In this chapter the author attempts to relate the concepts of weak state, strong-weak society, and state capability (roughly equated with ‘governance’) in two democratic countries: the Philippines (weak state and strong society) and Papua New Guinea (weak state and weak society).
In the process he compares state and society in the Philippines and Papua New Guinea; and looks at ongoing attempts at reforming the state in the two countries. He concludes that hostility to state strengthening in the Philippines and its acceptance in Papua New Guinea may be explained by perceptions of greater representativeness and state legitimacy in Papua New Guinea, notwithstanding lower levels of compliance.
For some time students of comparative politics, as well as aid donors and international agencies such as the World Bank, have been concerned with the poor performance of many states in Africa, Asia, the Pacific, Latin America, and more recently Eastern Europe and Central Asia. This has been reflected in the proliferation of such terms as ‘weak states’, ‘collapsed states’, and (going back some years) ‘broken-backed states’. Since the appearance of Joel Migdal’s Strong Societies and Weak States (1988) the ideas of weak states and strong (and weak) societies have been employed frequently in analyses of Southeast Asian and island Pacific polities, where state–society relations have long been a focus of scholarly and policy-oriented attention.
This chapter attempts to relate the concepts of weak state, strong–weak society, and state capability (which might be roughly equated with ‘governance’) in two countries of the region, both democratic, one with a weak state and strong society (the Philippines) and one with a weak state and weak society (Papua New Guinea). Essentially, it asks how far similarities and differences in the configuration of states and societies can explain similarities and differences in state capabilities. It finds – employing Migdal’s (1988:32–3) indicators of social control – that in both countries compliance is weak and participation is in some sense strong, but, paradoxically, that legitimation appears to be higher in Papua New Guinea than in the Philippines. This finding is related to questions of who occupies the institutions of the state and how representative the interests of those occupants are; it is suggested that the apparently higher level of legitimation in Papua New Guinea might be explained in terms of its greater ‘representativeness’, given the absence of major class or ethnic cleavages and the frequent turnover of state office holders. While at any point in time a substantial section of the population may be indifferent to the state, seeing it as largely irrelevant to their lives, over time virtually all of the population feel they have a stake in the state, as an institution which delivers public goods and services. This appears to be consistent with the trend of recent constitutional reforms and proposals in the two countries, which have tended to be state-limiting in the Philippines and state-strengthening in Papua New Guinea.
At first glance, the Philippines and Papua New Guinea are not obviously comparable. The Philippines, with a well-educated and internationally mobile population of around 70 million, has experienced over four centuries of colonial rule, a relatively long period of integration into the world economy, and half a century of independence, and now sees itself as an ‘Asian Tiger’ cub. Papua New Guinea, on the other hand, once referred to as ‘the last unknown’ (and more recently described somewhat cryptically by its national airline as ‘like no place you’ve never been’) had a comparatively brief contact with the outside world before achieving independence in 1975, and its roughly four million people, who speak some 800 different languages, are still predominantly subsistence farmers, with comparatively low levels of formal education and technical skills.
Notwithstanding these differences, in political terms the two countries share a number of more or less common features.
Both enjoyed a relatively smooth and amicable transi-tion to independence, without, therefore, creating the sorts of ‘heroic’ leaders, parties or armed forces that have emerged in countries which had to fight for their independence.
Although the Philippines inherited, from the US colonial regime, a presidential-style system while Papua New Guinea chose (largely from a reading of Westminster-style African models) a parliamentary system, unusually for the Asia–Pacific area both countries – with the comparatively brief exception of the Philippines under President Marcos (from martial law in 1972 till his downfall in 1986) – have maintained robustly democratic systems.
With the same (but qualified) exception for the Philippines under Marcos, both countries:
have competitive political party systems, though parties have tended to be fluid, for the most part not sharply ideologically differentiated, and often rooted in the politics of personality or region (Wurfel 1988:93–106; Landé 1965, 1996; Villanueva 1996; May 1984 [see chapter 5 above]),
have held regular and (albeit with some manipulation) genuinely contested elections,
have experienced regular and constitutionally mandated changes of leadership,
have maintained quite strong traditions of judicial inde-pendence and – at least in rhetoric – popular support for the rule of law, and
have enjoyed a free and active press.
While both countries – particularly the Philippines, but to a minor extent also Papua New Guinea – have seen the military exercise political influence, both have substantially upheld a tradition of Huntingtonian military professionalism (Huntingdon 1957) and neither has succumbed to a military coup (though some people interpreted the 1986 ‘People Power Revolution’ in the Philippines as a military coup to restore democracy – incorrectly in my view).
Both have embraced political decentralisation, with limited success (though Papua New Guinea is currently in the process of reversing this – ostensibly in the name of further decentralisation) (Tapales 1993; May and Regan with Ley 1997).
Both countries have maintained relatively open, capitalist economies, in which government has played an important developmental role (Golay 1961; Doronila 1992; de Dios and Fabella 1996; Connell 1997; May 1997b).
Both countries have experienced armed regional separatist rebellions (the Philippines in the Muslim south and to a lesser extent in the northern Cordilleras; Papua New Guinea on Bougainville), which they have attempted to deal with by a mixture of military and political means (Che Man 1990; May 1990; Wesley-Smith 1992; May 1996).
If this list of similarities does not impress, then one might consider the comparison of the Philippines and Papua New Guinea against, say, that of the Philippines with any of its Association of Southeast Asian Nations neighbours and that of Papua New Guinea with such other South Pacific Forum member states as Fiji, Samoa or Tonga.
With so many apparent similarities in what might be termed ‘political performance indicators’, it would seem reasonable to expect to find some parallels in political institutions and/or ‘political culture’. To the extent that these can be measured, to what extent is this so?
There seems to be general consensus that the Philippines state, in terms of its capacity ‘to penetrate society, regulate social relationships, extract resources, and appropriate or use resources in determined ways’ (Migdal 1988:4), is relatively weak. Its weakness may be traced to the country’s colonial history. ‘The state’ in the Spanish colonial period was a loose structure, essentially Manila-centred and substantially reliant on the clergy and private enterprise to sustain the colonial enterprise beyond Manila. Apart from the Muslim sultanates in Mindanao and Sulu, and the few bastions of Muslim influence in the Visayas and Luzon which were quickly overthrown, pre-colonial society was politically fragmented into largely autonomous barangays (hamlets). Spanish colonialism effectively created a sense of Filipino identity, promoting the growth of prominent local families, in part through intermarriage and the integration of immigrant Chinese. By the end of the Spanish period there was a well-developed, locally based, ilustrado (élite) class. During the American colonial period this class consolidated its position through political leadership in the democratic system promoted by the Americans (the Philippine literature commonly uses the term ‘élite democracy’) and economic dominance based on land ownership and the translation of this into equity in a burgeoning industrial sector.
In the post-independence period, social tensions, particularly between import-substitution manufacturing interests and those supporting an export-oriented development strategy, not only hampered coherent policy making but created the political instability which facilitated the election of Ferdinand Marcos in 1965. However, the increasingly authoritarian regime of President Marcos did little to improve the capability of the state to achieve its leader’s objectives (except perhaps the objective of enriching the Marcos family); indeed, it might be argued that Marcos became entrapped in a kleptocracy of cronies in much the same way as previous governments had been constrained by self-seeking élite competition. The People Power Revolution of 1986 only partly changed this, the Aquino regime being heavily dependent on winning the support of local élites just as Marcos had been (indeed in some places the local élite simply switched allegiance from Marcos to Aquino). Only under President Ramos (1992–98) does there appear to have been some strengthening of the state (see Miranda 1993).
What is interesting about the Philippines, however, is that even under Marcos, the institutions of the state remained largely intact: apart from the early years of martial law, elections were held (albeit being heavily manipulated), parliament met (though it was dominated by Marcos loyalists and much policy making was done by presidential decree), and the judiciary maintained a measure of independence. Moreover, political parties operated (albeit under various constraints), and an active press and radio were tolerated. Thus, when the challenge to Marcos came in 1986 (significantly, as the result of an election), the opposition had leadership, organisation, and the means of mobilising popular support; in the event, it was also able to call on a degree of Huntingtonian military professionalism. (The contrast with what happened in Burma two years later is striking.)
Pre-colonial Papua New Guinea was also intensely fragmented into small-scale political entities, with little overarching organisation or hierarchical leadership. (Traditional society in Papua New Guinea has commonly been referred to as ‘stateless’ and ‘acephalous’.) The state was very much a colonial creation and for much of Papua New Guinea’s population the experience of colonial administration was brief (in some parts, less than two decades). Although the colonial administration has been described as highly centralised, district officers frequently exercised considerable autonomy. After independence in 1975, separatist movements on Bougainville and elsewhere resulted in the establishment of a system of provincial government, decentralising political power and in the process inevitably creating tensions between provincial politicians and national Members of Parliament (MPs).
A legacy of the latter years of the colonial experience has been the widespread view of ‘the state’ as a provider of goods and services. With limited mobilisation of resources outside the government sector, access to those goods and services is best achieved by occupying the institutions of the state. Politics is thus seen as a major form of bisnis, with elections eagerly contested (despite a high cost of contesting, frequently upwards of forty candidates compete, in a first-past-the-post poll, in 89 single-member open constituencies of around 25 000 to 50 000 voters), and around 50 to 55 per cent of members losing their seats at each election. Since 1972, all governments have been coalitions, and in every parliament since 1977 government has changed in mid term as a result of shifts within coalitions. The consequent concentration on factional dealing and pork barrelling has contributed significantly to the inability of the Papua New Guinea state to commit itself to policies with a longer-term perspective, and to the growth of corruption and nepotism. The state’s ability to deliver services and to maintain its presence, especially in more remote rural areas, is limited by the capacity of its bureaucracy, availability of resources, and difficulties of terrain. The outbreak of armed rebellion, beginning in 1988, in Papua New Guinea’s relatively prosperous North Solomons (Bougainville) Province, primarily over unsatisfied demands by landowners for a larger share of revenue from the huge gold and copper mine at Panguna, tested the coercive powers of the state; after almost a decade, the conflict has not been fully resolved.
If the ‘state’ is sometimes an ambiguous and contested concept, ‘society’ is analytically even more difficult. ‘Society’ is used here, loosely, to refer to what Woods (1992:77) has described as ‘patterns of political participation outside of formal state structures and one-party systems’. In this sense, there is an overlap between ‘society’ and what below I refer to as ‘regime’.
The ‘strength’ of society in the Philippines lies in two, largely opposing, directions: the existence of essentially locally based patrons, caciques, dynasties or warlords (all four terms, and others, recur in the Philippines literature) on the one hand, and on the other, the existence of a vibrant ‘civil society’ which includes a mind-boggling array of non-governmental organisations (NGOs), a strong labour movement, a powerful church (especially the Roman Catholic Church) network, a highly politicised academic community, and a strong Left incorporating worker, peasant, church and other sectoral groups and a Communist Party. The former, with some rotation of players, have over time substantially penetrated and appropriated the resources of the state; the latter (‘civil society’) has been less successful in capturing government, but has asserted itself in other ways, including ‘everyday politics’ (Kerkvliet 1990) and armed insurgency, and in these ways exercised a restraining influence on an élite-dominated state (most dramatically in the overthrow of Marcos in 1986).
In Papua New Guinea, on the other hand, ‘society’ is arguably weak: traditional social groups are small and often in conflict; the labour movement (in a country where formal sector employment is relatively small and dominated by government) is at a low level of development. NGOs are embryonic (and having been partly co-opted by government following the June 1997 elections seem less likely to act as a restraint on the state). Christian churches play an important role socially but are not particularly active politically and are divided by denomination. And political parties lack a mass base. Yet the high rate of turnover of members of parliament, and the necessity for national MPs to deliver to their local constituencies, suggest that civil society exercises some influence over the state.
In terms of Migdal’s matrix (1988:35, Table 1.1), Papua New Guinea would seem to be a clear case of weak state, weak society; but it is doubtful whether it can be described, as Migdal labels this cell, as ‘anarchical’. Indeed, in terms of the classic criteria – regularity of elections, number of changes of government, and the constitutionality of those changes – it has been a remarkably stable polity.
Having provided these thumbnail sketches of one country with a weak state and strong society and one with a weak state and weak society, where do we go from here? Can we explain similarities and differences in state capabilities in terms of these characteristics?
An obvious approach would be to examine government performance in several comparable policy areas as a measure of state capability. This might reveal, for example, that both states, for more or less similar reasons (to do with the difficulties of placing long-term development strategies above short-term gains) have been poor managers of their economies; that both, for more or less similar reasons (to do with the limited reach of the state) have had difficulties maintaining law and order, and so on. I would expect, however, that on any measure the Philippines would have a better record on policy formulation and a somewhat better record on policy compliance. But I suspect that the more detailed the examination, the more difficult it would become to measure capability, to reconcile the indicators of compliance, participation and legitimation, and to disentangle state from society.
This suspicion appears to have been borne out in several chapters in Dauvergne (1998), which have concluded from studies of particular countries that the ‘strength’ or ‘weakness’ of the state can vary both from one policy area to another and one region to another, and, within a single policy area or region, over time (see, for example, Kerkvliet 1998). If this is the case (and there is little doubt that it is), do the concepts of state and societal strength or weakness retain any analytical usefulness?
Some of us, at least, would feel that, despite the inherent problems of definition, the ideas elaborated by Migdal in 1988 have provided sufficient analytical insights that we should be wary of throwing out the conceptual baby with the proverbial bath water. But some development of the initial analytical framework is clearly needed. Two conservationist strategies might be considered.
One would involve the compilation of an index of overall ‘state strength’ (and perhaps an index of ‘societal strength’), along the lines of the Freedom House Index of democracy or the Wall Street Journal/Heritage Foundation Index of Economic Freedom, which would make it possible to locate states (and societies) along a scale of strength–weakness according to the characteristics identified by Migdal (1988) or some other basket of indicators. While such an index might find a ready market, my impression is that, like the Freedom House Index and the Index of Economic Freedom, in the process of aggregation it would conceal as much as it revealed.
A second strategy would be to pursue the difficult task of clarifying the relationship between state and society (a task which Migdal and others have already addressed in Migdal, Kohli and Shue 1994, and Migdal 1997).
Thus, to return to the Philippines–Papua New Guinea comparison: it has been argued that the Philippine state has been dominated by a landed capitalist élite (Villacorta 1994) which has largely (though not exclusively) used the state to promote its own social and economic interests; as against this, the contrasting societal forces of a strong democratic ethos, arguably grounded in the Philippines pre-hispanic culture but certainly nurtured under American colonial rule, and locally based power structures which penetrate but do not coincide with the Philippine state, have for the most part exercised fairly effective restraint over autocratic tendencies in the Philippine state. In consequence, with reference to Migdal’s indicators of state strength, policy making is vigorously debated and legislative programmes (somewhat less vigorously) enacted but implementation and compliance is frequently weak – especially (in areas such as land reform, forestry, and environmental management) when the interests of local power brokers do not coincide with nationally legislated policies. The inability of successive Philippine governments, notwithstanding pressures from the International Monetary Fund, to enact much needed tax reforms is a striking case in point.
On the other hand – and apparently contrary to Migdal’s (1988:32) notion of ‘[i]ncreasing levels of social control’ – participation, on almost any measure, is quite high, with a wide range of civil society organisations active in formal and informal politics, and public demonstrations and strikes (welga) over policy issues reasonably frequent and largely tolerated by government. Nevertheless, although elections are well contested and voter turnout is comparatively high, political patronage remains important in the continuing dominance of prominent families – for example, see Soriano (1987) on the first post-Marcos election – and there is widespread popular cynicism about politics, especially as directed towards ‘traditional politicians’ (trapos). Correspondingly, popular perceptions of the state’s legitimacy (legitimation) in the Philippines appear to be mixed. The common claim that ‘Filipinos respect a strong leader’ suggests a high degree of state legitimation, but evidence for this is weak. President Marcos held on to power, from 1972 to 1986, only through the declaration of martial law and systematic state repression in the face of growing opposition. Successive governments have faced armed insurgencies from the Communist New Peoples Army and the Muslim Moro National Liberation Front and Moro Islamic Liberation Front (and for a while also the Cordillera Peoples Liberation Army); and after the 1986 change of government President Aquino faced a series of attempted coups by rebel fac-tions of the armed forces. Moreover, strong public reaction to moves to amend the constitution to enable President Ramos to seek a second presidential term (see below) suggests that even a successful and generally popular president, who appears to have significantly raised perceptions of state legitimacy in the Philippines, exercises a limited mandate.
In Papua New Guinea, the state has been more broadly representative, and in the absence of the sort of major class or ethnic cleavages which have characterised many other societies, and with a relatively high turnover of members of parliament, it is arguable that no substantial social group (except women as a gender group) feels itself to be systematically excluded from the possibility of sharing in state power. State and society are thus more broadly integrated. Participation is, in this sense, comparatively high, and is reflected in the large number of candidates contesting elections, the high voter turnout, the demands made on members of parliament by their constituents, and ultimately, the relatively high turnover of members.
A notable feature of state–society relations in Papua New Guinea, on the other hand, is the very low level of development of formal civil–society organisations (apart from churches) between the local community and the state. The integration of state and society and the high rate of participation in politics, however, makes for the ‘pork-barrelling’ style of politics noted above. With all national governments in Papua New Guinea’s political history being coalitions – with generally weak, poorly disciplined, and not ideologically sharply differentiated parties – candidates have increasingly stood as independents (in order to maximise their bargaining position, should they be elected, in the process of government formation) and elected MPs engage in an ongoing process of bargaining to maximise the benefits from office to themselves and their line (as noted above, every Papua New Guinea parliament since independence has changed government in mid term as a result of shifts within coalitions).
As I have argued elsewhere (May 1997b) [see chapter 14 above], employing Ekeh’s (1975) notion of ‘primordial’ and ‘civic’ realms, this is not an environment favourable to sound policy making and compliance. Problems of compliance are further exacerbated by the limited capabilities of many administrators, which inhibits government service delivery, and the frequency of political interference in administration at the local level. In such a context there is a good deal of cynicism about politics and politicians, and perceptions of state legitimacy tend to be low amongst those who are not currently recipients of state benefits. But while, at any time, most people have little regard for the state, there has not been (apart from Bougainville) the sort of direct opposition posed to the state that the Philippines has faced with the Moro National Liberation Front and the New People’s Army, since those who at present have poor access to the state and its benefits might, at the next election, be the winners. Thus, paradoxically, with a weak state and a weak society Papua New Guinea has enjoyed a generally higher level of state legitimation and political stability than the Philippines.
This Philippines–Papua New Guinea comparison suggests two propositions of more general application. First, consistent with the approach of defining the state in institutional terms and seeing it, in neo-Weberian terms, as mediating conflicting claims and establishing social values, it is necessary to distinguish the state from those who, at any time, occupy or control the institutions of ‘the state’. In a research project on Regime Change and Regime Maintenance in Asia and the Pacific, we have attempted to accommodate this by distinguishing between the state (defined in institutional terms), government (those who occupy the legislative and executive offices of the state at any time), and regime (as in popular usage – for example, ‘the Marcos regime’ – those who at any time occupy the institutions of the state or exercise significant continuous political influence over it). In practice, governments may change relatively frequently (in democratic systems principally as the result of regular elections) without any change in the state; regime change, on the other hand, occurs less readily and usually involves a dramatic event, such as a military coup, revolution, or foreign invasion. Changes of regime, unlike changes of government, generally imply changes in state structures. In the Philippines, the declaration of martial law marked a regime change (the ascendance of President Marcos, backed by the military, a political machine – the Kilusang Bagong Lipunan (KBL), New Society Movement – and a ‘crony’ business élite) and a change in the state (the suspension of the constitution). The ‘People Power Revolution’ of 1986, and subsequent passage of a new constitution, brought another change of regime (and state). The Philippine state under the authoritarian rule of Marcos was very different, in its relations with society, from the essentially democratic Philippine state under Aquino or Ramos. In contrast, there have been several changes of government in Papua New Guinea, but without a fundamental change of regime the role of the state in society has been essentially unchanged (though there have undoubtedly been differences in governing style).
Second, having established who occupies the institutions of the state, it is necessary to ask how representative that group is. There is little doubt, for example, that the Philippine state under Marcos placed considerable weight on the interests of the Marcos family and its supporters (‘the Marcos regime’). While many would argue that the Philippine state under Aquino and Ramos was still dominated by the interests of a landed capitalist oligarchy, few would dispute that under Aquino and Ramos the Philippine state was representative of a much broader range of interests (though many Muslims would argue that no Philippine state has really represented their interests).
In much of the writing about the state – even when the diversity of society is recognised – there is a tendency to assume that ‘nation building’, the creation of a sense of national identity (and thus identification with the nation-state) amongst socially diverse populations, can be achieved as a by-product of ‘modernisation’ or through specific policies of national ideology. The evidence to support such a view, particularly in the Asia–Pacific region, however, is weak. Most states are systemically biased in their representativeness, whether in privileging the interests of dominant ethnic, religious, class or cultural majorities (as in Fiji’s ‘paramountcy of Fijian interests’, or Malaysia’s ‘New Economic Policy’) or in making concessions to minority groups (as in the creation of an Autonomous Region of Muslim Mindanao and the recognition of indigenous ‘ancestral domain’ in the Philippines). In many states, moreover, there are significant groups who regard themselves as permanently denied the possibility of exercising political power at the national level (the Muslims in the Philippines, and perhaps workers, peasants and indigenous cultural communities provide examples). Sometimes such groups simply disengage from the state (compare May 1982); where they are sufficiently large and can be organised politically there is always the possibility of insurrection against the state.
This has implications for the assessment of state strength or weakness: different elements of society may judge indicators of state strength differently. The mining industry in the Philippines, for example, may see delays in the passage of mining regulations as evidence of state weakness where an indigenous minority may see the state’s concern over ancestral domain (a major cause of this delay) as evidence of state strength. Similarly, a government’s performance on decentralisation may be judged to be strong or weak depending on whether one takes the perspective of a national planner or a provincial governor. Performance on land reform may be judged differently by a land owner and a peasant. Performance on export-oriented industrialisation may be judged differently by workers and the owners of capital. And, ultimately, a state may be judged ‘strong’ if it successfully counters an insurgency, regardless of the moral justification for such action. It is for this reason that democratic theory emphasises such factors as turnover and tolerance of minorities.
In this context (perceptions of whose interests the state represents), it might be instructive to briefly consider and compare recent debates in the Philippines and Papua New Guinea concerning constitutional reforms.
Following the fall of President Marcos in 1986, a Constitutional Commission was created to draft a new constitution. The new constitution was ratified in 1987. One outcome of the Marcos experience was a new provision to limit terms of public office. Under the 1935 constitution presidential office was limited to two terms. It is generally believed that it was the ineligibility of Marcos to stand for a third term in 1974 that precipitated his declaration of martial law and suspension of the constitution in 1972. A new constitution, pushed through in 1973, removed this constraint. Under the 1987 constitution a president was granted only one term (of six years). Senators are allowed two consecutive terms of six years and members of the House of Representatives three terms of three years. In 1997, with President Ramos’s term of office due to end in 1998, Ramos supporters campaigned to ‘change the charter’ to allow the president to recontest in 1998 (and to extend the allowable term of other offices). Despite Ramos’s considerable popularity, the move provoked heated opposition from across a broad political spectrum, with accusations that Ramos intended to reimpose martial law. The president’s popularity plummeted, and the move was defeated. (Earlier proposals during the Ramos presidency for a shift from a bicameral, presidential system to a unicameral, parliamentary system, ostensibly in the interests of greater efficiency and democracidity, were similarly opposed as covert moves to prolong Ramos’s term in office.) A second aspect of the 1987 constitution concerned the role of political parties. Prior to martial law the Philippines had had a broadly two-party system, though the two major parties were not sharply differentiated ideologically, were élite-dominated, and membership was fluid. (Marcos himself had stood in the presidential election of 1965 as a Nacionalista Party candidate after being denied nomination by his chosen Liberal Party.) Under martial law, parties were initially banned and then allowed to re-establish under restrictive conditions; Marcos’s KBL dominated politics from its establishment in 1978 till 1986, when it virtually collapsed. In the wake of 1986, some commentators confidently predicted the re-emergence of a competitive two-party system (see, for example, Landé 1987:31–2). Amongst Filipinos, however, there was a good deal of hostility towards ‘traditional politics’ and the party system. This was reflected in the 1987 constitution (Art. VI S.5(1), (2)) which provided that a fifth of the membership of the House of Representatives should be elected through a party-list system of registered national, regional and sectoral parties or organisations. As a transitional measure (for three consecutive terms after ratification of the constitution), however, half of the seats allocated to party-list representatives were to be filled ‘by selection or election from the labour, peasant, urban poor, indigenous cultural communities, women, youth, and such other sectors as may be provided by law, except the religious sector’. In a detailed study of the party-list system as it operated during the eighth congress (1987–92), Violeta Corral described the objectives of the party-list system in the following terms:
The party-list provision was proposed to open up the political system to a pluralistic society through a multiparty system … The party-list system … in effect, equalises political power such that the traditional two-party system in Philippine politics is dismantled … The party-list system, with its transitory provision of sectoral representation, institutionalises people power and broadens participatory democracy. [Corral 1993:6–7]
Corral’s study suggests that the provisions may not have yielded the results hoped for by their proponents – even apart from the fact that during the first two post-1987 congresses all seats were allocated to sectoral representatives, with some members representing more than one sector – but the system has been retained and the first election under the full party-list system is planned to take place in 1998. (Further evidence of hostility towards ‘traditional politics’ was the inclusion in the 1987 constitution’s ‘Declaration of Principles and State Policies’ of a statement (S.26): ‘The state shall guarantee equal access to opportunities for public service, and prohibit political dynasties as may be defined by law’; to date no such legislation has been enacted.) A third element of the post-1986 reform in the Philippines was a major effort at political decentralisation through a new Local Government Code (1991). Although progress in some areas has been slow (and notwithstanding a belated concern that decentralisation might deliver political power into the hands of local warlords and political oligarchs), decentralisation has gathered pace over recent years. A significant aspect of this decentralisation has been the formal involvement of NGOs.
In all three areas the predominant rationale for reform has been to prevent the re-emergence of a dominating central state. When in 1992 Singapore’s President Lee Kuan Yew advised Filipinos that what they needed was discipline rather than democracy, he was quickly reminded, by President Ramos, of the Philippine’s ‘ill-fated flirtation with authoritarianism’ (see Far Eastern Economic Review 10 December 1992:29).
In contrast, much recent debate in Papua New Guinea has been concerned with ways of strengthening the state. At independence Papua New Guinea adopted an essentially Westminster constitution, which included provision for removing the prime minister or other ministers by a parliamentary vote of no confidence. In 1991, following repeated votes of no confidence in successive prime ministers, the constitution was amended to give an incoming government a grace period of 18 months (previously only six months) before facing a vote of no confidence, in the interests of political stability. In contrast to the opposition to charter change in the Philippines in 1997, there was no significant opposition to this amendment. In a second significant initiative, prior to the 1987 national elections in Papua New Guinea, concern over the growing number of candidates contesting elections (and the consequent decline in the percentage of the vote gained by winning candidates) prompted a move to raise the candidate deposit from K100 (then roughly $US90) to K1000. Although this proposal (which would have given Papua New Guinea, with average per capita GNP of $US1130 in 1993, one of the highest candidate deposits in the world) was dropped, on legal advice, it was successfully revived in 1992 and retained in 1997, when there were suggestions that it be increased to K2000. (Despite this, the number of candidates has continued to rise.) A third topic of discussion has been possible ways to strengthen the party system, with the aim of promoting accountability, discipline and stability; favoured measures have been legislation to discourage ‘party hopping’ and public funding for political parties. There have also been proposals (which were to have been put into effect in 1997) to shift from a first-past-the-post electoral system to a preferential voting system. In 1997, leading opposition figure, Sir Rabbie Namaliu, left the opposition to join the government in order, he said, ‘to strengthen the government and provide good governance to the people’ (National 16 December 1997). With respect to political decentralisation, the pre-independence Constitutional Planning Committee in Papua New Guinea recommended a system of decentralisation to provincial governments. Although initially dropped from the independence constitution, provincial governments were put into place in 1976–77 following separatist demands from Bougainville. Early opposition from members of the National Parliament and local government councillors, both of whom saw the new political tier as a threat to their local power bases, resulted in several attempts to shift power back to the centre. In 1995, the Organic Law on Provincial Government was replaced by a new Organic Law on Provincial Governments and Local-Level Governments, which abolished the elected provincial assemblies. Although the rationale for the provincial government reforms was stated in terms of greater decentralisation (increasing the powers of local-level governments) the general consensus has been that the new legislation represents a recentralisation of authority to the national government. Other evidence of centralist tendencies might include the increasing use of the army in law and order operations and the frequent expression, in political circles, of admiration for models of ‘social control’ drawn from Singapore, Indonesia and Malaysia. While such centralist tendencies might be partly explained as attempts by sitting politicians to consolidate their authority and perpetuate their terms in office, the fact is that there has been little public opposition to them and, indeed, a good deal of popular support.
The general hostility towards state-strengthening reforms in the Philippines and apparent widespread acceptance of the need for state-strengthening measures in Papua New Guinea is consistent with the suggestion above that, notwithstanding a relatively low level of compliance, the state in Papua New Guinea enjoys greater legitimation than that in the Philippines, and that this may be explained by the closer integration of state and society, or greater representativeness of the state, in Papua New Guinea.
Migdal (1994:14–15) has argued for a ‘new “anthropology of the state”’. In his earlier study, Migdal (1988:xvi) characterised the literature on the Third World, critically, as generally falling into two categories: one which ‘often remain[ed] enmeshed in the intricacies of social life at the local level’, with scant attention to the state; the other which focused ‘on life amongst the most influential elements’, which by implication tended to produce overly state-centred accounts. It is doubtful whether this criticism can be sustained for either Papua New Guinea or the Philippines. Studies of politics in Papua New Guinea – where state and local politics are deeply interpenetrated – have often been criticised for being too ‘anthropological’ and insufficiently engaged with grand theory. To a lesser extent the same ‘criticism’ might be levelled at those who write about Philippine politics, in which the state and local oligarchies are often difficult to disentangle. Insofar as they have focused on the complex interaction of state and society, however, the literatures on Philippine and Papua New Guinea politics provide a fertile ground for comparative studies of ‘the state’.
 This paper was presented at a workshop at The Australian National University in 1977 and was published in P. Dauvergne (ed.), Weak and Strong States in Asia Pacific Societies (1998).
 The Philippines fought its revolution, and produced its heroes, in the nineteenth century struggle for independence from Spain, but with no great practical relevance for politics in the second half of the twentieth century (which is not to say that the country’s revolutionary heroes are not still revered).
 The latest available Freedom House index rates Papua New Guinea (on a scale of 1 [most free] to 7 [least free]) 2 for Political Rights and 4 for Civil Liberties, and the Philippines 2, 3.
 On the Philippines see Carbonell-Catilo, de Leon and Nicolas (1985) and Santiago (1991).
 See, for example, chapters by Selochan and May in May and Selochan (1998).
 ‘The state’ is used here – as in Migdal (1988:xiii fn. 2, 19–20) and in other contributions to this volume – in an essentially institutional, Weberian sense. This broadly corresponds with the usage employed by Stepan (1978:xii), where the state is seen as ‘the continuous administrative, legal, bureaucratic and coercive systems that attempt not only to structure relations between civil society and public authority in a polity but also to structure many crucial relationships within civil society as well’. For further discussion see Migdal (1994:11–18, 1988, 1997); Stepan (1978:chapter 1); Goulbourne, ed. (1979); Evans, Rueschemeyer and Skocpol, eds (1985); Hall (1986); Mann (1986); Bratton (1989); Navari (1991); and Evans (1995).
 See, for example, Villacorta (1994), Rivera (1996) and Kerkvliet (1998). For a more detailed discussion of state and society in the Philippines see Wurfel (1988:chapter 3); Hutchcroft (1991); Timberman (1991); Hawes (1992); Gutierrez, Torrente and Garcia (1992); McCoy (1994); Rivera (1994). Also see Machado (1972) and Carroll (1994).
 See, for example, Mackenzie (1987), May (1989).
 For a more detailed discussion of state–society relations in Papua New Guinea see Dinnen (1998) and Standish (1994).
 See, for example, the report to the Constitutional Planning Committee, on central–provincial government relations, by William Tordoff and Ronald Watts (Tordoff and Watts 1974:2/2).
 In a parliament of 109 seats there are also twenty provincial seats, of around 25 000 to 400 000 voters.
 For a more detailed discussion, see, for example, Bratton (1989), White (1994), Diamond (1994), Hutchful (1995-96).
 Migdal (1988:279-86) seems to have had this possibility in mind in including an appendix on (quantitatively) ‘Assessing Social Control’.
 Dickson-Waiko (1994:134) records that during 1987–91 just over five per cent of 104 bills and resolutions introduced into Congress (some, admittedly, addressing the same subject) were enacted into laws.
 A more systematic analysis of legitimacy would need to distinguish, at least, between performance-based and moral or ‘democratic’ concepts of legitimacy.
 This observation was made during the Workshop on Weak and Strong States in Melanesia and Southeast Asia (12–14 August 1997), by Michael Ong of the Australian Parliamentary Research Service. It is not obvious whether the low level of civil society development reflects a low level of (‘modern’) political development generally, the satisfactory working of electoral representation, or simply a high level of political indifference at the national level.
 Compare the comments of Kabutaulaka and Dauvergne (1997) concerning the administrative incapacity of the state in the Solomon Islands.
 See also Calvert, ed. (1987:248), Fishman (1990:428) and Lawson (1991).
 Migdal (1998), for example, speaks of the ‘naturalisation’ of the state, as a ‘shared sense that the state is as natural as the rivers and the mountains’. Similarly see Migdal (1988:16). It is not clear, however, how much ‘coherence’ states must have in order to survive; sometimes, indeed, they survive through the political disinterest of their citizens, or through the superior coercive force of the national government. Like families, states may be dysfunctional, yet survive.
 Kerkvliet (1998) recognises this in proposing that ‘perspective’ be added to Migdal’s (1988) indicators of state strength. But while compliance, participation and legitimation are, at least in principle, quantifiable, perspective is inevitably subjective. Kerkvliet (fn.22), moreover, sets aside consideration of Muslim separatism in the Philippines on the grounds that it ‘challenged the legitimacy of the very idea of the Philippine nation’, unlike the two political legitimacy crises he considers, ‘which accepted the nation but challenged the legitimacy of those governing it’.
 It is notable that the Philippines, which is generally regarded as a weak state, felt able to make constitutional provision in 1987 for autonomous regions in Muslim Mindanao and the Cordilleras, while Indonesia, generally regarded as a strong state, has refused to consider substantive autonomy to East Timor, Irian Jaya or Aceh, on the grounds that such action could precipitate a fragmentation of the Republic.
 For a summary of, and commentary upon, this debate see Bolongaita (1995).
 In June 1997, as votes were being counted in the national election, Papua New Guinea’s Electoral Commissioner Reuben Kaiulo announced that he would prepare and publish a list of all newly elected members showing their political party affiliation according to their nomination form, so that any member changing party allegiance could be identified. Kaiulo proposed to distribute this list to the chief ombudsman, the police commissioner, churches, and the press (Independent 27 June 1997). Legislation to prevent ‘political turncoatism’ has also been introduced into the Philippine Congress, but to date has not been enacted.