Printer Friendly

On the comparison of presidential and parliamentary governments.

Comparison of the French and American presidencies has been made more difficult than it needs to be by confusion over the essential differences between presidential and parliamentary government. Classic textbook definitions, such as the aphorism that in parliamentary government the executive must be supported by a parliamentary majority while presidents are chosen through national election, confuse rather than clarify Franco-American comparisons. Even worse are definitions based on the language of constitutional provisions; any workable distinction must be based on actual behavior rather than legal prescriptions.

However misused, the terms "parliamentary government" and "presidential government" are unavoidable. What is the truly fundamental difference between them? Parliamentary government is a democratic regime in which the executive and the legislature ultimately must agree on policy. The two branches are synchronized, so to speak, like two gears that mesh. If the two branches are out of "synch," something will be done to compel policy agreement between them. How this harmonization is brought about, which branch tends to dominate the other, and the composition and designation of the executive branch will vary from country to country. Temporary failure to agree leads to a short-term crisis; prolonged and continuous failure to agree indicates that the system no longer works properly and is due for reform or replacement. This is a behavioral, not a constitutional or structural definition. It is how the political actors behave, rather than their constitutional, partisan, or parliamentary environment that sets the definition of parliamentary government. Those environmental elements are, of course, crucial in producing parliamentary government and determining whether and how it works, but no particular aspect, such as the written constitution, parliamentary procedure, or party organization in itself defines or identifies parliamentary government.(1)

Behavior of political actors in presidential government is very different. The executive and legislative branches are free to disagree with each other; any prolonged domination of one branch by the other, normal in parliamentary regimes, is exceptional and abnormal. Prolonged preponderance of one branch by another signifies evolution toward democratic parliamentary government, or more ominously, presidential authoritarianism. Not only do many important policies require agreement between the branches, but that agreement results from voluntary cooperation rather than forced synchronization. While failure to reach agreement could mean failure to accomplish policy goals, failure does not produce a governmental crisis. Indeed, both sides may find it politically expedient to agree to disagree. It is true that some measures, budgets for example, must be enacted to keep the system going. But even then the chief executive is hardly likely to micromanage budget items as executives are able to do in many parliamentary regimes. The tenure of the executive does not depend on support from the legislature; by the same token, the legislature is not required to conform to executive directives in order to keep the system in operation.

For presidential government to operate, three institutional patterns are indispensable. The first is that the chief executive be designated through an effectively democratic national election (not necessarily a direct popular election). Neither a hereditary monarch nor a president selected without participation of the people would possess the necessary legitimacy for authentic presidential government.(2) Except for impeachment under extraordinary circumstances or removal for disability, the chief executive can remain in office for a fixed term unless removed earlier by a national electoral process such as recall by popular vote. Electing the chief executive is a necessary, but not sufficient, requirement for presidential government.

A second indispensable pattern for presidential government is that the chief executive work with shifting majorities in the parliamentary bodies. What would happen if a parliament were to produce a consistent, cohesive majority party or coalition? If that majority were led by a member of parliament, that member would be a de facto parliamentary chief executive with or without the title of "prime minister," and the power of the ostensible chief executive, even if elected, would evaporate. An executive who opposed such a leader and party would not be able to function as was the unhappy experience of President Andrew Johnson. On the other hand, if the elected executive were the effective leader of a consistent, cohesive majority party or coalition, that chief executive would be far more powerful than the American president. The legislative body would then be synchronized with the executive which would fulfill the definition of parliamentary government.

In other words, the existence of a consistent, cohesive parliamentary majority will cause either an elected chief executive or the leadership of parliament to dominate the opposite branch, relationships consistent with parliamentary government but not with presidential government. The executive and legislature under presidential government must be free to disagree with each other without one branch able or willing to overcome the other. Obviously this condition win not exist if either the elected chief executive or the legislative leaders can ordinarily compel the two branches to concur on policy decisions.

To summarize, it follows that for a system to be defined as presidential government, (1) the chief executive has to be democratically elected, (2) the chief executive has to work with shifting majorities in order to have programs adopted, and (3) the executive and legislature need to be free to disagree on policy without provoking a constitutional crisis. Although parliamentary governments may fulfill one or two of these conditions from time to time, they are not transformed into authentic presidential governments unless and until they regularly fulfill all three conditions.

Among the parliamentary democracies, government by shifting legislative majorities does indeed occur from time to time, especially in the Scandinavian countries. In theory one would expect shifting majorities to be even more likely in a multi-party system than in a two-party system. Shifting majorities make possible a step in the direction of presidential government, but only a step. Election of the chief executive by the people is also only a step, and no more than that, in the same direction. Unfortunately, the word "president" seems to mesmerize political scientists. When confronted with systems in which an elected president exercises real power but in which the government is responsible to parliament, they assume that the regimes are "presidential" because the chief executive is called "president" Instead of confining "presidential government" to its proper sense, they invent a mythical category to be added to the classic dichotomy of presidential and parliamentary government.

Maurice Duverger popularized this trend with his invention of "semi-presidential" government. These regimes not only have governments responsible to parliament but also have elected presidents with significant constitutional powers.(3) According to Duverger, countries falling under this definition have included the Weimar (German) Republic during 1920 to 1933, Austria since the constitutional reform of 1929, Ireland since the Constitution of 1937, Iceland since the Constitution of 1945, France since the Constitution of 1958, and Portugal since the Constitution of 1976. He concedes that the actual, as opposed to the formal, power of the president depends on politics. Duverger identifies the French, Finnish, and, to a lesser extent, Icelandic heads of state as powerful presidents while the rest are figureheads. Whether or not the president exercises effective power, Duverger classifies these regimes as "semi-presidential" because their constitutions provide for a two-part executive branch with the division of power between the two varying with the country and over time. Thus Duverger's "semi-presidential" regime is based on constitutional provisions rather than the behavior of the executives involved or the functions they perform. This is ironic because Duverger's famous study of political parties teaches us that party organization and electoral laws, for example, are more important than constitutions in determining the behavior of a political system.(4)

Duverger initiated the new category in order to facilitate cross-national studies of democratic executives, including the American presidency, a, praiseworthy objective:

Confrontation of this model with that of a

presidential system [i.e., the United States] makes it possible to clarify

not only the operation of regimes in the seven countries listed above but

also the operation of the regime in the United States. The major purpose of

this discussion is to open the way to such comparative analysis, the

development of which should be fruitful.(5)

Good intentions do not always produce good results. Duverger's alleged third category has led to confusion in comparative executive studies by filling a need that does not exist. Unfortunately, many writers have followed Duverger although they usually confine the proposed third class to the truly powerful presidents. Thus Kent Weaver and Bert Rockman classify France as "hybrid," Matthew Shugart and John Carey classify that country as "premier-presidentialist," while for Gabriel Almond and G. Bingham Powell France is a "mixed parliamentary-presidentialist regime."(6) Why this confusion on the part of the most brilliant and scholarly political scientists? This difficulty is well illustrated, albeit inadvertently, by Arend Lijphart:

When the president is head of government, the system is mainly presidential,

but when the prime minister assumes this role, the system shifts to

parliamentarism. We are inclined to classify the Fifth Republic as mainly,

albeit not fully, presidential. We defer, however, to the prevalent view

that it should be labeled a semi-presidential regime.(7)

The word "role" is the key. What is a "role"?

The basic unit of political structure is the individual role. A role is a

regularized pattern of behavior, established by one's own expectations and

actions and those of others.... But describing a political structure spells

out a network of roles; individuals have positions in which they are

expected to act in certain ways and regularly do so.(8)

In other words, roles involve expected behavior rather than possession of a particular title such as president or prime minister. This is made clear by Almond and Powell when they define the relationships of political structures to the functions they perform:

The analysis of specific political structures and their relationship to

political functions makes it possible to describe and compare very

different political systems. We defined our concept of political system, and

our typology of political functions, in a way that means that all political

systems must include performance of these functions. We can therefore

compare political systems by observing which political structures perform

the different functions in them. The great advantage of an explicitly

structural-functional approach is that it enables us to avoid the confusion

between the formal goals of structures and the political functions actually

performed by them.(9)

Whether there really are universal functions, and how they might be defined, involves controversies that need not be settled here. The essential point to keep in mind is that you cannot tell what role a political actor is playing, and what function his or her office is performing, from the title of that role or office. Although that is basic political science, even the outstanding political scientists who have valiantly defended this enlightening insight do not always follow their own advice. Official titles and constitutional rhetoric may block their view of political realities.

When an executive directs a government that is responsible to a parliamentary body, that executive is playing the role of a parliamentary political executive. In the enormous majority of cases, the individual playing that role has a title such as prime minister, premier, chancellor, or president of a council of ministers. In Fifth Republic France, the role of parliamentary political executive is normally assumed by the President of the Republic. He performs the functions normally carried out by prime ministers in other democratic regimes. If the majority of the National Assembly is able to support a government opposed to the president, then the executive who actually has the title of prime minister takes over the function of parliamentary political executive. France does not alternate between presidential and parliamentary regimes; it is a parliamentary system because the executive and the legislature are compelled to agree with each other just as in Britain or Germany.(10) Only in the United States, where the president is free to disagree with Congress and yet still runs the government, can it be said that the regime is fully "presidential."

France does not become "presidential" when the president is effective chief executive, nor does it become "parliamentary" when the prime minister directs the executive. Either executive may assume the role of a chief executive whose acts are synchronized with the parliament; the functions performed rather than the officer's title determine the officer's role and therefore the nature of regime. Is this a mere professorial quibble about terminology? Not really. Let me give two illustrations and then the major consideration for French-American comparison. First, from Powell's prize-winning study of contemporary democracies:

Except in Chile (and potentially in France), the very weak party systems in

most presidential countries, as measured by low party-group linkages and

high volatility, made it easier for minority presidents to deal with their

legislatures. As Duverger suggests, weak parties increase presidential


Is France really an exception to Duverger's rule? Presidential authority in France has indeed depended on an increase in party strength rather than party weakness (contrary to De Gaulle's expectations). During the period of minority government in 1986 to 1988, President Mitterrand's authority required the reliable support and solid discipline of a Socialist parliamentary group linked closely to the national party organization. Once it is understood that France is indeed a parliamentary system, and not some bastardized presidential regime, the French exception disappears and the paradox is resolved. Weak parties may well be consistent with authentic presidential government, as Duverger has suggested, although one may wonder whether American parties have carried that virtue to an excess.

Presidential government may conceivably emerge in some contemporary parliamentary countries. Proposals to establish authoritative elected executives in the Netherlands and Israel would permit the Dutch monarch and the Israeli president, respectively, to continue as heads of state while the prime minister would be the effective chief executive without the title of "president."(12) Enactment of such a measure seems unlikely in the Netherlands, but in March 1992 the 12th Israel Knesset did pass the Law for the Direct Election of the Prime Minister.(13) This law was actually applied in the election of May 1996. Whether the elected Israeli prime minister will work with shifting majorities and remain in office even when his or her policy initiatives are voted down, a condition required for presidential government, remains to be seen. If an elected prime minister, Dutch or Israeli, were to behave like an American president, would some political scientists insist that the system was still parliamentary just because the executive was called a prime minister? I should hope not!

The Dutch kingdom cannot use the title of president while the Israelis want to retain an impartial head of state. So the effective chief executive has to be the prime minister even in the event of evolution toward presidential government. It would then be awkward to use the term presidential government for such regimes. "Separation-of-powers" comes to mind but as M.J.C. Vile demonstrated in his monumental work, separation of powers means too many different things to different people.(14) As long as such a regime is not called "parliamentary" or "semi-presidential," this nomenclatural problem can be left to the future.

Unlike Israel or the Netherlands, France already has an elected president who, more often than not, exercises real power. Contrary to Duverger, a strict dichotomy between parliamentary systems and presidential government is essential for effective comparison between the French and American presidencies. Since the French people enacted direct election of the president in 1962, I have scrutinized French government and politics for any indications of authentic presidential government. To label France "semi-presidential" or "mixed" confuses the issue. If we start with a clean parliamentary slate, behavior that fulfills any or all of the conditions for presidential government will be much easier to detect. Such hybrids as semi-presidentialism or alternating spasms of presidentialism and parliamentarism serve only to befog the mind.

This is not to say that a powerful elected president in a parliamentary system makes no difference. Direct election of a powerful president certainly can make a big difference, but that difference remains within the parameters of parliamentarism.

A simple contrast between parliamentary and presidential systems suggests

substantial homogeneity within each type of system. A closer examination,

however, reveals that policymaking structures and processes in parliamentary

systems can vary tremendously across countries and over time. Indeed,

comparing parliamentary systems and the American separation-of-powers

system is less a matter of comparing apples and oranges than of comparing

apples with other fruits.(15)

One could hardly have said it better. An elected president in the role of parliamentary political executive is indeed an unusual but significant variation of parliamentary government. Although this regime should not be confused with presidential government, the politics of presidential elections along with other behaviors associated with the presidency render this French "fruit" most appropriate to compare with the "apple" of American presidential government. Even though the French president functions in a parliamentary context, the presidential office does indeed have peculiarities of its own that differ from those of the prime minister's establishment and customs; the development of these presidential characteristics can and should be compared to the long-established American presidency to ascertain parallels and differences.

In addition to an elected president, France has had a period of minority government (1988-93). Whether or not one accepts the rather broad conception of minority government applied by Kaare Strom, minority government does frequently occur in parliamentary systems.(16) Minority government could be a step in the direction of presidential government because it sometimes involves policy making with the support of shifting majorities. In the absence of an elected chief executive, evolution toward presidential government could not very well take place. But if minority government were to be combined with an elected executive who continues to dominate the government but relies on shifting majorities in the parliament to enact legislation, a regime could approach presidential government.

Whether minority government in France under the aegis of an elected president was anything more than what the French call an accident de parcours (accident along the way), remains to be seen. But more than enough has happened in the French presidency to provide abundant material for comparison with the American chief executive. The last thing needed for such comparison is semi-presidential confusion.

This is not to say that France or other countries should adopt presidential government. That's their business. As students of politics, we need to be able to recognize authentic presidential government when and if it should occur, as well as compare the parliamentary presidencies with our own presidential regime.


(1.) Has anyone else ever offered this definition of parliamentary government? It would be very helpful if someone has, so that I do not have to bear the burden of originality.

(2.) It is true that the Twenty-fifth Amendment to the U.S. Constitution makes possible a nonelected president in the event that a vice-presidential vacancy is filled by a presidential appointment confirmed by Congress. Gerald Ford became president through this procedure without challenge to his legitimacy. For a number of reasons, this circumstance does not undermine the requirement that a president be elected. First, this was a purely stop-gap procedure which no one expects to become routine. Second, once he became president, Gerald Ford was the obvious Republican candidate for election as president, a position which maximizes a president's influence. Third, a political system that has functioned for over 200 years can put up with occasional discrepancies that might upset a less well-established regime.

(3.) See any of the following by Maurice Duverger: Echec au roi (Paris: Albin Michel, 1978), pp. 21-56. "A New Political System Model: Semi-presidential Government," European Journal of Political Research 8, no. 2 (1980): 165-87, or "Systeme presidentiel et systeme semi-presidentiel," in La presidence en France et aux Etats-unis, Jean-Louis Seurin, ed. (Paris: Economica, 1986), pp. 347-58.

(4.) Maurice Duverger, Political Parties: Their Organization and Activity in the Modern State, trans. Barbara and Robert North (New York: Wiley, 1954), pp. 352-3, 392-412. Originally published as Les partis politiques (Paris: Armand Colin, 1951). This is one of the great classics of political science; Duverger's theories on political party organization and electoral systems are still actively debated by contemporary political scientists.

(5.) Duverger, Systeme presidentiel, p. 350.

(6.) Kent Weaver and Bert A. Rockman, "Assessing the Effects of Institutions," in Kent Weaver and Bert A. Rockman, eds., Do Institutions Matter? Government Capabilities in the United States and Abroad (Washington, D.C.: Brookings Institution, 1993), p. 12, n. 21; Matthew Sobert Shugart and John M. Carey, Presidents and Assemblies: Constitutional Design and Electoral Dynamics (Cambridge, England: Cambridge University Press, 1992), pp. 23-4; Gabriel Almond and G. Bingham Powell, Comparative Politics: System, Process and Policy, 2nd ed. (Boston, MA: Little, Brown, 1978), p. 239.

(7.) Arend Lijphart, "Separation of Powers and Cleavage Management" in Weaver and Rockman, Do Institutions Matter?, pp. 307-8.

(8.) Almond and Powell, Comparative Politics, p. 53.

(9.) Ibid., p. 55.

(10.) Although monocratic executives are customary, plural executives do exist and even flourish. In Finland the president and prime minister share the functions of parliamentary political executive while in Switzerland an all-party council is customary. Whether a collegial executive would actually survive in presidential government is debatable, but we won't know unless it is tried.

(11.) G. Bingham Powell, Jr., Contemporary Democracies: Participation, Stability and Violence (Cambridge, MA: Harvard University Press, 1982), p. 152.

(12.) Myron A. Levine, "Is a Presidential System for Everyone? Some Reflections on the Dutch Rejection of an American-style Presidency," Presidential Studies Quarterly 18, no. 2 (1988): 277-81.

(13.) For a complete version of the new Constitutional provisions, see "Basic Law: The Government,"Israel Yearbook and Almanac 47 (1993): 279-92.

(14.) Weaver and Rockman, "Assessing the Effects of Institutions," p. 11; M.J.C. Vile, Constitutionalism and the Separation of Powers (London: Oxford University Press, 1967), pp. 1-20, 315-50.

(15.) Weaver and Rockman, "Assessing the Effects of Institutions," p. 19.

(16.) Kaare Strom, Minority Government and Majority Rule (Cambridge, England: Cambridge University Press, 1990), pp. 6, 58.
COPYRIGHT 1997 Center for the Study of the Presidency
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1997 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Kaminsky, Elijah Ben-Zion
Publication:Presidential Studies Quarterly
Date:Mar 22, 1997
Previous Article:George Washington's pre-presidential statesmanship, 1783-1789.
Next Article:Unified government, the two presidencies thesis, and presidential support in the Senate: an analysis of President Clinton's first two years.

Related Articles
Social Democrats say presidential election should be called \'snap\' election.
Chair of Parliamentary Committee for Constitutional Legislation explains why presidential election has been scheduled for July 23, 2009.
Ar Namys party to decide whether to nominate a candidate to run in presidential election.
President Bakiev proposes to double number of international observers for presidential elections.
Speaker Tagaev: We all want presidential elections to be transparent and fair.
Abbas says on course to hold in January parliamentary and presidential elections.
Osama Al-Ghazali Harb says willing to run for president.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters