Disclaimer. Don't rely on these old notes in lieu of reading the literature, but they can jog your memory. As a grad student long ago, my peers and I collaborated to write and exchange summaries of political science research. I posted them to a wiki-style website. "Wikisum" is now dead but archived here. I cannot vouch for these notes' accuracy, nor can I say who wrote them.
Andeweg. 2000. Consociational democracy. ARPS 3:509-36.
An excellent review of the consociational genre, addressing its main arguments and criticisms, as well as the causes and consequences of consociational democracy. A focus on Lijphart's work, since Lijphart has been the primary advocate for consociational and consensus democracy. Though it is not always clear in Lijphart's writings, consociational and consensus democracy are different; Andeweg begins by clarifying these differences.
Lijphart's early work (1977) spoke of "consociational" democracy, defined by four conditions: a grand coalition, proportionality (not just in elections, but in everything, e.g. cabinets, parliament, civil service, etc), mutual vetoes, and segmental authority (i.e. each social segment has its own sphere of authority, either territorially or functionally).
Lijphart later (1984, 1999) broadened this into the concept of "consensus" democracy, defined along ten (broadly institutional) lines, fitting into two dimensions. The executives-party dimension includes oversized cabinets, separation of powers, multiparty system, proportional representation, and corporatist interest group system; the federal-unity dimension includes bicameralism, decentralization/federalism, a rigid constitution, judicial review, and an independent central bank.
The four aspects of consociationalism are largely behavioral and are broadly defined; "they may find expression in the rather specific institutional arrangements of consensus demoracy, but they are not confined to these mechanisms" (p 513). "The difference between them is that consociationalism is the stronger medicine: while consensus government provides many [institutional] incentives for broad power sharing, consociationalism requires it" (Lijphart 1989).
Perhaps most significantly, consociationalism seems to assume a deeply divided society, while consensus democracy is advocated for any societal type, divided or not.
Nonetheless, most of the debates have concerned consociationalism, not consensus democracy.
Consociational theory began as an attempt to explain how democracy remained stable in the deeply divided societies of Switzerland, Austria, and other places. Over time, consociational democracy was discovered in several other countries, at least for a brief historical period: Colombia, India, Lebanon, Czechoslovakia, etc. In fact, the fact the consociationalism is often temporary raises the question of whether it is only a transitional regime type--and once it has tempered the effects of deep societal divisions (or once cross-cutting cleavages have developed), societies evolve beyond it into something more majoritarian.
Eventually, however, consociationalism became a normative theory. Lijphart was a participant in some of the commissions seeking to develop a democratic alternative to apartheid in South Africa, for example.
Lijphart's cases have been criticized on many grounds. For example, he presents Dutch society as deeply divided into pillars, but there may actually be considerable cross-cutting cleavages (Van Schendelen 1984); same with Switzerland. He's also been accused of mischaraterizing the institutional arrangements in some places. For example, the Swiss Federal Council is not really a grand coalition, since the coalition members do not see themselves as (or act as) leaders of a specific societal segment and since many decisions are made by majoritarian referendums, not accommodating elites (Barry 1975).
It's vague what many concepts mean. For example:
It's unclear where consociationalism comes from. This doesn't prevent consociationalism from being theorized about, but if Lijphart wants to convert consociationalism into a normative theory, then he should certainly be more explicit about where it comes from. (See Table 2).
Lijphart was most concerned that the mutual vetoes created by consociationalism could lead to indecisiveness, inefficiency, and ultimately deadlock (1977). But despite these short-term policy disadvantages, consociationalism would have long-term success in preserving stability.
Critics have focused instead on the quality of democracy (e.g. Lustick 1979; Rose 2000). "Absence of opposition, a predominance of elites, and mass political apathy do not suggest democratic vitality" (p 530). Lijphart has responded that "there is nothing in consociationalism that true democrats have to be ashamed of" (1985). He backs this up by citing Dahl (1971), who ranked countries according to their degree of democracy (polyarchy). Polyarchy has two dimensions, competition and inclusiveness. Although Lijphart's critics worry about competition, the inclusiveness of consociationalism ensures it a high ranking in Dahl's measure. This inclusiveness leads Lijphart to argue that consensus democracy is "kinder, gentler" democracy, based on women's representation, political equality, voter turnout, and so on (1999). But even when it comes to competition, Dahl looked at the right to run for office, free and fair elections, and the responsiveness of institutions to votes. Possibly excepting the latter criterion, there is no reason that consociational regimes cannot meet the other criteria of competitiveness.
Research on similar subjects