Photo by Al Jazeera English/Wikimedia Commons
By Anthony Bell |
It is not surprising that the recent spate of political clashes between the Pakistani civilian government and the military has generated widespread speculation among policymakers that a military coup could transpire in Islamabad. Pakistan has experienced four military coups and thirty years of military rule since the nation achieved independence in 1947. The Pakistani military is indisputably the most powerful political actor in one of the world’s most dangerous countries due to the nation’s political instability, violent internal strife, and its enduring rivalry with a more powerful India. In the civil-military literature on Pakistan, there are an abundance of explanations for the military’s political hegemony, the causes of military takeovers, and the consequences of its dysfunctional civil-military relations for democracy. Rather than examining why and how the military intervenes in politics, this paper will look at the other side of the coin: why and how the military chooses to disengage from politics.
Given the scholarly focus on military interventions in Pakistan, there has been little rigorous inquiry into why and how its four military regimes have ended, and even less investigation into the pivotal role the military has served in facilitating each transition and the implications this has for the wider body of civil-military relations theory. The central research question of this paper is: Why does the Pakistani military return to the barracks? It will explore the underlying causes that have led the military to abdicate from direct rule, arguing that withdrawals are best explained by the divergence of interests between the leaders of the military regime, who seek to preserve their grip on political power, and the active leaders of the armed forces, who seek to protect and advance the corporate interests of the military. This breach is the key motivation for the military to return to the barracks. The opportunity for withdrawal is triggered when the regime encounters political difficulties consolidating its authority, which often backfires by rejuvenating civilian opposition and requires the regime to turn to the military to enforce its will. So long as the domestic opposition to the regime does not pose a direct threat to the military’s core interests and there are viable civilian alternatives available, the military will seek to return to the barracks rather than incur the costs of repression.
This paper begins with an overview of Pakistan as a case study and a review of the civil-military relations literature related to military interventions and military withdrawals. It will then outline the civil-military scholarship related to Pakistan and provide an overview of its civil-military relations. This is followed by a short background outlining Pakistan’s political history after independence. Next, it presents a detailed examination of the conditions that have led to the end of Pakistan’s four military regimes in 1968, 1971, 1988, and 2008 and illustrates how the divergence of interests between the active leaders of the armed forces and the leaders of the military regime eventually caused the former to prefer a return to the barracks. The paper then analyze the commonalities and patterns of behavior in these cases to explain how and why this divergence of interests takes place and concludes by reviewing the causes of military withdrawal in Pakistan, providing theoretical implications, and offering paths for further research.
The Case of Pakistan’s Praetorian Military
Pakistan is an exceptionally rich case study for civil-military relations for several reasons. First, Pakistan has proven resilient in the face of the waves of democratization over the past few decades that contributed to widespread changes in civil-military relations in previously coup-prone corners of the world. Indeed, the basic question of civilian control in Pakistan remains a highly contentious political issue today. Second, Pakistan’s political system is characterized by its strong praetorian military, which has repeatedly overthrown weak civilian governments. The country has experienced four military regimes that have ruled the country for over thirty years under Ayub Khan (1958-1969), Yahya Khan (1969-1971), Zia ul-Haq (1977-1988), and Pervez Musharraf (1999-2008) (See Annex 1). Third, Pakistan is a critically important state for U.S. policymakers. Pakistan is a strategic mid-tier regional power that possesses nuclear weapons. It has served as a frontline state during the Cold War and reemerged as such in the War on Terror as a hotbed of Islamist extremist groups including al-Qaeda. Lastly, Pakistan’s turbulent relationship with its powerful nuclear-armed neighbor India has produced several major wars, countless skirmishes, and enough saber-rattling to foster a near-perpetual crisis atmosphere in the subcontinent.
While the Pakistani military is extremely politicized, it has remained a highly professional organization with a significant degree of internal discipline and cohesion. The most powerful political and military position in Pakistan belongs to the Chief of Army Staff (COAS), which is currently filled by General Raheel Sharif. The COAS heads the Army Combatant General Headquarters (GHQ) in Rawalpindi and represents the Army in the Joint Chiefs of Staff Committee (JCSC), a body charged with inter-service coordination. At the very top of the Army’s hierarchy, the COAS has a broad portfolio of military and non-military duties, the latter including managing the Army’s relations with the civilians, brokering deals amongst the political parties, and supervising the military’s policy fiefdom (especially in foreign affairs). The COAS always led Pakistan’s military takeovers and subsequently served as the head of state or returned the military to the barracks (See Annex 2). Besides coups, the military’s principal method for interfering in the civilian political process is through the Inter-Services Intelligence (ISI) directorate. The ISI exists outside of the formal military command structure but informally serves under the COAS. The ISI has a domestic political-wing that monitors and intervenes in domestic politics on the military’s behalf.
Evolution of the Theoretical Approaches: Military Interventions to Military Withdrawals
Military intervention in politics and military disengagement from politics are two sides of the same theoretical problem. These dependent variables are similarly rooted in the basic civil-military problématique: societies create militaries for protection but in doing so endow their militaries with the power to threaten their societies. The crux of the issue revolves around how civilian governments can exercise control over militaries that have the strength to usurp their authority, or conversely, how civilians can regain control after being kicked to the curb by their militaries. The majority of civil-military scholarship has focused on developing and refining one side of this problem: the causes of military intervention. The effort to locate the source of military interventions has produced a large and contentious literature which first emerged from the debate over the relationship between military professionalism and civilian control.
The coup d’état is the apex of military intervention and therefore its study became an early central tenet in civil-military relations theory. Many scholars including Huntington (1957, 1968), Finer (1962), Nordlinger (1977), Thompson (1975), Welch (1976), and Luttwak (1979)—to name only a few—have addressed the causes, typologies, and frequency of military coups. They have identified numerous and complex webs of casual factors behind military coups, which are commonly separated into opportunities and motives. Coup opportunities arise from a weakening of the civilian regime, whether it has lost legitimacy through acts of corruption, electoral fraud, or other forms of political mischiefs, or perhaps a general ineptitude at governing. Opportunities can also be fashioned from broader conditions in the country such as a floundering political process, internal unrest, external security threats, or economic crises. In terms of motivations, coups are nearly always justified by their perpetrators to have been committed on behalf of the national interest. Such claims, however, are commonly dismissed as camouflage to hide the military’s pursuit of narrower sets of interests. Some scholars contend that the military may be motivated to stage a coup to defend or advance a particular socioeconomic class or sectional group in society. Another possible motivation is the personal interests of the military leadership who seek greater political power, glory, or riches for themselves after seizing power. Many authors have recognized one of the strongest motivations for coups is to preserve the military’s corporate self-interest, which includes the preservation of the military’s autonomy, hierarchical discipline, internal unity, organizational prestige, access to resources, privileges, and other aspects. Despite the abundance of coup studies, deciphering the precise causes behind a coup is extremely difficult. Apart from the sheer array of possible causal factors, none of these explanations are mutually exclusive. Nevertheless, the traditional theoretical focus on coups has fallen by the wayside in contemporary civil-military relations due to the decline of coups in many parts of the world after the 1980s as many countries stabilized along either democratic-paths, as did much of Latin America, or along military-authoritarian paths, as did much of the Middle East.
Compared to the abundant attention theorists have given to studying interventions, the other side of the coin dealing with military withdrawals is astonishingly underdeveloped. As David Pion-Berlin observes, “Although for almost every military intervention one can find a military withdrawal, the former has been the subject of literally hundreds of major studies while the later has not.” Rather than looking at what causes a military to rush from the barracks and storm the palace, this perspective is concerned with what causes a ruling military to hand the palace back to the civilians and return to the barracks. Withdrawals, unlike coups, almost never transpire by sheer force of arms. Militaries have the ability to intervene in politics and seize power because they possess a monopoly on force in their society. Civilians do not have reciprocity to forcibly remove the military from power. In this regard, their relationship is decisively one-sided. Despite the military’s ability to seize and hold onto power, scholars have noted that military regimes are unusually short-lived forms of authoritarian government. Ruling militaries nearly always withdraw and do so with a degree of acquiescence—but usually not with much enthusiasm. Lastly, it is important to draw a distinction between what is meant in this paper by military withdrawal, disengagement, and “returns to the barracks.” These terms relate to the military’s abdication from overt rule—the self-dismantlement of the military regime—it does not necessitate the military’s full retreat from politics nor an acceptance of civilian supremacy.
There are few studies directly concerned with military withdrawal. Most of the literature has developed as secondary findings within the traditional focus on interventions and remains scattered and under-theorized. Perhaps the earliest and most important work to touch on military disengagement is S.E. Finer’s The Man on Horseback. In the course of his inquiry, Finer observed that militaries sometimes terminate their rule, restore civilians to power, and subsequently “return to the barracks.” According to Finer, a return to barracks occurs due to the combination of three conditions: “The disintegration of the original conspiratorial group, the growing divergence of interests between the junta of rulers and those military who remain as active heads of the fighting services, and the political difficulties of the regime.”
Alfred Stepan offers a useful framework for understanding the complex division of responsibilities and interests that arise in military regimes. He unpacks “military regimes” into several parts. The first component is the military-as-government, which refers to the top military leaders who directly wield the political power of the state which typically comprises the ruling general or junta and a small clique of key officers and civilians. The second component is the military-as-institution, which denotes the majority of the officer corps who for the most part remain responsible for “normal” military duties. Stepan contends that if the military-as-government precipitates a crisis or otherwise endangers the corporate interests of the military-as-institution; the latter might be willing to overthrow the former to redress the problem. Even then, the military must feel it is in its best interest to transition power to civilians rather than just reconstitute another military-as-government.
Eric Nordlinger’s study on praetorianism focuses on interventions and the political sociology of the officer corps. He briefly touches on the motivations behind disengagement, suggesting that sometimes militaries prefer the “relative political purity” of the barracks to the unpleasant governing process. Noting the military regimes are characteristically short-lived, he identifies three paths by which military regimes are supplanted by civilian governments: civilian pressure which forces the military from power (extremely rare), military countercoups which return power back to civilians (somewhat rare), and voluntary disengagements (the most common). Nordlinger identifies the military’s corporate interests as among the most important reasons behind a voluntary withdrawal.
In terms of literature primarily concerned with military withdrawals, Christopher Clapham and George Philip diagnose factionalism as a major problem for military regimes. In the long-term, political intervention is corrosive to military unity. They posit that all military regimes face a common political problem in preserving their ruling interests with their limited capacity for political organization and management. Claude Welch, exploring the conditions for successful “long-term” military disengagement from politics, contends that withdrawals are based on the military’s corporate interests and broader dynamics such as domestic conflict and the economy. Many of the works on military withdrawal dovetail nicely with the comparative politics literature on transitions from authoritarianism. Most of these approach the matter of military withdrawal from the normative standpoint of how civilians can best muzzle a politicized military and return it to the barracks. Yet as Welch points out, “No single, coherent school of thought exists regarding the causes and consequences of returns to the barracks and the pathways of liberalization and democratization.” Some scholars have poked and prodded at the issue of diverging interests in military regimes within the context of disengagement but there have been few direct inquiries and even fewer substantive case studies into its potential theoretical applications for civil-military relations.
What Causes the Pakistani Military to Return to the Barracks?
In the civil-military literature on Pakistan, there are ample explanations for the military’s political interventions and the causes and consequences of the country’s frequent interventions. Many of the independent variables scholars have offered to explain military interventions in Pakistan fail to account for military withdrawals (and why repeated interventions transpire). The first and most prominent explanation offered is the weakness of the country’s civilian political parties and democratic institutions compared to the physical and political strengths of the military. Second is that the military receives the lion’s share of the national budget in addition to the enormous business empire it has amassed. The third argument focuses on the external security environment, primarily Pakistan’s fierce rivalry with India and the existential threat Islamabad feels from its more powerful neighbor. Unfortunately, these arguments predict stability for military rule in Pakistan. They cannot explain the variation in military rule over time and why the military returned to the barracks in 1971, 1988 and 2008 or its bungled efforts to hand power to the civilians after staging a countercoup in 1969. Surveying the explanations for the instability in Pakistan’s regimes, Paul Staniland concludes, “The unanswered puzzle is why Pakistan has not become a military ruled bureaucratic-authoritarian state despite three sustained attempts to build precisely such a regime…Scholars assume that the army simply could not maintain its rule, but comparisons make it clear that military-backed authoritarianism can be enduring.”
The literature covering military withdrawal in Pakistan is quite small, paralleling the small segment the study of “returns to the barracks” occupies in the field as a whole. Mazhar Aziz uses path dependence and a historical institutionalism framework to explain how early political arrangements and military interventions compounded to shape civil-military outcomes, diagnosing that the corporate interests of the military are key. Hassan Askari Rizvi offers a comprehensive investigation into the military’s role in politics, coups, and the character and policies of the military regimes. Rizvi does discuss the causes of military withdrawals in Pakistan, but unfortunately his basic framing of the phenomena is somewhat inelegant. Moreover, a host of independent variables used to explain military withdrawals in other countries do not fit Pakistan’s often peculiar circumstances. For instance, a common cause of withdrawals is rampant factionalism within the military. Yet every coup and withdrawal in Pakistan has been led by the active head of the Army, supported by the top brass, and received cooperation down the command structure. As Staniland notes, “…the army enters into and withdraws from politics cohesively.” There has never been a serious coup attempt in Pakistan emanating from junior officers, spontaneous unit uprisings, or a non-military party. Although the military is extremely politicized, it has remained a highly professional organization with a significant degree of internal discipline and cohesion.
Accordingly, this paper utilizes the works of Finer and Stepan to disaggregate a military regime into two echelons: the regime or military-as-government led by the political heads of the Army and the military or military-as-institution led by the active leaders of the armed forces. This paper argues that withdrawals in Pakistan are best explained by the divergence of interests between the leaders of the military regime, who seek to preserve their grip on political power, and the active leaders of the armed forces, who seek to protect and advance the corporate interests of the military. The breach between the two echelons is the key motivation for the military to stage a return to the barracks. The opportunity to withdraw is triggered when the regime encounters political difficulties consolidating its authority and faces off against a revitalized civilian opposition which requires the regime to wield to the military as an instrument of repression. So long as the domestic opposition to the regime does not pose a direct threat to the military’s core interests and there are viable civilian alternatives available, the military will seek to return to the barracks rather than incur the costs of repression.
Pakistan’s Historical Context
Pakistan’s armed forces are a legacy from the British Indian Army that was inherited upon the nation’s independence in 1947. At that time, the military was instilled with a tradition of civilian supremacy from the colonial administration. Pakistan’s violent separation with India and the dispute over the Kashmir region, which led to the indecisive Indo-Pakistan War of 1947-48, served as the first test for the Pakistani Army. The early period of post-independence politics in Pakistan were dominated by the Muslim League led by Muhammad Ali Jinnah and Liaquat Ali Khan, who respectively served as Governor General and Prime Minister. The League had been a potent force within the political environment of British India because it had served as a mass platform for the Muslim minority. While committed to democracy, the League’s leaders believed they were entitled to govern the country they had forged but discovered religion was no longer a salient rallying point once the nation had separated from India. Pakistan was a fragmented nation with intense political cleavages along provincial, ethnic, sectarian, tribal, and socioeconomic lines. Furthermore, West Pakistan and East Pakistan were separated by a thousand miles of hostile Indian-territory. The country was dominated principally by Muhajirs and Punjabis of West Pakistan who filled the upper-echelons of the League, the other political parties, the civilian bureaucracy, and the officer corps. A major dilemma emerged in the late 1940s which would become the central political problem in Pakistan for several decades: the Muhajir and Punjabi politicians became reluctant to initiate democratic processes because it would guarantee an enormous permanent shift in national political power to the numerically superior Bengalis of East Pakistan. With West Pakistan certain to lose out to East Pakistan in any fair electoral contest, the country’s political development stalled. For over a decade after independence, the politicians could not reach a final agreement on a constitution and other electoral arrangements. The League fragmented into opposing factions. Instability amongst the political parties led to a rapid turnover in leadership, with seven prime ministers rising and falling between 1951 and 1957 before the first military coup occurred in 1958.
The Ayub Khan Regime: The Countercoup
After seizing power in 1958, COAS General Ayub Khan promoted himself to Field Marshal and Supreme Commander of the Armed Forces. General Muhammad Musa replaced Ayub as the COAS and was widely seen as a compliant apolitical officer. More than subsequent military rulers, Ayub kept the military at arm’s length in the day-to-day running of the country, largely relying on the bureaucracy to administrate. Nevertheless, a number of senior officers did join Ayub in the military regime, becoming ministers, governors, and wearing other political hats. Replacing the parliamentary government, Ayub introduced “Basic Democracy” and the 1962 Constitution. Basic Democracies were local apolitical party-less councils meant to bestow legitimacy on Ayub’s rule by electing him president in 1960 and 1965. Over time, he relied less and less on his staff and corps commanders for policy advice and more on civilians, like his foreign minister Zulfikir Ali Bhutto, who became one of Ayub’s closest confidants. In addition to providing the military a relatively high degree of autonomy from the military-as-government, Ayub provided the armed forces with large increases in defense expenditures and secured enormous U.S. military aid packages for modernization.
The first visible splits between Ayub and the military took shape in 1964-65. Ayub, Bhutto, and others in the military-as-government saw the window of opportunity for a favorable resolution of the Kashmir dispute was closing and began plotting a military solution, over the objections of the military-as-institution. Ayub delegated the planning for the operation to a joint civilian-military cell, which was dominated by Bhutto’s men from the Foreign Office and some intelligence officers. The influence of Bhutto and the Foreign Office on military planning undercut the military’s chain of command and threatened its autonomy and unity. The regime’s final plan was a half-baked scheme that would unfold in two phases. The first phase called for as many as 30,000 “guerrillas” to infiltrate across the Line of Control (LOC) into Indian-controlled Kashmir and instigate an uprising. In the second phase, an Army division would swiftly cross the LOC and seize strategic areas before the Indians could react. COAS Musa Khan and other senior Army officers raised objections with Ayub over the plausibility of the plan’s success and the risk high of escalation, which Ayub dismissed. The Army was confident that if India felt it was losing the fight in the Vale of Kashmir, it would launch a massive counterattack across the international border in the Punjab, a fight the Army knew it was unlikely to win. The COAS and the GHQ, as one general later wrote, viewed the entire ordeal as “…a bastard child, born of the liaison between the Foreign Minister [Bhutto] and General Malik [Division Commander in Pakistani Kashmir].” Accordingly, the military high command shirked in supporting planning for the war. The GHQ delegated responsibility for the Kashmir operation’s planning and execution solely to Malik’s divisional headquarters and made no effort to notify its combat commanders or otherwise prepare the Army and the other services for war.
The Second Kashmir War in 1965 ended with Pakistan’s acrimonious defeat. After Pakistan achieved some success in Kashmir, India launched an armored counterattack into the Punjab and caught Islamabad’s forces there entirely unprepared. The military-as-institution became disenchanted with Ayub and Bhutto’s mishandling of the war and the ceasefire agreement that resulted in no change of territory. Senior combat commanders and mid-level officers were extremely dissatisfied with the poor generalship displayed by the GHQ. Musa Khan was replaced as COAS by General Yahya Khan, one of Ayub’s protégées. The breach between Ayub’s regime and the military rapidly diverged after the war and was soon exacerbated by the regime’s declining political support and the rise of popular political parties. After the war, Bhutto split with Ayub and departed from the regime. He formed the Pakistani People’s Party (PPP) in 1967 on a populist, socialist, and anti-Ayub platform which gained a large following among the Punjabi majority in West Pakistan. In East Pakistan, the Awami League (AL), a Bengali nationalist mass political party with a separatist agenda, was also on the rise and further threatened the military regime’s legitimacy.
The shell game of Ayub’s Basic Democracy soon collapsed upon the emergence of popular opposition parties demanding a return to parliamentary democracy. The heightened agitation by civilians, however, was not the decisive blow to the Ayub regime. As Stepan points out, “the loss of civilian support must somehow be transformed into a tangible cost or a direct threat to the military-as-institution.” The political opposition increasingly fanned civil unrest, which led Ayub to seek to ratchet up repression by imposing martial law and calling out the Army in early 1969. For the military, violent suppression of the mass protests, especially in the Punjab where it recruited the majority of its officers and enlisted personnel, came at a high risk to its corporate interests by threatening its unity and prestige. Given the massive breach between interests of the military and those of the regime, the military no longer saw its best interests vested in the perpetuation of the Ayub regime and became willing to sacrifice the military-as-government to quell the unrest.P58FP Army Chief Yahya Khan, with the consensus of the Army brass, refused to enforce martial law on behalf of Ayub and in effect staged a bloodless countercoup. Having lost the support of the active leaders of the Army, Ayub subsequently resigned his office and handed power to Yahya.
The Yahya Khan Regime: The Generals’ Revolt
General Yahya Khan assumed the presidency from Ayub in March 1969. He formed an entirely new military-as-government that included far more dual-hatted officers than under Ayub. Yahya promoted his close associate General Abdul Hamid Khan as COAS. Yahya moved quickly to address many of the underlying political issues that had plagued Pakistan’s domestic political order since independence in order to extradite the military from direct power. He abrogated Ayub’s 1962 Constitution and announced free elections would be held in late 1970 for a National Assembly, which would draft a new constitution and guarantee East Pakistan proportional representation with West Pakistan for the first time. The regime and the military felt there was some risk to the military’s long-term corporate interests by returning the civilians to political power, and especially enfranchising the Bengalis. Yet these fears were placated by intelligence reports throughout the campaign that showed a roughly even split among the various opposition parties in each wing of country. The regime believed it could act as an arbiter among the divided parties in a deadlocked legislature to protect the military’s long-term interests.
The military regime’s plan to withdraw was entirely derailed by the election results. The Awami League attained an outright majority in the legislature, sweeping all but two of the seats in East Pakistan. Bhutto’s PPP won roughly two-thirds of the seats in West Pakistan. After the elections, the AL issued six hardline demands on regional autonomy including East Pakistan’s right to form its own military. Bhutto refused to form a government with the AL. In March 1971, after negotiations between the parties and the regime failed, Yahya declared the National Assembly would be indefinitely delayed. East Pakistan descended into open revolt and the AL soon declared the region’s independence as Bangladesh. Yahya, supported by the Army (minus the few Bengali regiments which mutinied), unleashed a brutal military crackdown in the East. The situation devolved into a full-scale civil war in which hundreds of thousands of civilians were killed by the military over the course of nine months. India invaded East Pakistan in November 1971 to support the Bangladeshi rebels in response to the massive flow of Bengali refugees into India triggered by the fighting. Pakistan retaliated by launching an invasion from West Pakistan into India in early December. Within several weeks, India successfully routed the Pakistanis along the two fronts, captured a third of the Pakistani Army that was trapped in the East, and forced Yahya to capitulate.
Days after the dismemberment of Pakistan and the catastrophic military defeat to India, Yahya and his cadre indicated that the regime would remain in power. At that point, many of the Army formation commanders and senior headquarters officers stepped in. The Yahya regime had been a complete disaster not only for the country, but for the military as well, as morale and prestige hit rock-bottom and unity frayed. Acting on the demands of many lower commanders, Chief of General Staff (CGS) Lt. Gen. Gul Hassan, who occupied the tier directly under the COAS, sent an explicit threat to the regime: unless Yahya and the military-as-government resigned en masse the next day, tanks would roll into the capital and unseat them. COAS Gen. Hamid Khan, who was seen by his subordinates as a Yahya lackey, assembled the senior officer corps to see if he would be an acceptable replacement—an idea floated by several in the regime—but he was rejected. Yahya, the military-as-government, and Gen. Hamid were forced to resign in late December. Once the top echelon of the regime was removed, Gen. Hassan was appointed acting COAS and the Army quickly arranged for Bhutto and the PPP to assume the reins of power and returned the military to the barracks.
The Zia Regime: The Fortuitous Plane Crash
General Zia-ul-Haq and the Army seized power in 1977 after Bhutto rigged parliamentary elections and attempted to have the military impose martial law to quash opposition protests. Early in his regime, Zia made considerable efforts to make the military-as-institution a stakeholder in his regime. Zia had not truly established himself as the head of the Army before the coup, and in the first few years of his rule was respectful of the senior officer corps and solicited their policy views. As Rizvi claims, “The key to the invulnerability of Zia-ul-Haq was the support he enjoyed from the senior Army commanders.” Furthermore, plenty of incentives were given to the military in terms of increased defense expenditures and personnel pay, as well as huge tracts of agricultural land, residences, bank loans, and other side perks.
The military favored free elections and a return to democracy in the early 1980s. Nevertheless, it was content to go along with Zia’s efforts to legitimize his government so long as it did not threaten the military’s self-interests. A referendum was held in 1984 that gave Zia a five-year term as president. Like Ayub, Zia oversaw general elections for a sham non-party legislature the following year and handpicked a civilian politician, Muhammad Khan Junejo, as his Prime Minister, after which martial law was finally lifted. Zia had miscalculated Junejo and the legislature, which proved to be more dynamic than anticipated and unwilling to act as the rubber stamp Zia wished. The relationship between Zia and Junejo caused rising political divisions within the military-as-government while revived opposition parties such as the PPP began clamoring for the end of military rule and a return to parliamentary democracy. The military’s corporate interests became tangled in the Zia-Junejo feuding, as they quarreled over cabinet appointments, military spending and perks, promotions for the top-brass, and Zia’s dual-hatted role as he clung to the ceremonial title of Chief of Army Staff (whose actual responsibilities rested with the VCOAS).
The military not only resented Junejo’s forays into its autonomy, but also Zia’s meddling in its internal affairs and his inability to protect the military’s corporate interests from his own faux civilian government. The period of collaboration between the military-as-government and the military-as-institution was at an end. As Zia civilianized his rule, he distanced himself from the military-as-institution. Where once Zia had shrewdly dabbled in the promotions and assignments of the top brass, by the late 1980s his interference in the Army’s hierarchy had evolved into blatant nepotism, a disregard for the rigid promotion system, and an undermining of the chain-of-command. Zia surrounded himself with civilians and loyal officers. He even wed his children with some of those senior officers to build kinship alliances. The officers that Zia did not trust were denied promotions to key postings or subsumed into the ever expanding number of active and retired officers running the civilian bureaucracies where they could amass fortunes for themselves but held no command. Zia rotated three VCOAS during his tenure, ending with Gen. Mirza Aslam Beg, who was appointed in 1987. The turnover of personnel at the top and the passage of time required Zia to promote relatively junior officers to senior posts. He also stopped seeking political input from senior officers. Zia’s meetings with the corps commanders and top-brass became more and more infrequent and devoid of discussion and debate. Zia undercut Beg and GHQ’s command, preferring to meet with the younger corps commanders and even lower officers in private where he would dole out patronage.
The political tensions within Zia’s regime, and between the regime and the military, came to a head in the summer of 1988. In April, an ISI munitions dump used to arm the Afghan mujahedeen exploded in the heavily populated military capital of Rawalpindi, killing hundreds of civilians. To stem public outrage, Junejo demanded the resignation of the head of the ISI and the directorate’s former head then serving as CJCS, perhaps the two generals with whom Zia was the closest. In May, Zia responded by dismissing Junejo and dissolving the legislature. He did so without consulting with the VCOAS or corps commanders, and then ordered the weary Army to seize key buildings and arrest the civilian leaders of Zia’s own government. Despite the political setbacks and rising unrest, Zia began making plans to reconstitute his government with new elections and stay in power. In August 1988, Zia, along with most of his closest military aides and the U.S. Ambassador to Islamabad, were killed when their C-130 crashed shortly after departing from a tank demonstration where much of the Army high command was in attendance. Upon Zia’s death, the leadership of the country passed to VCOAS General Beg. Within hours of Zia’s death, Beg, the corps commanders, and the GHQ staff reached a unanimous decision to return to the barracks. A civilian caretaker government was appointed to oversee free elections held in November 1988 in which the PPP under Benazir Bhutto, ZA Bhutto’s daughter, regained power.
The Musharraf Regime: Pulling the Plug
Pakistan’s fourth military regime occurred under COAS General Pervez Musharraf, who seized power in October 1999 after the civilian government under Nawaz Sharif attempted to fire Musharraf amid rising civil-military tension over Sharif’s peace overtures to India and the military’s subsequent provocation and defeat in the Kargil War. In its formative years, the Musharraf regime’s relationship with the military preceded much like the early years of the Zia regime. Defense spending increased, as did side benefits for the officer corps. Musharraf was deferential to his corps commanders and mindful to safeguard the military’s corporate interests from his political endeavors. Musharraf anointed himself president and held faux elections for parliament with his own political party, the PML-Q, while he crushed the PPP and the offshoot of the Pakistani Muslim League loyal to Nawaz (PML-N).
By late 2006, however, Pakistan was besieged by surging militant violence pouring out of the Afghan border area where the Army was waging a war against rural tribes aligned with al-Qaeda and the Pakistani Taliban. Amid the backdrop of the escalating extremist violence, the regime began to unravel in 2007 in the run up to Musharraf’s presidential reelection and parliamentary elections as the military withdrew its support for the government. The judges Musharraf handpicked for the Supreme Court started to challenge the legality of the military regime, mainly his unconstitutional dual-hatted role as President and COAS. In response, Musharraf suspended the highly-respected Supreme Court in March, igniting riots across the country that lasted until his successful reelection in October. As protests demanding a return to democracy gripped the major cities, Musharraf acquiesced to the domestic and U.S. pressure to allow the PPP and PML-N to contest the next parliamentary elections in early 2008.
After the Supreme Court threatened to invalidate his reelection due to his position as COAS, Musharraf declared a national emergency, abrogated the constitution, suspended the court again, and ordered the Army into the streets to quell the unrest. Divisions between Musharraf and the military rapidly grew during the two-month emergency and the subsequent run-up to the parliamentary elections. The senior officers were likely extraordinarily unhappy with Musharraf’s decision to have the Army forcibly step in to protect his political fortunes. Several weeks into the emergency, Musharraf stepped down as COAS and retired from the Army—the first of Pakistan’s military rulers to relinquish the title and leave the military while in office—and was succeeded by General Ashfaq Parvez Kayani. In concurrence with the corps commanders, Kayani sought to refurbish the military’s unity and prestige, and he signaled that the Army would no longer jeopardize its corporate interests for Musharraf, who would have to survive politically on his own laurels. Kayani ordered the hundreds of active duty officers assigned to civilian positions back into military service. To prevent Musharraf from politicking to his commanders, Kayani placed a ban on officers taking meetings with Musharraf or partaking in any political activity. The next month, hundreds of retired senior Army officers demanded Musharraf’s resignation—a clear signal he had lost the support of the military. Days before the February 2008 parliamentary elections between Musharraf’s PML-Q party and the opposition parties, Kayani ordered the Army and intelligence agencies not to interfere with or otherwise rig the polls. While Musharraf was still chief executive, his regime was on political life support and the Army was pulling the plug on the besieged military-as-government and returning itself to the barracks.
Without the military’s backing, Musharraf’s party was trounced at the ballot box by the PPP and PML-N. Suddenly, Musharraf faced near-certain impeachment from the new hostile parliament and likely criminal charges from the Supreme Court. Musharraf considered removing Kayani and dissolving the legislature, but he lacked supporters within the Army after Kayani had the handful of commanders still believed loyal to the beleaguered president reassigned. In August, Kayani reportedly informed Musharraf it was time for him to resign rather than fight the impeachment charges. In exchange the Army would secure him immunity from prosecution.P85FP With no leg to stand on, Musharraf resigned from the presidency and was replaced by President Asif Ali Zardari, Benazir Bhutto’s widower and the leader of the PPP. In this instance, the military returned itself to the barracks and was content to sit idly by and watch Musharraf’s regime crumble, confident its corporate interests were best served—at least for now—under a civilian government.
Inferences: What Causes Pakistan’s Returns to the Barracks?
The Pakistani military has played a crucial role in ending the nation’s four military regimes, and facilitating a return to the barracks and the transition to civilian government. Many scholars, however, have accepted returns to the barracks in Pakistan as an inevitability given the immediate circumstances surrounding them: a disastrous defeat on the battlefield (1971), the sudden death of the ruling general (1988), and widespread civil discontent (1969, 2008). Although these were certainly proximate causes, none of them alone were ultimately sufficient to force the military to withdraw. Rather, these factors presented the Army with opportunities to return to the barracks, and they mattered to the extent that each pushed apart or pulled together the interests of the active leaders of the armed forces and the leaders of the military regime. These events and their political contexts altered the military’s underlying preference for how best to safeguard its corporate interests from reinforcing the status quo regime to selecting an alternative regime and staging a return to the barracks. Assessing Finer’s three conditions for military withdrawals, the disintegration of the original conspirator group and the political difficulties of the regime can be seen as sub-variables of the divergence of interests between the military-as-government and the military-as-institution. Furthermore, there are several commonalities and patterns in these cases to explain how and why the divergence of interests occurs.
The Vice of Origin Problem: In the aftermath of a coup, the military regime’s authority to govern is pro tempore. There is a fusion among the military and the newly-formed regime with shared interests and common goals, which allow the two to operate in relative harmony compared to the ousted civilian government. The threat to the military’s corporate interests which triggered the coup is righted by the establishment of the military regime. In due time, the military rulers recognize they cannot rely on the military as a blunt instrument to provide legitimacy by force and expect it to repress domestic opposition in the long-term. The regime begins down a rocky path to legitimize itself through sham liberalization efforts. So long as the military’s corporate interests are secured, it lends support to the regime’s effort to legitimize itself. The rifts between the military and the regime are still young and shallow, but the regime’s political difficulties at achieving legitimacy begin to drive a wedge between the two.
Rewards and Punishment: Pakistan’s military regimes, like their civilian counterparts, have all suffered from severe control problems. Military and civilian leaders’ efforts to control the military have generally been all carrot and no stick. The regimes are quick to provide economic incentives to the military by increasing defense spending, securing foreign arms suppliers, increasing pay, and providing side benefits. Neither military nor civilian leaders have engaged in elaborate coup-proofing measures, doubtlessly because such efforts would draw the ire of the military which would face a threat to its unity and monopoly on force. While both military and civilian leaders have frequently tinkered with the promotion system and top-brass assignments to install loyalists, these efforts prove inadequate. Moreover, wholesale liquidations or mass purges of the officer corps, a common feature of military regimes elsewhere, have not occurred in Pakistan. Without many sticks, there is clearly a stark limit to how far the regime can motivate the military to do its bidding through building informal personalistic-alliances and providing economic incentives when the military is faced with a high cost to its corporate interests.
Regime Maintenance: An authoritarian military regime only seizes and remains in power by the use or threat of violence. Once its military becomes unwilling to repress on the new regime’s behalf or takes other actions towards achieving a return to the barracks, it is committing an act of shirking essentially equal to the original coup. Furthermore, by illegally seizing power when the military’s corporate interests were threatened, the regime itself has set a strong precedent for the military to do so again under such conditions. While all of Pakistan’s coups have been relatively bloodless, there is often a brief period afterward when the military is willing to crackdown for the new military regime in order to reestablish calm. Over time, however, the military becomes gradually wearier of performing widespread regime maintenance at the expense of its corporate interests. At the same time, the regime’s efforts to achieve legitimacy often begin to backfire as political opposition revives and greater repression is required. The military begins to shirk in protecting the regime from domestic political crises and officers form a preference to return to the barracks.
Conclusion and Implications
The Pakistani military is motivated to return to the barracks for the same reason it leaves the barracks: to protect its corporate interests. Theorists have generally assumed that the basic civil-military problématique ends after a coup. The military then rules as a unified actor or the military acts as a perfect agent of the military regime. Yet this is clearly not the case in Pakistan. To draw on agency theory, the military-as-government becomes the substitute principal for the civilian authorities while the military-as-institution remains the agent. Obviously, the distinction is blurry because both principal and agent are wearing uniforms and often appear to act in concert. Yet, beneath the surface, their interests are on different paths as the active leaders of the armed forces seek to protect the military’s corporate interests while the leaders of the military regime seek to protect its grasp on political power. In the long run, these preferences prove divergent. Further research is needed on military withdrawals from politics, which remains considerably underdeveloped and has fallen to the back of the research agenda in recent years. Theorists should recognize that politics and agency problems in civil-military relations pervade even under military regimes when the “civilian” principal is actually the military-as-government. Although comparative politics has recognized that schisms among the military elite as the leading cause of death for military regimes, they have not fully drawn out the civil-military dimensions. For policymakers, an enhanced understanding of Pakistan’s dysfunctional and complex civil-military relations will help craft better policy towards U.S. military aid and balancing the dealings with the current civilian government and the military.
Anthony Bell is a recent graduate of the Security Studies Program at Georgetown University’s Edmund A. Walsh School of Foreign Service. He is currently an Instructional Assistant at the George Washington University and an intern in the Office of the Secretary of Defense (Policy), Office of African Affairs.
 See Larry J. Diamond, “Is Pakistan the (Reverse) Wave of the Future?” Journal of Democracy 11, no. 3 (July 2000).
 The Army is the foremost branch within the armed forces, a primacy which in part stems from the continental nature of Pakistan’s security threats that requires a sizable ground force and reflects its role as the vessel for seizing political power. While the military services are relatively autonomous from one another, the Navy and the Air Force have always been treated as junior partners in the Army’s conduct of political and military affairs and thus junta-style governments have not been seen in Pakistan. Ayesha Siddiqa-Agha, Military Inc.: Inside Pakistan’s Military Economy (London: Pluto Press, 2007), 60-61.
 The military’s three service chiefs and the Secretary of Defense (who is almost always a three-star Army officer) are members of the JCSC, a body charged with inter-service coordination and providing military expertise to civilian leaders. The council is headed by the Chairman of the Joint Chiefs of Staff Committee (CJCSC), who also serves as the Supreme Commander of the Pakistani Defense Forces and the senior military adviser to the civilian government. While the CJCSC is technically the nation’s highest-ranking military officer, he has no command authority, and in practice is regarded as ceremonial figure and subordinate to the service chiefs. Nawaz, Crossed Swords, 20-25.
 C. Christine Fair, “Why the Pakistan Army Is Here to Stay,” International Affairs 87, no. 3 (2011): 583.
 Schofield and Zekulin, “Appraising the Threat of an Islamist Military Coup in Post-OBL Pakistan,” Defense & Security Analysis, 320.
 For instance, the ISI has financed small religious and conservative political parties ahead of elections to draw votes away from the major parties. The ISI has also been accused of paying bribes to politicians, buying votes, and other nefarious political activities. Mark J. Roberts, “Pakistan’s Inter-Services Intelligence Directorate: A State within a State?” Joint Force Quarterly¸48, no. 1 (1st Quarter 2008).
 As S.E. Finer notes, “Instead of asking why the military engage in politics, we ought surely to ask why they ever do otherwise.” S.E. Finer, The Man on Horseback: The Role of the Military in Politics (New York: Praeger, 1962), 5. Peter D. Feaver, “The Civil-Military Relations Problématique: Huntington, Janowitz, and the Question of Civilian Control,” Armed Forces & Society 23, no. 2 (Winter 1996).
 Peter D. Feaver, “Civil-Military Relations,” Annual Review of Political Science 2 (1999). See Samuel P. Huntington, The Soldier and the State (Cambridge: Harvard University Press, 1957), 84. Samuel P. Huntington, Political Order in Changing Societies (New Haven: Yale University Press, 1968), 194-195.
 Feaver, “Civil-Military Relations,” 218.
 See Ekkart Zimmermann, Political Violence, Crises, and Revolutions: Theories and Research (Boston: G.K. Hall, 1983).
 Finer, The Man on Horseback, 35.
 See Huntington, Political Order in Changing Societies, 1968; Guillermo A. O’Donnell, Philippe C. Schmitter, and Laurence Whitehead, eds., Transitions from Authoritarian Rule: Prospects for Democracy (Baltimore: Johns Hopkins University Press, 1986; José Nun, “The Middle Class Military Coup Revisited,” in Armies and Politics in Latin America, ed. Abraham F. Lowenthal and Samuel Finch (New York: Holmes & Meier, 1986).
 Bruce W. Farcau, The Coup: Tactics in the Seizure of Power (Westport, CT: Praeger, 1994), 26. See also: William R. Thompson, The Grievances of Military Coup-makers (Beverly Hills: Sage Publications, 1973); Martin C. Needler, “Military Motivations in the Seizure of Power,” Latin American Research Review 10, no. 3 (Autumn 1975).
 Feaver, “Civil-Military Relations,” 218.
 David Pion-Berlin, “Retreat to the Barracks: Recent Studies on Military Withdrawal from Power,” Journal of Interamerican Studies and World Affairs 32, no. 1 (Spring 1990): 137.
 Finer, The Man on Horseback, 191.
 Alfred C. Stepan, Rethinking Military Politics: Brazil and the Southern Cone (Princeton, NJ: Princeton University Press, 1988). Stepan’s dichotomy between the military-as-government and military-as-institution is essentially a refinement of the distinction Finer drew when he differentiated between the “junta of rulers” and the “active heads of the fighting services.”
 Stepan, Rethinking Military Politics, 30-33; 71.Although sometimes termed the “fighters,” this category can also include the military bureaucracy and the general staff. Feaver, “Civil-Military Relations,” 215.
 Stepan, Rethinking Military Politics, 76. Reinterpreting Stepan’s concepts into agency theory, it is possible to comprehend the military-as-government as the principal and the military-as-institution as the agent. Many authors erroneously conflate the military to be a perfect agent for the military regime.
 Eric A. Nordlinger, Soldiers in Politics: Military Coups and Governments (Englewood Cliffs: Prentice-Hall, 1977), 4.
 Nordlinger, Soldiers in Politics, 60.
 Alongside a lack of desire to remain in power, the difficulties of governing, restoring the military’s reputation, achievement of the intervention’s objectives, and confidence that successor governments will not violate the military’s interests. Nordlinger, Soldiers in Politics, 145-147.
 Christopher Clapham and George Philip, eds., The Political Dilemmas of Military Regimes (Worcester: Billing & Sons, 1985), 5. Maniruzzaman examined a large number of cases of military disengagement looking at the different withdrawal process and their impact on the duration of the civilian governments. Talukder Maniruzzaman, Military Withdrawal from Politics: A Comparative Study (Cambridge: Ballinger, 1987).
 Claude E. Welch, No Farewell to Arms: Military Disengagement from Politics in Africa and Latin America (Boulder: Westview Press, 1987). Pion-Berlin, “Retreat to the Barracks,” 141.
 See O’Donnell, Schmitter, and Whitehead, eds., Transitions from Authoritarian Rule: (1986); Juan J. Linz and Alfred C. Stepan, The Breakdown of Democratic Regimes (Baltimore: Johns Hopkins University Press, 1978); Barbara Geddes, “What Do We Know About Democratization after Twenty Years?” Annual Review of Political Science 2 (1999).
 Claude E. Welch., “Military Disengagement from Politics: Paradigms, Processes, or Random Events,” Armed Forces & Society 18, no. 3 (Spring 1992): 324.
 See Veena Kukreja, Civil-Military Relations in South Asia: Pakistan, Bangladesh, and India (New Delhi: Sage, 1991). Saeed Shafqat, Civil-Military Relations in Pakistan: From Zulfikar Ali Bhutto to Benazir Bhutto (Boulder: Westview, 1997). Ayesha Siddiqa-Agha, Military Inc.: Inside Pakistan’s Military Economy (London: Pluto, 2007).
 Paul Staniland, “The Poisoned Chalice: Military Culture, Contentious Politics, and Cycles of Regime Change in Pakistan.” MIT Working Paper, 2009, 5-8.
 Ayesha Jalal, The State of Martial Rule: The Origins of Pakistan’s Political Economy of Defence (Cambridge: Cambridge University Press, 1990). Ayesha Siddiqa-Agha, Military Inc.: Inside Pakistan’s Military Economy (London: Pluto Press, 2007).
 Staniland, “The Poisoned Chalice,” 5-8.
 Ibid., 8.
 Mazhar Aziz, Military Control in Pakistan: The Parallel State (London: Routledge, 2008).
 Hasan Askari Rizvi, Military, State, and Society in Pakistan (New York: St. Martin’s Press, 2000), 28-33.
Rizvi considers instances when the ruling general legally ends martial law and attempts to legitimize the military regime by taking off their uniform as genuine withdrawals rather than treating these efforts as smokescreens like most scholarly work on civil-military relations and authoritarianism. This essentially blurs the distinction on the regime type and the occurrence of military withdrawals. Indeed, the regime periods in question (Ayub 1962-1968; Zia 1985-1988) would meet Geddes’ basic definition of a military regime, which is “governed by an officer or retired officer, with the support of the military establishment and some routine mechanism for high level officers to influence policy choice and appointment.” See Barbara Geddes, Paradigms and Sand Castles: Theory Building and Research Design in Comparative Politics (Ann Arbor: University of Michigan Press, 2003), 73.
 See Staniland, “The Poisoned Chalice” 4-9.
 Ibid., 5.
 This consistency has been accomplished by an extralegal decision-making council comprised of the COAS, the GHQ staff, and the corps commanders, totaling about thirty of the most senior Army generals. These meetings, which have become institutionalized over the years, are a key mechanism for achieving a consensus within the senior officer corps on the direction of the military’s political activities, including whether or not to overthrow the civilian regime. This process is significantly more formalized and tiered than the standard model for military coups in other countries where ad hoc coalition-building process occurs among like-minded officers. Schofield and Zekulin, “Appraising the Threat of an Islamist Military Coup in Post-OBL Pakistan,” Defense & Security Analysis, 315. Bruce W. Farcau, The Coup: Tactics in the Seizure of Power (Westport: Praeger, 1994), 5.
 Shuja Nawaz, Crossed Swords: Pakistan, Its Army, and the Wars Within (Karachi: Oxford University Press, 2008), 22.
 Stephen P. Cohen, The Idea of Pakistan (Washington, DC: Brookings Institution Press, 2004), 47-48.
 Stephen P. Cohen, “Pakistan: Army, Society, and Security,” Asian Affairs 10, no. 2 (Summer 1983): 3-4.
 Ian Talbot, Pakistan: A Modern History (London: Hurst, 2009), 162.
 Akbar Zaidi, “State, Military and Social Transition: Improbable Future of Democracy in Pakistan,” Economic and Political Weekly 40, no. 49 (December 2005): 5173.
 Shafqat, Civil-Military Relations in Pakistan, 26.
 In Huntington’s chapter on praetorianism and political decay, he heaps praise on Ayub Khan: “More than any other political leader in a modernizing country after World War II, Ayub Khan came close to filling the role of a Solon or Lycurgus or “Great Legislator” on the Platonic or Rousseauian model.” Huntington, Political Order in Changing Societies, 251. Shafqat, Civil-Military Relations in Pakistan, 35.
 Stephen P. Cohen, The Pakistan Army (Berkeley: University of California Press, 1984), 127. Nawaz, Crossed Swords, 174-175. Kukreja, Civil-Military Relations in South Asia, 78-79.
 Kukreja, Civil-Military Relations in South Asia, 90-96.
 Nawaz, Crossed Swords, 195.
 Rizvi, Military, State, and Society in Pakistan, 104-105. Nawaz, Crossed Swords, 200.
 Nawaz, Crossed Swords, 230.
 Talbot, Pakistan, 177.
 Nawaz, Crossed Swords, 206.
 General Gul Hassan quoted in Nawaz, Crossed Swords, 207.
 Nawaz, Crossed Swords, 207-240
 Ibid., 239.
 Shafqat, Civil-Military Relations in Pakistan, 58-66.
 Rizvi, Military, State, and Society in Pakistan, 113-120.
 Alfred Stepan, “Paths toward Redemocratization: Theoretical and Comparative Considerations,” in Transitions from Authoritarian Rule: Prospects for Democracy, ed. Guillermo O’Donnell, Philippe C. Schmitter, and Laurence Whitehead, vol. 3 (Baltimore: Johns Hopkins University Press, 1986), 77.
 Nawaz, Crossed Swords, 242-245.
 Stepan, “Paths toward Redemocratization,” 76-77.
 Nawaz, Crossed Swords, 243. Rizvi, Military, State, and Society in Pakistan, 119-120.
 Nawaz, Crossed Swords, 251-254 and 258.
 Apart from paring down defense spending, there was unease that a national government dominated by Bengali politicians would want the West Pakistan-oriented military to provide more than a token constabulary force in East Pakistan, which had been left almost completely undefended in the 1965 Indo-Pakistan War, and would add Bengalis to the officer corps, which comprised about 55 percent of the population but only five percent of the officers. For ethnic divisions in the Army, see Cohen, The Pakistan Army, 32-54.
 Nawaz, Crossed Swords, 258-260.
 Sumit Ganguly, Conflict Unending: India-Pakistan Tensions since 1947 (New York: Columbia University Press, 2001), 57-61.
 Ganguly, Conflict Unending, 61.
 Rizvi, Military, State, and Society in Pakistan, 142.
 Cohen, The Pakistan Army, 73. Shafqat, Civil-Military Relations in Pakistan, 79.
 Rizvi, Military, State, and Society in Pakistan, 181.
 Nawaz, Crossed Swords, 366; 379.
 Nawaz, Crossed Swords, 380-382.
 Rizvi, Military, State, and Society in Pakistan, 200.
 Brian Cloughley, War, Coups, and Terror: Pakistan’s Army in Years of Turmoil (New York: Skyhorse, 2008), 48.
 Nawaz, Crossed Swords, 386.
 Nawaz, Crossed Swords, 386-387.
 Rizvi, Military, State, and Society in Pakistan, 202.
 Steve Coll, Ghost Wars, (New York: Penguin Press, 2004), 42. Nawaz lays out the mysterious circumstances surrounding the plane crash and the subsequent post-crash investigations. It is widely thought that Zia’s death was an assassination. See Nawaz, Crossed Swords, 393-405.
 General Beg had “coincidentally” that day turned down two offers from Zia to join him abroad his aircraft after the tank trials, as did a number of corps commanders, who all subsequently headed to the GHQ. Nawaz, Crossed Swords, 394.
 For further history, see Babar Sattar, “Pakistan: Return to Praetorianism,” in Coercion and Governance: The Declining Political Role of the Military in Asia, ed. Muthiah Alagappa (Stanford: Stanford University Press, 2001), 399-401.
 Cloughley, War, Coups, and Terror, 48-52.
 Ayesha Siddiqa-Agha, Military Inc.: Inside Pakistan’s Military Economy (London: Pluto Press, 2007), 99
 Ganguly, Conflict Unending, 127-129.
 The United States threatened to cut off the billions of dollars in operating funds and equipment it was providing the military. Laura King, “Quitting army a risky proposition for Musharraf,” Los Angeles Times, November 15, 2007.
 Tim Johnson, “Pakistan military retreats from Musharraf’s influence,” McClatchy, January 18, 2008.
 Carlotta Gall, “Military Retirees Demand Musharraf’s Resignation, The New York Times, February 8, 2008.
 Ahmed Rashid, Descent into Chaos: The United States and the Failure of Nation Building in Pakistan, Afghanistan, and Central Asia (New York: Viking, 2008), 390.
 “Pakistan army to ask Pervez Musharraf to resign,” Daily Telegraph, August 8, 2008.
 Rashid, Descent into Chaos, 394.
 For example, Shafqat writes, “The Pakistani experience revealed that the military withdrew from politics, not voluntarily, but under conditions of extreme distress or defeat.” Shafqat, Civil-Military Relations in Pakistan, 167.
 Alfred Stepan, “Paths toward Redemocratization: Theoretical and Comparative Considerations,” in Transitions from Authoritarian Rule: Prospects for Democracy, ed. Guillermo O’Donnell, Philippe C. Schmitter, and Laurence Whitehead, vol. 3 (Baltimore: Johns Hopkins University Press, 1986), 76-77.
 Stepan, Rethinking Military Politics, 31.
 Sattar, “Pakistan: Return to Praetorianism,” 397-399.
 See James T. Quinlivan, “Coup-proofing: Its Practice and Consequences in the Middle East,” International Security 24, no. 2 (Autumn 1999).
 For instance, Prime Minister ZA Bhutto promoted Zia ul-Haq to COAS in 1976 over more senior officers because he was perceived to be more politically pliable to Bhutto’s civilian government. Zia later led the coup against Bhutto and engineered his execution. Prime Minister Nawaz Sharif promoted Pervez Musharraf to COAS in 1998 for similar reasons, only to be overthrown and exiled by Musharraf the next year. Nawaz, Crossed Swords, 337-338, 500-501.
 See Finer, The Man on Horseback, 192-193.
 For further discussion and case study, see Terence Lee, “Military Cohesion and Regime Maintenance: Explaining the Role of the Military in 1989 China and 1998 Indonesia,” Armed Forces & Society 32, no. 1 (October 2005).
 Finer, The Man on Horseback, 193.
 One major exception to this was the military’s ruthless crackdown in East Pakistan in 1971. Yet from the perspective that the possible breakup of Pakistan was a major threat to the military’s corporate interests—defending the state’s territorial integrity being its primary mission—can at least partly explain its willingness to engage in outright butchery on behalf of the Yahya regime. Shafqat, Civil-Military Relations in Pakistan, 61.
 Feaver makes such a case, but does not link it to returns to the barracks, see Peter Feaver, Armed Servants: Agency, Oversight, and Civil-Military Relations (Cambridge, MA: Harvard University Press, 2003), 292-293.
 See Barbara Geddes, “What Do We Know About Democratization after Twenty Years?” Annual Review of Political Science 2 (1999).