It is tempting to conclude from this overview that each country’s experience is explicable in terms of its particular historical and cultural circumstances, and to proceed directly to the individual country studies.

Certainly the range of civil-military interactions seems to be greater than that among the states of Africa and Latin America, a factor which might be at least partially explained by wide variety of colonial experiences.^[14] Nevertheless, some common patterns, and some contrasting patterns, invite comparison.

Three countries – Burma, Indonesia and Pakistan – experienced fairly conventional military coups in which the army intervened after several years of fractious parliamentary politics, ostensibly to restore ‘political order’. In Burma the army reinstated civilian politics after two years but soon after again intervened and has remained in power since, becoming one of the modern world’s most durable military regimes. In both Burma and Indonesia the military had played a prominent part in the achievement of independence and soldiers had played an early role in government. In both countries, having intervened decisively, the military consolidated its position by expanding into civilian administration and business and by establishing a military-dominated political party. Both regimes have maintained strong central control, repressing opposition (especially on the ethnic peripheries), and both have had a poor record in terms of civil and political liberties.^[15]

But there the similarities end. In Indonesia at least some of the trappings of a democratic system have been largely maintained, with three effectively state-approved parties contesting elections (which have been consistently dominated by the military-backed Golkar); fairly purposeful policy making has achieved an impressive rate and reasonable distribution of economic development, and since the late 1960s a fairly high degree of political stability has been maintained. This has contributed to a degree of performance legitimacy that has enabled President Suharto to remain in power for almost thirty years, despite criticisms of what Filipinos might have labelled croneyism and frequent predictions of his regime’s imminent demise. In contrast, Burma abandoned any pretence of participatory politics after 1962 and has waged an ongoing war against non-Burman ethnic groups as well as, for some time, a communist insurrection. These factors, coupled with a record of economic performance which by 1987 had reduced Burma to one of the world’s poorest countries, and a high degree of political repression, has severely undermined the legitimacy of the regime. This culminated in the unsuccessful popular uprising of 1988, from which emerged a more repressive military regime. In both cases the lack of pronounced divisions within the military (once Burma had effectively purged the army of its non-Burman elements) has been a factor in regime maintenance, though in Burma in 1988 it looked for a while as though a people power movement along the lines of that in the Philippines two years earlier might force a regime change with military acquiescence. Explaining the differences in regime performance is more difficult, though the serious ethnic cleavages which independent Burma inherited from the colonial period probably imposed greater obstructions to national unity than Indonesia’s (not inconsiderable) ethnic diversity, and it is difficult to avoid the conclusion that Burma’s opting for virtual economic isolation largely accounted for the disastrous economic record which denied any claim the military regime might have made to legitimacy based on performance.

In Pakistan, also, a politicised military intervened ostensibly to restore political order. But after a decade as martial law administrator, General Ayub Khan became elected president and what Pakistan has seen since is an increasing interpenetration of military and civilian politicians, compounded by ethnic divisions, and a succession of regimes on both sides of a mid point on Bebler’s (1990) proposed ‘militocracy’/’civilocracy’ continuum. And there seems to be nothing to suggest that this pattern will change substantially.

In Bangladesh, on the other hand, the military initially intervened not to restore order among fractious politicians but to remove an increasingly authoritarian civilian regime. And having gained power the military proceeded to civilianise and democratise Bangladesh politics. Factions of the military again intervened, however, and though there were suggestions that Bangladesh was moving towards a fused system similar to Indonesia’s dwifungsi, opposition to the authoritarianism of the Ershad regime instead led in 1990 to a popular uprising to restore democracy (though for how long remains to be seen).

In the two South Asian cases, as also in Thailand, the military (or factions of the military) has emerged as one of several key players in a fluid political system. Having expanded its role into civil administration, business and politics, and having formed linkages with non-military players (including linkages along established ethnic/regional and class lines), the military seems likely to continue to play a role in a broadly civilian-military mixed system, the nature of the role varying over time according to the political and economic performance of the government of the day. Much the same might be said of Korea, where an initially authoritarian civilian regime was overthrown by popular uprising and the military stepped in to impose order. Since 1961 Korea has experienced a series of mixed military-civilian, civilian-military governments, alike in their tendencies towards authoritarianism, though civil society seems to have become stronger since the 1980s.

In Thailand, and perhaps Korea, there seems to be some validity in the general proposition that military intervention is less likely as societies become more complex and the middle class expands; the proposition seems less relevant to Pakistan and Bangladesh – despite the often-cited common military professionalist heritage of British colonialism.

Fiji presents another example of decisive military intervention, but in this case not so much to restore ‘political order’ – since Fiji had enjoyed a considerable period of orderly parliamentary government – as to maintain ethnic Fijian (and chiefly Fijian) dominance. Once this had been achieved, by introducing a new constitution and holding new elections which returned coup leader Rabuka as prime minister, civilian rule was restored and further military intervention seems unlikely.

The Philippines under Marcos presents one of a number of cases of an authoritarian, repressive regime (yet one which largely preserved the formal semblance of democracy – elections, parties, a legislature and judiciary, a reasonably free press) in which the military played a relatively minor role. As in Bangladesh, the military’s substantive entry into politics came in support of popular demands for the restoration of democracy. Having played a part in the removal of Marcos, elements of the military clearly saw themselves as having a continuing role in government, but notwithstanding a series of unsuccessful coup attempts the model of military professionalism was substantially maintained. Thus, what has to be explained in the Philippines – as in Papua New Guinea, where despite occasional rumours of an imminent coup military intervention has never been attempted – is why successful coups have not occurred. In both countries most of the classic preconditions and motives for coups have been present: imbalance between the military and civil political institutions and at least periods of arguable political instability, threatened corporate interests of the military, and personal ambition; factionalism within the military has also existed, though not on the same scale (and without the obvious ethnic or class divisions) that has been ex-perienced elsewhere. Both countries inherited strong traditions of professionalism, but in that they were no different from Fiji or Pakistan. An attractive line of explanation perhaps lies in the vitality of civil politics in both countries – a vitality which in the Philippines even the repressive regime of President Marcos failed to stifle – and in the sheer logistical difficulties of maintaining centralised control. But in varying degrees both these arguments might be applied to other cases (for example, Pakistan and Indonesia) in which coups have occurred.

Indeed the case studies in this volume produce little to support systematically any of the common ‘explanations’ for military intervention, although elements of all such explanations can be invoked. In explaining the individual cases, history (especially concerning the role of the military in the colonial regime and its part in a struggle for independence) is obviously important, as is ethnicity in some cases (notably Burma and Fiji) and factionalism within the military (for example, Indonesia, Bangladesh).

On the question of performance, also, generalisation is difficult. In terms of economic performance, military or military-civilian fused regimes have performed well in Korea and, to an extent, Indonesia (though perhaps not as well as non-military regimes in the region such as Singapore and Malaysia), but have performed poorly in Burma and Bangladesh (though no more poorly than the civilian administration of the Philippines under Marcos); Thailand’s record (as in many other respects) is mixed.

In terms of political performance, measured against the three criteria listed above – competition, participation, and civil and political liberties – there is stronger evidence of a military/non-military divide, but again the evidence is not clear cut. Comparing countries, Burma and to a lesser extent Indonesia have performed poorly against all three criteria, as have Thailand, Pakistan, Bangladesh and Korea under military rule. In Fiji, also, during the brief period of military rule there was a decline in political competition and a deterioration of civil and political liberties, though not to the extent experienced in the Asian states. On the other hand, the essentially civilian regime in the Philippines under Marcos also performed badly against the competition, and civil and political liberties criteria and, with some qualifications, against the participation criterion, for at least part of the period of the Marcos administration. Within the region, the civilian governments of Singapore, Malaysia and Tonga also have far from unblemished records.

As a rough comparative measure, the nine countries covered in this volume, together with nine other Asian and Pacific countries, are ranked below (Figure 1) on the basis of the 1994 Freedom House ‘Comparative Survey of Freedom’ (the two Freedom House gradings, for political rights and civil liberties, ranked on a scale (best to worst) of 1-7, have been averaged; those with a rating of 1-2.5 are categorised by Freedom House as ‘free’; those scoring 3.0-5.5 as ‘partly free’ and those above 5.5 as ‘not free’). The Freedom House ratings are not beyond question (it is not obvious, for example, why Papua New Guinea is classed as ‘partly free’, below Western Samoa and South Korea), but they are probably the most widely accepted measure available of comparative freedom, and thus of the degree of democracy (or relative ‘democracidity’). They show the two long-time military-dominated regimes of Indonesia and Burma at the bottom of the list, along with Brunei and several communist states; most of the rest (including the two states – the Philippines and Papua New Guinea – in which coups have either failed or not been attempted) are grouped around the middle of the range, with Bangladesh and Papua New Guinea performing better and Thailand and Pakistan worse – but all outranking the civilian regimes in Singapore and Malaysia. South Korea alone is listed (contentiously, perhaps) among the ‘free’ countries.

Figure 1: Freedom House, ‘Comparative Survey of Freedom, 1994’ 1.0 (Australia) 1.5 South Korea 2.0 (Western Samoa) 2.5 3.0 Bangladesh, Papua New Guinea 3.5 Philippines, Fiji 4.0 Thailand, Pakistan, (Tonga) 4.5 (Malaysia) 5.0 (Singapore) 5.5 6.0 6.5 Indonesia, (Brunei) 7.0 Burma (North Korea, PRC, Vietnam) (Source: Freedom Review 25(1) 1994:14-15).

What is more pertinent, however, is how changes in regime within a single country affect political performance. Here the evidence is less opaque, but still not unambiguous. In general, military intervention has resulted in restrictions on both competition and participation and, sometimes with a lag but usually increasingly, in limitations on civil and political liberties. The arguable exceptions are Thailand in 1932, Korea in 1960-61 and Bangladesh in 1975, where the military ostensibly intervened to restore civil and political liberties and increase competition and participation, though even among these cases (notably Korea) it may be argued that the tendency to democratisation was shortlived.

It should also be observed, however, that the impact of military intervention on different parts of the population is uneven. Typically, the impact of military intervention is heaviest on those most actively engaged in politics, and these are often (but not always) a social as well as political elite. When military intervention does something to restore ‘political order’ and promote economic development, large segments of the population may perceive themselves (as the proponents of bureaucratic authoritarianism and its variants once argued) to be better off. It is this, perhaps, that helps explain the longevity of the Suharto regime and the acceptance, by much of the population, of martial law in the Philippines in 1972. Similarly, it has been argued by some that the Fiji coups of 1987 were welcomed by most ethnic Fijians as a reassertion of the paramountcy of Fijian (over Indo-Fijian) interests (although Lawson’s analysis below suggests that this is an oversimplification). The broad question of who gains and who loses from military intervention has seldom been adequately addressed, either for the larger civil society or for those within the military itself.

Beyond these restricted comparisons, generalisations are hazardous. Nevertheless several low-level generalisations suggest themselves.

First, by virtue of their monopoly (or at least dominant control) over the means of coercion, and frequently because they are a relatively coherent organisation in a fragmented society, militaries can play a major role in bringing about changes of regime, not just in fluid political situations (such as in Burma in 1958 or Indonesia in the mid 1960s) but in fairly stable ones (Fiji in 1987 [though the 1987 coups were essentially regime maintaining], Korea in 1960-61). They may also play an important role in forestalling changes of regime (as in Burma in 1988).

Second, in ‘explaining’ military intervention, it is evident that the relative strength of civil and military institutions, larger divisions in society, corporate and factional interests of the military, personal ambitions, and external factors may all be relevant in different proportions, but none provides a reliable indicator of military intervention (as the Papua New Guinea and Philippines cases show).

Third, while a shift along the continuum from civilian to military regime is not strongly correlated either with economic performance or with the degree of democracy, there is, not surprisingly, substance to the general proposition that military regimes are oriented more towards maintaining ‘order’ – against which criterion, however, they perform variably, with Indonesia and Thailand providing polar examples of regime stability – and to maximising their corporate (or perhaps more correctly their collectively individualised) interests, than to promoting the liberal democratic values of competition, participation, civil and political liberties, and more egalitarian distribution of wealth.

Fourth, although these case studies provide varying instances of military withdrawal, the general conclusion seems to be that having once intervened military leaders are likely to seek to maintain a political role, either as guardians, with the implication that further interventions are likely, or by the interpenetration of the interests of military and civilian personnel in politics, civil administration and business. This conclusion, which is amply recognised in a growing body of literature on the morphology of civil-military regimes, suggests there is scope for further research in at least two major areas of civil-military relations. One of these concerns the role of the military in civilian administration and in the military/civilian borderland of paramilitary, internal security, and law and order type operations.^[16] The other has to do with the involvement of militaries institutionally, and soldiers individually, in business. In both these areas, the almost universal tendency towards expansion of the role of the military suggests the possibility of gradual change in regime type without major discontinuities in government.

The military seems likely to continue to play an important role in the politics of the countries of Asia and the Pacific, notwithstanding predicted tendencies towards democratisation. To comprehend that role it will be increasingly necessary to shift the focus of research from the military per se to the activities of soldiers in the complex of military-civil relations. It is towards this endeavour that our volume is directed.