Propaganda as Signaling

In a lovely piece published nearly three decades ago, the French historian Paul Veyne noted that much “propaganda” art throughout history has been “without viewers.” His key example was Trajan’s Column:

“Trajan’s Column Panorama” by Juan Francisco Adame Lorite. Licensed under CC BY 3.0 via Wikimedia Commons.

Not far from the Forum in Rome, Trajan’s Column raises its shaft thirty meters. Spiraling around it is a sculpted frieze whose 184 scenes and one thousand figures illustrate, like a cartoon strip, the conquest of Dacia by Trajan. Except for the first two spirals, viewers cannot make out these reliefs. Archaeologists study them with binoculars. Moreover, nobody would want to itemize this repetitious swarm or try to follow the account of military campaigns declaimed by the conquest of barbarian villages whose name or place on the map was unknown. Historians explain Trajan’s Column as a work of “imperial propaganda”. That shows how much a shortsighted rationality, one that cannot distinguish between expression and information, keeps its prestige even to our day, when it brings something to “society” or states what this thing is assumed to “bring to society”. We may however doubt that the Romans of Trajan’s time looked very much more at the reliefs, materially invisible, than today’s Romans and that they rushed to this spectacle to go around the Column twenty-three times with their noses in the air. The Column does not inform people; it simply lets them see the evidence of the greatness of Trajan faced with time and the weather. In the same way, at the summit of the Behistun Rock, Darius the Great had a monumental inscription engraved in three languages to the glory of his reign. This inscription was not meant to be read: it is located at the top of a peak, and only eagles or mountain climbers suspended on their ropes could read it (p. 3)

Behistun Rock inscription by Hara1603. Licensed under Public Domain via Wikimedia Commons.

The lack of clear visibility of the great works of imperial art is so common, according to Veyne, it is hardly noticed, and when noticed it is explained as if it were some embarrassing technical fault. (The sculptor of the Vendôme column, a “faithful imitation” of Trajan’s column constructed for Napoleon, published a book to explain the meaning of the reliefs, since these could not be seen from below; he thought the book “might be helpful” given their invisibility. One wonders how many people read the book). Though Veyne did not use this terminology, his argument was that such works were a form of costly signaling:

The Column expresses the glory of Trajan, just as the heavens (which it is useless to itemize star by star) express the glory of Jahweh. In both cases there have to be far too many stars and far too many sculpted scenes. The expression of a superiority is only undoubted when it is excessive (p. 3)

Imperial art was thus not a way to transmit some specific ideological content to “legitimate” the social order, and its political force did not depend on any understanding of its meaning:

What the Column bears as ideology is the right it claims to exist, just as, in a country submitted to an authoritarian regime, loud-speakers diffusing official discourse in the streets count more for their omnipresence than for what they broadcast. Trajan’s Column is propaganda of a sort but not because of its imagery. It is such for its presence and for the power expressed by its redundancy (p. 11)

The same was true, mutatis mutandi, of most (especially political) rituals, whose meaning, painstakingly reconstructed by the anthropologist, is only dimly recognized, if at all, by participants. Much ritual activity, in Veyne’s view, was “conduct without belief,” as the title of the article had it, not because participants actively disbelieved or resisted what was said in and through the ritual but simply because they did not have the foggiest idea about its meaning:

Their multiplicity of meaning and the feeble intensity of the meaning most generally received make these ceremonials a behavior that functions at only about ten percent of their energy[,] and that meaning is not the one involving their content and what their creator intended. It is not the words of the Marseillaise that matter, when the day-nursery is inaugurated with music (p. 13)

Veyne’s point was not that official rituals are necessarily unconvincing or incredible, but that most of the time most people just aren’t paying enough attention to “get” the official message. (This point should not be a surprise to anyone who has ever given a lecture to a crowd of indifferent students). The meaningful content of official ritual and art matters much less than its limited emotional charge and oppressive bulk. For example, the Stalin cult required enormous resources to maintain, and much specialist energy was devoted to its symbolic construction. Yet despite the vast efforts of historians to understand the iconography of the cult (see, e.g., Jan Plamper’s The Stalin Cult), most Soviet citizens only engaged with cult productions superficially and without enthusiasm, hardly in the way required for any detailed understanding of their meaning. The Stalin cult was powerful primarily because Stalin was powerful, not because it had some specially designed symbolic content that most people “got.” A similar thing could be said of the later rituals of Soviet power, described with such care in Christel Lane’s The Rites of Rulers, and intentionally designed, on her account, to impress Soviet citizens with a wide variety of explicit values. Yet there is little evidence that the symbolism of these rituals had any but the most passing influence on the values and attachments of Soviet citizens to the political order; to the extent that they worked to “legitimate” the regime (as Lane claimed in a 1984 piece), such legitimation was predicated on the “signal” of the permanence and power of the CPSU, which was soon to vanish. As Veyne puts the key point more poetically:

We must […] be careful not to infer from the ceremonial of coronation of kings, for example, what monarchy is and what is thought of it and to bring grist to the mill of the ideological analysis of symbols. This ceremony does not show us the real visage of monarchy: it is merely a portrait by a court painter. The subjects of the king in all probability think something different of the monarchical regime. Even more probably, they think less of it: every portrait painter embellishes, interprets and defines the features of the model. (p. 14).

As long time readers will know, I think the “signaling” view of official ritual and propaganda is more often closer to reality than the “indoctrination” view, so Veyne’s ideas cater to my prejudices. Yet the “signaling” view of propaganda goes against the commonsense view that authoritarian ritual and propaganda “works” insofar as it indoctrinates or educates. (For example, my students are enormously resistant to it; I sometimes think that there must be some evolutionary benefit to believing that other people always believe what they are told, given my difficulties in convincing them otherwise). So I was interested to learn of new piece forthcoming in Comparative Politics, fittingly entitled “Propaganda as Signaling,” by Haifeng Huang, which provides further evidence for the signaling view.

The study looks at students exposed to mandatory political education courses in a Chinese University. These courses are seen as a pain:

Chinese students in general and even many instructors regard such courses as nuisances, i.e., rituals that they dislike but have to observe. Students also typically regard the courses as useless for their future careers. When asked how they treated the political education courses, only 8.0% of the students surveyed in the study reported they somewhat actively studied for the courses, with the rest acknowledging that they listened to lectures only casually, did not listen to lectures at all, relied on cram sessions to prepare for exams, or simply skipped some classes (p. 9)

The clever bit of the study exploits the fact that, since nobody likes these boring courses, student performance should thus depend primarily on their incentive and ability to maintain a high GPA. Conditional on academic standing, family income, Communist party membership, and the like, their satisfaction with the government should thus be unrelated to their recall of propaganda content. And indeed this is what Huang finds: people who do better at recalling propaganda content glorifying the CCP (the “good students”, let’s say) are no more likely to be satisfied with the government than people who do not recall such content. But the “good students” do appear to show a diminished willingness to challenge the CCP through dissent actions. They are more likely to believe, in Huang’s view, that the government is strong, even if it is not good.

Now, Huang does not dismiss the possibility of propaganda as indoctrination, though such socialization into regime values would happen in the more lively public sphere and by more indirect means. Moreover, the exact mechanism by which the better students believe that the government is strong is not altogether clear from the paper. (Perhaps they are more likely to attribute their boredom to the government’s ability to compel their attention, and thus draw inferences about the government’s strength? Perhaps their “ability” means they are simply more likely to form accurate beliefs about the government’s strength, irrespective of their feelings of satisfaction?). But the basic point is surely correct: power “legitimates” power in this case. Indeed, I suspect the very popularity of nationalist symbolism in the wider Chinese public sphere shows the same thing: it’s not that the government is powerful because it has been able to craft some very specific nationalist narrative that cleverly appeals to people’s values, but its ability to project strength makes the nationalist narratives a bit like Trajan’s column: a reminder of the CCP’s apparent permanence and overwhelming strength.

Electoral Parodies

I've been reading a lot about various dictators this summer. And again and again I am struck by the existence of what one might call “electoral parodies” in many dictatorships: abuses of electoral rules and standards so blatant and obvious that they cannot be interpreted as anything other than mockery. Here is a good example, from the Dominican Republic:

In 1941, as the new ally of the United States against the Axis Powers, Trujillo felt obliged to extend the democratic facade by creating an opposition party, and so the Trujillo party was formed. But Trujillo was the presidential candidate of both the new creation and his old official party. Under the Trujillo Party banner, he received 190,229 votes and as the candidate of the Dominican Party he polled 391,708; the total of both parties, 581,937, meant that Trujillo had again received 100 per cent of the vote (Wiarda, Dictatorship and Development: The Methods of Control in the Dominican Republic, p. 66.)

Others include such things as releasing the results of the election before it takes place, receiving 100% of the vote from more than 100% of the voters (another Trujillo specialty), declaring victory while failing to announce any vote totals, and so on. But my all-time favorite is this story, from Haiti. François “Papa Doc” Duvalier had been first elected to the Presidency in 1957 for a six year term in a rigged but reasonably competitive election, and had since consolidated his power. On 30 April 1961, there was a scheduled congressional election. Although Duvalier was not supposed to be on the ballot,

[f]ew voters considered it unusual that the name Dr François Duvalier was printed at the bottom of each and every ballot … Late that evening rumour spread that Duvalier – with two years to go on his current term – had declared himself reelected for an additional six years because his name had appeared on the ballot. On 4 May Attorney General Max Duplessis declared to the Electoral Board – called the Census Committee – that Duvalier indeed had been voted another term in office. Crowds were organized to collect before the palace and applaud this view. … … In three days the Census Committee convened, agreed, declared President Duvalier re-elected, and announced that he received more than 1.3 million votes … … The new legislature ratified the election (14 May 1961), and Duvalier responded: ’I accept the people’s will because being a revolutionary I do not have the right [not? sic] to hear the people’s voice (Diederich and Burt, Papa Doc: Haiti and its Dictator, pp. 169-170)

The sheer chutzpah of this “election” is a thing of beauty. Voters voted Duvalier into office without realizing that they had done so! He wasn’t supposed to even be up for election!

I’ve written before about elections without choice, but this stuff still fascinates me. Clearly, much of what is going on in these “elections” is about signaling – showing who is boss by doing something outrageous and getting away with it – but I am struck by the degree to which people who should know better still interpret them as somehow “legitimating” performances. There is something like a “third person effect” going on here, as if analysts thought that electoral parodies must have a greater impact on gullible others than they do on themselves. Yet electoral parodies are not “democratic facades” that legitimate regimes in the eyes of ill-informed citizens or commentators from abroad, but “democratic burlesques” that ridicule any possible standards of democracy, and are clearly experienced as such by most citizens and international audiences.

The Duvalier case is especially interesting, since it is so absurdly brazen, and if anything contributed to the “delegitimation” of the Duvalier regime in the eyes of international audiences. Unlike in the Trujillo case noted earlier, the “re-election” of Duvalier did not come about because of pressure from abroad; in fact the US did not know it was going to happen, and made it clear through its ambassador (the aptly-named Robert Newbegin, who was supposed to inaugurate a new era of Haitian-American relations after the expulsion of the previous American Ambassador, Gerald Drew, in 1960) that it considered it illegal. As the end of Duvalier’s original term approached in 1963, the US even put out the message that it would welcome a military coup to oust him, since it considered Duvalier’s legal term to be over. (Though various ineptly planned coup attempts and a number of poorly executed invasions from the Dominican Republic did take place over the next few years, Duvalier survived them all. The man was lucky, and his opposition was almost comically disorganized.) Most other nations in the Organization of American States, though generally reluctant to intervene in Haiti, agreed that Duvalier was an embarrassment, and his “re-election” a sham.

Worse, not only did Duvalier further strain his (already strained) relations with the US by insisting on his “re-election,” but he did not even gain any of the benefits of normally rigged elections, like (biased) estimates of opposition or government strength, or the ability to make a more or less plausible claim of popular support, since nobody knew that he was going to be on the ballot; it wasn’t supposed to be a presidential election! From the point of view of contemporary analysts, it looked like an unforced error; why couldn’t Duvalier just have waited for two years to steal the next scheduled presidential election in a proper way? It’s not as if Duvalier had no genuine supporters; the “quiet doctor,” underestimated by the traditional Haitian elite at every turn, appears to have skillfully appropriated the noiriste and populist mantle of Estimé, and could count on at least some genuine support in the countryside. And he could play the nationalism card to perfection, presenting Haiti as the injured victim of US racism and imperialism while extorting large amounts of aid by playing on US fear of communism. (He would occasionally invite the Polish or Czechoslovak legations for a chat, or let communist student unions stage the odd protest, just to show the Americans the dangers of not agreeing to his demands; the man had a singular talent for blackmail). But the “election” actually endangered the flow of aid (which, it should go without saying, was in large measure used for pure patronage purposes or simply stolen). It seemed to be more like an impulsive f*** you than a rational attempt to retain power.

My best "rationalistic" guess for why Duvalier nevertheless staged this electoral parody is that given the totally uninstitutionalized Haitian political context he might have thought that waiting two more years to steal an election might provide his enemies with too many opportunities to overthrow him; better to flush them out by an unexpected, enraging move like this than to give them time to prepare. And it certainly changed the political opportunity structure, since it entailed the cancellation of the scheduled presidential election of 1963, and thus foreclosed the option to battle Duvalier on an electoral terrain, however uneven; perhaps by forcing the opposition to adopt the techniques of the underground, he was moving the fight to a place where he had greater advantages, though this was a high-risk gamble. 

And yet all actors involved were forced to speak, with a straight face, the language of legitimacy, laws, rights, and popular sovereignty. Here, for example, is a bit Diederich and Burt quote from the Attorney General’s ruling, a masterpiece of this nonsensical rhetoric:

It is essential that immediately, even before counting the votes obtained by the candidates, we insist upon the principle which, far from being new, constitutes the very essence of national sovereignty. This principle can be defined: the members of the electorate, when presented with an electoral decree or law which looks to them incomplete, have the right to fully manifest their will to complete the law or decree and to designate a civil servant or a group of civil servants whose election was not foreseen. The intangibility of the principle once established, it will be easy for us to understand the the electorate of Port-au-Prince Arrondisement (used to establish legal precedent) has acted within its full sovereign power by designating François Duvalier for a new term under the title of President of the Republic. (Diederich and Burt, Papa Doc: Haiti and its Dictator, p. 169)

The language of legitimacy coordinates expectations even as it is being obviously flouted. Hypocrisy may be the tribute that vice pays to virtue, but this is not even hypocrisy; it’s more like trolling.

The Irrelevance of Legitimacy

(As I intimated a while ago, I’ve grown weary of the concept of legitimacy. This is an experiment in thinking about how one might understand political life without recourse to this idea, or with a very different version of it).

Both everyday and academic explanations of uprisings and revolutions tend to make heavy use of the concept of legitimacy. For example, a common argument suggests that some of the regimes of the Middle East (Egypt, Tunisia, Libya) collapsed in part because they had long forfeited their legitimacy due to the abusive ways in which they treated their own people, whereas others (Morocco, the absolute monarchies of the gulf) are likely to weather the current crisis because their governments are still considered legitimate by the majority of their populations. (A very sophisticated version of this argument can be found in this piece by Jack Goldstone). More generally, I often come across arguments for the view that some action or discourse “legitimates” certain forms of power and thus helps sustain it, or conversely that the breakdown of particular power relations can be explained (at least in part) by pointing to the fact that people have ceased to consider them “legitimate.” Yet I find most of these explanations for the maintenance or breakdown of regimes unsatisfactory. They seem to amount to little more than saying that regimes (or, more generally, relations of domination) endure so long as they are accepted by the ruled, and when they are not, they don’t. But this is not obviously true.

For one thing, it is not empirically clear that “acceptance” needs to be very deep to sustain many forms of domination and oppression. Consider the variety of ways in which we might say that someone accepts their domination. For example, a person might sullenly submit to some oppressive institutional order because of his or her inability to imagine a different one; or (more commonly) because of his or her inability to mobilize collective action in favor of some alternative order (they face a coordination problem); or because despite the fact that a different institutional order would be better for a large group, it is individually “rational” for individuals to defect from collective efforts to change the current institutional order (they face a standard “prisoner’s dilemma”); or because the institutional order so shapes his or her interests and identity that they find challenges to the order against their “long-run” interests (their interests and those of the order are ultimately aligned, though they still think of themselves as oppressed); or (very rarely) because they think that the institutional order that dominates them is right and just. Many obviously oppressive social orders are not believed to be right and just by majorities of the dominated and yet they endure for a very long time, at least if we believe studies like James C. Scott’s Domination and the Arts of Resistance or Weapons of the Weak, which thoroughly document the fact that peasants in many agrarian societies do not come to accept their domination as rightful and just in any sense. The same is true of most authoritarian regimes, where preference falsification is often rampant, as I indicated in passing in my cults of personality post. Yet it is also clear that many people in such orders also accept their domination in a loose sense of the term: they conform publicly, they vote, they don’t rebel, they even contribute to the maintenance of the oppression by denouncing others or taking advantage of opportunities to climb the social ladder at the expense of others, and so on. Should we say that these social orders are considered “legitimate”? I say no: they should not be considered legitimate in any empirically relevant sense (let alone in any normatively relevant sense, but that’s another story).

The basic (and probably correct) intuition behind the use of the concept of legitimacy to explain the endurance of oppressive regimes or social relations is the idea that power – and more specifically, relationships of domination - cannot normally be sustained by private incentives (payments and threats) alone; “something more” is necessary if domination is to endure for any length of time. Domination is involved in relationships that are typically contrary to the interest of at least one of the parties, and hence it is likely to be resisted whenever the opportunity presents itself. Domination thus often requires repression to sustain the relationship, but repression is costly and often ineffective over the long run if the relationship is supposed to induce the cooperation of the dominated in some productive endeavour; hence domination needs to be “legitimated,” i.e., needs to be based on a set of shared and relatively stable beliefs that enable those who benefit from relationships of domination to direct the actions of those who are in subordinate positions with a minimum of repression even when such direction is against the interests of the latter. When domination is actually sustained in the long run, the argument goes, it must be because those who are dominated somehow accept their domination, however grudgingly, and in particular because they believe certain things about the people or institutions that dominate them (e.g., that the powerful have a “right” to command, or that they have a special sort of charisma, or that certain institutions represent the natural order of things). Conversely, if domination breaks down, then it must be because the dominated have stopped believing these things. (Note that I am talking here about “empirical” legitimacy, not “normative” legitimacy: I am interested in the role the concept of legitimacy plays in explaining domination, not in whether particular relationships of domination are legitimate in some interesting normative sense. Empirical legitimacy claims to be about the beliefs that people actually do have, not about the beliefs that they ought to have).

But this general understanding of how domination requires legitimacy is, I suspect, incorrect or at least fundamentally confused. Though belief may on occasion help sustain domination, the idea that domination is always sustained by (shared) belief is not true. At the very least, the majority of the mechanisms that sustain relationships of domination over the long run are not reducible to beliefs in the rightness or charisma or naturalness of certain people or institutions. For example, we ought to distinguish between a belief in a lack of alternatives (which may make people sullenly conform to a social order they deem oppressive), and a belief in the rightness of a social order, or between the idea that some of the mechanisms through which people are dominated are “hidden” and the idea that these power relationships are considered to be legitimate.

In order to make some progress on thinking about this problem, it may be useful to take a somewhat lengthy detour into Max Weber’s Economy and Society, where Weber rigorously tries to define the idea of empirical legitimacy in order to explain what constitutes a “social order” or a regularity of social action (feel free to skip the next 3-4 paragraphs if this is not your cup of tea). This is still the standard understanding of legitimacy in the social sciences (though it is not the only possible one, other conceptualizations of legitimacy typically draw on it), so it is worth examining in some detail (and improving on it, if possible). At the very beginning of the work, Weber analyzes how different sorts of reasons for action (“micromotives,” to use Schelling’s term) give rise to and disrupt different kinds of social order (“macrobehaviors” or patterns of social action). According to him, there are only three kinds of social order (I draw here on Habermas’ interpretation of Weber’s thought in his Theory of Communicative Action, especially I.ii.1, pp.157-185 and I.ii.4, pp. 254-271; the relevant passages of Weber are mostly in Economy and Society I.i, especially sections 4-6):

1. In habitual orders, social regularities emerge and are kept in existence through the unthinking inertia of everyday activity (“habitual action”); reasons do not play a motivating role in their creation. Such orders are not stable, however, to the introduction of reflection; when people think about what they are doing, they may sometimes act differently, transforming their social order into an interest-based or a legitimate order.
2. In interest-based orders (such as markets, though Weber does not think that markets are sustained purely by interest), social regularities emerge from the mutual adjustment of the activities of more or less instrumentally rational actors engaging in tactical and strategic activity in pursuit of their various interests. Reasons here play a motivating role in the emergence of social regularities, but only privately: each actor has his or her own (different) reasons for acting as he or she does in pursuit of his or her interests, yet social regularities still emerge from the private adjustments each actor makes to his or her behavior in light of the actions of others. In modern game-theoretic terms, the social order is an equilibrium of some game, given people’s (potentially heterogeneous) private incentives for action. Weber thinks, however, that pure interest-based orders are generally unstable, and are thus often stabilized by what we might call “shared” or “public” reasons, which transform interest-based orders into “legitimate” orders properly speaking.
3. In legitimate orders social regularities emerge from the shared acceptance by agents of certain reasons for action or inaction. These reasons are usually understood as normative constraints on the kinds of courses of action that agents might privately consider, such as for example beliefs about the validity, justice, fairness, or virtue of particular actions or norms. Here reasons produce social regularities not through their being held privately (as in interest-based orders) but through their being shared or “public” reasons that can rule out of bounds or override, so to speak, certain kinds of private reasons for action (e.g., reasons to revolt). Weber goes further and identifies three basic ideal kinds of legitimate order: the traditional order, where the legitimating reasons refer to the supposed naturalness of an institution (people fail to imagine alternatives; this is the legitimate equivalent of a habitual order); the charismatic order, where the legitimating reasons refer to the special qualities of a person or persons from whom rules issue; and the legal-rational order, where the legitimating reasons refer to the special qualities of a set of rules (which can themselves be used to generate further rules). These ideas are rather abstract, but the basic point is, I think, generally comprehensible. In a traditional order, when one asks the question “why do you obey/submit/follow somebody’s commands?” the expected answer is “because that is the way we do things here” (not “because otherwise I will be beaten over the head with that stick”). Similarly, in a charismatic order, the expected answer is “because my leader or messiah told me,” and in a legal-rational order the expected answer is “because (other) rules authorize it” or “because the rules are just (i.e., in accordance with the rules or principles of justice).”

But what does it mean for reasons for action to be shared or public? The obvious interpretation (and the one that Weber seems to prefer) is that reasons are shared and public to the extent that many agents believe the same thing about why they or their rulers should or should not act in particular ways. In game-theoretic terms, we might say that in legitimate orders social order emerges not from the private adjustment of behavior by participants in a game according to their private reasons for action (the expected "utility" they might derive from acting in one way rather than another), but from the shared beliefs of participants about the rules of the game, which limit the strategies available to them: a legitimate order is simply one in which enough participants accept the rules of the game as rules (and not merely as constraints imposed on them by the action of others in equilibrium). Empirical legitimacy enables a social order to economize on coercion both by consistently narrowing the range of the possible strategies open to actors and by clearly signaling any violations of the  rules (and hence enabling violators to be more easily punished). This sort of legitimacy is also conceptually related to trust: a government that is illegitimate is one that violates the (shared) rules consistently enough to lose the trust of the population.

One might want to say that to the extent that some people believe in the legitimacy of a social order (that is, believe that the rules structuring the social order, or the persons who have authority to produce such rules, are somehow the right rules, or the right persons), that social order will be made more “resilient” to changes in the habits or private reasons people may or may not have for conforming to the demands of the social order in question (e.g., rewards/penalties and the probability of physical punishment for nonconformity). How else should we explain the enormous investment that dominant groups make in the deployment of what we might call (I’m trying to use this term neutrally) “ideological” resources (persuasive arguments, sophistic arguments, propaganda, cults of personality, the mobilization of scientific or other authoritative discourses to “naturalize” certain institutions or practices, etc.)? How else should we think of these efforts to “legitimate” particular social orders, if not by seeing them as efforts to change people’s beliefs (by means fair or foul, good-faith persuasion or underhanded manipulation) so that they conform better to a particular social order? But then: should we really say that when Robert Mugabe holds a rigged election in Zimbabwe, for example, he is attempting to legitimate his rule?

The key to a different understanding of this problem lies in taking seriously the idea that supposedly “legitimating” beliefs are shared. What matters for (what we normally call) charismatic legitimacy, for example, is less the belief that some particular person is a demigod but that this particular answer to the question “why do you submit/obey/do X/etc.?” becomes expected in some group, and that not giving the expected answer singles one out for sanctions, exclusion, and other bad things. And just as we can have the spectacle of charismatic legitimacy without charisma, we should be able to see legal-rational “legitimacy” in the midst of corruption, or (I suspect) traditional “legitimacy” even when there is widespread awareness of the newness of tradition. Legitimacy is thus underpinned primarily by signals, not beliefs: those who do not provide the appropriate answer in the right circumstances (or rather, those who do not provide a credible answer) identify themselves as violators. Even the corrupt bureaucrat gives lip service to the law, whatever he or she may believe privately, and the irreligious king still takes seriously the traditional liturgical forms.

“Legitimate” social orders thus function like a signaling system in which rulers and ruled, dominator and dominated, both provide (credible) signals of their commitment to particular rules or persons or existing practices; and insofar as the system “works” it produces authority, i.e., it identifies certain people or rules or forms of speech as precisely those people or rules or forms of speech that one is expected (by other people subject to the social order) to follow or use in particular circumstances. But “credible” signals are not always “true” signals; that a signal – a particular answer to the question, “why do you submit/obey/ do X?” – is taken as credible by a relevant receiver does not mean it actually reflects some deep belief about the rightness or justice of the system (though it of course may). Credibility in a signaling system may be achieved in many ways, only some of which involve any sort of belief in the content of the signals.

What matters [for explanations of social change] are the conditions under which alternative legitimacy claims can emerge as focal points for new signaling systems, or under which the signaling equilibrium is disrupted. Here traditional explanations based on legitimacy have somewhat more to say: there are many common conditions under which, for example, the failure of authorities to demonstrate a credible commitment to norms of justice to which a population is committed produces anger and in turn triggers activities that reduce the costs of coordinating signals of commitment to a different social order (a different set of rules or persons). And from this point of view, “ideological” investments (or such practices as blatantly rigged or fraudulent elections, or unbelievable cults of personality) are thus useful not so much because they make people believe in the rightness or special qualities of particular people or institutions, but because they prevent the emergence of alternative focal points for legitimacy claims - e.g., because they destroy the common knowledge necessary for collective action, or because they are too salient for other foci of collective justification to easily emerge (they colonize public space).

But if this analysis is even partially correct, then it seems to me that legitimacy in the traditional sense (as beliefs in the rightness of people or institutions) is irrelevant to the explanation of political phenomena such as revolutions. Legitimacy still matters normatively – we want to live under social orders that are just or fair – but not so much for explaining social change. 

Nauseating Displays of Loyalty (Towards a General Theory of Sycophancy and Related Phenomena)

An anonymous reader points me to a very interesting paper by Victor Shih on "'Nauseating' Displays of Loyalty: Monitoring the Factional Bargain through Ideological Campaigns" (Journal of Politics 2008, vol. 70(4) pp. 1177-1192 [ungated]):

Autocrats, as factional patrons, only find out the true loyalty of clients during a serious political challenge, when they are least able to enforce the factional bargain. In autocracies with norms against cults of personalities, public, exaggerated praises may constitute an alternative way for clients to signal loyalty credibly. By suffering the social cost of being despised by others, sycophants credibly signal their affinity to a particular leader, thus deterring factional rivals from recruiting them into an alternative coalition. This article develops a measure of such displays of loyalty in China through content analysis of provincial newspapers between 2000 and 2004. OLS and PCSE estimations are used to inquire whether provincial faction members were more likely to echo an ideological campaign launched by their patron. Further analysis explores whether faction members in rich and poor localities echoed the campaign in different ways. The findings suggest that ideological campaigns function as radars that allow senior leaders to discern the loyalty of faction members.

The argument here is in interesting contrast to what I was trying to say in the post on cults of personality. The problem with cults of personality is that the "signal" of loyalty the dictator gets from followers is often uninformative: if everyone says that the dictator is a god, then the dictator cannot distinguish who is loyal (who will stand by him in a crisis, or at least not rebel if given the opportunity) and who is not. Mere praise in such circumstances is "cheap talk." So the leader has a incentive to develop some ways of making praise costly if it is to serve as a signal of loyalty (where loyalty is understood as a certain level of commitment to support the dictator, or at least not to support challengers). But where can this cost come from?

In my post on cults of personality, I argued that the cost comes precisely from the very dynamics of the strategic situation: because the dictator knows that the extravagant praise is uninformative as a signal of loyalty, he demands ever more bizarre performances, and in particular demands that one denounce those who show insufficient enthusiasm for the ever more bizarre performances. To the extent that most people do find it costly to deny reality and denounce others (especially if those others are friends and family), the signals retain some information about the level of commitment of the population to support him, or at least to acquiesce in his rule (given also the costs of not praising the dictator). The level of extorted praise serves as a gauge of the effectiveness of extortion. (Especially when the extorted praise includes denunciations of others: this is what it means in practice to support the dictator, i.e., to be loyal. It has little to do with liking the dictator).

To be sure, as Bernard Guerrero notes here in an interesting response, it is possible that what happens is that you get a sort of "arms race," where ever more bizarre performances are required as old performances lose their information content (because everyone eventually does them). Yet it does not necessarily follow that the signals from the cult lose all their informational value immediately; and as many dictators well know, a cult of personality has to be constantly refreshed. Propaganda is never-ending work. Moreover, even if the cult does not work well as a gauge of support, it can still produce loyalty directly (if some fraction of those exposed to it come to believe in the leader's charisma, which increases their commitment to support him) and it can prevent coordination, so that even if people actually hate the dictator, the cult still prevents them from plotting to overthrow him because they can't gauge other people's feelings. (For a somewhat different if related take on this, emphasizing the ways in which cults implicate the population into supporting the ruler even when they do not actually believe in the leader's charisma, see Lisa Wedeen's superb piece on Syria's Hafez al-Assad and his cult of personality, also recommended to me by a reader. The anecdote that opens the piece is priceless).

Which of these functions of the cult of personality as a tactic of power (gauging loyalty, producing loyalty, and preventing coordination) is most important is a complex question, whose answer probably depends on particular features of the strategic situation facing the dictator. (I'm writing a paper on the topic, so I hope to come to more definite conclusions in the future). I suspect, however, that the direct production of loyalty is the least important function; it seems exceedingly unlikely that calling Assad pere "the World's greatest dentist," as a friend told me apparently happened in Syria in the 80s, was ever seriously intended to persuade people of his charisma. Moreover, I think (for reasons that will become clearer in a second) that perhaps cults of personality are most useful to the dictator when he fears revolutionary threats (threats from outside the ruling elite) more than he fears coups (threats from inside the ruling elite), perhaps because he has been able to sufficiently consolidate his power at the expense of this elite. (Though there's a chicken-egg problem here, for the cult of personality also seems useful as a tactic to consolidate power, as it appears happened in Mao's China and Qaddhafi's Libya). There is after all a tension between the loyalty-gauging and the coordination-prevention uses of the cult, because the cult works best to prevent coordination when the costs to not praising the dictator are much higher than the costs of praising him, whereas it works best to gauge loyalty when the costs of praising him are not insignificant (though both costs could be and normally are high: not praising may entail jail or worse, but praising may entail denouncing loved ones or engaging in humiliating behavior). This means that the dictator may wish to relax the cult if he needs to gauge the loyalty of his close followers (who will help him against his people) more than he needs to prevent coordination among them. One might add that dictators don't always need very precise knowledge of the level of loyalty of the general population (and at any rate there are often other indicators of their likely level of loyalty, like protests, informers, surveys, the level of unemployment, etc.), in which case the coordination prevention and loyalty production functions of the cult becomes more useful vis a vis the general population than its loyalty-gauging uses.

Shih's paper nevertheless helps us understand how mechanisms similar to the cult of personality can help autocratic leaders gauge the loyalty of their close followers (not so much of the population as a whole). His focus is on the "ideological campaigns" that one sees in many communist countries, and especially in China, such as the "Three Represents" campaign during Jiang Zemin's tenure (opening the party to businessmen), or the "Harmonious Society" campaign that is still going on. Such campaigns typically present the thoughts of some particular leader as some momentous and utterly brilliant contribution to philosophy, and they constitute a standing invitation to sycophants, who say things like this:

‘Comrade Jiang Zemin’s thought concerning the "Three Represents" is like a giant building that overlooks the whole situation and contains rich content and deep meanings. It is a creative usage and development of Marxist theory and is strongly theoretical, scientific, creative, and practical. (Yang Yongliang, the vice-secretary of Hubei, quoted by Shih).

But how is this sort of thing useful to leaders? The problem a leader faces here is that he needs to cultivate his supporters by paying them in various forms; but until the chips are down, he does not necessarily know who will in fact help him in such circumstances, because there are no regular opportunities to test their loyalty (like elections in democracies), and after a crisis he may not be around to punish actual disloyalty. So the leader really does need to gauge the loyalty of his clients if he fears potential revolt from below or attacks from other factions, but even extravagant praise does not reliably indicate a credible commitment to support him in times of crisis.

Shih argues that in modern China (post-Mao) extravagant praise has retained its informational value as a signal of loyalty precisely because top leaders have supported norms against cults of personality (a norm that existed before Mao consolidated his power and which was supported by the top leadership after he died as a preventive measure against attempts to concentrate power in similar ways). When there is a norm against cults of personality, the stigma of violating it (and being known as a groveller) is a sufficient cost to ensure that the "praise" really is a credible signal of loyalty to a patron, especially when there are few other options to provide credible signals of loyalty (like, e.g., providing business opportunities for the leader's family or extending extravagant "hospitality" to the leader when he comes to visit your city). The norm seems to exist not only or even at all to prevent concentrations of power, but because top leaders gather useful information from its violation. So leaders launch "ideological" campaigns (like the "Three Represents" campaign) in order to see who will violate the norm against cults of personality.

This is a very clever piece of research. The key fact that Shih exploits to support his thesis is the degree of variation in the extent to which ideological campaigns are echoed by party newspapers around China. In particular, he shows that during the "Three Represents" campaign, newspapers in provinces linked to Jiang Zemin's clients were much more likely to echo it than other newspapers, but only if the province apparatchiks had few other means to signal support. So party newspapers in richer provinces (like Shanghai) which could offer Jiang other signals of support (like business opportunities for his family members or special hospitality when he came to visit) were less likely to exhibit "nauseating" displays of loyalty (the phrase comes from one of the people Shih interviewed, and reflects the anti-cult of personality norm current in today's China) than party newspapers in poorer provinces (which were more dependent on central government support), allowing Jiang to keep tabs on the loyalty of his poorer clients. And in provinces which were not linked to his faction, there were far fewer nauseating displays of loyalty. (One could quibble with a few things. For one, I am unsure how good Shih's measures of whether a province's leaders could be said to be part of Jiang's faction are. But I'm no China specialist. And there is a question as to how useful those extreme displays of loyalty really are to the leader).

The more interesting general point that comes out of these sorts of studies, for me, is how little traditional ideas about "legitimacy" matter for explaining support in all sorts of regimes. Support seems explainable in many cases as a result of signalling equilibria, whereas the traditional Weberian ideas about traditional, charismatic, and rational legitimacy seem to play little role. In fact, I have a hunch - not well developed - that one could understand what is traditionally called "legitimacy" in terms of various sorts of signalling equilibria, and not much would be lost. But that would require a much longer post to explain, and perhaps a paper.

Qaddafi's Chickens

[In 1977] the Libyan leader suggested that in order to achieve self-sufficiency every Libyan family had to raise chickens in the home. The cages and birds were imported and, for an obligatory fee of fifty-seven Libyan dinars ($150 at the 1977 exchange rate), were distributed by the government to Libyans. To many city dwellers in small apartments raising chickens in their kitchens was a difficult if not impossible affair. The result was that many ate the birds and found other uses for the cages.

...General Qaddafi's declaration that Libya must achieve food self-sufficiency was justification enough for his aid[e]s to institute that controversial plan of raising chickens in the home. The Libyan leader found the idea novel enough to encourage its implementation. On another occasion the General commented on the high cost of new automobiles. Soon after, the government agency entrusted with importing and selling cars to the public began to import only used cars and ironically sold them at new car prices. The policy was reversed only after a great number of people complained. He remarked about the proliferation of Western musical instruments in the country. The result was the gathering and burning of musical instruments. While driving through an area in the suburbs of the city [of] Benghazi  he wondered whether the area would be suitable for agriculture. Within a month all residential buildings in that area were demolished. 

...On the whole Qaddafi is rarely precise about the type of policy he desires and prefers to see the potential policy implemented before he intervenes and modifies it. Even The Green Book is general enough to permit different interpretation and experimentation by the revolutionary committees. Ultimately, however, all policies need the blessing of General Qaddafi. (Mansour O. El-Kikhia, Libya's Qaddafi, p. 106).

This looks like a variant of the signalling process that contributed to the great famine in China. Qaddafi gets a harebrained idea that presents a profit opportunity for his ruling coalition (note the $150 mandatory fee, which seems rather high for 1977, payable, one supposes, into the pockets of the well connected). The idea is therefore vigorously implemented, despite its evident absurdity, and just as quickly discontinued once profits diminish. Similarly with other policies: every weird idea that passes through Qaddafi's lips apparently presented both an opportunity for signalling support and (with the possible exception of the burning of musical instruments) for profit, at least for those with ties to the ruling elite in Libyan society. (Which, judging from El-Khikia's book, published in 1997, was extremely narrow; indeed, the book's appendix basically lists every one of its members at the time).

Many of these policies were "justified" by the "ideology" of the "Green Book." I suppose there are people out there, other than Qaddafi, who take the Green Book seriously. I've even briefly skimmed a good article carefully examining Rousseau's influence on Qaddafi's thought. But unlike the case of Marxist ideas in communist countries, it is abundantly clear from El-Kikhia's book that this "ideology" has primarily, if not exclusively, served as a signalling medium. There are few real ideologues in Libya, only careerists. Belief is mostly irrelevant, since the ideology is incoherent and impossibly vague, and its interpretation depends entirely on Qaddafi's whims. Its only real use is as an instrument of control: Qaddafi gets to decide which performances by competing factions within the revolutionary committees count as sufficiently loyal, which seems to encourage an escalation of zeal (especially in the absence of rewards for slowing down the implementation of absurd ideas). And he can test these policies by gauging which implementations are popular and which ones aren't - i.e., which interpretations of his words can generate oppositional collective action and which ones cannot - without committing himself to any particular interpretation of the policy (since he is the sole authority for their interpretation).

But how do you arrive at this point? El-Kikhia tells a story of institutional destruction: Qaddafi suspends all laws leaving only his dicta and their interpretations by a fluid network of committees (none of which is ever certain of Qaddafi's favor) as the only means of coordinating collective action within the "state." (He apparently suspended the laws in 1974. I couldn't quite believe it, but apparently there is nothing quite like law in the Libya described by El-Kikhia - no real courts for the settlement of disputes according to norms, though I suppose this may have changed since 1997. In fact, I'm not sure it makes sense to speak of a Libyan state in the full Weberian sense of the word, at least given what I've read in this book). Yet this only deepens the puzzle: for Qaddafi didn't start as the kind of ruler who could "suspend" all laws, or whose every passing fancy precipitated a cascade of costly signalling on the part of people wishing to benefit materially from his rule. He had to work towards this point - taking power away from his partners within the Revolutionary Command Council that seized power in 1969, stacking the institutions of the state with family and tribal loyalists, and keeping them all guessing by purging them at irregular intervals. And even then, he still had to deal with coup attempts.

A recent paper by Milan Svolik ("Power Sharing and Leadership Dynamics in Authoritarian Regimes," AJPS 53:2, 2009, pp. 477-494) suggests some possibilities. The key insight exploited by Svolik is that members of a ruling coalition (always necessary for control of the state) delegate some power to the dictator to coordinate activity for their mutual benefit (e.g., extracting revenues and sharing them), but the dictator can augment this power by means of actions that are not always observable by the coalition. (In delegating power, after all, the coalition surrenders some control over information). In order to prevent this, the coalition can threaten a coup, but the threat is never wholly credible because failed coups are very costly (you can easily be killed, or in the best case scenario exiled to Outer Mongolia), and members of the coalition can never be sure of what actions the dictator has taken (or failed to take) in response to the threat: dictators lie easily, and can hide inconvenient information. Svolik derives two possible scenarios from these ideas: one in which dictators are constantly threatened, and easily removed by coups (the vast majority of cases: most dictators do not survive their first five years in office) because the coalition is (rightly) suspicious of any moves by leaders to amass power, and one in which they basically last forever (like Qaddafi), barring external intervention or other "exogenous" shocks, like popular revolutions (which are very rare: of 303 dictators lasting more than one day in office and removed by "nonconstitutional" means Svolik examines, 205 were removed in a coup d'etat, and only 32 in a popular uprising, with 30 more stepping down by popular pressure to democratize. Other leaders died in their sleep or were succeeded by "constitutional" means like hereditary succession; these are not counted among the 303 noted here).

I suspect that one of the means by which dictators can amass power vis a vis a ruling coalition is to encourage (consciously? unconsciously? does it matter?) the use of "ideologies" (the name is too grand: signalling languages, perhaps?) whose interpretation they can personally monopolize or near-monopolize as means of coordinating collective action, in lieu of existing norms and institutions, whose interpretation may be more easily controlled by members of powerful elites. The process may start small, with ideas that draw on popular aspirations or customs, and take hold in particular institutional niches. Qaddafi starts with Nasserism and Arab nationalism, which had wide appeal, but Nasserism proves unwieldy, and its interpretation not easily monopolizable. (For one thing, Nasserism in Libya involved the creation of a political party that could draw on other sources of authority for the interpretation of the norms that were to guide collective action, and hence enabled members of the ruling coalition to credibly threaten Qaddafi). But this initial move was not obviously threatening to members of the coalition in Libya, who may have been genuinely attracted to pan-Arabism and Nasserism. But by using Nasserism to disrupt the older institutional order of the monarchy (which was, after all, nothing but forms of coordinated action in light of shared expectations) Qaddafi narrowed the range of people who could authoritatively interpret norms and ideas that could guide coordinated collective action. This gave him an opening to disrupt Nasserism in turn with the "popular committees," which could be more easily used to "weed out those who did not conform to his thinking" (El-Kikhia, p. 54), since the interpretation of the norms guiding collective action there was more easily monopolized by him, even as this development could be presented as a sort of "evolution" or "deepening" of the revolution.

With each step, there is a narrowing of the plausible interpreters of the "signals" that can serve to coordinate collective action, until (with the revolutionary and later the "cleansing" committees) we reach a sort of maximum monopoly on the interpretation of norms and expectations for organizing collective action. (Only Qaddafi can tell what is and what isn't in accord with the norm, and only Qaddafi is believed among interpreters of the norm).

I'm not sure this argument is entirely clear, or right for that matter. So use with care.

[Update 4/30/11: fixed some small grammar problems]

Careerists and Ideologues in China's Great Leap Famine

From the department of perverse incentives, a new APSR paper by James Kai-Sing Kung and Shuo Chen ["The Tragedy of the Nomenklatura: Career Incentives and Political Radicalism during China's Great Leap Famine," vol 105, pp. 27-45]:

A salient feature of China's Great Leap Famine is that political radicalism varied enormously across provinces. Using excessive grain procurement as a pertinent measure, we find that such variations were patterned systematically on the political career incentives of Communist Party officials rather than the conventionally assumed ideology or personal idiosyncrasies. Political rank alone can explain 16.83% of the excess death rate: the excess procurement ratio of provinces governed by alternate members of the Central Committee was about 3% higher than in provinces governed by full members, or there was an approximate 1.11% increase in the excess death rate. The stronger career incentives of alternate members can be explained by the distinctly greater privileges, status, and power conferred only on the rank of full members of the Central Committee and the “entry barriers” to the Politburo that full members faced.

This seems to me to tie into the “signalling” theme of the last post on cults of personality (which proved surprisingly popular). The problem here appears from the point of view of the people who want access to power and privilege: how can they signal sufficient commitment to the leadership so that they are rewarded with power and privileges?

Here is what Kung and Chen argue happened in China. In the hierarchy of the CCP, the three highest levels are politburo members, full members of the central committee, and alternate members of the central committee. The politburo is tiny – about 20 people. (This is, we might say, the highest level of the “winning coalition”). In Mao’s time, most of them were founding members of the CCP, had gone through the Long March, or had otherwise participated extensively in guerrilla activities before 1949. Generally speaking, it was thus very difficult for anyone who did not have these experiences to enter the politburo at the time. But it was possible to move from alternate membership to full membership in the Central Committee, a larger body of about 300 or so people (the exact size of the Central Committee has varied over time); and this move brought substantial material and status benefits – more offices, opportunities for patronage, etc. Yet in order to move from alternate to full membership, one had to give sufficient indications of commitment and reliability. In this case, Mao indicated that rewards would come to those who signalled credible radicalism, and credible radicalism could only be signalled by excessive grain procurement, leading to famine.

The Great Leap provided these party officials [alternate members of the Central Committee] with a rare, extraordinary opportunity to respond to Mao’s unambiguous signal that radical behavior would be duly rewarded. The evidence clearly shows that even after controlling for the idiosyncrasies of individual provincial leaders and variations in local conditions, the alternate members were, as a group, indeed more likely to act radically. Our findings thus substantially challenge the reigning assumption that ideology is the main source of bureaucratic radicalism in totalitarian regimes. (P. 43)

But since these full members could not move any further up the hierarchy (the only people who could enter the Politburo at the time were those who had been important in guerrilla warfare or had been through the Long March), once they reached the top they became less ideological:

The idea that career incentives matter is further bolstered by the provocatively counterintuitive finding that radicalism declined among those bureaucrats who, although still having room to move further up the career ladder [to the politburo], nonetheless lacked the necessary “prerevolutionary credentials” to do so, at which point most apparently became satisfied careerists rather than revolutionary zealots. (P. 43).

An interesting question is how a dictatorship moves from the signalling equilibrium where crazy radicalism is rewarded to the signalling equilibrium where other things (e.g., “measured economic performance”) are rewarded, as China has moved. Indeed, it seems to be a common though not universal pattern in communist (and perhaps other) dictatorships: a period of radical policy, with high levels of repression and ideological “investment” (Mao, Stalin, Ulbricht) is often followed by a period characterized by lower levels of ideological fervor, less “proactive” repression, and more emphasis on the provision of material benefits for both the “selectorate” (members of the party) and the rest of the population (Deng, Khrushchev, Honecker). (These material benefits need not consist in economic growth per se – it may be just an emphasis on economic security for the majority of the population and further material privileges for the party, as in East Germany). Totalitarian dictatorships seem to turn into careerist hierarchies concerned with preserving the material privileges of its elites and preventing revolution from below through economic “bribes.” Why?

Kung and Chen seem to think that this simply depends on the character of the dictator: the key difference between Mao’s China and modern China is that Mao was crazy and his successors were not, to put the point bluntly. (I’m putting words in their mouths, but the basic point is simply that Mao was ideologically committed to a crazy vision of communism while his successors, starting with Deng, were more committed to a pragmatic model of economic development). The consequence is that the system remains susceptible to economic disaster, even though it is doing well today:

[I]n the absence of political checks and balances on the dictator, he can easily misuse the same career incentives that have been employed to promote economic growth [in the post-Mao period] under the same conditions of centralized personnel control by the nomenklatura and economic decentralization, leading in this case to economic disaster. (P. 43)

But this seems unsatisfactory to me, though there is probably some truth in the idea. Here are a couple of alternative theories (or rather, sketches of theories). First, following an interesting argument by Kurt Weyland (2008, gated link), one might think that dictators, like all leaders (but even more so: they are an “epistemic bottleneck”), are cognitively constrained; they simply implement whatever policy is seen to be “effective” in their milieu given their objectives (which may include building up the status of the country in the international arena, an objective that we may assume both Mao as well as later Chinese leaders held, and which involves pursuing policies that they believe strengthen the economy). In other words, they emulate those [countries, leaders] they trust, but do not really know what will work (in fact, nobody really does); this accounts for the fact that policies get adopted as “models” and transferred from one country to another sometimes rather quickly. In the 50s, radical agricultural collectivization and other such policies were thought to be “effective” among  Chinese communist leaders (as they had been thought to be effective among Soviet leaders slightly earlier); later they became discredited, but “market-based” policies became popular. As long as relatively good policies are thought to be “effective” in the dictator’s milieu, centralized  dictatorships with the sort of personnel policies that China has will do relatively well, as the dictator (or tiny ruling group) can effectively reward supporters for the implementation of the policy. But if disastrous policies again become popular in the ruling group’s milieu, then the dictatorship will do badly.

But perhaps what happens is that in demanding credible signals of commitment from the upper levels of the hierarchy, the dictator necessarily gets the unprincipled careerists. (This would not come as a surprise to Machiavelli, among other theorists of autocracy: beware of flatterers). Imagine you have a population of principled and unprincipled upper-level party members. The principled party members mostly agree with the dictator, but not 100%; because they are principled, they have their own interpretation of whatever doctrine they all claim to espouse. And they are unwilling to compromise; they have, as we say, “principles.” By contrast, the careerists are willing to say and do anything for the sake of advancement. When the dictator demands radical policy, the people without principled commitments jump at the opportunity, whereas the principled members of the hierarchy get disproportionately punished for demurring or having independent thoughts. (The Bukharins get purged, for example). Over time, the upper level of the hierarchy fills up with careerists. But when the dictator dies, the careerists prefer not to have to do so much counterproductive signalling, and they are now in a position to select the next dictator. So they tend to go for people who are likely to protect their material interests rather than true ideologues, and as a side effect the dictatorship lowers the level of repression and becomes more focused on providing economic goods for both the party organization and the rest of the population. (Also, insofar as they lower the level of repression, they now need to provide material benefits in order to avoid challenges from outside the party organization).

I’m not sure this is right; I would imagine that one would have to first establish whether or not totalitarian dictatorships (high ideological investment, high proactive repression) do reliably turn into post-totalitarian or authoritarian dictatorships (low ideological investment, low proactive repression, a focus on material “bribes”). If it is right, I suspect that this sort of success eventually runs out: without political competition or ideological commitment, the state (or the party) decays into a pure patronage organization staffed by careerists. This seems to have happened in the Soviet Union, though there the problem was compounded by the reliance on central planning (which is to corruption as clouds are to rain); could it also happen in China?

[Update 4/4/11: Added Kung and Shen's title, corrected some obvious typos].