公 法 评 论 你们必晓得真理，真理必叫你们得以自由。
Cite as: 113 Harv. L. Rev. 633
THE NEW SEPARATION OF POWERS
This essay in comparative constitutional theory considers whether an American-style separation of powers should serve as a model for other countries. Professor Ackerman argues against the export of the American system in favor of an approach based on the constitutional practice of Germany, Italy, Japan, India, Canada, South Africa, and many other nations. According to this model of "constrained parliamentarianism," the constitution should not create an independently elected presidency to check and balance a popularly elected congress. Instead, it should authorize a prime minister and her cabinet to remain in power as long as they can retain the support of a democratically elected chamber of deputies. Constrained parliamentarianism tries to check the power of the cabinet and the chamber, however, by granting independence to a variety of other checking institutions, including a constitutional court. Professor Ackerman argues that this model offers a more promising path to constitutional development than the American approach. He shows how it can generate a variety of institutional strategies that better serve the three great principles that motivate the modern doctrine of separation of powers ¾ democracy, professionalism, and the protection of fundamental rights.
[T]he Federalist Constitution has proved to be a brilliant success, which unitary nation states and parliamentary democracies all over the world would do well to copy. I give it most of the credit for the fact that ours is the wealthiest, most technologically advanced, and most socially just society in human history, not to mention the fact that we have with ease become a military superpower . . . . The rest of the world is quite rightly impressed with us, and it is thus no accident that the United States of America has become the biggest single exporter of public law in the history of humankind. Almost wherever one looks, written constitutions, federalism, separation of powers, bills of rights, and judicial review are on the ascendancy all over the world right now — and for a good reason. They work better than any of the alternatives that have been tried.
Perhaps Steven Calabresi’s triumphalism is typical today, but it contrasts sharply with previous American attitudes. A half-century ago this country stood even taller in the world than it does now. As the only great power escaping massive destruction during World War II, America’s moralistic pretensions were at their apogee. Yet its constitutional prescriptions were a good deal more discriminating. To be sure, the United States supported written constitutions, bills of rights, judicial review — and, on occasion, federalism. But the separation of powers?
American influence reached its zenith in post-war Japan — with General MacArthur’s legal staff presenting a draft constitution to the Japanese within a ridiculously short space of time. For all the rush, the draftsmen did not propose an American-style separation of powers. In particular, they did not require Japan to embrace an American-style presidency as part of the price of its defeat. There emerged instead a distinctive regime-type: one that I will call "constrained parliamentarianism." As in Great Britain, Japan’s Prime Minister and his Cabinet must retain the confidence of the Diet to remain in office. But, in contrast to the Westminster model, the Japanese Parliament is not fully sovereign. Its legislative powers are limited by a written constitution, a bill of rights, and a supreme court.
Nor did the Americans impose a strongly bicameral legislature — featuring an upper house checking and balancing the lower with full Madisonian vigor. The Japanese House of Representatives plays the dominant role in selecting the Cabinet. Although the upper House of Councillors has significant powers, it is not the constitutional equal of the lower House. Call this the "one-and-a-half house solution."
The story is the same in Germany — though the Americans and other Allies played a less heavy-handed role, letting German jurists and politicians call most of the shots. Fresh from their experience with Adolf Hitler, nobody was in the mood for an elected presidency. Once again the result was constrained parliamentarianism, with a one-and-a-half house solution.
The Italians were still more in control of their constitutional pro- cess — and we shall see that they created a very interesting variation on the one-and-a-half house theme. Nonetheless, the Italian variation fits comfortably within the basic framework of constrained parliamentarianism. Like the Germans, the Italians were entirely unprepared to build a presidential platform upon which future Mussolinis might vie for (democratic) preeminence.
But times have changed, as the words of the younger Professor Calabresi remind us. Especially since 1989, American jurists have become big boosters of the American Way at constitutional conventions everywhere. When they arrive at the scene, however, their intellectual preeminence is by no means assured. To the contrary, American jurists regularly encounter vigorous competition from French and German constitutionalists, who also operate as cheerleaders for their native constitutional traditions.
Political scientists have played a more edifying role. When constitutional conventions have turned to them, modern-day framers have heard something more than triumphalist success stories packaged as the American, French, and German "models" of constitutional government. They have been rewarded with some useful tools for critical reflection on fundamental constitutional choices.
Most notably, the choice between the parliamentary and presidential systems of government. While American legal scholars content themselves with pietistic references to Montesquieu and Madison, modern political scientists deign to consider the way alternative systems have actually worked in the world. Their research is a precious resource for anyone who wishes to reflect upon the future of the separation of powers.
Nonetheless, it is inadequate. For starters, the political scientists have largely focused on a single issue: should constitution-writers follow England in concentrating lawmaking power in a single parliamentary institution, or should they follow the United States and France in separating lawmaking authority among democratically elected rivals? This is an important question, but it is not the only one. The separation of powers involves not only presidents and parliaments, but also the constitutional status of courts and administrative agencies. As we shall see, the resolution of the first separation issue has non-obvious implications for the others, and vice versa.
I also propose to refine the normative terms of the debate. "Liberal democratic constitutionalism" is not a unitary concept but a placeholder for congeries of different values, coexisting in deep tension. To illuminate these complexities, I shall be navigating between methodological extremes. On the one hand, I appeal to a range of political ideals in assessing alternative forms of separation. Without clear normative orientation, talk of the separation of powers degenerates into facile constitutional engineering. The very idea of institutional "efficiency" is completely empty unless it is linked to more substantive ends. On the other hand, this is not a philosophical essay on the foundations of political legitimacy. My primary concern is to illuminate the complex ways in which institutional arrangements serve as concrete expressions of ultimate ideals, not to philosophize about the ideals themselves. I make only those conceptual distinctions that seem absolutely necessary for thoughtful institutional assessment, thereby begging hosts of philosophical questions. This will (rightly) prove annoying to some readers, but all I can say is Sorry, a single essay can’t do everything.
More concretely, I return repeatedly to three legitimating ideals in answering the question, "Separating power on behalf of what?" The first ideal is democracy. In one way or another, separation may serve (or hinder) the project of popular self-government. The second ideal is professional competence. Democratic laws remain purely symbolic unless courts and bureaucracies can implement them in a relatively impartial way. The third ideal is the protection and enhancement of fundamental rights. Without these, democratic rule and professional administration can readily become engines of tyranny.
I take up each of these ideals in turn in the three parts that follow. Part I considers separating the power of democratic lawmaking among different branches and introduces two theses that will recur throughout this essay. The first thesis is negative and cautions against the export to other countries of an American-style separation among house, senate, and presidency. Although this system has worked well enough at home, it has proved nothing less than disastrous abroad. We should reject Professor Calabresi’s invitation to transform it into one of the bright lodestars of the new millennium.
My second thesis is more constructive and distinguishes this essay from traditional critiques of American lawmaking arrangements. Generally, English-speaking critics of American separationism have looked to Great Britain as the source of a competing model of democratic government. The modern British Constitution famously concentrates lawmaking power in the House of Commons, giving the Prime Minister and her Cabinet effective control over the legislative agenda. The real-world operation of this "Westminster model" has provided critics with a club to batter American self-confidence. Given the British success in avoiding the inexorable slide into tyranny predicted by Madison and Montesquieu, perhaps we should give up on the very idea of separation of powers?
My message is different. I reject Westminster as well as Washington as my guide and proffer the model of constrained parliamentarianism as the most promising framework for future development of the separation of powers. Not only has this model set the terms of the post-war settlement with the Axis powers, but it also characterizes many of the more successful democratic regimes that have emerged from the dissolution of the British Empire. For all their differences, the constitutions of India, Canada, and South Africa fit within the broad contours of the basic model. Moreover, the success of the German Constitution has inspired other countries, most notably Spain, to use it as a reference point in their own transitions from authoritarianism.
Constrained parliamentarianism, then, is a rising force in the world, and there is much to be learned from its practical operation over the past half-century. There is no reason to suppose, however, that any existing regime has hit upon the best way to constrain parliamentary government. Here is where my title’s promise of a "new" separation of powers enters. Although I reject an American-style competition between house, senate, and presidency, I believe we have only begun to tap the separationist potential of constrained parliamentarianism. Part I elaborates this point by considering how the lawmaking powers of parliament may be constrained by other institutions of democratic self-government, including popular referenda on the national level and the representation of provincial governments in federal systems.
Part II takes up the same negative and positive themes in assessing separationism’s potential contributions to professionalism in the judiciary and the bureaucracy. On the negative side, American performance again comes under the microscope. Although the American system has been quite successful in fostering an independent and professional judiciary, the same cannot be said of its impact on the bureaucracy. The ongoing competition between House, Senate, and Presidency for control over the administrative apparatus has created an excessively politicized style of bureaucratic government, transforming the executive branch into an enemy of the rule of law.
In contrast, the ongoing interaction between parliamentarianism and public administration holds the promise of a more constructive relationship between democracy and professionalism. But again, I think it is a big mistake to leave the relationship between parliament and the bureaucracy subject entirely to the unwritten constitution. I propose the explicit constitutional construction of two distinct branches to assure that bureaucratic government redeems its central claims to integrity and expertise in regulating for the public interest.
Part III turns to the question of fundamental rights and complicates the conventional wisdom that links American-style separation to their protection. Even for partisans of laissez-faire, there is much to fear from the patterns generated over time by the dynamic interaction of president, house, and senate. On a more constructive note, I urge constitutionalists to transcend their traditional court-centered focus. A supreme court for the protection of fundamental rights is, without question, an essential component of the model of constrained parliamentarianism. But it should be supplemented by separate non-judicial institutions concerned with the more effective protection of rights of democratic participation, on the one hand, and the realization of fundamental commitments to distributive justice, on the other.
By the end of the essay, we shall be moving far beyond the now-standard recipe of one-and-a-half houses, a bill of rights, and a constitutional court. My aim is to show how constitutional framers can elaborate the basic idea of constrained parliamentarianism in a rich variety of ways to achieve a complex set of political objectives. Though particular parts of my scheme have precedents in one or another existing system, my overall proposal is more than the sum of its parts and, when viewed as a whole, may seem quite novel. Much more comparative study, and analytic work, is required before my scheme can become the basis for serious practical proposals.
For the present, I offer it up in a more speculative spirit. I hope to encourage comparative constitutional law to transcend its present condition of naïve boosterism and to engage in genuine transnational conversation about the future of Western constitutionalism. What are its intellectual and institutional resources? How may they combine into different patterns that promise better performance? If my elaboration of constrained parliamentarianism gets the conversational ball rolling, provoking counter-proposals from partisans of different legal traditions, this will be fine with me.
If it also helps American constitutional lawyers place their own tradition in critical perspective, so much the better. Although I will be very critical, I should not be misread to suggest that Americans should junk their system of separation of powers as it has evolved over the past two centuries. Even though its pathologies are many and serious, the rituals of confrontation between the President, the House, and the Senate are by now second-nature to Americans, providing a grammar of legitimation that has allowed citizens to define, and sometimes to decide, matters of fundamental importance over the generations. Given its deep roots in American culture, it would be rash to suppose that we can invent a better system out of whole cloth.
But it is one thing for Americans to try to make the best of their evolving scheme of checks and balances, quite another to hold it up as an inspiring beacon for liberal democrats everywhere.