In lieu of an abstract, here is a brief excerpt of the content:

  • Presidents vs. ParliamentsComparing Democratic Systems
  • Donald L. Horowitz (bio)

In "The Perils of Presidentialism" [Journal of Democracy 1 (Winter 1990): 51-69], Professor Juan Linz makes the claim that parliamentary systems are "more conducive to stable democracy" than are presidential systems. "This conclusion," he continues, "applies especially to nations with deep political cleavages and numerous political parties." This theme forms a leitmotiv in Professor Linz's recent works, has been picked up by other scholars, and runs the risk of becoming conventional wisdom before it receives searching scrutiny.

Linz argues that the presidential office introduces an undesirable element of winner-take-all politics into societies that need mechanisms of conciliation instead. A presidential candidate is either elected or not, whereas in parliamentary systems many shades of outcome are possible. Moreover, a directly elected president may think he has a popular "mandate," even if he has been elected with only a small plurality of the vote, perhaps even less than 40 percent. The potential for conflict is accordingly enhanced.

Conflict is promoted, in Linz's view, by the separation of powers that divides the legislature from the president. The fixed term of a separately elected president makes for rigidity between elections. By contrast, parliamentary systems are able to resolve crises at any time simply by changing leaders or governments. Separate presidential election also produces weak cabinets and fosters electoral contests in which extremists either have too much influence or the whole society becomes polarized.

This is a powerful indictment, supported by an abiding concern for [End Page 73] the stability of precarious democratizing regimes. Linz's claims, however, are not sustainable. First, they are based on a regionally skewed and highly selective sample of comparative experience, principally from Latin America. Second, they rest on a mechanistic, even caricatured, view of the presidency. Third, they assume a particular system of electing the president, which is not necessarily the best system. Finally, by ignoring the functions that a separately elected president can perform for a divided society, they defeat Linz's own admirable purposes.

Presidentialism and Political Instability

As frequent references to Brazil, Colombia, Venezuela, and Chile attest, Linz believes that presidentialism has contributed to instability in Latin America. If, however, his focus had been on instability in postcolonial Asia and Africa, the institutional villain would surely have been parliamentary systems. Indeed, Sir Arthur Lewis argued 25 years ago in his lectures on Politics in West Africa that the inherited Westminster system of parliamentary democracy was responsible for much of the authoritarianism then emerging in English-speaking Africa. What Lewis emphasized was the winner-take-all features of the Westminster model, in which anyone with a parliamentary majority was able to seize the state.

Lewis's understanding conforms to that of many Africans seeking to restore democratic rule. The most impressive efforts at redemocratization, those of Nigeria in 1978-79 and again at the present time, involve adoption of a presidential system to mitigate societal divisions. Under the parliamentary system inherited at independence, a cluster of ethnic groups from the north had managed to secure a majority of seats and shut all other groups out of power. This game of total inclusion and exclusion characterized Nigerian politics after 1960, precipitating the military coups of 1966 and the war of Biafran secession from 1967 to 1970. By choosing a separation of powers, the Nigerians aimed to prevent any group from controlling the country by controlling parliament.

Now it is possible that parliamentary systems helped stifle democracy in Africa while presidential systems helped stifle it in Latin America, but there are grounds for doubt. Linz refers to the emergence of conciliatory practices in the presidential systems of Colombia, Venezuela, and Brazil, but he dismisses them as "deviations." Chile under Salvador Allende, on the other hand, is regarded as closer to the norm, with presidentialism exacerbating social conflict. Yet at least some research by Arturo Valenzuela suggests that, before Allende, many Chilean presidents actually bolstered centrist, moderating tendencies. The experience of the presidency in the United States, where the presidency was invented, is also explained away as "an exception." Consequently, Chile's exacerbated conflict is traced to its presidency, while the moderated conflict of the...