I’ve never thought much of the concept of political legitimacy, and a recent rereading of Seymour Martin Lipset’s Political Man has reminded me why. In theoretical discourse on political stability and change, legitimacy is the Ouroboros, the mythical serpent locked in a circle as it eats its own tail. We appeal to legitimacy when we need to explain the persistence of political arrangements that defy our materialist predictions, and when those arrangements do finally collapse, we say that their failure has revealed a preceding loss of legitimacy. In statistical terms, legitimacy is the label we attach to the residual, the portion of the variance our mental models cannot explain. It is a tautology masquerading as a causal force.

In the chapter of his 1960 classic that got me thinking about this topic again, Lipset writes (emphasis mine) that

The stability of any given democracy depends not only on economic development but also upon the effectiveness and the legitimacy of its political system. Effectiveness means actual performance, the extent to which the system satisfies the basic functions of government as most of the population and such powerful groups within it as big business or the armed forces see them. Legitimacy involves the capacity of the system to engender and maintain the belief that the existing political institutions are the most appropriate ones for the society.

Lipset goes on to argue that this concept has predictive power; if we know how legitimate a system is, we can anticipate whether or not it will survive challenging times, or what social scientists sometimes refer to as “exogenous shocks.” To illustrate this point, he contrasts the fate of various democracies in Europe in the face of the Great Depression. “When the effectiveness of various governments broke down in the 1930s, those societies which were high on the scale of legitimacy remained democratic, while such countries as Germany, Austria, and Spain lost their freedom, and France narrowly escaped a similar fate.”

Voilà, right? I mean, going into the 1930s, I’m sure every astute social observer could have told you that those four countries were the ones where “belief that the existing political institutions are the most appropriate ones for the society” was weakest; that citizens in neighboring countries did not harbor similar doubts; and that those variations in beliefs would largely determine the trajectories European countries would follow through the coming storm. Otherwise, this remarkably accurate after-the-fact prediction would be nothing more than common hindsight bias.

More damning, though, are Lipset’s brief comments on Thailand. After discussing mid-century Europe, he writes that

From a short-range point of view, a highly effective but illegitimate system, such as a well-governed colony, is more unstable than regimes which are relatively low in effectiveness and high in legitimacy. The social stability of a nation like Thailand, despite its periodic coups d’etat, stands out in sharp contrast to the situation in neighboring former colonial nations.

Looking back from 2012, that view of Thai politics seems almost laughably wrong-headed. In the past few decades, the regime that Lipset identified as an exemplar of social stability founded on high legitimacy has been wracked by periodic waves of mass unrest and separatist insurgency. The latest and still-ongoing wave of mass upheaval has now lasted nearly a decade and concerns precisely the question at the core of Lipset’s definition of legitimacy: are the existing institutions the most appropriate ones for Thai society? Based on their voting patterns and participation in mass protests, it now seems clear that many Thais think not, but an oligarchical alliance of monarchists, militarists, and well-to-do urbanites still has enough power to resist attempts to fully dislodge the old regime.

To be fair to Lipset, I suspect few observers of Thai politics in the early 1960s would have foreseen the ruptures that seem inevitable with hindsight (and if this article from a May 1996 issue of Time is exemplary of the information available at the time, it’s easy to see why). But then, that’s really the problem, isn’t it? If we can’t reliably observe legitimacy or know that it’s crumbling until people behave in ways that show it has, what value is it adding to our theories of political change?

In a terrific working paper that disassembles the problem far more thoroughly than this blog post, political scientist Xavier Marquez accepts that legitimacy may have some value as a summary concept in casual discussions of politics, but he shows that it just doesn’t work as an element of explanatory or even normative political theory. As Xavier puts it, explanations of political stability that appeal to legitimacy “are strictly speaking tautological: they do not so much explain stability as restate the problem of stability in different terms.” If you’re interested in comparative politics, you really should read the whole thing, but the following passage (emphasis mine) nicely summarizes his reasoning:

To the extent that the concept of legitimacy appears to have some explanatory value, this is only because explanations of social and political order that appeal to legitimacy in fact conceal widely different (and often inconsistent) accounts of the mechanisms involved in the production of obedience to authority and submission to norms. Very often legitimacy works as a residual concept, a sort of virtus dormitiva that is used to explain the persistence of social and political order wherever obvious coercion or material incentives appear unable to account for its stability. But like most such residual concepts, it tends to hide the wide variety of mechanisms that actually sustain social order, including epistemic deficits, collective action problems, signalling conventions, emotional attachments, and cognitive biases. I thus suggest that explanatory social science would be better off abandoning the concept of legitimacy for more precise accounts of the operation of these mechanisms in particular contexts.

Maybe I’m being too harsh on the concept. Analogizing to physics, maybe legitimacy is more like the dark matter of political development, a substance we cannot observe but whose existence we can infer from the otherwise strange behavior of human particles in visible political systems.

The problem with that analogy is that the theoretical models we have of social systems are nowhere near as well-developed and specific as the ones physicists have used to infer the existence of dark matter. No one has seen dark matter, but physicists can and have used careful observation of many related phenomena to develop a fairly sharp idea of what it is (and isn’t). For now—and maybe forever—social scientists have nothing that even comes close. Until we can find a way to reliably observe preferences and beliefs across a wide variety of cultural contexts, appeals to legitimacy are going to keep us stuck in a pre-scientific world, where things can be true because they just make sense.