Jump to ContentJump to Main Navigation
Tempting FateWhy Nonnuclear States Confront Nuclear Opponents$

Paul C. Avey

Print publication date: 2019

Print ISBN-13: 9781501740381

Published to Cornell Scholarship Online: May 2020

DOI: 10.7591/cornell/9781501740381.001.0001

Show Summary Details
Page of

PRINTED FROM Cornell University Press SCHOLARSHIP ONLINE (www.cornell.universitypressscholarship.com). (c) Copyright University of Cornell University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in Cornell for personal use. date: 21 September 2021

(p.159) Appendix B Additional Cases

(p.159) Appendix B Additional Cases

Tempting Fate

Paul C. Avey

Cornell University Press

The Correlates of War database reports sixteen discrete wars in nuclear monopoly. These wars serve as the basis for general discussion of conduct of war in nuclear monopoly in the conclusion chapter. Five of these were addressed in the case study chapters: the Korean War (chapter 4), the War of Attrition and the October War (chapter 3), and the Gulf War and the Iraq War (chapter 2). In this appendix I briefly discuss the remaining wars: the 1956 Sinai War; the 1956 Soviet-Hungary War; the 1965 Vietnam War; the 1967 Six Day War; the 1979 and 1987 wars between China and Vietnam; the 1982 Falkland Islands War; the 1982 war over Lebanon; the 1987 war over Angola; the 1999 Kosovo War; and the 2001 Afghanistan War. I focus on the portions identified as interstate wars. Conflict and instability after the end of organized interstate hostilities between states are beyond the scope of my analysis.

Sinai War (1956)

The 1956 Sinai (or Suez) War pitted Egypt against Israel, France, and nuclear-armed Great Britain. The dispute centered on control of the Suez Canal. Following the coup that deposed Egypt’s King Farouk, the new Egyptian government sought to assert its control over the important waterway that cut through its territory. Recognizing its limited options, the British government agreed in October 1954 to remove British troops by 1956. On June 13, 1956, the last British soldiers left the Suez Canal Zone. President Gamal Abdel Nasser nationalized the Suez Canal Company a little over a month later, on July 26. The move, on top of other Egyptian rhetoric, alarmed officials in London. The British prime minister Anthony Eden then conspired with French and Israeli leaders to retake the canal. The plan called for Israeli forces to attack Egypt in the Sinai, strengthening Israeli borders. Britain and France would then demand an end to the fighting and call for (p.160) both sides to withdraw from the area. Next, they would introduce troops to seize the canal. Israel dutifully attacked on October 29, followed by the planned Anglo-French ultimatum on October 30. When fighting continued, British and French aircraft attacked Egypt, followed by the introduction of ground troops on November 5.1

Discussion of nuclear weapons in the war typically center on Soviet nuclear threats. The general consensus is that those threats had little influence on British and French decision making.2 Rather, US pressure—including economic coercion—compelled the European powers to reverse course and withdraw their forces by the end of December, handing Nasser an important political victory. The British nuclear monopoly relative to Egypt gets little attention. The British had developed a capable bomber force by 1956 that could deliver nuclear weapons, and British bases in the region could have been used as staging areas for nuclear strikes against Egypt.3

The British homeland, its nuclear arsenal, and its military faced no danger throughout the war. Without the prospect of military defeat, there were minimal benefits of nuclear use. Any nuclear-related costs—and Britain endured economic and political costs for the conventional military action alone—would therefore loom large. The lack of military danger to Britain from the conduct of the war is evidenced in several ways.4 To begin with, all the fighting took place outside British territory. Moreover, the Israeli Defense Forces (IDF) conducted the bulk of ground operations and were involved in the most intense fighting. Egyptian military performance was uneven, with some units fighting tenaciously and others quickly abandoning their positions. As a whole, though, Egyptian forces were poorly coordinated across the theater and were unable to execute counterattacks, allowing the IDF to make steady progress. The Egyptian air force posed little threat. Initial British-French attacks destroyed large numbers of Egyptian aircraft on the ground. From October 31 to November 2 the Egyptians lost more than 150 aircraft. Egypt then withdrew forty aircraft to bases out of Anglo-French range, but this removed their ability to engage in the fighting.5 Anticipating the Anglo-French attack, Nasser ordered the military to fall back from the Sinai on November 1 to meet the new threat. The withdrawal was poorly coordinated, though, and the retreat quickly turned into a rout. As Kenneth Pollack notes, “only one Egyptian battalion returned from the Sinai intact and capable of engaging in combat operations.”6

The Anglo-French invasion faced little opposition. The Egyptians reasonably concentrated their defense to protect Cairo from an expected British assault. However, the two allies targeted Port Said and Port Faud, massing a British infantry division, airborne brigade, and marine commando brigade alongside a French airborne division, parachute battalion, and mechanized brigade for the assault. The supporting naval force consisted of six aircraft carriers with modern jet aircraft. Out of position, with their air forces immobilized or destroyed, the Egyptian troops were woefully overmatched. For (p.161) example, the Egyptian force at Port Said consisted of two reinforced battalions of reservists and four self-propelled guns.7 British and French paratroopers seized key objectives on November 5 and easily beat back Egyptian counterattacks. On the sixth, British forces began amphibious landings against little opposition; most Egyptian defenders had fled. Total British and French losses from the operation were 26 killed and 129 wounded.8

Soviet Invasion of Hungary (1956)

The conflict began with popular demonstrations in Budapest on October 23.9 The protesters called for the return of ousted former leader Imre Nagy and demanded a series of economic and political reforms. They also sought the end of any Soviet presence in the country. Fighting erupted between government forces and the demonstrators, quickly spreading across most of the country. Rebels, in many cases joined by Hungarian troops, seized key installations such as radio stations, party headquarters, and the Ministry of the Interior. By October 28 the CIA reported that rebel forces were “in control of most of Hungary outside of Budapest.”10 Nagy returned to power on October 24 and initially requested the support of Soviet troops to help restore order. Soviet forces in the country had already been mobilizing as the Soviet leadership debated their response. On October 25 Soviet forces fired on demonstrators outside the Hungarian parliament, killing sixty. The Soviet forces subsequently disengaged from the fighting, awaiting Moscow’s decision, but mobilization outside Hungary continued. The Nagy government openly identified with the rebels by October 29 and called for the withdrawal of all Soviet forces on October 30. Hungary subsequently announced its intention to withdraw from the Warsaw Pact.

The Soviet leadership decided on October 31 to use the Red Army to restore Soviet influence.11 While the political effect of Hungary leaving the alliance would have been large for the Soviet Union—Soviet leaders worried that the loss of Hungary would destabilize the Soviet position in Eastern Europe and possibly even lead Hungary to enter the American orbit—the immediate military danger that Hungary posed to the Soviet Union during the fighting was nonexistent.12 The Soviets invaded on November 4 with a massive force of two hundred thousand troops and some twenty-eight hundred tanks that engaged and overwhelmed Hungarian freedom fighters and military troops.13 The entirety of the fighting took place on Hungarian territory, with no danger to the Soviet state. The Hungarian freedom fighters and allied military forces fought with defensive and guerrilla tactics; there was no capability of any major offensive against the Soviet Union. While the Hungarian forces fought determinedly, “hurling Molotov cocktails and even themselves at the turrets and treads of the Russian tanks,” they were simply outmatched.14 The Soviets suffered (p.162) modest losses during the campaign. Precise casualty estimates vary, but the Correlates of War reports 720 Soviet battlefield deaths, compared to 926 Hungarian battlefield deaths.15 That coding includes the interstate war portion of the conflict and does not include civilian casualties or losses earlier during the initial uprising. Some estimates place those in the thousands.

Vietnam (1965–1973)

In 1919, in the wake of the Great War, a young Vietnamese man named Nguyen Ai Quoc, who would later go by his more famous moniker Ho Chi Minh, traveled to Paris. He had hopes of presenting the US president, Woodrow Wilson, with Vietnamese desires for greater political power and freedoms under French rule, appealing to Wilson’s calls for greater selfdetermination.16 Less than thirty years later, following another world war, the Vietnamese and the United States found themselves on opposing sides. The United States was initially hesitant to back French efforts to maintain control of Indochina. Yet the US position shifted as the struggle became embroiled in Cold War dynamics. Indeed, by the early 1950s the US was no longer a reluctant French patron but instead a strong proponent of French efforts to counter the growth of communism, and with it fears of Soviet influence, in Vietnam. The French defeat at Dien Bien Phu in 1954 finally led to French withdrawal.17

The United States remained involved as Vietnam split into North and South, what would become respectively the Democratic Republic of Vietnam (DRV) and the Republic of Vietnam (ROV).18 The United States opposed national elections, fearing Ho Chi Minh had effectively seized the nationalist mantle and could out-organize noncommunist opposition. Instead, the US backed the staunch anticommunist Ngo Din Diem’s effort to create a viable South Vietnamese state. Throughout the 1950s and into the 1960s, a Communist insurgency, the Viet Cong, supported by the North Vietnamese government in Hanoi, gained strength and controlled sizable parts of the countryside. By the end of 1963 US involvement included sixteen thousand military advisers, though some fought alongside ROV units. Despite US support, Diem failed to build popular support and was overthrown in November 1963. In June 1964 North Vietnamese troops began directly participating in Viet Cong operations, and in September the first North Vietnamese regiment moved down the Ho Chi Minh Trail to South Vietnam.19 President Lyndon Johnson and his advisers saw little chance of victory but believed that the United States could not afford to walk away. The fear was that abandoning the ROV would call into question US credibility globally. In a series of decisions in 1964 and 1965, then, Johnson dramatically escalated the US presence and combat operations. US ground operations continued until August 1972, and large-scale air operations ended with the (p.163) termination of the Linebacker II bombing campaign on December 29. After a brief cease-fire, North Vietnam launched an assault on the South. South Vietnam continued to be unable to stand on its own despite the years of US involvement. After Saigon fell in April 1975, Hanoi consolidated control of the newly unified Vietnam.

The United States homeland or interests in regions outside Southeast Asia faced no danger throughout the war. This is not to minimize the American losses suffered. Yet the entirety of the fighting took place in and around Vietnam, thousands of miles from American territory. The Viet Cong relied primarily, though not exclusively, on guerrilla tactics, while the North Vietnamese Army would engage in conventional and mechanized military operations. Importantly, though, they were unable to project power to threaten American interests outside the region. Their stated limited aims to unify the country were therefore credible. While they inflicted losses, they were unable to decisively defeat the US military or cause the US homeland or key positions in Europe and the Pacific to be left defenseless. The sheer number and scope of military engagements make a full review of the combat beyond the scope of this chapter. In brief, North Vietnamese forces inflicted major losses on South Vietnamese forces but struggled to notch any battlefield victories against the United States.20 The sustained Rolling Thunder bombing campaigns by US forces proved ineffective against guerrilla operations. When North Vietnam adopted more conventional operations, the Linebacker I and II bombing campaigns did help contribute to North Vietnamese flexibility in negotiations that ended the US presence.21

Hanoi was ultimately able to achieve a political victory by inflicting losses on the United States that were out of proportion to American interests in the conflict. Years of inconclusive fighting in a country many saw as not vital to US security took its toll. Especially after the Tet Offensive in 1968, opposition to the war grew within the US public and ultimately among the soldiers asked to fight for an unpopular cause with little apparent prospect of success. The US military generally remained effective in combat but began to suffer a decline in morale and discipline.22 While not discounting American losses, it is clear that the Vietnamese lost more relative to their American opponents. From 1959 to 1975 the United States lost 58,178 killed and over 300,000 wounded (150,000 of which were hospitalized). Estimates are more difficult for the North Vietnamese and Viet Cong. Using conservative figures leads to approximately 730,000 North Vietnamese troops and Viet Cong killed, along with 65,000 North Vietnamese civilians.23

Six Day War (1967)

As noted in chapter 3, the inclusion of the Six Day War as a conflict in nuclear monopoly is debatable. The Dimona nuclear reactor came online in (p.164) 1964, and two years later Israel likely had the ability to produce weapons-grade fissile material. By early 1967, US intelligence estimated that Israel could construct a nuclear weapon in six to eight weeks.24 Avner Cohen’s authoritative work on the Israeli nuclear program concludes that it was not until immediately prior to the war that “Israel ‘improvised’ two deliverable nuclear explosive devices.”25 Egyptian leaders were aware of Israeli nuclear progress and understood after 1967 that Israel had nuclear capabilities. Yet in May–June 1967 it is doubtful that Egyptian leaders believed Israel possessed a functioning nuclear device.26 I nevertheless include a more detailed discussion here as a check against excluding a potential relevant case.27

The road to the Six Day War began with escalating Israeli-Syrian tensions and subsequent Egyptian troop deployments into the Sinai Peninsula in May 1967.28 More provocatively, on May 17–18, Nasser requested that UN forces withdraw from the area, and on May 22 he announced that Egypt would close the Tiran Straits to Israeli shipping.29 On May 17 and May 26, Egyptian aircraft overflew the Dimona reactor, generating alarm within Israel.30

Despite these moves, there is little evidence that Nasser or other Arab leaders intended to launch a war against Israel. If conflict came, the Egyptians would fight defensively. Egypt failed to undertake offensive preparations, with the final operational plan calling for a forward defense of the Sinai.31 Moreover, as noted in chapter 3, the Egyptian military was unreliable because its leader, Field Marshal Muhammad Abd al-Hakim Amer, treated the armed forces as his “own personal fiefdom,” where personal loyalty was more important than military competence.32 Though Egypt had a modest advantage in troop numbers and military platform quality, it lacked a decisive conventional advantage to overcome its other deficiencies. As President Lyndon Johnson told Israel’s foreign minister Abba Eben on May 26, “Our best judgment is that no military attack on Israel is imminent, and, moreover, if Israel is attacked, our judgment is that the Israelis would lick them.” In case he had not been clear, he added that “you [Israel] will whip the hell out of them.”33 Nasser’s motives appeared to center on overturning the post-1956 Suez War status quo, attaining a propaganda victory to offset Egyptian setbacks in Yemen and elsewhere in the region, and deterring further Israeli action against Syria or future action against Egypt.34 As Nasser told UN Secretary-General U Thant on May 24, Egypt had “achieved its goal by returning to pre-1956 position, with one difference: that they [the Egyptians] were now in a position to defend their country and their rights.”35

Similarly, Syria and Jordan were in no position to launch major conventional offensives against Israel. Israeli and Jordanian forces were relatively evenly matched in numbers and equipment, even with the bulk of the IDF engaged with the Syrians and Egyptians. Jordan planned for a forward defense of the West Bank. The one limited offensive element in its planning centered on capturing part of Jerusalem in the expectation that Israel would seize large parts of the West Bank elsewhere.36 “Amman’s major objective (p.165) during the Six Day War was simply to survive intact,” Pollack tersely concludes.37 The Syrians enjoyed a quantitative superiority along the Golan Heights, but two decades of political turmoil had taken its toll on the Syrian army. Its troops were poorly equipped, trained, and led. The Syrians focused primarily on the defense of the Golan. Even after being told (incorrectly) that Egyptian forces had routed the IDF, the Syrians launched only a few uncoordinated air strikes and staged a single limited offensive that Israeli settlers stopped largely unassisted.38

Egyptian planning prior to the war did include the possibility of air strikes against the Dimona nuclear facility. This appears to contradict my argument that the NNWS should avoid targeting the NWS’s nuclear arsenal. The Egyptian planning, which Cairo was never able to implement, is not a major challenge to my argument, though. To begin with, Egypt was not planning to target Israel’s nuclear arsenal. Egypt was likely unaware that Israel had nuclear weapons, or of the precise location of those weapons. Also, my argument focused on the difficulty of eliminating an opponent’s nuclear arsenal. This is discrete from a situation when a state believes the opponent has no nuclear weapons to begin with. In that case, a strike against the opponent’s ability to produce the necessary material can eliminate the future nuclear danger without any risk of immediate nuclear retaliation. Finally, Egypt did not intend to start a war over Dimona, but if war did occur, then Dimona would be an attractive target.39

The Egyptian deployment to the Sinai and ongoing tensions with Syria created serious problems for Israel. Israeli leaders had previously identified the closing of the Tiran Straits as a “red line” that would necessitate war, and feared potential attacks against Dimona.40 More basically, though, Israel could not afford to maintain mobilization indefinitely to offset Egyptian moves. As Zeev Maoz notes, during the crisis “one-fifth of Israel’s labor force was mobilized. The Israeli economy came to a screeching halt.”41 The ongoing crisis was creating an intolerable domestic situation for Eshkol, who resigned the post of defense minister (he retained the prime minister position) on June 1, with Moshe Dayan selected to take up the post.42 In addition, several in the government saw an opportunity to avoid a possible diplomatic defeat and substantially alter the military-political climate in the region.43

Israel launched a series of strikes beginning on June 5 that resulted in a decisive victory. The conduct of the war accords with my argument: there was little danger to Israeli territory, regime survival, or its nuclear arsenal during the fighting. Indeed, Israel dramatically increased its territory, to take control of the Golan Heights, the West Bank, the Gaza Strip, and the Sinai Peninsula. The Israeli attack began with Israeli Air Force strikes that caught Egypt by surprise, essentially eliminating the Egyptian air force on the ground.44 Israeli ground forces quickly advanced in the Sinai. While individual Egyptian units at times fought admirably, they were unable to (p.166) respond to the fast-moving Israeli charge. Amer ordered a general retreat on the afternoon of June 6 that quickly turned into a rout.45 On June 5, Amman ordered a more robust offensive than previously planned in response to erroneous Egyptian claims of military success.46 That offensive never progressed, though, as Israeli counterattacks on the fifth and sixth occupied the Jordanian forces in the West Bank. At 10 p.m. on June 6, King Hussein ordered the Jordanian military to retreat. Though the king rescinded the order, the Israeli advance and Jordanian confusion created a hopeless situation for Jordan by the morning of June 7.47 As noted above, the Syrians launched only a single and ineffective ground offensive during the first few days of the war. Combat was limited to an occasional Syrian air raid and artillery exchanges with the IDF. Israel began its major offensive against Syrian positions in the Golan on June 9 after the defeats of Egypt and Jordan. Syrian troops fought determinedly at times but undertook no offensives and indeed failed to even reposition forces or launch counterattacks in service of their broader defensive posture.48

China versus Vietnam (1979 and 1987)

The Correlates of War codes two wars between China and Vietnam in nuclear monopoly. The 1979 war was the larger and deadlier fight, after which low-level fighting continued throughout the 1980s. I discuss both wars together because of the continuous hostility during the period. Tensions between Vietnam and China had been increasing throughout the 1970s, brought on by the end of US-Vietnamese fighting and Chinese realignment toward the United States. The Vietnamese role in Southeast Asia was steadily increasing, culminating in the Vietnamese invasion of Cambodia in late 1978. Vietnamese persecution of ethnic Chinese living in Vietnam also increased. Vietnamese assertiveness and growing ties to the Soviet Union alarmed Beijing. The Chinese invasion, dubbed the “Punitive War,” sought to inflict military and civilian damage in an effort to deter further Soviet and Vietnamese expansion in the region.49

The conduct of the war is congruent with my argument’s expectations. The fighting took place almost entirely on Vietnamese rather than Chinese territory. Vietnam was on the defensive throughout the conflict, relying on both conventional and guerrilla means. The military performance of the Chinese People’s Liberation Army (PLA) was found wanting in a number of cases. While both sides endured losses, Vietnamese forces likely suffered more. In addition, China inflicted significant damage on Vietnamese infrastructure and civilian assets while suffering no comparable losses. In short, there was never any major danger to China.

The initial Chinese attacks on February 17 caught the Vietnamese by surprise and quickly broke through the Vietnamese front lines. Progress then (p.167) slowed. In the east, PLA units struggled against determined militia resistance and rugged terrain. They ultimately succeeded in taking their main objectives of Cao Bang and Dong Dang by February 25.50 In the west, the assault against the provincial capital Lao Cai proceeded methodically against entrenched Vietnamese forces. Superior Chinese numbers allowed the Chinese to capture Lao Cai on February 20–21.51 The Chinese inflicted substantial losses on Vietnamese militia and People’s Army of Vietnam (PAVN) forces, though Chinese units suffered as well. One battalion of the 308th PAVN Division, 3rd Battalion, 460th Regiment, conducted a limited incursion into Yunnan to attack a PLA position in China on February 23. But there was no sustained PAVN offensive into Chinese territory.52 After the seizure of Lao Cai in Vietnam, the recently arrived PLA 149th Division then moved to seize Sa Pa. Enemy action, rain, and difficult terrain slowed the advance. The Vietnamese countered, with the 316th PAVN Division putting up a determined blocking action in what was some of the most intense fighting of the war. After a week of continuous action, the 149th had lost 420 soldiers, while the PAVN 316th and supporting units lost an estimated 1,398 killed, 620 wounded, and 35 captured.53

Chinese forces in the east then moved to take the main target of Lang Son. As Xiaoming Zhang notes, that was the route that “Chinese imperial armies had historically used to invade Vietnam.” The city controlled key rail and road networks that could threaten Hanoi less than 140 kilometers away.54 After pausing to regroup, China launched the offensive on February 27 with seven divisions, totaling roughly eighty thousand troops. The fighting was again intense, but by March 1 PLA forces were shelling Lang Son.55 The Vietnamese ordered their forces to fall back on March 2, and on March 4 Chinese forces crossed the Ky Kung River to capture the southern portion of the city. Casualties mounted on both sides, but the Vietnamese suffered greater losses. In addition, China had turned Lang Son “into a ruin.”56

On March 5 Chinese leaders announced they had “achieved the expected objectives” and would “withdraw all troops back to Chinese territory.”57 Beijing had planned for only a short campaign, but the intensity of Vietnamese resistance and PLA struggles likely contributed to China’s decision.58 China’s withdrawal announcement did not end the fighting. Over the next two weeks PLA forces engaged in various battles against dispersed Vietnamese forces. As the editorial of the Vietnamese party journal Nhan Dan put it on March 7, Vietnam would allow Chinese withdrawal, but that did not mean providing a “red carpet exit.”59 Indeed, PAVN forces managed to rout the PLA’s 448th Regiment when the latter engaged in an illconceived operation to gain military experience.60 Beyond the military engagements, the PLA operational commander Xu Shiyou “ordered PLA troops to destroy everything they could along their way home.”61 The damage was extensive. The war ended on March 16 when Chinese forces completed their withdrawal.

(p.168) Precise estimates of the losses during the monthlong fighting are difficult to come by. The 2010 Sarkees and Wayman Resort to War (the basis for the COW dataset) reports 13,000 Chinese and 8,000 Vietnamese battlefield deaths.62 This makes the 1979 war the only one in the COW dataset in which the NWS suffered greater losses. Other reports list approximately 10,000 battlefield deaths on each side, while Xiaobing Li lists 26,000 Chinese casualties (the Vietnamese number for Chinese killed) and 37,300 Vietnamese troops killed.63 This does not include civilian losses, which were almost exclusively inflicted on the Vietnamese side. In any event, the losses to China, while not insignificant, did not threaten the destruction of the PLA and were close to Vietnamese military losses.

Military clashes along the Sino-Vietnamese border continued for the next decade. Fighting again took place primarily in Vietnam. China launched a series of limited offensives in 1980 to capture key positions in the Luojiaping Mountains along the border. In 1981 PLA operations concentrated on small areas along both the Guangxi and Yunnan borders. Vietnamese counterattacks failed to dislodge Chinese forces, so that the two sides remained locked in a low-level confrontation with occasional Chinese artillery bombardments in 1982–1983.64 In April 1984 the PLA launched a series of offensives in the Laoshan area supported by heavy artillery bombardment. Vietnamese counterattacks in June and July ended in failure. For the next several years PLA forces rotated in and out of the area in an effort to provide combat experience to the troops.65 As Zhang notes, Chinese troops wondered “why they had to fight for hills on the Vietnamese side of the border in ‘self-defense.’”66 Vietnamese sapper commando units conducted occasional raids beyond Chinese lines. From 1979 to 1986 Vietnamese aircraft overflew Chinese airspace on at least twelve occasions, but most such incursions lasted only a few seconds or minutes.67 For the most part, Vietnamese operations after 1984 were limited to attacks against Chinese encroachments into Vietnamese territory. There was no major offensive launched into Chinese territory, although the border itself was contested.68 Sarkees and Wayman code the simmering dispute as having escalated to a war from January 5 to February 6, 1987, with eighteen hundred Chinese and twenty-two hundred Vietnamese battlefield deaths.69 Chinese-Vietnamese relations improved at the end of the decade, in part due to the end of the Cold War, and Chinese troops withdrew and returned to China in 1992.70

Falkland Islands (1982)

The United Kingdom fought a nonnuclear-armed opponent once again in April 1982 when Argentina invaded the Falkland (Malvinas) Islands. The fighting was intense at times, and the British fleet faced real danger from (p.169) Argentine aircraft. Yet the conduct of the war posed little threat to Great Britain. My argument does not predict that the NNWS will avoid trying to harm the NWS. After all, the NNWS must have some conventional strategy that can outlast or inflict losses so that the NWS will negotiate. In the end, though, the fighting took place far from the core British territory, the Argentines used limited means and did not (because they lacked the ability) go beyond their limited aims, and the British suffered modest losses.71 The actual conduct of the war, despite the geographic disparity in favor of Argentina, testifies to the British conventional advantages.

The roots of the dispute traced back to the British occupation of the islands in 1833 that ended control of the islands by what would become Argentina.72 The intensity of the dispute had waxed and waned over the decades. In December 1981 the leaders of the Argentine military junta, President Leopoldo Galtieri, Admiral Jorge Anaya, and Brigadier General Basilio Lami Dozo, decided to invade the islands. A desire to distract from domestic problems drove the decision. As Amy Oakes notes, there is “a considerable degree of scholarly consensus regarding the degree to which the junta was influenced by the rising social unrest when it planned to invade the Falklands.”73

Though the focus in this appendix is on the conduct of war rather than nuclear views prior to fighting, the nature of the conflict warrants a somewhat lengthier treatment.74 Importantly, the junta did not initially expect Great Britain to respond with military force at all. If Britain did not fight, this would necessarily rule out the use of British nuclear weapons. As Oakes concludes, the “simple truth is that Argentina’s leaders would not have considered an invasion if they thought the United Kingdom was prepared to go to war over the islands.”75 Galtieri later stated that “such a stormy reaction as was observed in the United Kingdom had not been foreseen.”76 Argentine diplomats in London and New York reported to Foreign Minister Nicanor Costa Méndez prior to the war that Britain would likely impose economic sanctions and sever diplomatic relations, but would avoid military action.77 Several British actions prior to the invasion reinforced this view. Most notable was the British decision to remove its only semipermanent naval presence, the HMS Endurance, from the region.78 Prime Minister Margaret Thatcher’s decision to use military force surprised Argentina and even some British and Americans.

The Argentine strategy, doubting a British military response, centered on a bloodless operation to capture the Falklands, present Britain with a fait accompli, and seek negotiations. “Occupy to negotiate” was the basic objective.79 Initial plans called for withdrawing the bulk of the invasion force and leaving behind a five- to seven-hundred-man garrison to maintain order.80 As Richard Thornton concludes, as late as March there was no “concept, let alone plan, to defend the Malvinas against a British attempt to recapture the islands.”81

(p.170) Perceptions of American neutrality, if not support, also worked to embolden Galtieri and his lieutenants. Relations between the United States and Argentina warmed with the election of Ronald Reagan, buoyed by a shared anticommunism. Some felt Argentina had become a “privileged ally” to the United States.82 “We expected that the US government would act as a real-go between, a real neutral friend of both parties interested in the full implementation of the UN Charter,” Méndez stated. The United States might even lean on Great Britain to avoid any military action as it had done in 1956 during the Suez War.83 The United States did initially push for a negotiated solution. Ultimately, though, the “special relationship” led the Americans to side openly with the British, dashing Argentine hopes.

The Argentines pursued their planned limited-aims offensive. They could do little more than take the islands, because they lacked significant power-projection capability. Their claims of limited intentions thus had high credibility. Argentine forces seized the islands on April 2 in an operation that resulted in no British deaths. The British estimated Argentina lost five dead and seventeen wounded.84 With the onset of winter looming, and with it any chance of British military operations, Argentina succeeded in presenting Britain with a fait accompli and sought negotiations.

Once it was clear Britain would not negotiate, the Argentines undertook few defensive preparations. They neglected to extend the runway at Port Stanley to enable the deployment of several types of aircraft, forcing them to fly from the mainland, which reduced combat capability. The Argentine commander, General Mario Menéndez, dispersed his troops in static positions ill-suited to fending off a British attack. Argentina’s best units remained deployed along the Chilean border.85

The British forces engaged were qualitatively superior to their Argentine counterparts.86 Argentina had the advantage of fighting much closer to home, though. Buenos Aires sought to inflict sufficient damage on the British task force to deny Britain the ability to retake the islands or, at the least, make the effort to retake the islands too costly. While this would weaken British power projection abilities, it would not leave Britain defenseless. Argentina’s task was aided by the limited air support that the British fleet could muster, and British surface vessels were particularly vulnerable to the French-made long-range Exocet missiles. US estimates were cognizant of the challenges that Britain would face and the potential for British losses. Nevertheless, as a US National Security Council briefing report noted on April 28, “Britain has the means—whatever Argentina does—to isolate the islands, disable the airstrip, and attack the defenders, who are likely to run short of supplies in three weeks.”87

In any event, the conduct of the war ended up inflicting only modest losses on British forces, with Argentina sustaining relatively larger losses. On May 1, the British task force executed an attack against various military targets to convince the Argentines a landing was imminent. (p.171) The British engaged four Argentine Mirage III fighter-attack aircraft, destroying two while losing one Harrier aircraft. Britain suffered minor damage to surface ships, although in one case the Argentine planes barely missed the destroyer Glamorgan with two-thousand-pound bombs. Later, the British intercepted two Canberra light bombers, destroying one. The response was actually much less than Argentina had intended. Argentina dispatched fifty aircraft from land and their aircraft carrier, the Vienticinco de Mayo, to assault the British forces. Yet a third of the aircraft were forced to turn back when they failed to connect with airborne refueling tankers. Thirty planes did manage to reach the Falklands, but only six managed to locate British forces.88

The Argentine navy recognized their vulnerability to British submarines and proceeded cautiously. The British located and disabled the Argentine submarine Santa Fe on April 25. After briefly moving toward the Falkland Islands, by May 2 Argentine surface ships had turned back toward the mainland.89 In what became a contentious incident, the submarine HMS Conqueror torpedoed the cruiser Belgrano. The World War II–era Belgrano sank within an hour, killing 321 Argentine sailors.90 The Argentine navy subsequently refused to venture forth for the duration of the conflict, remaining within twelve miles of the Argentine coast.91 That policy had the virtue that it minimized Argentine naval losses, which would otherwise have been higher. It also accounts for the discrepancy in the number of surface vessels damaged between the two opponents. Simply put, the Royal Navy was engaged throughout the fight, the Argentine navy was not.

The Argentine air forces proved the most dangerous for the British. The most dramatic success came on May 4. Exploiting a gap in British low-level aircraft defenses, two Super Étendard aircraft each launched one Exocet missile at the British task force. One struck and disabled the destroyer HMS Sheffield, which later sank.92 After British troops began landing, forcing British ships to operate near the islands, Argentine aircraft managed to sink several British ships, including the frigates Ardent and Antelope (May 21 and 23–24), the logistic landing ship Galahad (June 8), the cargo ship Atlantic Conveyer (May 25), and the destroyer Coventry (May 25).93 Several more sustained damage. Importantly, though, Argentina failed to hit either British aircraft carrier or troop transports prior to the landings. As D. George Boyce notes, British aircraft armed with Sidewinder AIM-9L missiles “forced the Argentine pilots to deliver their bombs from a low altitude without adequate time for defusing—which resulted in the large number of Argentine bombs which hit their targets but failed to explode.”94

The air attacks were taking a significant toll on Argentina, though. It is doubtful their air force could have sustained the fight much longer. During the week of May 21 alone, Argentina lost twenty-one planes.95 (p.172) Indicative of the direction the fighting was going, on May 25 US Secretary of State Alexander Haig implored Thatcher “not to try to crush the Argentines.”96 British and Argentine accounts of total aircraft losses differ. The British Ministry of Defense reported they faced 120 fast jet aircraft, along with numerous other aircraft. Argentina claims it deployed eighty-one Mirage IIIs, Vs, and A4 Skyhawks. The British report the destruction of 109 Argentine aircraft of all kinds, including thirty-one Skyhawks and twenty-six Mirage jets. Argentina reports the loss of only thirty-four Mirages and Skyhawks—though even that smaller number would account for more than 40 percent of the force Argentina claims to have deployed.97 The British lost a total of five Harriers to ground fire, none in air-to-air combat.98

Land engagements ended in decisive British victories. The fighting was intense at times—involving aircraft, artillery, and light armor alongside infantry maneuvers—but resulted in few British casualties. On April 25, British soldiers retook South Georgia to the south and east of the Falklands with little resistance. On May 15, a British special forces raiding party surprised one hundred Argentine defenders at Pebble Island, destroying “eleven Pucara turboprop ground support aircraft, an ammunition dump, and other installations before departing.”99 British forces landed at San Carlos, East Falklands, on May 21. Argentine troops did not seriously contest the landings. Indeed, throughout the campaign Argentina failed to mount any counterattacks against the British advance. British mobility and superior firepower overwhelmed the Argentine defenders. The main assault on Port Stanley began on June 12; Argentina surrendered on June 14.100 “Even without the word ‘unconditional,’ the surrender was total and comprehensive,” writes Freedman.101

In this environment—an initial belief that Britain would not oppose the invasion, a limited-aims offensive against an isolated target, and subsequent fighting that posed little danger to British territory or nuclear forces—it would be surprising that the junta discussed the British nuclear arsenal at all. Yet they did just that. Based on interviews with former officials, T. V. Paul reports that Buenos Aires “considered the chances of Britain using its nuclear forces against Argentina, in the event of its losing the conventional battle.”102 This reflects a basic costs-benefits logic. According to one Argentine account, a West German official remarked after the war that it was best for Argentina they had not done more damage to the British fleet. “Queried about this apparent contradiction, he elaborated: ‘otherwise, Mrs. Thatcher’s government would have resorted to the use of nuclear weapons against the mainland.’”103 While it is unclear whether that particular exchange occurred, former Ministry of Defense official Michael Quinlan recalls that Thatcher “would have been prepared actually to consider nuclear weapons had the Falklands gone sour on her.” In particular, had Britain lost an aircraft carrier, Thatcher told Quinlan, she “would have been (p.173) willing to face up to the real eventuality of [nuclear] use.”104 It is unlikely Britain ever came close to using nuclear weapons, given the conventional asymmetry and low danger to Britain throughout, but the basic logic in these accounts is consistent with the framework developed in this book: nuclear use is more likely to be considered as the military benefits increase and begin to outweigh the associated costs. In the end, Lawrence Freedman writes in the official British history of the conflict that during his research he “found no references to any consideration of nuclear employment. This was never taken seriously as a realistic possibility.” He adds that “while there was never any thought of strategic nuclear use the possibility of tactical nuclear use was less readily dismissed.” The British leadership also took pains to transfer nuclear weapons onboard surface naval vessels to the carriers, which had more robust safety measures, and eventually move them back to Great Britain.105

In addition, similar to the other cases examined in this book, there is some evidence that the NNWS discounted the likelihood of nuclear use because it believed external actors would constrain the nuclear opponent. As Paul notes, Argentine officials believed that “the US and USSR would have prevented it if the British threatened to use nuclear weapons in a small conventional theater.”106 Argentina could also point to global public opinion against nuclear weapons—heightened during the 1980s amid renewed Cold War tensions and debates about impending American intermediate nuclear force deployments to Europe—as a further restraint against nuclear use.107 Argentine officials went so far as to raise the issue publicly. “I don’t think a country with nuclear arms will use them against a country that doesn’t have them,” the head of the Argentine National Atomic Energy Commission Castro Madero argued on May 28. Echoing language used by Stalin and Mao to deter nuclear threats, he went on to characterize nuclear discussions as a “psychological action” against Argentina.108 For their part, the British were aware of potential psychological advantages of conventional and nuclear strikes. During a British cabinet meeting on April 16, Thatcher highlighted that although “there was in reality no intention of attacking the Argentine mainland, there might be some military advantage in the Argentines being afraid of that; the fact that the Vulcans were being given conventional bombing practice in Scotland was in any case likely to become known…. Though the Vulcans were associated in the public mind with their long-standing nuclear role, there was of course no question of their carrying nuclear weapons in the present context.”109 Left unexplained was what might lead the “present context” to change.

There is evidence Argentine leaders believed that their own nuclear weapon might offset the British nuclear advantage. As Thornton concludes, for many at the time, “a nuclear weapons capability would permit Argentina to deal with Great Britain over the Falkland Islands dispute from a position (p.174) of equality, if not strength.”110 To be sure, at most Argentina explored a nuclear device, and Brazil likely loomed larger in Argentine calculations.111 That did not stop CIA analysts from worrying during the conflict that “the Argentine leadership might somehow calculate that the chances for a favorable outcome would have been greater if Argentina possessed a nuclear weapons capability.”112 Argentina provided reasons for such concerns. As the agency noted later that year, during the war “Buenos Aires asserted publicly that its adherence to nonproliferation rules had placed it at a clear disadvantage…. Buenos Aires [claimed it] could not continue to accept a discriminatory situation that denies Argentina the legitimate use of nuclear materials for its national defense.” The report concluded that Argentine military leaders probably “believe that if their country had possessed nuclear weapons … the British would not have been so quick to send so large an expeditionary force against them.”113 Julio Carasales, a former senior Argentine foreign affairs official, would later acknowledge that the Falklands War “caused some Argentine citizens, for the only time in Argentinean history, to want the country to possess nuclear weapons…. The fear that their [British nuclear weapons] mere presence inspired put the Argentinean forces at a disadvantage. More than one Argentinean thus considered that the outcome could have been different, or at least the defeat would not have been so humiliating, if his country had possessed nuclear weapons, even without using them.”114

A very large danger to Britain could generate sufficient benefits from nuclear use to offset any associated costs. Yet the nature of the participants and conduct of the war meant that the danger to the United Kingdom was low, and Argentina could gamble that Britain would not resort to nuclear strikes. The Falklands were British territory, but they were located nearly eight thousand miles from the British homeland. Their contribution to the British economy or strategic position were minimal, the islanders were not granted full British citizenship, and Britain had been reducing its presence in the South Atlantic for several years.115 The entire fight took place on and in the immediate vicinity of the islands, which were approximately four hundred miles from Argentina.

War over Lebanon (1982)

On several occasions in the late 1970s and early 1980s Israeli forces attacked individuals associated with the Palestine Liberation Organization (PLO) who were residing in Lebanon. At the same time, Syria had intervened in the Lebanese Civil War in 1976, occupying Eastern Lebanon and attempting to maintain order.116 This put Syrian and Israeli units in close proximity, and the two occasionally collided, fighting briefly in April 1981, after which Syria deployed some surface-to-air missile units to Lebanon.117 Limited (p.175) attacks against the PLO proved insufficient from Israel’s perspective, and so Israeli leadership, spearheaded by Defense Minister Ariel Sharon, elected for a larger operation. In particular, Israel sought to eliminate the PLO presence in Beirut. The assault would necessarily bring Israeli Defense Forces (IDF) into contact with the Syrian military, which Israel sought to expel from Lebanon.118 The immediate catalyst for the war came on June 3 when the Abu Nidal organization, a splinter group of the PLO, gravely wounded Israel’s ambassador to the United Kingdom.119 Israel launched Operation Peace for Galilee on June 6; Israeli forces would remain in Lebanon until May 2000,120 their operations there directed largely at various non-state actors. In this section I focus on the portion of the fighting against Syria, which constituted a war between two states.

The initial Israeli invasion consisted of two major advances to attack the PLO and engage Syria. The Israeli cabinet was reluctant to authorize military operations against Syrian forces, and so the IDF sought to threaten Syrian forces to provoke a response.121 During a series of clashes around Ayn Zhaltah on June 8, the Syrians inflicted only minimal damage but managed to delay the IDF. As Kenneth Pollack concludes, that delay “was one of the most important factors in preventing the complete destruction of the Syrian army in Lebanon.”122 That same day, Israeli forces also attacked a Syrian task force at Jazzin. On June 10 the IDF broke through Syrian defenses in the Bekaa Valley and proceeded methodically northward. In addition to the ground fighting, Israel systematically dismantled Syrian air defenses in Lebanon and easily defeated Syrian Air Force (SAF) efforts to contest the skies. Through September the SAF lost eighty-six Soviet-made MiGs to the Israeli Air Force without destroying a single Israeli aircraft.123 Israel consistently defeated Syrian troops but was unable to completely rout its Syrian opponents, which generally retreated in good order. As a result, Syria remained a factor in Lebanon.124

The conduct of the war resulted in little danger to the NWS, consistent with my argument. Syria’s President Hafez Asad cautiously observed the Israeli invasion and sought to avoid overtly provoking the Israeli forces. Once fighting began, Syria fought primarily on the defensive, setting ambushes against advancing Israeli units at various places. The fighting took place in Lebanon; Syria did not threaten Israeli positions in the Golan Heights or the Israeli homeland. Syrian aims centered primarily on maintaining the status quo of their position in Lebanon, avoiding a massive military defeat, and guarding against any possible Israeli advance on Damascus itself. Though Syrian forces fought with determination at various points, they inflicted only modest losses on the IDF. Pollack reports the grim relative tally: “the Syrians lost 1,200 dead, 3,000 wounded, and 296 prisoners in addition to 300–350 tanks, 150 APCS [armored personnel carriers], nearly 100 artillery pieces, twelve helicopters, 86 aircraft, and 298 SAM [surface-to-air missile] batteries. Against the Syrians during 6–25 June, the (p.176) Israelis suffered 195 killed and 872 wounded in addition to 30 tanks lost (with another 100 damaged) and 175 APCS destroyed and damaged.”125

War over Angola (1987)

The war over Angola was part of the broader Angolan civil wars. Angolan revolutionaries long contested Portuguese rule. The ongoing conflict contributed to a coup and popular revolution in Portugal that overthrew the fascist dictator António Salazar. Angola then achieved independence in 1975. Several different groups fought for control of the country. The Correlates of War provides one interstate war number but two separate start and end dates, breaking the conflict into two separate interstate wars. This is unusual, but there are a number of other wars that are part of longer ongoing conflicts (e.g., China versus Vietnam and Egypt versus Israel) that are similar. I include the 1987 war here to avoid arbitrarily excluding a case.

The first interstate war began in October 1975 when, as Michael Clodfelter writes, “outside intervention had rapidly turned what was basically a tribal war into an international affair.”126 That phase ended in February 1976 and involved no nuclear-armed states. The Soviet Union backed Cuba and Angola—in particular the Popular Movement for the Liberation of Angola (MPLA)—but limited its involvement to aid and advisers. Internal and unconventional fighting continued, with Angola and Cuba facing a determined guerrilla resistance from the National Union for the Total Independence of Angola (UNITA) operating primarily in southern Angola. This suited the interests of South Africa, which controlled Namibia and sought “to deny a southern Angolan sanctuary to Namibian insurgents and to maintain a buffer against the Angolan regime.”127

In August 1987 a new Angolan offensive pressed into southern Angola. The offensive was the beginning of the second interstate war, which COW codes starting on August 4, 1987, when South Africa made the decision to undertake a sizable intervention to halt the offensive. The war occurred between, on the one side, Cuba and Angola (with Soviet-supplied equipment and advisers), and on the other, nuclear-armed South Africa. The war ended in June 1988, with a formal cease-fire on August 5 of that year. The Tripartite Agreement signed on December 22 committed the Cubans and South Africans to withdraw from Angola.128

South Africa faced little danger throughout the war. Rough indicators of the balance of power understate the South African advantage. For example, South Africa’s per capita GDP was only slightly larger than those of Cuba (1.5:1) and Angola (1.9:1). Yet the 1985 Military Balance concluded that “South Africa remains the only African country capable of significant force projection operations against her neighbors.” Although Angola (p.177) might be capable of conventional operations against South Africa, even with Soviet and Cuban assistance Angola “is stretched to her limits containing the forces of UNITA and could not also defend against a major South African offensive.”129 As Narang concludes, “South African defense and air forces were both quantitatively and qualitatively superior to their primary regional threats, even with the deployment of Soviet surface-to-air missile batteries in the region.”130 Cuban forces were capable, but they were operating thousands of kilometers from their home and so posed little danger to South African territory.

The fighting itself took place in Angola, far from South African territory. The initial Angolan offensive advanced from the strategic town of Cuito Cuanavale into UNITA territory before being stopped in a series of conventional battles near the Lomba River by smaller South African and UNITA forces. In one particularly lopsided engagement on October 3, South African forces killed over six hundred Angolan troops and destroyed or captured 127 tanks, armored cars, and other vehicles, at a cost of one South African killed and five wounded.131 The advance then “turned into a headlong retreat over the 120 miles back to the primary launching point at Cuito Cuanavale,” writes Chester Crocker, an American diplomat at the time, who would help negotiate an end to the fighting.132 The South Africans harassed the retreating forces the entire way. On November 15 Cuban leaders decided to reinforce the beleaguered Angolan forces to prevent a deeper UNITA–South African advance. Cuba increased its troop strength in Angola and rushed reinforcements to Cuito Cuanavale.133 The arrival of Cuban reinforcements stabilized the defenses, though domestic South African political constraints, which prevented Pretoria from committing large numbers of reinforcements, simplified the defensive effort. The battle essentially ended in March, with South African forces shifting to a defensive posture in the area.134

As Cuito Cuanavale ended, Cuban forces moved to threaten southwest Angola. Cuba’s Fidel Castro hoped that this would put pressure on South Africa and aid the Cuban position in negotiations.135 As Peter Liberman notes, “Castro warned at the time that South Africa risked ‘serious defeat’ and hinted at an offensive into Namibia.”136 Despite the bluster, “Cuba never seriously contemplated a decisive military showdown with Pretoria,” concludes Stephen Weigert. Moreover, “Castro had secretly agreed with Moscow that Cuban troops would not cross the Angolan/Namibian border.”137 Cuban-Angolan and South African forces instead fought a series of small engagements in Cunene Province. Neither side gained a decisive advantage, and combat effectively ended following bloody air and ground clashes on June 26–27.

South Africa possessed only a rudimentary nuclear capacity. As the chief of the South African Defense Forces from 1985 to 1990, General Jan Geldenhuys, recalled, “Invasions were seen as slight possibilities, (p.178) adventurous transgressions of borders on such small scale that nuclear capability never came into the picture.”138 South Africa gave little thought to using its nuclear weapons against military or civilian targets. “No offensive tactical application of nuclear weapons was ever foreseen … as it was fully recognized that such an act would bring about nuclear retaliation on a massive scale,” writes Waldo Stumpf.139 Rather, Pretoria contemplated using the nuclear weapons in a “catalytic” manner to generate outside involvement, particularly from the United States. Andre Buys, the chair of the strategy group for the state arms procurement and production agency (Armscor) recalled that only if all efforts to elicit support failed, then “the last step would … be to threaten to use nuclear weapons on the battlefield in self-defense.”140

Nuclear weapons played only a minor and indirect role in South African thinking during the Angolan War. South Africa reopened its Kalahari test site in 1987, though the precise date is contested. The activity was limited to Armscor building a hangar above a test shaft, pumping out water, and checking the shaft’s readiness.141 Buys told Liberman in 1999 that the decision was made because “for the first time the government started considering the possibility that we might lose the war militarily.” Stage two of South Africa’s nuclear strategy—covert signaling or secret acknowledgment of the nuclear arsenal—would “come into operation once we were confronted by a serious and escalating military threat. We got close to that in 1987 … in Angola.”142 As one South African counterintelligence officer noted, “we knew satellites would see the whole thing … Soviet and Western intelligence were suddenly convinced we were serious about nuclear weapons and the West began to put pressure on the Soviets to get the Cubans to withdraw from Angola.”143 It is debatable if the chain of events worked out this way; as noted, Castro sought to avoid a major conflict but sought some form of battlefield victory to assist in negotiations. In any event, congruent with my argument’s predictions, the danger to South Africa was minimal throughout the war against Angola and Cuba.

Kosovo (1999)

The Correlates of War identifies the primary participants in the Kosovo War as the nuclear-armed United States against nonnuclear Yugoslavia (Serbia). Violence erupted in the Yugoslavian province of Kosovo in March 1998 following the killing of twenty-four ethnic Albanians by Serbian police on February 28.144 Serbian efforts to assert control resulted in the displacement of tens of thousands of people.145 US mediation efforts stabilized the situation briefly but ultimately collapsed as the Kosovo Liberation Army (KLA) was able to reconstitute itself, leading to Serbian redeployment of its forces.146 President Slobodan Milošovic´ of Yugoslavia (p.179) rebuffed US demands to cede Serbian control of Kosovo. In response, NATO launched Operation Allied Force on March 24, 1999. Air strikes continued until June 9, when Serbia agreed to a peace proposal. The combination of mounting costs, Russian pressure, and the reduction in US demands put forth in the G8 (Group of 8) foreign ministers peace proposal led to Serbian acquiescence.147

There was no danger to the nuclear weapon state. The fighting was entirely fought in Yugoslavian territory. No NATO ground troops were used, though there is debate whether the threat of a ground invasion contributed to Serbian concessions.148 Regardless, the campaign was fought entirely with NATO naval and air strikes, which typically operated out of range of effective Serbian counter-fire. The Correlates of War lists two American and five thousand Serbian battlefield deaths.149 The Serbian goal to maintain rule over Kosovo and not expand the conflict were credible because Serbia could not do more; Serbia struggled to even interfere with NATO operations over its own homeland. As Phil Haun notes, Serbian strategy was limited to inflicting “combat losses on NATO aircraft and aircrew, making it either too costly for NATO to continue air operations or, at a minimum, creating tension among NATO countries that might cause a fissure in the alliance.”150 In sum, the fighting was entirely on NNWS territory, the NNWS had limited and defensive aims, and there were very low losses to the NWS.

Afghanistan (2001)

As of this writing, US combat operations continue inside Afghanistan. The United States (along with Britain and other allies) acted in response to the terrorist attacks by al-Qaeda on September 11, 2001, that destroyed the two towers of the World Trade Center in New York City and part of the Pentagon. The United States initially demanded that the Taliban government of Afghanistan hand over al-Qaeda’s leaders, including Osama bin Laden, and shut down al-Qaeda training camps.151 The Afghan government refused US demands. The interstate war phase of the conflict began on October 7, 2001, with US air strikes and ended on December 22, 2001, with the installation of the interim Afghani government led by Hamid Karzai.152

The danger to the United States and the United Kingdom from Afghanistan was minimal. To be sure, al-Qaeda had managed to coordinate an operation that constituted the worst attack on US soil since Pearl Harbor. Yet the interstate war with Afghanistan involved fighting far from the nuclear weapon states. The major US involvement initially was special operations forces and air strikes that assisted Northern Alliance ground forces that opposed the Taliban. Taliban and al-Qaeda fighters were unable to defeat moderately skilled opponents that had American support, (p.180) though they were able to resist effectively against unskilled opponents.153 Prior to the attack, concludes Haun, the “probability of U.S. victory against Afghanistan … was high and cost of fighting relatively low.”154 During the war “sixteen Americans had died in defeating the Taliban, 15 of them in (predominantly air) accidents or in friendly fire incidents…. Taliban losses were uncounted but numerous.”155 In sum, the interstate portion of the war was fought on NNWS territory against an adversary that fought defensively.


(1.) Tony Judt, Postwar: A History of Europe since 1945 (New York: Penguin, 2005), 294–97.

(2.) Richard K. Betts, Nuclear Blackmail and Nuclear Balance (Washington, DC: Brookings Institution Press), 62–65; Todd S. Sechser and Matthew Fuhrmann, Nuclear Weapons and Coercive Diplomacy (Cambridge: Cambridge University Press, 2017), 225–26.

(3.) Mark S. Bell, “Beyond Emboldenment: How Acquiring Nuclear Weapons Can Change Foreign Policy,” International Security 40, no. 1 (Summer 2015): 101–2, 114–15; Judt, Postwar, 296–97.

(4.) The discussion in this paragraph and the next draws from Kenneth M. Pollack, Arabs at War: Military Effectiveness, 1948–1991 (Lincoln: University of Nebraska Press, 2002), 27–47.

(8.) Pollack, 39. Egyptian losses against the British-French-Israeli forces are estimated at one thousand killed, four thousand wounded, and six thousand captured. The Correlates of War (May 2018) estimate twenty-two British battlefield deaths. See conclusion chapter.

(9.) This paragraph draws on Jonathan Haslam, Russia’s Cold War: From the October Revolution to the Fall of the Wall (New Haven, CT: Yale University Press, 2011), 168–73; Vladislav Zubok, A Failed Empire: The Soviet Union in the Cold War from Stalin to Gorbachev (2007; Chapel Hill: University of North Carolina Press, 2009), 115–19; and Charles Phillips and Alan Axelrod, eds., Encyclopedia of Wars, vol. 2 (New York: Facts on File, 2004), 580–81.

(11.) This reversed an October 30 decision not to intervene. See Zubok, Failed Empire, 115–17.

(12.) On concern with the United States see Haslam, Russia’s Cold War, 171; Zubok, Failed Empire, 117.

(13.) Phillips and Axelrod, Encyclopedia of Wars, 2:581; Meredith Reid Sarkees and Frank Whelon Wayman, Resort to War: 1816–2007 (Washington, DC: CQ Press, 2010), 106–7.

(14.) Michael Clodfelter, Warfare and Armed Conflicts: A Statistical Encyclopedia of Casualty and Other Figures, 1494–2007, 3rd ed. (Jefferson, NC: McFarland, 2008), 576.

(15.) The 720 number matches the number of killed (669) and missing in action (51) provided by Clodfelter and Phillips and Axelrod. See Sarkees and Wayman, Resort to War, 105; Clodfelter, Warfare and Armed Conflicts, 577; and Phillips and Axelrod, Encyclopedia of Wars, 2:580.

(16.) Fredrik Logevall, Embers of War (New York: Random House, 2013), 3–4.

(17.) Logevall, chaps. 5–21; and John M. Schuessler, Deceit on the Road to War: Presidents, Politics, and American Democracy (Ithaca, NY: Cornell University Press, 2015), 60–63.

(20.) On Vietnamese victories over South Vietnamese forces see Caitlin Talmadge, The Dictator’s Army: Battlefield Effectiveness in Authoritarian Regimes (Ithaca, NY: Cornell University Press, 2016), chap. 3. For discussions of US engagements see Jasen Castillo, Endurance and War: The National Sources of Military Cohesion (Stanford, CA: Stanford University Press, 2014), 187, 194–202, 207–9; Clodfelter, Warfare and Armed Conflicts, 716–38.

(21.) Robert A. Pape, Bombing to Win: Air Power and Coercion in War (Ithaca, NY: Cornell University Press, 1996), chap. 6.

(23.) Clodfelter, Warfare and Armed Conflicts, 761–62. Using different start and end dates, Sarkees and Wayman, Resort to War, 111, report 58,153 US battlefield deaths.

(p.226) (24.) Avner Cohen, Israel and the Bomb (New York: Columbia University Press, 1998), 232; Vipin Narang, Nuclear Strategy in the Modern Era: Regional Powers and International Conflict (Princeton, NJ: Princeton University Press, 2014), 283.

(27.) Including the case in fact biases the results against my argument, because aggregate capability indicators discussed in the conclusion chapter code Israel’s Arab opponents as having significant capabilities relative to Israel.

(29.) Zeev Maoz, Defending the Holy Land: A Critical Analysis of Israel’s Security and Foreign Policy (Ann Arbor: University of Michigan Press, 2006), 80–81; Mearsheimer, Conventional Deterrence (Ithaca, NY: Cornell University Press, 1983), 143–44; Roland Popp, “Stumbling Decidedly into the Six Day War,” Journal of Cold War Studies 62, no. 2 (Spring 2006): 285.

(30.) Avner Cohen, “Cairo, Dimona, and the June 1967 War,” Middle East Journal 50, no. 2 (Spring 1996): 191 203–6; Narang, Nuclear Strategy in the Modern Era, 285–86.

(31.) Pollack, Arabs at War, 58–62. For general reviews that Nasser did not seek to provoke war but took a series of escalatory steps see Ben D. Mor, “Nasser’s Decision-Making in the 1967 Middle East Crisis: A Rational-Choice Explanation,” Journal of Peace Research 28, no. 4 (November 1991): 359–75; Popp, “Stumbling Decidedly,” 293–98; Cohen, “Cairo, Dimona, and the June 1967 War,” 200; and Richard B. Parker, The Politics of Miscalculation in the Middle East (Bloomington: Indiana University Press, 1993), 43, 49, 89–96.

(32.) George W. Gawrych, The Albatross of Decisive Victory: War and Policy between Egypt and Israel in the 1967 and 1973 Arab-Israeli Wars (Westport, CT: Greenwood, 2000), 12–13; Pollack, Arabs at War, 58; Risa Brooks, “An Autocracy at War: Explaining Egypt’s Military Effectiveness, 1967 and 1973,” Security Studies 15, no. 3 (July–September 2006): 412, 416.

(33.) Quoted in Popp, “Stumbling Decidedly,” 304. The full conversation is available at Memorandum of Conversation, May 26, 1967, FRUS 19641968, vol. 19, https://history.state.gov/historicaldocuments/frus1964-68v19/d77.

(34.) On these points see Cohen, Israel and the Bomb, 263–65; Stephen M. Walt, The Origins of Alliances (Ithaca, NY: Cornell University Press, 1987), 96–98; Gawrych, Albatross of Decisive Victory, 5–6; Narang, Nuclear Strategy in the Modern Era, 285.

(35.) Thant-Nasser UN Memcom, May 24, 1967, in Parker, Politics of Miscalculation, 231. See also Anderson to Rusk and Johnson, June 2, 1967, at 236.

(43.) For the argument that Israel was aware of its conventional superiority and the crisis created an opportunity for Israel to eliminate potential threats see Popp, “Stumbling Decidedly,” 299–308; and Mearsheimer, Conventional Deterrence, 150–53.

(49.) M. Taylor Fravel, Strong Borders, Secure Nation: Cooperation and Conflict in China’s Territorial Disputes (Princeton, NJ: Princeton University Press, 2008): 63; Andrew Scobell, China’s Use of Military Force: Beyond the Great Wall and the Long March (Cambridge: Cambridge University Press, 2003), 120–25, 135–37; Xiaobing Li, A History of the Modern Chinese Army (Lexington: University Press of Kentucky, 2007), 251–53; Henry J. Kenny, Shadow of the Dragon: Vietnam’s (p.227) Continuing Struggle with China and Its Implications for US Foreign Policy (Washington, DC: Brassey’s, 2002), 53, 98–100.

(50.) Xiaoming Zhang, Deng Xiaoping’s Long War: The Military Conflict between China and Vietnam, 1979–1991 (Chapel Hill: University of North Carolina Press, 2015), 94, 99, 108–9; Li, History of the Chinese Army, 254.

(51.) Charles Phillips and Alan Axelrod, eds., Encyclopedia of Wars, vol. 3 (New York: Facts on File, 2004), 1061; Li, History of the Modern Chinese Army, 255; Zhang, Deng Xiaoping’s Long War, 100–101.

(52.) Zhang, Deng Xiaoping’s Long War, 113n132. Phillips and Axelrod report that Vietnamese forces staged a “counteroffensive into Chinese territory” to which “Chinese defensive forces responded quickly and drove this incursion back, but there was widespread shock over the temerity of the invasion.” Phillips and Axelrod, Encyclopedia of Wars, 3:1061. It is unclear if Phillips and Axelrod are referring to the 3rd Battalion, 460th Regiment, attack. In any event, Vietnam does not appear to have “invaded” China but rather conducted a limited assault directly across the border with a small force on one occasion to disrupt the Chinese advance.

(58.) As Zhang notes, Beijing “granted operational autonomy to regional commanders but kept the duration and space of the fight under the command of the central leadership in Beijing. Deng Xiaoping was determined to avoid having the invasion turn into a quagmire for China.” Zhang, Deng Xiaoping’s Long War, 71. See also Scobell’s account, which highlights concerns that a larger war could lead to escalation with the Soviet Union as well. Scobell, China’s Use of Military Force, esp. 125–29.

(61.) Quoted in Zhang, 112. Xu primarily commanded forces in the eastern part of the fighting next to Guangxi; Scobell, China’s Use of Military Force, 132, 127.

(62.) Sarkees and Wayman, Resort to War. The May 2018 COW dataset reverses the figures to report eight thousand Chinese and thirteen thousand Vietnamese battlefield deaths. This change would strengthen my argument.

(69.) Sarkees and Wayman, Resort to War, 141–42. The May 2018 COW dataset has the same start and end dates but reports twenty-two hundred Chinese battlefield deaths with one thousand for Vietnam.

(71.) The Correlates of War lists 255 British battlefield deaths, with nearly three times as many Argentine battlefield deaths at 746.

(72.) T. V. Paul, Asymmetric Conflicts: War Initiation by Weaker Powers (Cambridge: Cambridge University Press, 1994), 147; Richard C. Thornton, The Falklands Sting: Reagan, Thatcher, and Argentina’s Bomb (Washington, DC: Brassey’s, 1998), 11.

(73.) Amy Oakes, “Diversionary War and Argentina’s Invasion of the Falkland Islands,” Security Studies 15, no. 3 (July–September 2006): 449, see also 441; Paul, Asymmetric Conflicts, 159–65.

(74.) The case is sometimes cited as evidence that the nuclear nonuse norm is sufficient to explain nonnuclear weapon state belligerency or that nuclear weapons cast little to no shadow (p.228) in international politics. For example, see Robert Farley, “The Long Shadow of the Falklands War,” National Interest, September 8, 2014, http://nationalinterest.org/feature/the-long-shadow-the-falklands-war-11224?page=show; T. V. Paul, The Tradition of Non-use of Nuclear Weapons (Stanford, CA: Stanford University Press, 2009), chap. 7; Ward Wilson, “Doubts about Nuclear Deterrence, Part III: Yom Kippur and Falkland Islands,” Arms Control Wonk, February 6, 2013, https://www.armscontrolwonk.com/archive/206263/ward-wilson-wednesdays-part-3/.My argument, by contrast, is that this is a case where one expects a limited nuclear shadow because of the conventional imbalance and conduct of the war but that one should not generalize beyond similar cases. The benefits of nuclear use for the NWS being low, the NNWS will have greater confidence that any costs associated with nuclear use will be sufficient to constrain the opponent. In other words, a limited nuclear shadow here does not imply a limited nuclear shadow when the NNWS is more conventionally capable or creates larger dangers to the NWS.

(75.) Oakes, “Diversionary War,” 456. See also Thornton, Falklands Sting, 106, 113; and Ruben O. Moro, The History of the South Atlantic Conflict: The War for the Malvinas (New York: Praeger, 1989), 2–3.

(80.) Thornton, Falklands Sting, 105–6; D. George Boyce, The Falklands War (New York: Palgrave Macmillan, 2005), 35.

(84.) Bruce W. Watson and Peter M. Dunn, eds., Military Lessons of the Falkland Islands War: Views from the United States (Boulder, CO: Westview, 1984), 138; Thornton, Falklands Sting, 69; Boyce, Falklands War, 39–40.

(87.) Memorandum for Mr. William P. Clark, National Security Council Meeting Agenda Item—the Falkland Islands Dispute, April 28, 1982, in Jason Saltoun-Ebin and Andrea Chiampan, “The Reagan Files: The Falklands Crisis,” August 5, 2011, http://www.thereaganfiles.com/820428-state-paper-for-nsc.pdf. Note too Secretary Haig’s comments at the subsequent NSC meeting on April 30: “unless Argentina softens on sovereignty, the British will go ahead and do some damage.” Earlier that month, the Joint Chiefs of Staff had highlighted British superiority in naval capabilities but, owing to geographic proximity, Argentine advantages in air capabilities. The Joint Chiefs noted that should Britain dispatch additional nuclear attack submarines it could seek to “engage the Argentine Navy to gain control of the seas off the Argentine coast.” Haig comment: Minutes of a National Security Council Meeting, April 30, 1982, FRUS 1981–1988, vol. 13, https://history.state.gov/historicaldocuments/frus1981-88v13/d195; Joint Chiefs comment: Background paper for the Chairman, Joint Chiefs of Staff, circa April 15, 1982, https://nsarchive2.gwu.edu/NSAEBB/NSAEBB374/.

(88.) Thornton, Falklands Sting, 184–85. A third Mirage was lost to friendly fire as it returned to base.

(90.) The official British history reports 321 Argentine deaths. Writing from an Argentine perspective, Ruben Moro lists “368 seamen who went down with the General Belgrano. ” See Lawrence Freedman, The Official History of the Falklands Campaign, vol. 2, War and Diplomacy (London: Routledge, 2005), 293; Moro, History of the South Atlantic Conflict, 141.

(91.) William J. Ruhe, “Submarine Lessons,” in Watson and Dunn, Military Lessons, 9; Norman Friedman, “Surface Combatant Lessons,” in Watson and Dunn, Military Lessons, 21–24.

(92.) The Sheffield would sink on May 10 while being towed outside the combat zone. Boyce, Falklands War, 110; Thornton, Falklands Sting, 207–8.

(p.229) (93.) Dates reflect the day of the main strikes, not necessarily the day the vessel sank. All dates taken from Lawrence S. Germain, appendix, in Watson and Dunn, Military Lessons, 150–67.

(96.) Quoted in Thornton, 233.

(97.) Germain, appendix, in Watson and Dunn, Military Lessons, 166–67. The International Institute for Strategic Studies reports that Argentina had 240 air force and naval combat aircraft in 1982 and only 125 in 1983. See IISS, The Military Balance (London, 1982), 92–93; IISS, Military Balance (London, 1983), 99–100.

(98.) Germain, appendix, in Watson and Dunn, Military Lessons, 166. The British lost additional aircraft to accidents.

(107.) On the INF debate and renewed antinuclear movements see, for example, Leopoldo Nuti, Frederic Bozo, Marie-Pierre Rey, and Bernd Rother, eds., The Euromissile Crisis and the End of the Cold War (Washington, DC: Woodrow Wilson Center, 2015).

(109.) Minutes of Meeting Held at 10 Downing Street, Defense and Oversea Policy Committee, April 16, 1982, United Kingdom National Archives, CAB 148/211, 33. This meeting is discussed in Sechser and Fuhrmann, Nuclear Weapons and Coercive Diplomacy, 168. Despite Thatcher’s statement, there is evidence that the British government did explore possible action against the mainland; see Hannah Kuchler, “Britain Considered Bombing Argentina,” Financial Times, December 27, 2012, https://www.ft.com/content/fdd00d54-4dcb-11e2-a0fc-00144feab49a.

(110.) Thornton, Falklands Sting, 20, also 81–82, 90, 229.

(111.) Philip Bleek codes Argentina as exploring a nuclear device from 1978 to 1990: “political authorization to explore (but not pursue) a nuclear weapons option or formal linking of atomic research to defense agencies … [or] engaged in proliferation-relevant activity but not with sufficient intensity to merit coding pursuit.” See Philip C. Bleek, “Why Do States Proliferate? Quantitative Analysis of the Exploration, Pursuit, and Acquisition of Nuclear Weapons,” in Forecasting Nuclear Proliferation in the 21st Century: The Role of Theory, vol. 1, ed. William C. Potter and Gaukhar Mukhatzhanova (Stanford, CA: Stanford University Press, 2010), 168–69. On the Argentine nuclear program and role of Brazil see T. V. Paul, Power versus Prudence: Why Nations Forgo Nuclear Weapons (Montreal: McGill–Queen’s University Press, 2000), esp. 104–5.

(112.) Warning Meeting—Nuclear Proliferation, Central Intelligence Agency, April 27, 1982, https://www.cia.gov/library/readingroom/docs/CIA-RDP83B01027R000300040028-4.pdf.

(113.) Argentina’s Nuclear Policies in Light of the Falklands Defeat, SNIE 91-2-82, September 8, 1982, Central Intelligence Agency, Digital National Security Archive, U.S. Intelligence on Weapons of Mass Destruction: From World War II to Iraq, item no. WM00268.

(114.) Julio C. Carasales, “The So-Called Proliferator That Wasn’t: The Story of Argentina’s Nuclear Policy,” Nonproliferation Review 6, no. 4 (Fall 1999): 62. See also Paul, Power versus Prudence, 105.

(p.230) (117.) Zeev Maoz, Defending the Holy Land: A Critical Analysis of Israel’s Security and Foreign Policy (Ann Arbor: University of Michigan Press, 2006), 176.

(120.) The May 2018 Correlates of War data report the start date of the Israeli-Syrian portion of the war as June 6, 1982, and the termination as June 10, 1982. Sarkees and Wayman, Resort to War, date the conflict as April 21 to September 15, 1982. On May 2000 see Maoz, Defending the Holy Land, 171.

(121.) On the internal Israeli political dynamics see Maoz, Defending the Holy Land, 174–206, and Pollack, Arabs at War, 524–29.

(122.) Pollack, Arabs at War, 531, also 529–31.

(124.) Syrian combat forces continued to operate in Lebanon until 1990, mostly against Lebanese militias but also occasionally against Israeli forces. See Pollack, Arabs at War, 548.

(125.) Pollack, 540. Clodfelter reports Syrian losses from June 6 to September 3 as 1,350 killed, 4,800 wounded, and 450 tanks. Clodfelter, Warfare and Armed Conflicts, 630.

(126.) Clodfelter, Warfare and Armed Conflicts, 603; Peter Polack, The Last Hot Battle of the Cold War: South Africa vs. Cuba in the Angolan Civil War (Philadelphia: Casemate, 2013), 17.

(127.) Peter Liberman, “The Rise and Fall of the South African Bomb,” International Security 26, no. 2 (Fall 2001): 60.

(128.) Edward George, The Cuban Intervention in Angola, 1965–1991: From Che Guevara to Cuito Cuanavale (London: Frank Cass, 2005): 202–3, 243–46, 249, 254–55; Clodfelter, Warfare and Armed Conflicts, 604; Polack, Last Hot Battle, 184. COW May 2018 Directed Dyad List of Wars.

(132.) George A. Crocker, High Noon in Southern Africa: Making Peace in a Rough Neighborhood (New York: W. W. Norton, 1992), 360. See also the accounts in George, Cuban Intervention in Angola, 202–10; Stephen L. Weigert, Angola: A Modern Military History (New York: Palgrave Macmillan, 2011): 86–89; and Polack, Last Hot Battle.

(139.) Waldo Stumpf, “South Africa’s Nuclear Weapons Program: From Deterrence to Dismantlement,” Arms Control Today 26, no. 10 (December 1995 / January 1996): 5.

(141.) Liberman, 57–58. Most sources date the test in 1987, although it is unclear if this was prior to the August South African decision to launch Operation Moduler in response to the Angolan offensive. Other sources date the activity to June–October 1988 or 1986. See Liberman, 58n43; Narang, Nuclear Strategy in the Modern Era, 213; David Albright, “South Africa and the Affordable Bomb,” Bulletin of the Atomic Scientists 50, no. 4 (July/August 1994): 45; Stumpf, “South Africa’s Nuclear Weapons Program,” 6.

(145.) Phil Haun, Coercion, Survival, and War: Why Weak States Resist the United States (Stanford, CA: Stanford University Press, 2015), 115.

(148.) Andrew L. Stigler, “A Clear Victory for Air Power: NATO’s Empty Threat to Invade Kosovo,” International Security 27, no. 3 (Winter 2002/03): 124–57; Daniel R. Lake, “The Limits (p.231) of Coercive Airpower: NATO’s ‘Victory’ in Kosovo Revisited,” International Security 34, no. 1 (Summer 2009): 83–112; Haun, Coercion, Survival, and War, 130–31.

(149.) Sarkees and Wayman, Resort to War. Serbia reported 642 soldiers and 114 police killed from NATO action. See Clodfelter, Warfare and Armed Conflict, 583.

(152.) Sarkees and Wayman, Resort to War, 155–56. Reiter, Stam, and Horowitz also code October 7 and December 22 as the start and end date for the international phase of the conflict. The May 2018 updated COW data alter the dates to September 15 and November 15, 2001. It is unclear why this change was made.

(153.) Stephen D. Biddle, “Allies, Airpower, and Modern Warfare: The Afghan Model in Afghanistan and Iraq,” International Security 30, no. 3 (Winter 2005/6): 161–76.

(155.) Clodfelter, Warfare and Armed Conflicts, 767. COW estimates four thousand Afghani battlefield deaths. (p.232)