At a recent workshop a group of scholars analyzing the post-electoral state of a particular party system expressed concern about utilizing the term âparty fragmentation.â Typically employed to describe an increase in the effective number of political parties represented in a legislature, it conveys, according to these scholars, a negative connotationâ“a partition into âfragmentsâ, a certain brittleness. This claim holds that the term stems from a traditional preference for majoritarian institutions, which are more decisive, over more plural ones. Instead, these critics assert, plural representation is the very basis of democracy and an increase in the number of parties reflects the direct inclusion of more social sectors. To them, this does not represent “fragmentation” but a move towards a better form of democracy, a âconsensual democracyâ.

How valid are these claims? Can greater pluralismâ“expressed in a greater number of partiesâ“be equated with greater democratic quality?

If we value âsurvivabilityâ of the regime we must ponder the effect that the number of parties has on political stability. For if the proliferation of parties affects regime stability, opening the possibility of its degeneration into a non-democratic state, or a state of anarchy, what does this tell us about the virtues of increasing the number of parties? Would it not be right to consider this a âfragmentationâ?

Linz reminds us that, in presidential regimes, extreme multiparty systems exacerbate conflict between the executive and the legislature, which in the absence of an institutional âescape valveâ can lead to a breakdown, given the fixed nature of presidential terms. Extreme multiparty systems can also worsen polarization, diminishing the effectiveness of democratic government and leading to an opposition that encourages âirresponsibility and the politics of outbidding, culminating in the collapse of the center of the political spectrumâ (Coppedge 2012, 96). Fixing the point of inflection at which the number of parties becomes âextremeâ is difficult, and there are other factors that interact to determine the fate of a regime. But these are tendencies that can be plausibly posited to be likely as the number of parties increases.

From Arrowâs theorem we know that there is a tradeoff between social rationalityâ“understood as the ability to reach collective decisions that are coherentâ“and the concentration of power. When actors are many and their preferences heterogeneous, the probability of reaching collective decisions diminishes. Institutional rules that foster the aggregation of interests have the virtue of working against this tendency. They are not counter to pluralism, understood as diversity, as all groups can be represented. As Pitkin holds, representing means acting in the interest of the represented, in a manner responsive to them, and democratic mechanisms enable that this happens accountably. On the other hand, institutional rules that foster the disaggregation of interests must account for the difficulties they entail in terms of collective action.

To the degree that an increase in âveto pointsâ favors gridlock, the resulting stasis has implications for the quality of democracy. The inability of a regime to adequately respond to and satisfy the needs of its constituents erodes its legitimacy and, ultimately, its popular support. The resulting sense of malaise can lead to perceptions of government unfairness and erode the public trust. Under these circumstances, citizens may be more willing to dispense with democratic institutions when a messianic savior, or the military, offer deliverance through direct intercession.

So, while the notion of increased pluralism and government by consensus are intuitively appealing, they do harbor dangers for democracy. These dangers, as we have seen, relate to the stability and quality of democratic regimes. But, more fundamentally, the notion of consensus is dangerous in itself because it is indeterminate. What exactly do we mean by consensus? How is it reached?

Conceptually, these dangers increase with the number of political parties, which is why an increase in their number is not necessarily an unqualified good, and it is proper to refer to the phenomenon as fragmentation.