norm is not rooted in international law, which provides many loopholes due to definition weaknesses (Schaller, 2005). In spite of weak international law, strong social norms can exist (Percy, 2007a: 367–397). The social anti-mercenary norm is a case in point, as it has developed considerable strength. Its emergence is intertwined with the emergence of the modern state. During the 19th century, non-state actor violence was increasingly delegitimized. The authority over the use of force was monopolized by the state. Mercenaries employed forces outside the state structure, which led to their criminalization. This taboo remained influential during the last century (Thomson, 1994: 96). Although mercenaries never disappeared completely, they did not play a crucial role in 20th-century warfare.

Over the last two decades, a new market for force has developed alongside the state system (Avant, 2005: 310). Private Military and Security Companies (PMSCs) are offering force and force-related services in several conflict zones across the globe. In Iraq, for instance, approximately 30,000 armed guards protected personnel and convoys, while in Afghanistan, between 18,500 and 28,000 perform similar armed services (Rimli and Schmeidel, 2007: 15). In short, commercial non-state actors selling armed services were present in war zones. One of the main reasons why the anti-mercenary norm is being violated again is the advantages these actors provide to their clients. PMSC services bought on the market are considered to be more cost-efficient than in-house solutions (Carafano, 2008: 242). Moreover, PMSCs are able to relieve regular troops of maintenance and support duties. However, despite this widespread practice, no actor openly challenged the norm. On the contrary, the differences between mercenaries and PMSCs are emphasized in order to keep PMSCs from being subsumed under the regulatory scope of the anti-mercenary norm. The UK Parliament’s Foreign Affairs Committee argued that PMSCs are tasked not to provide military services but to provide protective services. Hence, they cannot be considered mercenaries (Foreign Affairs Committee, 2002). Although such services are very difficult to differentiate, the interpretation that PMSCs provide defensive tasks while mercenaries engage in offensive operations has taken hold (Pelton, 2007: 109; Percy, 2007b: 227–228). This interpretation was recently corroborated by the Montreux Agreement, ¹³ which has been signed by 25 states. PMSCs are considered market actors, permitted to use force in self-defence. However, the anti-mercenary norm does not differentiate between different types of force (Petersohn, 2014). This begs the question of how defensive use of force by market actors became legitimate again.

The social anti-mercenary norm, in contrast to the weak legal norm, is not formally embedded in a negotiation system. However, the debates on both norms cannot be not entirely separated and merge into each other. In line with Hypothesis 1, the normative changes were initiated by norm violations, yet the UN was an important arena for the following debate (Krahmann, 2012). Crucial proponents of change are certainly the USA and UK, as they are home to the largest PMSC markets. However, they were not alone; outsourcing military services to the private sector is a global trend (Dunigan and Petersohn, 2015). As anticipated by Hypothesis 2, the norm challenge was successful. During the 1990s, the use of PMSCs was less widespread, and mainly weak states relied on their services (Musah and Fayemi, 2000). At that point in time, the firms were still often labelled as mercenaries. However, the discourse changed in the early 2000s when the USA and UK started to employ PMSC services extensively. Increasingly, the firms were called private military or private security companies, and associated with defensive services. Their role in supporting legitimate governments was emphasized, and mercenaries were increasingly associated with offensive operations (Petersohn, 2014). The UN Special Rapporteur on the use of mercenaries, Enrique Ballesteros, was strongly opposed to the new development, and tirelessly warned about the dangers of the new mercenary trend and the emergence of paramilitary actors (Ballesteros, 1997, 2001). However, his actions were not able to prevent the incremental change of the norm.

The vagueness of the old norm, ¹⁴ as well as the interest of strong actors to curtail the regulatory scope of the anti-mercenary norm, resulted in an incremental change. Although the norm was maintained rhetorically, the exemption of PMSCs restricted its regulatory scope. The weakening of the old norm has been facilitated by the extent of PMSC deployment since 2003 and by repeated acknowledgement of PMSCs as legitimate violent actors. Only a decade after their emergence, the applicatory scope of the 200-year-old norm became considerably restricted, weakening it accordingly.

Case 6: norm against unrestricted submarine warfare

Submarines have been used in warfare since the 19th century. Already at the Washington Conference in 1922 and the London Conference in 1933, the main maritime powers – the USA, the UK, France, Italy and Japan – set out to regulate the use of this weapons system (Goldman, 1994: 293–294). Over 30 states signed the 1922 and 1933 agreements, including Russia and Germany. The agreements stipulated that the laws of maritime warfare also applied to submarines. Most importantly, commercial ships could only be attacked and destroyed if the crew could be brought to safety. In the absence of these circumstances, an attack was prohibited.

During the Spanish Civil War, the norm displayed remarkable robustness. When Italy violated the norm, several states enacted sanctions, despite the fact that the norm was not embedded in an institutionalized negotiation system and each state had to bear the cost of the sanctions alone (Legro, 1997: 31–61). Regardless of earlier support for the norm, during WWII, unrestricted submarine warfare became the rule rather than the exception. The WWII norm violations resulted in the norm’s decline, as the norm was not embedded in an institutionalized negotiation system. Norm challengers were among the strongest naval powers at the time, and had large submarine fleets at their disposal. In 1940, the German empire declared unrestricted submarine warfare around the British islands in reaction to the armament of British cargo ships. Later, Germany waged unrestricted submarine warfare against all cargo ships supplying Great Britain. This set an escalation spiral into motion; when the USA entered the war with Japan in 1941, it immediately declared unrestricted submarine warfare (Legro, 1997: 31–61). No other state supported the maintenance of the norm.

The norm was very precise. Its goal was to limit the extent of submarine warfare; complex clauses were avoided and the description of how to deal with civilian cargo ships or naval warships was detailed and precise. Deviations from the rules were prohibited by the norm. However, changes in ship manufacturing, underwater navigation and weapons technology took place that required accommodation (Goldman, 1994: 157 and 212). ¹⁵ As a consequence, the norm challengers succeeded to abolish the norm against unrestricted submarine warfare. In contrast, merely delimiting the norm’s scope was not possible due to its high precision.