Lijphart: The wave of power-sharing democracy

Lijphart. 2002. The wave of power-sharing democracy. In Architecture of Democracy: Constitutional design, conflict management, and democracy. Andrew Reynolds, ed. Oxford: Oxford University Press.

The General Problem Area

The book's editors asked Lijphart to write on this topic: "Constitutional Design in the Third Wave." Lijphart amended this topic so he could write on his usual theme, consociational (power-sharing) democracy. Why: because "ethnic divisions have replaced the cold war as the world's most serious source of violent conflict," and Lijphart is convinced that, in deeply divided societies, constitutional provisions for power sharing are vital.

Lijphart's arguments assume familiarity with the consociational/consensus type of democracy that he has advocated in several previous publications. In particular, see Lijphart's Patterns of Democracy.

What the Experts Agree On

When it comes to engineering a constitution for a divided society, experts tend to agree that

• Democracy is harder in divided societies
• Getting democracy started in a divided society is harder than maintaining it after it's been consolidated
• Two key ingredients, or two "primary characteristics" of "consociational democracy":
  • Sharing executive power (grand coalitions)
  • Group autonomy (segmental authority)
    • esp. in areas of education and culture
  • Two secondary characteristics that have less widespread acceptance:
    • mutual veto power
    • proportionality

Critiques of Power-Sharing and Autonomy

Problems of Power-Sharing

Complaint: Power sharing isn't democratic enough. Grand coalitions conflict with idea that you must have a strong opposition. These states fail the "turnover" tests of whether a gov't is democratic.

• Response: these critiques assume that only majoritarian democracies are democratic. Too narrow a conception of democracy.

Complaint: Power sharing doesn't work. It results in deadlock and democratic breakdown, e.g. Cyprus 1963 and Lebanon 1975.

• Response: Lijphart gives a much longer list of those that succeeded.

Complaint: Power sharing removes incentives for compromise/moderate behavior (Donald Horowitz). It has only incentives to coalesce, not compromise.

• Response: Lijphart contradicts this claim without giving a strong argument in his defense.

Problems of Regional Autonomy

Complaint: Regional autonomy is a slippery slope leading to secession/partition.

• Response: Would forced centrism be better? Cites Gurr's empirical study (MAR): autonomy can help manage regional conflicts.

Complaint: Regional autonomy strengthens intraethnic cohesiveness, raising likelihood of interethnic conflict.

• Response: Again, Gurr's study refutes this.

A General Problem

Complaint: Both halves of the prescription (i.e. power sharing and autonomy) are native to Western experiences and are not suitable for multi-ethnic societies elsewhere.

• Response: Lebanon, Malaysia, Colombia, elsewhere

Practical Guidelines Derived from Consociational Theory

Power-sharing vs. majoritarian debate:

• Warning: pure majoritarianism (one-party cabinet, plurality voting, etc) is dangerious
• Even majoritarian institutions that reward centrism--e.g. preferential or runoff voting--won't please a minority ethnic group. PR is better.

Presidential-parliamentary debate:

• Presidential elections: zero sum. PR is better.

Decentralization, federalism, and autonomy debates:

• Lipset: recommends that federal boundaries cross ethnic boundaries, because "democracy needs cleavages within linguistic or religious groups, not between them." But aren't you more likely to get these intra-group cleavages by putting all the French in Quebec (e.g.)? Otherwise, you would expect a French Front party to pop up. By putting them all in Quebec, you would expect competition within Quebec to create these cleavages Lipset speaks of.

Electoral system design in new democracies:

• PR is better. Closed-list PR is better than open-list PR. Electoral thresholds shouldn't be too high (probably not over 3%.