Administrative Expertise. Let me start by arguing that the capacity of modern governments to govern requires industrialization, but it does not necessarily require capitalism. Although capitalism played a decisive role in the Industrial Revolution, states can manage the processes of industrial production without depending on capitalists. However, state enterprises probably cannot be managed as efficiently as private enterprises which means that government support for a market system probably assures higher levels of productivity. There is also a reciprocal relationships between governance and capitalism in the sense that (1) the survival of democracy (i.e. a vigorous constitutive system) depends on the strength of a state's civil society in which a free enterprise system seems to be a necessary component and (2) the viability of market-based industrial capitalism depends of the ability of government to provide necessary resources and controls.

An unregulated capitalist market system will, assuredly, destroy itself, but effective regulation by the state of powerful industrial corporations requires an effective and honest bureaucracy. Corrupt officials who can easily be bribed by those with wealth and influence are unable to sustain effective controls in a market economy. Moreover, when public officials become managers of state enterprises, the temptation to misuse power to accumulate wealth becomes even stronger and leads to corruption and mismanagement.

The tasks facing modern bureaucracies are more complex and difficult to handle than were the simpler burdens carried by pre-modern bureaucrats. To the degree that a state manages industrial production and marketing activities through public enterprises, the administrative problems faced by its public servants are magnified. However, liberal states that rely heavily on the private sector and market forces to sustain the economy and promote industrialization also confront many negative (para-modern) consequences of industrialization: they need to regulate the private sector in order to maintain its efficiency and fairness and to prevent monopolies from emerging, and they must also replace traditional social structures for managing such by-products of industrialism as urban poverty and crime, environmental destruction and the illnesses it produces.

To be effective, therefore, a modern bureaucracy must be both technocratic and authoritative: officials need expert knowledge and training in order to understand and cope with complex technological issues, and they must also have enough authority to act promptly and appropriately to deal with local situations and sudden crises. Unqualified officials lack the knowledge required to deal with new and complex problems, but even highly qualified bureaucrats without the authority to take appropriate action become frustrated, angry and ineffectual. However, too much bureaucratic power produces arbitrary and irresponsible action. Maladministration -- whether due to pervasive corruption and abuse of power, or incompetence, laziness, and powerlessness -- generates growing popular suspicion and distrust of government, resistance to administrative controls, the collapse of democracy and economic catastrophes. Ultimately, revolutions, ethnonational revolts and pandemic violence results. Note that the processes of industrialization that both empower and challenge modern states can also be exploited by the enemies of a state to facilitate their anti-state activities.

Good public administration, therefore, is essential both to promote industrialization and to manage the consequences of industrialization -- and it is far more important for modern societies than the equivalent processes in non-modern (agrarian and pastoral) societies. Moreover, I believe that only democratic institutions based on a popularly based polyarchy (constitutive system) can maintain the bureaucratic structures needed to organize and police an industrialized society.#11 In order to succeed, the constitutive system needs to achieve a precarious balance between having too much and too little control over the hierarchic (bureaucratic) apparatus of governance. The basic need is for a constitutive system rooted in popular sovereignty that is able to monitor and control the bureaucracy without trying to second guess every administrative decision -- bureaucrats need enough freedom of action to administer effectively provided they are also well guided by non-bureaucrats (i.e., the constitutive system) (Riggs, 1997b??).

Political Control: Alternative Models. If the most efficient way to promote industrialization and economic growth requires a bureaucracy subject to democratic control, then the most important political choices open to any country are those that affect its capacity to control its bureaucracy and to employ competent and honest public administrators. Moreover, the more industrialized a country becomes, the more tools its public officials acquire both for effective administration and for the abuse of their power, and the more necessary it is to have a constitutive system that can reliably sustain and control a powerful bureaucracy.

Political scientists and lawyers often emphasize a variety of institutional options and non-institutional contexts that affect the capacity of a state to achieve this goal. Because of the exceptional American experience, they are likely to consider presidentialism as a viable alternative to parliamentarism or even, in some cases, as preferable. I do not claim that my preference for the parliamentary option is demonstrably valid, but I do think it is plausible enough to deserve serious consideration. I shall, therefore, set forth the arguments that have lead me to conclude that, in most cases, a democratic state founded on some form of parliamentarist constitutional system is more likely than a presidentialist regime to succeed in the complementary goals of sustaining and controlling an adequately motivated and qualified bureaucracy which can cope effectively with the problems generated by industrialization.

This claim does not conflict with the proposition that non-institutional differences of culture, history, size, economic level and growth, ethnic homogeneity/heterogeneity, etc. all play important roles: they provide contexts for evaluating the institutional considerations. However, I cannot deal with these factors here and they have already received a great deal of attention. By contrast, the available institutional choices have received less attention. Moreover, since decision-makers can control constitutional and institutional factors more easily than they can manipulate ecological variables, they acquire strategic importance. To move from the level of abstract theorizing to that of empirical evidence, let us now take a look at some countries that have, historically, embraced the presidentialist and parliamentarist forms of governance.

The American Exception . An important reason for the widespread belief that presidentialist regimes can effectively manage the problems of industrialization can be attributed to the American "success story." If the U.S. economy has done so well then surely, we may conclude, presidentialism can work as well as the parliamentarism found in all other industrialized democracies. However, two lines of reasoning will question this conclusion. First, I think the U.S. economy could have been better managed had a parliamentary model been adopted in this country. Second, the American polity is truly exceptional -- virtually all presidentialist regimes are far less successful than the U.S. in coping with industrial problems. What any new state choosing a presidentialist model can expect to happen is far more likely to resemble the experience of most Latin American countries, or Nigeria, the Philippines, and, actually, the Russian Federation today.

To assess the actual performance of American bureaucrats in managing the problems of industrialism, consider first the difficulties they face because of the separation of powers: they are often uncertain whether to follow guidelines set by the President, or those created by Congressional committees -- and they always face uncertainties whenever a judge declares a law unconstitutional. Because Congress, under the separation-of- powers principle, must decide a host of questions for which floor debate in plenary session is impossible, effective decision-making power devolves to committees, sub- committees, and often, sub rosa, to appointed staff members of these committees. When top bureaucrats testify at hearings organized by a sub-committee, they act like the chief executive of a corporation meeting with its Board of Trustees, or a university president confronting his or her Board. Guidelines set by the President, members of his "cabinet", or even his executive office, appear to have only secondary relevance to the decisions taken by bureau chiefs.

Superficially, one might suppose that the authority granted to a president would assure policy coordination at the chief executive level. Traditionally, the main device for helping a president manage the bureaucracy and for integrating its activities and public policies, is known as a "cabinet". However, the actual dynamics of a presidentialist "cabinet" is quite different from that of a parliamentarist cabinet -- it is, indeed, confusing that we have only one word for two quite different institutions. I would like to refer to the group of ministers in a parliamentarist regime as a "true cabinet," whereas its counterpart in presidentialist systems might be referred to as a "so-called cabinet." Since this terminology is cumbersome, and may be criticized as pejorative, I shall simply put "cabinet" in quotations marks when referring to the so-called cabinets that exist under presidentialism.

In fact, top level coordination by a "cabinet" is virtually impossible, whereas the integration of public policies in a cabinet is almost assured. The former is composed of a coterie of political appointees, in competition with each other, whose tenure depends concurrently on the president's pleasure, and their ability to win the support of the legislators whose committees control their policy domains. This often generates conflicts of interest that make any "cabinet" more of an arena for combat than a coordinating mechanism.

By contrast, the constraints of parliamentary accountability compel members of a cabinet to coordinate their diverse interests when formulating policies so as to enhance the likelihood that they will, collectively, remain in office. Rifts within a parliamentary cabinet are likely to produce defections among members of the supporting majority in Parliament, leading to a possible vote of no confidence. Thus the dynamics of parliamentarism promote integration among members of a cabinet, whereas the dynamics of presidentialism foster internal cleavages in a "cabinet."In this context, turf battles between rivals in both the "cabinet" and the congress compel bureaucrats to take sides, in alliance with supporting groups in Congress and the "cabinet," as well, of course, as with organized private sector interests. This creates issue networks: including what are often called "sub- governments" or "iron triangles". Critics familiar with parliamentary governments view these difficulties as peculiarly American without, I think, recognizing their constitutional basis (Rose ??). However, widespread criticisms of the resulting malintegration of public policy in America are quite correct in their assessment of the damage it causes. Because they tend to blame the bureaucrats rather than the system that motivates these internecine conflicts, critics attack the victims rather than the structural causes of this dysfunctional situation.

In order to exercise effective executive leadership in a context where such cleavages prevail, American presidents have increasingly sought to plug the gap by creating large executive offices staffed by "in-and- outers," persons who rotate in and out of office as loyal supporters of each successive president. However, their transient status and, often enough, their lack of relevant experience and their personal rivalries, means that they soon find themselves doing battle with subordinates in the career bureaucracy. The separation-of-powers principle, therefore, not only plagues bureaucrats because it imposes multi-headed authority over them, but it also curtails the capacity of government to guide and discipline the American bureaucracy.

The Judiciary, as a third branch of government in presidentialist regimes, adds a further source of confusion. Because neither the President nor Congress has decisive authority over the other, the constitutional Charter is treated as the ultimate source of legitimacy for all governmental policies, as noted above. In practice, this means that judges can invalidate any act of government as "unconstitutional," generating a paradox. Such decisions may weaken bureaucrats who are conscientiously attempting to implement official policies, but it may also strengthen them irrationally by providing pretexts for idiosyncratic conduct on the premise that any given policy may be judged unconstitutional at some future time. Since decisions by lower courts can be overruled by higher courts, some policies hover in limbo until the Supreme Court finally makes an authoritative ruling on their constitutionality.

In other presidentialist countries, unfortunately, the bureaucratic situation is worse. All such regimes share the structural dispersal of authority that causes the malintegration of administration by rival bureaucratic groups -- both professional and clientelistic in character - - who necessarily respond to a range of competing centers of authority in the constitutive system, including the Congress and its committees, the President and his "cabinet" and executive office and, finally, the Courts. Since the problems generated by industrialization call for highly integrated and effective administration of the complex policies needed to cope with this range of problems of modernity, we need to ask whether parliamentarist regimes are not better equipped to manage the apparatus of government. Before doing that, however, we need to consider the composition of bureaucracies as an independent factor that also affects the ability of any democracy to manage the complex problems generated by industrialization.

Recruitment and Motivation. Two different but linked questions need to be answered. First, we must ask why, in some respects, public administration in America is more effective than its counterparts in other presidentialist regimes; and second, why can bureaucrats in most parliamentarist regimes administer more successfully than their opposite numbers in all presidentialist regimes, including the U.S. To answer these questions, we need to consider some important differences between bureaucracies. They hinge on two key variables: (1) how bureaucrats are recruited and (2) how they are treated after they receive their appointments. The answers, in both cases, are affected by modernity -- and especially by industrialization. They are also deeply affected by constitutional design -- what works well in parliamentarist regimes cannot work under presidentialism. In order to develop a democratic government that can manage the problems posed by industrialization, it is important to understand these matters.

Consider, first, how bureaucrats are recruited. Two types of criteria are important -- there are others I shall not discuss. These criteria involve the qualifications of individual bureaucrats, and their attitude toward the regime, i.e. to support or betray it. The relative importance of these criteria varies contextually: in some contexts loyalty is more important than qualifications, but in others, the reverse is true. Both criteria are always important but their relative salience varies: when loyalty is more important, rulers favor patronage in order to appoint officials whose support can be counted on; but when administrative problems are very salient, merit is emphasized.

In traditional monarchies, patronage was usually more important than merit, but large-scale empires create complex problems for whose solutions merit becomes more decisive. On the eve of the industrial revolution, both criteria were in use: the former in China and other empires, the latter in smaller kingdoms, including those of Europe. When the British consolidated their imperial possessions in southern Asia, they found that the prevalent patronage system that had long prevailed in England would not suffice -- accordingly, they borrowed the Chinese system and introduced training and examinations in order to establish the Indian Civil Service, a prototype of the merit system. Subsequently they reformed the domestic bureaucracy in Great Britain, establishing the Administrative Class as a ruling mandarinate, adapting the Chinese principles to the requirements of an industrializing country. Similar practices were also developed in the other European industrializing states. Thus merit replaced patronage throughout Western Europe during the 19th centuries, mainly after 1850. By contrast, all the contemporary presidentialist regimes retained patronage as a basis for bureaucratic appointments -- with the exception of the United States. It did not, however, copy the European (Chinese) "mandarin" type of system but, instead, invented a new kind of merit-based career system (for reasons I shall explain below) while also retaining patronage for important top-level positions.

Two major considerations affect the motivation of individuals toward their functions and duties as public officials: first, the way they are compensated for their work and, second, their status (security of employment). In all pre-industrial societies, where agriculture and pastoralism provided the main economic support system, states could scarcely afford to pay salaries that would support public officials and maintain their prestige. Instead, therefore, they offered inadequate stipends and authorized bureaucrats to keep part of what they collected for their work or they gave them land and other perquisites from which they could derive supplementary income. This type of compensation is prebendary and we may conclude that, regardless of regime type, all pre-modern bureaucracies were mainly prebendary. Because prebendary income is independent of the state, prebend-holders are also more autonomous -- they can afford to be more independent, a factor that makes personal loyalty to the rulers more important.

The industrial revolution brought about a great increase in national wealth, permitting rulers to increase taxation and, thereby, to start paying bureaucrats salaries that would more adequately support the life style they demanded. Salaries had a major impact on the criteria for recruitment: loyalty became less decisive after salaries replaced stipends -- not only would officials become more dependent on the state as their source of livelihood, but by surrendering prebends, they reduced their autonomy of action. Since the industrial revolution also made life more complicated, the expertise and experience of officials also become more important. These two changes reinforced each other, making technical competence more important than loyalty in all modern (industrialized) countries.

The second motive for bureaucrats is based on career expectations -- the more secure an appointment, the more valuable it becomes for officials. Insecure tenure leads officials to worry about their futures and, therefore, to conspire with each other to protect their jobs, and also to spend time planning and thinking about what they would do in case they were discharged. By contrast, tenured officials are able to increase their experience and capabilities while in office and dedicate themselves to their work more single- mindedly.

When emphasis on merit prevails in the recruitment of officials, regimes tend to offer long-term security (tenure) as an incentive to attract more qualified applicants and to keep them focussed on their work. Accordingly, industrialization enhances the need for well qualified public officials, and it also provides the resources needed to support tenure plus salaries, but not everywhere. For reasons to be explained later, in some states bureaucrats are given salaries but not tenure. They remain patronage appointees with indefinite terms of office.

Unfortunately, we have no generally accepted terms for officials who find themselves in this situation. I refer to them as retainers, modifying an earlier usage in which the staff of rulers (and rich persons) were referred to as "retainers." In pre-modern European usage, a king's retainers became officials as monarchies expanded their domains and powers. It seems appropriate, therefore, to use this rather old word in a somewhat new way: namely, to mean a patronage salaried appointee working without tenure. In practice, retainers often kept their jobs for a long time, sometimes even passing them to their children on a quasi- hereditary basis. We can think of retainers, therefore, as long-term public employees without formal tenure in office, but with reasonable expectations that they could keep their appointments (Riggs, 1994b).

By contrast, of course, in states where industrialization lags, the resources available to pay adequate salaries to long-term tenured employees remain inadequate, and one may suppose that the need for sophisticated public policies also lags, permitting patronage to survive even if salaries replace prebends as a basis for rewarding bureaucrats. These motivational factors can now be linked to the constitutional structure and power position of public bureaucracies.

Bureaucratic Power. In traditional polities, so long as religious beliefs and the potency of supernatural powers prevailed, bureaucrats could never replace rulers. Exceptionally, they might conspire to create a new ruler vested with supernatural powers, but the notion that appointed officials could dominate a polity was not viable. Moreover, in all pre-industrial societies, social structures and the economy were largely self-sustaining -- the functions of governance were limited and auxiliary but not fundamental. Officials with prebends are also conservative - - they prefer not to jeopardize their autonomous income by conspiring to make any radical changes in the system.

Modernity changed that in two ways: first, the spread of secularism undermined the sacred authority of kings and made it seem reasonable for one group of men to replace another group as rulers -- the sacred foundations of health and wealth would not, thereby, be undermined. Second, and perhaps even more decisively, industrialization not only made bureaucratic performance of administrative tasks more necessary, but it gave powerful new tools to appointed officials, especially military officers. During a time of severe crisis, they could use their control over the weapons of violence to subdue opposition and establish themselves as rulers.

Moreover, civil servants dependent on salaries have little choice but to obey the new rulers and, indeed, many of them share the dissatisfaction of military officers with failing regimes to the degree that their own income and status has been jeopardized. Increasingly, also, as industrialization advances, the services of experienced civil servants become increasingly indispensable, even in very poor countries. Consequently, even rulers who might distrust their loyalty feel constrained to retain their services.

Under modern conditions, therefore, administrative failures increase both the motives for bureaucrats (led by military officers) to seize power, and the resources, both military and organizational, that enhance the capacity of public officials to seize power. At the same time, of course, regime failures increase popular discontents and generate support for any groups able to dislodge current political elites, including revolutionary movements that would remove current public officials as well as the ruling elites. Normally, officials who recognize this possibility are strongly motivated to join military groups intent on preempting a revolution by means of a coup.

The likelihood of a successful coup is affected both by the design of constitutional systems and the structure of bureaucracy. (The relative immunity of single-party regimes arises for special reasons discussed in note no. 5) Empirical evidence suggests that presidentialist regimes are far more likely than parliamentarist ones to experience coups (Riggs, 1993). The theoretical arguments advanced here provide a rationale for this observation: the fused power available to parliamentarist regimes enable them to manage a powerful and effective bureaucracy based on the "mandarin" principles discussed above. Such officials are often permitted to run for political offices and, of course, they exercise great influence on regimes, especially as advisers to Cabinet members. (Dogan, 1975). However, they do not stage coups to seize power -- perhaps because they accept the status quo as both legitimate and advantageous and, also, because parliamentary regimes, with fused power, are able to maintain effective political control over them.

By contrast, in almost all presidentialist regimes, constitutional collapses have occurred and public officials, led by military officers, have seized power and ruled directly. The great exception is the United States where, despite major crises such as the Civil War and Great Depression, the Constitution and the regime have both survived. An understanding of the structure of their bureaucracies helps us understand both the normal case and the exception. The main difference, as I shall show, involves the relatively great power of bureaucrats in all presidentialist regimes except the United States.

Retainers. I think the most important difference is that the officials found in these countries are mainly retainers, i.e., patronage appointees who retain their appointments indefinitely. These retainers are powerful enough to prevent administrative reforms that would replace them with mandarin bureaucrats - - a transformation that would, assuredly, bring such long-term tested generalists to power in any presidentialist regime. In the U.S., exceptionally, retainers have been largely replaced, but not by mandarins: instead transients and functionists have taken office for reasons, and with consequences, that I explain below.

The inability of presidentialist regimes to maintain effective control over a retainer bureaucracy can be explained from both a top-down and a bottom-up perspective. From the top-down point of view, presidentialist regimes cannot reliable manage a mandarin bureaucracy -- it would reject the cross- pressures generated by a separation-of- powers system and intervene, perhaps non-violently, to assure coordination between the competing branches of government.

During the 19th century, most presidentialist regimes were not affected much by industrialization and remained primarily agrarian societies. The problems confronting their bureaucrats and the need for better remuneration had not reached critical levels. By the 20th century, however, the problems and resources generated by industrialism had led these countries to adopt increasingly complex public policies and to put their officials on salaries. In response to these changes, appointed officials, led by military officers, began to demand more job security and income. As retainers, their jobs were not secure enough and they had good reason to organize informally to protect and enhance their positions. During crises, they could establish conspiratorial cabals through which to seize power by a coup d'etat and displace the elected politicians. Mandarin bureaucrats in such situations would, I suspect, be able to maneuver non-violently to gain effective control over an ineffectual presidentialist regime, but retainers would not be so subtle, and would, therefore, be more likely to resort to violence. This also means that mandarin power would be exercised largely by civil servants, whereas retainers seeking power are more likely to succeed only under military leadership.

Even when elected politicians control the government in a presidentialist regime, the lack of training and inherent insecurity of retainers would, I surmise, lead to relative incompetence and turf battles likely to hamper effective public administration and limit the regime's ability to cope with complex new problems generated by industrialization. Because opportunities for the accumulation of wealth also increase with industrial development, and because foreign trade enhances the power of international corporations that are not easily controlled nor likely to be swayed by patriotism, both the leaders of a weak presidentialist regime and retainer officials often accept bribes for conduct that is not in the best interests of the people they govern.

We may, therefore, expect that corruption, the abuse of authority and internecine conflicts between rival factions and government agencies will prevail in presidentialist regimes and seriously hamper the handling of complex new problems generated by industrialization. Admittedly these propositions are speculative and need further research to be validated, but they suggest an important reason why presidentialist regimes, in general, are unlikely to handle industrialization in an effective way.

The Parliamentary Alternative. One line of analysis to test this reasoning is based on comparisons with parliamentary regimes. As noted above, all the industrialized democracies except the United States have parliamentarist forms of government. Two of the most spectacular success stories are those of Germany and Japan where, in the aftermath of their defeat in a devastating war, rapid industrial growth has taken place. In both cases, and despite the influence of American military government -- that would, presumably, have promoted presidentialist solutions, as it did in South Korea and South Vietnam -- parliamentarist regimes were installed and they have successfully managed rapid industrial growth. In both Korea and Vietnam, the areas under U.S. military control adopted presidentialist constitutions and both inherited mandarin traditions, first from China and, later, from Japan and France. In both cases, military coups terminated these regimes -- had their bureaucracies been better established, I suspect they would have taken control of these governments by non-violent means.

Rapid economic growth was, subsequently, promoted by military regimes in Korea that were able, for a while, to control a mandarin type bureaucracy. Eventually, the rising Korean bourgeoisie and the bureaucracy led movements that have restored a semi-presidentialist regime in South Korea. My guess is that this country, which already has a "prime minister", will evolve in a parliamentarist direction -- for more details see Riggs, (1996b). The failure of American support for South Vietnam's highly corrupt military regime was followed by the triumph of single-party rule -- it remains to be seen whether in that country, as in China, free enterprise and communism can co-exist, leading to rapid industrialization. In a third relevant case, Nigeria, its parliamentary model failed, leading to military dictatorship and the establishment of presidentialism. Despite the evident riches of this great African country, its economic plight is notorious. I would imagine that, even if (when?) democracy is restored, if it is presidentialist in character, its economy will continue to falter under the domination of a corrupted mandarinate inherited from British rule.

Such examples may be viewed as anecdotal and I cannot provide statistical evidence to support my views. However, I am persuaded that parliamentary democracies are, in general, more likely to provide the coordinated public services needed to handle the complexities of an industrialized (or industrializing) society, and I offer a theoretical argument to support my conclusion. The most important point, I think, is the capacity of a unified cabinet system of government to provide a coherent focus for the integration and management of a complex and powerful (mandarin) bureaucracy. Such a bureaucracy is, inherently, better qualified and motivated to administer highly interdependent and technologically advanced public policies than a non-mandarin bureaucracy. To state this opinion in reverse, a presidentialist regime cannot retain effective control over a mandarin bureaucracy and the less qualified retainer bureaucracies typical of presidentialism are, I believe, unable to provide the quality of public administration needed for industrialization to succeed.

In order to understand why presidentialism is incompatible with a powerful mandarin bureaucracy, consider the implications of the congressional role when the separation-of-powers prevails, as explained below. Since every congress must make up its own collective mind on all public policy matters, it must cope with an agenda that is far larger than that of any parliament. Because it is not possible to hold plenary debates on every issue, the delegation of power to committees and subcommittees is inescapable. This means, in turn, that the committees need a lot of technical advice to deal with the complex problems imposed by any industrial society. Bureaucrats, who need the help of these committees to get the laws and budgets they need, are only too glad to offer the advice most likely to win support for their own programs. From this symbiotic relationship, with the strong support of interested private organizations, sub-governments (iron triangles) emerge, leading to an administrative labyrinth that cannot be coordinated or rationalized. If the bureaucrats advising congressional committees were mandarins, able to reach effective agreements among themselves, they could surely dominate legislative decisions in such a congress.

In parliamentary systems, by contrast, there is almost no need for the micro-management of public administration by legislative committees because the assembly does not have the authority to establish public policy. Instead, it takes its cues from the cabinet, whose members as a planning group are assisted by the mandarin officials who typically serve as their chief advisers, always subject to the constraint that if they do not achieve consensus among themselves, they are likely to lose the confidence of their colleagues in the parliament. As for the members of parliament who are not in the cabinet, since they cannot exercise real power through program-oriented committees, their primary focus is on partisan concerns: their preoccupation with partisan matters overcomes their preferences on public policy issues. As a result, the incentives for bureaucrats and interest groups to lobby members of parliament is minimal and the ultimate authority of the cabinet is reinforced -- or, rather, it is not undermined by links between powerful bureaucrats and members of parliament.

If these premises are valid, we may conclude that parliamentary regimes are more likely than presidentialist ones to provide relatively well integrated and effective public administration in increasingly complex industrializing societies -- and their capacity to do so enhances the durability of constitutional democracy in parliamentary systems by comparison with what we find in most presidentialist regimes.

Oligarchy and a Semi-Powered Bureaucracy. This argument about the contrasting tendencies of presidentialist vs. parliamentary regimes is perfectly consistent with the contention that, exceptionally, a presidentialist regime, like the U.S., may do well and that, again exceptionally, a parliamentary system may perform badly. However, exceptional cases compel us to examine our generalizations to learn why they are not always decisive. In short, are there factors in the American case that enable why, despite its presidentialist constitution, its government has been able to cope with the problems of industrialization with at least moderate success. Although the U.S. has experienced near- fatal crises, its constitutional regime has survived for over 200 years (see Riggs 1988 and 1994a) In view of the new presidentialist systems now taking shape in the successor states of the Soviet and Yugoslav empires, it becomes doubly important to become aware of the normal fate of such regimes and to ask whether or not they could or should try to emulate the U.S. example.

The most important and general proposition to explain the American exception is that it has always been an oligarchic democracy, one dominated by its bourgeois constituencies at the expense of the impoverished and alienated minorities who fail to vote in its elections -- I cannot explain why that should be so, but some comments on this phenomenon can be found in Riggs, (1997a and b). Strong pressures exist in most presidentialist regimes to push then toward greater democracy, especially by adopting proportional representation and compulsory voting both of which, I think, have a centrifugalizing effect that erodes the regime's capacity to make decisions. The oligarchic results of its centripetal party system permits the American political system to continue to function.

Powerful support for the American system also derives from the relatrive weakness of its functionist/transient bureaucracy which enables it to administer public policies with moderate efficiency while resisting pressures to seize power. In fact, despite the persistence of anti-bureaucratic sentiments in American public opinion, public officials in the U.S. have administered well enough to provide substantial support for the success of industrialization and the solution of its most important problems. As a result, the American bureaucracy is semi-powered: it lacks the power potential of either a mandarin or a retainer bureaucracy. It is also technologically well trained and experienced, providing the relatively high level of competence required to succeed in its efforts to support and moderate the dynamic forces of industrialism.

The explanation can be found in the exceptional historical situations that enabled the U.S., alone among presidentialist regimes, to replace its retainers with a mixture of partisan transients (in-and-outers), career functionists, and private sector contractors. Both the transients and contractors play important roles in the administrative system, but they cannot ever conspire to seize power. As for the functionists, their specialized roles as career officers with external loyalties compete with their identity as bureaucrats, seriously blocking any conspiracies to seize power. The relative success of American public administration, combined with the oligarchic character of its political system, has also muted revolutionary sentiments and protest movements, helping to make the constitutive system viable i.e., the motives that might lead appointed officials to revolt are weak compared with those found in other presidentialist regimes.

I cannot explain here the complex historical events and forces that produced this peculiar mixture in the American bureaucracy -- they are discussed in some detail in Riggs, (1994b). Let me, however, say a few words about each.

Partisan Transients. Temporary government employees, have no expectation of long-term employment. When they enter the public service as bureaucrats, they know they will not retain their posts for any length of time -- less than 2.5 years is normal -- and they fully expect to return to private sector employment. Unlike retainers, therefore, they have no reason to conspire together to enhance their positions or extend their employment. Instead, their temptation is to develop private sector contacts to assure interesting and profitable positions after leaving their government jobs. Several scholars have written excellent analyses of this phenomenon, but they have emphasized its administrative implications and political causes rather than its consequences for the survival of presidentialist rules - - see, for example, Heclo, (1977).

The origins of this system can be traced to the administration of President Andrew Jackson who came to office in 1829. The patronage (retainer) system prevailed at that time, enabling new presidents to place their proteges in new offices, often over the heads of incumbents, but Jackson took the bold step of firing incumbents and replacing them with his followers (for details see White, 1994: 4-5, and Riggs, 1994b: note 31 on p. 133). This rotation system gained ground during succeeding administrations and fueled the reform movement that led, eventually, to the Pendleton Act of 1883 that institutionalized a peculiarly American form of career civil service system.

---

Updated: 11 June 2000 (first posted May 1997)