Huntington (1957), in a study based primarily on the history of the military in Western societies), elaborated what was widely accepted as the liberal democratic model of civil-military interaction. ‘[T]he principal responsibility of the military officer’, Huntington said, ‘is to the state’:
Politics is beyond the scope of military competence, and the participation of military officers in politics undermines their professionalism … The military officer must remain neutral politically … The area of military science is subordinate to, and yet independent of, the area of politics … The military profession exists to serve the state … The superior political wisdom of the statesman must be accepted as a fact (Huntington 1957:16, 71, 73, 76).
The idea of the subservience of the military to civilian authority, as Grundy (1968) has pointed out, follows a tradition going back to Plato. Huntington, however, challenged the simple identification of civilian control with democratic government, and military control with absolute or totalitarian government: the military may undermine civilian control in a democracy, he argued, acquiring power by legitimate processes, and within a totalitarian system the power of the military may be reduced by such means as creating competing military or paramilitary units or by infiltrating it with ‘political commissars’. ‘Subjective civilian control’, he concluded, ‘thus is not the monopoly of any particular constitutional system’ (ibid.:82). Huntington went on to distinguish five patterns of civil-military relations, based on differing relative degrees of military/anti-military ideology, military power, and military professionalism (see ibid.: chapter 4), but as evidenced in his later study (Huntington 1968), for Huntington military ‘intervention’ represented an essential breakdown of the liberal democratic political order.
While Huntington’s concept of military professionalism has remained influential, the spate of post-independence military coups in the new states of Africa and Asia from the late 1950s prompted a more critical examination of the relation between civilian government and the military. Some commentators, indeed, suggested that the presumed neutrality and separation of the military from politics was at best a Western concept, if not a complete fiction (see, for example Perlmutter 1980:119; Valenzuela 1985:142; Ashkenazy 1994:178). Not only did military intervention sometimes occur in response to the effective breakdown of democratic civil regimes – with the ostensible aim of restoring democracy, and often with substantial popular support – but in some new states, notably the communist ‘people’s republics’ and the ‘guided democracy’ of Indonesia’s President Soekarno, an alternative model of ‘democracy’ was espoused, in which the military was seen as an integral part of the political system rather than, as in Huntington’s formulation, an agency outside the political realm.
That a variety of political regimes, in which the pattern of relations between civilian politicians and the military covers a broad spectrum, should claim to be ‘democratic’ is testimony to the popularity of the term in international political discourse. Such popularity reflects the extent to which the term acts as an agent of political legitimation in a world where democracy is accepted, at least rhetorically, as a universal ‘good’. But can military regimes ever be described as democratic? Or, indeed, are they necessarily anti-democratic? Gallie’s (1956) formulation of democracy as an ‘essentially contested concept’ lends support to a relativist position, the extension of which is that democracy can mean all things to all people. As Hewison, Robison and Rodan (1993:5) point out, this effectively denies the possibility that any universal understandings can be reached and serves to ‘indemnify the most scurrilous of dictatorships and to undermine the legitimacy of democratic and reformist oppositions’. On the other hand, too narrow a definition, especially with respect to institutional forms, is unrealistic.
One way of dealing with this definitional problem is to acknowledge that regimes measure up differently against various criteria of democracy, and that the idea of a continuum from more democratic to less democratic is the most useful and meaningful approach to the problem of analysing and comparing regimes. Diamond, Linz and Lipset (1990:6-7), for example, define democracy in terms of three essential and generally accepted conditions: meaningful competition for government office; a high level of political participation; and a level of civil and political liberties sufficient to ensure competition and participation. They recognise, at the same time, that ‘countries that broadly satisfy these criteria, nevertheless do so to different degrees’ and that the ‘boundary between democratic and undemocratic is sometimes blurred and imperfect’ (ibid.:7; see also Dahl 1989:112; Hadenius 1992; Sørensen 1993; Lawson 1993).
For military rulers, however, the widespread association of democracy with civilian supremacy has created a particular crisis of legitimacy. A central pillar of modern democratic theory is the doctrine of constitutionalism which, in its simplest form, refers to limited government, a system in which any body of rulers is as much subject to the rule of law as the body of citizens. An important corollary to the democratic doctrine of constitutionalism is civilian supremacy (though this in itself is not a sufficient condition for democracy since, as Huntington pointed out, many non-democratic governments maintain civilian control over their military and police organisations). Democracy requires, therefore, not only that armed forces be subject to civilian control, but that ‘those civilians who control the military and police must themselves be subject to the democratic process’ (Dahl 1989:245). A fundamental principle of the democratic model of civilian supremacy in civil-military relations resides in the important distinction between the state and the legitimate government. It is to the latter that the military owes its primary allegiance, and any implicit distinction that the military might be tempted to draw between the goals of the government and those of the state must provoke a serious legitimacy problem (Harries-Jenkins and van Doorn 1976); this is so because the democracy model insists that the military’s power is legitimate only in so far as it has been endorsed by society as a whole and that its practical objectives are those set for it by the government of the day. Van Gils (1971:274) states this succinctly:
Under the conditions of pluralistic democracy, the relations between the armed forces and civilians are, at least theoretically, quite straightforward. Soldiers are public officials. They are not the embodiment of any particular set of values. They are not the chosen defenders of any specific social or political institution. They hold public office on the assumption that they will provide society with a specific set of services whenever society considers itself in the need of having such services performed.
This reflects the deeply embedded assumption of modern democratic theory, that it is the popularly elected government, and no other body or person, that is wholly responsible for deciding what policies are to be pursued in the name of the people. In so doing, the government is constrained by the limits to action set out under the law of the constitution, and is ultimately held accountable for its activities and decisions when it faces the judgement of the people at the polls.
But what if a constitutionally and popularly elected civilian government once in office abrogates the constitution and rejects the democratic values embodied in it (including genuinely competitive elections)? In such circumstances – which have been not uncommon in post-colonial states – the military may be the only entity within the country capable of reversing such a development and reinstating democratic government.
While contemporary democratic theory appears to be entirely at odds with the notion that the military has any role in unilaterally acting to ‘safeguard the national interest’, the most common justification for military intervention is just this. Such appeals to the national interest have frequently been coupled with references to some perceived crisis or threat involving the security of the state or serious economic or social problems. As Goodman (1990:xiii) observes for Latin America:
The frequent military ascension to power has often been motivated by a perceived need to save their nations from weak, corrupt, and undisciplined civilian leadership.
Numerous commentators on the role of the military in politics have observed the tendency of armed forces to justify their intervention in terms of the national interest, and thereby to identify themselves with the desiderata of nationhood. Most have been sceptical. Lissak (1976:20), for example, notes that the military can acquire a self image as guarantor of the fundamental and permanent interests of the nation, thereby arrogating to itself the requisite legitimacy to assume the right to rule. Similarly, Nordlinger (1970:1137-8) highlights the manner in which the military’s corporate interests can be defined, legitimised, and rationalised by a close identification with the interests of the nation, while at the same time portraying oppositional protests to their actions as ‘expressions of partial and selfish interests’.
Nevertheless, authoritarian rule is not exclusive to military regimes and, as the case studies in this volume illustrate, armed forces have played a role in pro-democracy regime transitions (see also Chazan et al. 1988; Goodman 1990; Rial 1990a). The critical factor for most commentators on civil-military relations concerns the intention of military rulers to return to the barracks.
To legitimise their intervention, military regimes commonly contend that their rule is only a preparatory or transitory (but entirely necessary) stage along the road to a fully democratic political system, and promise an early return to civilian rule, thereby recognising, Dahl (1989:2) argues, that ‘an indispensable ingredient for their legitimacy is a dash or two of the language of democracy’. In some cases, military rule has been justified ‘as necessary for the regeneration of the polity to allow for stable and effective rule’; military regimes have even portrayed their role as that of ‘democratic tutor’ (Huntington 1968; Nordlinger 1977:204-5). Yet once out of the barracks military rulers have seldom been anxious to relinquish power and even where there have been transitions back to civilian rule the armed forces have typically retained an involvement in politics and have been more likely to intervene again if dissatisfied with the performance of civilian governments.
Observing processes of transition from authoritarian military rule to democracy in Latin America, Goodman (1990:xiv) comments that, ‘successful transitions have utilised a process of incremental rather than immediate civilian control’; he goes on to suggest:
For democracy to take root in Latin America, both military men and civilian leaders must take on new roles…. Recognition that the military is one of the strongest formal institutions in societies that are in dire need of political and social coherence poses challenges to Latin American civilian leaders that are very different from those confronted by their developed-nation counterparts (ibid.:xiv; see also Stepan 1988; Rial 1990a, b and Varas 1990).
Goodman, however, is not explicit on the nature of these ‘new roles’, and other contributors to the same volume suggest that recently democratised regimes in Latin America remain vulnerable to ‘the rapid rebirth of military authoritarianism’ (Rial 1990b:289).
In Asia and the Pacific armed forces have played a role in both democratising and anti-democratic transitions, and though, as elsewhere, their tendency as rulers has been towards authoritarianism, patterns of civil-military relations and degrees of authoritarianism/democracy in governance have varied widely. Any attempt at understanding this variety must begin with an appreciation of the particular historical and cultural circumstances under which military involvement in politics has developed in different countries.
 In context, Huntington appears to equate ‘state’ with ‘government’; the significance of distinguishing ‘state’ from ‘government’ is discussed below.
 Also note von Clausewitz (1832/1968:405): ‘… subordination of the military point of view to the political is … the only thing which is possible’.
 For a recent statement of this theme, drawing primarily on US experience, see Johansen (1992).
 See, for example, Albright’s (1980) critique of Huntington’s ‘conceptual framework’ on the basis of the experiences of sixteen communist states. On civil-military relations in communist states, also see Perlmutter (1982) and Herspring and Volgyes (1978).