Regimes and Contention

CIAO DATE: 6/99

Regimes and Contention

Contentious Politics Series
May 6, 1998

Lazarsfeld Center at Columbia University

How do diverse forms of political contention—revolutions, strikes, wars, social movements, coups d’état, and others—interact with shifts from one kind of regime to another? To what extent, and how, do alterations of contentious politics and transformations of regimes cause each other? These questions loom behind current inquiries into democratization, with their debate between theorists who consider agreements among elites to provide necessary and sufficient conditions for democracy and those who insist that democracy only emerges from interactions between ruling-class actions and popular struggle. They arise when political analysts ask whether (or under what conditions) social movements promote democracy, and whether stable democracy extinguishes or tames social movements. They appear from another angle in investigations of whether democracies tend to avoid war with each other.

The same sorts of questions recur in studies of industrial conflict, where one school of thought opines that strikes represent breakdowns in bargaining that could be pursued more efficiently by other means, another school of thought argues that strikes entail compromises of labor with capital and thereby integrate workers unwittingly into capitalism, while a third view treats strikes as rational, essential means of struggle in competitive capitalism but not elsewhere. They dog every analysis of revolution, which must consider whether certain kinds of contention regularly promote revolutions as well as whether revolutions regularly generate certain kinds of contention. Yet we have no coherent theory of links between regime change and contentious politics. At least two obstacles bar the path to coherent theory: first, that the relationship between regime change and contentious politics is surely complex, contingent, and variable; second, that no codification of variation in regimes has commanded wide assent.

This paper will not unveil a general theory of regime change, of contentious politics, or of their interaction. It rests, in fact, on a set of premises that deny the possibility of a general, law-like theory in this domain:

although political change is causally coherent, it is also path-dependent
as a consequence, it is crucial to trace effects of existing precedents, models, practices, and connections on any particular sequence of changes
whole sequences and structures rarely or never repeat themselves
smaller-scale causal mechanisms do, however, recur in a wide variety of settings
explanation of changes in contention, in regimes, and in their interaction therefore has two components: 1) identification of crucial causal mechanisms, 2) analysis of how preceding and existing conditions affect the concatenation and sequence of those causal mechanisms
even at the unattainable limit of exhaustive explanation, a satisfactory account of interaction between regime change and contentious politics would not take the form of general laws for large sequences or structures but of constraints on combinations and sequences of mechanisms

This paper simplifies such an enormous agenda by singling out broad correspondences between regimes and forms of politics as indications of what must be explained. First the paper lays out a relevant map of regime variation. Then it surveys likely correlates and consequences of regime change with an eye to identifying causal mechanisms deserving further attention. It ends not with answers, but with suggestions for a research program.

How shall we map regimes? At first, Aristotle made it all seem vividly simple: “The true forms of government . . . are those in which the one, or the few, or the many, govern with a view to the common interest; but governments which rule with a view to the private interest, whether of the one, or of the few, or of the many, are perversions” (Barnes 1984: II, 2030). This reasoning led to a straightforward typology of all governmental forms:

Thus if a single ruler (a monarch) promoted his own self-interest instead of the common good, he became a tyrant; if an aristocracy similarly used governmental power exclusively for its own advantage, the regime became an oligarchy; and if the majority in a constitutional government likewise sought only their own benefit without regard to the commonwealth, their regime became a democracy.

According to Aristotelian principle, proper monarchy rested on rule by the best man, aristocracy on rule by the richest and best men, and constitutional government on rule by free men. (For Aristotle, ineluctable nature condemned women, like slaves, to inferiority.) Since the rich are usually few in number and the free poor many in number, reasoned Aristotle, as a practical matter aristocratic regimes generally mean rule by the few in the common interest, constitutional government rule by the many, likewise in the common interest. Perversions into tyranny, oligarchy, and democracy arise where rulers —one, few, or many—place their own interest above the common good. Democracy’s characteristic perversion, in this Aristotelian view, consists of discrimination by the governing poor against both the state’s collective interest and the interests of the rich.

To be sure, Aristotle recognized distinctions within his major types of regime, for example five types of democracy, of which the fifth

is that in which not the law, but the multitude, have the supreme power, and supersede the law by their decrees. This is a state of affairs brought about by the demagogues. For in democracies which are subject to the law the best citizens hold the first place, and there are no demagogues; but where the laws are not supreme, there demagogues spring up. For the people becomes a monarch, and is many in one; and the many have the power in their hand, not as individuals, but collectively . . . this sort of democracy is to other democracies what tyranny is to other forms of monarchy (Barnes 1984: II, 2050-2051).

In these circumstances, furthermore, demagogues often stir up the rabble to attack the rich and thereby seize power for themselves. In this way, democracy turns into tyranny. When he got to details, Aristotle allowed for plenty of transitions and compromises among his three pure types.

Aristotle proceeded repeatedly from ostensibly static categories to dynamic causal processes. In thinking through the effects of different military formats, for example, he offered a shrewd causal account:

As there are four chief divisions of the common people, farmers, artisans, traders, labourers; so also there are four kinds of military forces—the cavalry, the heavy infantry, the light-armed troops, the navy. When the country is adapted for cavalry, then a strong oligarchy is likely to be established. For the security of the inhabitants depends upon a force of this sort, and only rich men can afford to keep horses. The second form of oligarchy prevails when a country is adapted to heavy infantry; for this service is better suited to the rich than to the poor. But the light-armed and the naval element are wholly democratic; and nowadays, where they are numerous, if the two parties quarrel, the oligarchy are often worsted by them in the struggle (Barnes 1984: 2096-2097).

In the Politics, Aristotle confined his systematic discussion of political contention to revolutions, which meant forcible overthrow of regimes by ostensible subjects of those regimes. In passing, however, he also mentioned factional struggles, conspiracies, and collective resistance to state demands. In each case, he treated the form of regime as an outgrowth of the balance among local forces (notably among the rich, the middle class, and the poor) tempered by historical circumstance. He then explained contention as a joint outcome of that balance and the regime type, again tempered by historical circumstance. Without developing his observations at length, Aristotle clearly saw regimes as having their own characteristic forms of contention, and changes of regime as resulting largely from political contention. In contrasting regimes, different ruling coalitions pursued distinct strategies of rule, which altered the incentives and capacities of various constituted groups within the state to defend or advance their own interests by acting collectively. Aristotle explained struggles of his time by combining the perspectives of rationalists and structuralists, millennia before anyone used those labels (for those labels, see Lichbach and Zuckerman 1997).

Today’s students of contentious politics have not advanced much beyond Aristotle in this regard. Robert Dahl’s treatment of approximations to democracy, for example, has a distinctly Aristotelian air. As summarized in Figure 1, Dahl’s useful scheme distinguishes two dimensions of variation: inclusiveness, or the extent to which people under a given regime’s jurisdiction have the right to participate at all; liberalization, or the extent to which participants in the regime have rights to contest conditions of rule. Dahl adds to Aristotle a recognition of very inclusive regimes that allow little public contestation, which Dahl calls inclusive hegemonies. He also leaves a large open space among his four corner types, where we might locate a great many other regimes—for example the thinly-ruled nomadic empires, urban federations, composite dynastic states, and city-empires that governed much of Europe five hundred years ago.

Figure 1: Robert Dahl’s Classification of Regimes

What Dahl calls contestation enters his classification as a bundle of rights; at the liberal extreme 1) freedom to form and join organizations, 2) freedom of expression, 3) the right to vote, 4) eligibility for public office, 5) competition by political leaders for support, 6) alternative sources of information, 7) free and fair elections, and 8) institutions for making government policies depend on votes and other expressions of preference. Regimes vary enormously, as Dahl declares, “in the extent to which the eight institutional conditions are openly available, publicly employed, and fully guaranteed to at least some members of the political systems who wish to contest the conduct of the government” (Dahl 1975: 119; see also Lindblom 1977). His closed hegemonies accord such rights to no one, his competitive oligarchies extend them to a small elite, his inclusive hegemonies entertain no such rights, and his polyarchies open them to much of the population. Note that under the label “contestation” Dahl is speaking about institutionalized rights to opposition, not about the character or frequency of contention.

Non-institutionalized public contention enters Dahl’s story incognito, as demands (of unspecified form) that regimes remove causes of extreme inequality, as disputes in which one segment of the population appears to threaten the survival of another, as the formation of revolutionary oppositions, and as foreign conquest. His scheme therefore challenges us to specify the interaction between regimes and the rights embedded within them, on one side, and contentious politics that sometimes adopt rightful means and sometimes defy them, on the other. The work at hand includes relating regimes and regime change to prevailing distributions of 1) actors, actions, and identities in contentious politics, 2) conditions for emergence of contentious politics, and 3) trajectories and outcomes of contentious politics.

Samuel Finer’s posthumous History of Government provides another neo-Aristotelian handle for the classification of regimes. After stipulating that one can classify regimes along a territorial dimension (city, national, or empire), divide decision-making personnel into elites and masses, and distinguish decision-implementation by bureaucracies and armed forces, Finer ultimately settles, like Aristotle, for a focus on the social character of a regime’s ruling personnel. As represented in Figure 2, Finer identifies four pure types: Palace (monarch and following), Nobility (privileged class), Forum (segments or representatives of populace), and Church (priesthood). The diagram’s double-headed arrows portray likely paths of movement from one regime type to another and likely locations of mixed regime types.

Figure 2. Samuel Finer’s Typology of Regimes

Contention thrusts its way repeatedly into Finer’s accounts of particular regimes. Speaking of Italian city states, for example, Finer observes that thirteenth-century patriciates often closed their ranks to newcomers. “But as they did so,” he remarks,

They came under pressure from the less wealthy or newly wealthy elements demanding a due share in office; the so-called ‘democratic’ movement. These elements, characteristically, used their guild organizations to channel their pressure, so that the struggle looks like craft-guilds trying to break the political monopoly of the wealthier and more prestigious merchant-guilds. In Italy . . . these excluded elements formed themselves into sworn associations and called themselves the ‘People’—the popoloI—and tried to assert their claims by revolt. But what happened in Italy is but the paradigm case of what was occurring in much of urbanized Europe as the thirteenth century began to close: resistance to the oligarchy, violence, even revolution (Finer 1997, II: 954).

Pursuing other ends, however, Finer does not examine relationships—empirical or causal—among regime types, political transitions, and forms of contentious politics. This paper concentrates, in contrast, on asking how and why political contention varies from one regime type to another, and how contention interacts with movement from regime to regime.

Following the premises I laid out earlier, let us approach that pair of questions here in profound skepticism about the existence of neat correspondences between regime type A and action X, emergence process Y, or trajectory Z. On the contrary, we should search for rough empirical regularities in hope of accomplishing two distinct objectives: first, to specify what theoretically telling similarities and differences must be explained by any causal account of contention; second, to place firmly on the agenda how historically-accumulated models, memories, understandings, and social relations—for example, residues of the Mongol empire’s previous hegemony in a given region—affect the operation of contentious politics. The challenge is therefore to create two rough conceptual maps—one of regimes, the other of contentious politics—whose similarities and differences pose crucial questions of causation.

In meeting this challenge, we have deplorably little systematic analysis to build on. Analysts commonly recognize the concentration of social movements (narrowly defined) in parliamentary democracies, the vulnerability of weakened despotic regimes to revolution, the greater frequency of coups d’état where military forces exercise great autonomy, and a miscellany of near-tautologies such as the prevalence of strikes under industrial capitalism or the concentration of peasant revolts in large-landlord systems. But we have no well-established general mapping of variation in the forms and dynamics of contentious politics across the multiple types of state regime. Existing formulations, furthermore, suffer major weaknesses: first, little insight into interactions between contentious political processes and their settings, for example in the ways in which contentious politics incited by certain sorts of regime transforms those regimes; second, no effective account of interpretation, for example, in the interplay between understandings that pervade routine non-contentious politics and those that inform contentious claims. Much less, then, do we possess a dynamic causal account that explains interconnections between regimes and contention.

Let us therefore take a leaf from Aristotle, creating a simple taxonomy of regimes on the way to reasoning about variations, trajectories, and transformations of contentious politics. The term regime, in this context, refers to any distinctive configuration of a polity: connections among a government, members of the polity defined by their routine access to agents of that government, challengers consisting of constituted actors lacking routine access to governmental agents, and intermittent actors—outside governments, international organizations, third parties, and so on—based outside the zone of the government’s jurisdiction.

In order to make such a model fit the complexities of real political processes, we must complicate it: show the government as less like a unitary star and more like a galaxy, with multiple centers and hierarchies, often competing, rather than a single unitary point; vary the sharpness of the polity’s boundary; allow for jagged or blurred edges to the government’s jurisdiction; recognize that contenders (both members and challengers) vary in strength and coherence; note that a given individual or group within a government’s jurisdiction may belong to multiple contenders or none at all.

We must also put the model into motion, with the government shifting, contenders changing, and claim-making fluctuating. Finally, we must place polities within their historical and cultural settings, recognizing at a minimum that previous and adjacent forms of government provide powerful templates for the creation of new states; as a consequence, history and culture constrain the operation of ostensibly general processes such as repression and political mobilization. We are dealing with mutual claim-making and responses to claim-making among unequally powerful contenders in the presence of at least one government.

The simple polity model opens the way to a taxonomy of all regimes since Aristotle’s era. The taxonomy shifts away from the Aristotle-Finer emphasis on the identity of ruling classes to the Dahlian emphasis on political relations between rulers and ruled. The classification concentrates on relations between governments and polity members. It operates as a function of five dimensions:

state capacity (actual impact of state action on activities and resources within the state’s jurisdiction, relative to some standard of quality and efficiency): low (0) to high (1)

breadth of polity membership: ruler alone (0) to every person under a government’s jurisdiction belonging to at least one polity member (1)

equality in polity membership: radically unequal (0) to every person who belongs to a polity member has equal access to governmental agents and resources (1)

strength of collective consultation among polity members with respect to governmental personnel, policy, and resources, considered as a multiple of a) how binding that consultation is, b) how effectively that consultation controls governmental personnel, policy, and resources: from non-binding and ineffectual (0) to binding and determining (1)

protection of polity members and persons belonging to them from arbitrary action by governmental agents: no protection whatsoever (0) to complete protection (1)

Thus 10011 (high capacity, narrow polity membership, unequal polity membership, strong consultation, extensive protection) describes an idealized powerful oligarchy, or perhaps even a valid aristocracy in Aristotle’s view. The figures 11100 (high capacity, broad polity membership, equal polity membership, no consultation, no protection) describe an idealized totalitarian state, Aristotle’s worst dream of tyranny. The series 00000, finally, designates utter anarchy. All real governments fall somewhere between, with the average western capitalist country, relative to all states that have ever existed, scoring perhaps .75 on capacity, .80 on breadth, .75 on equality, .70 on consultation, and .85 on protection. Although I will make no effort at deriving precise measures of these five elements here, we can imagine history since Aristotle’s time as unfolding before an immense scoreboard that displays five fluctuating numbers for each state. The explanatory problem is then to identify and explain connections between those fluctuations, on one hand, and changes in the character, intensity, and trajectories of contention, on the other.

The five dimensions are logically distinct: to some extent we can analyze variation within each dimension independently. Nevertheless, they (or rather the causes embedded in them) interact so strongly that much of the logical space they imply is empirically empty. Low-capacity states, for example, rarely or never provide their polity members with extensive protection from arbitrary action by state agents. Nor do very broad polity membership, very unequal polity membership, and binding consultation of polity members long (if ever) cohabit. In general, it looks as though substantial increases of state capacity propel broadening of polity membership when the essential resources for the state’s operation come from the population within the state’s jurisdiction, because struggle over those resources lead to provisional bargains that establish mutual rights and obligations between state agents and providers of resources. Thus a whole theory of state transformation awaits articulation in the form of causal propositions linking the five dimensions. For now, however, the salient questions concern variation in contentious politics as a function of a regime’s location with respect to the five dimensions taken singly.

State capacity does not enter the definition of democracy, yet it strongly affects the chances for democratic processes. In principle, one could imagine broad political participation, relative equality of individuals or other social units, binding collective consultation, and protection in the absence of an enforcing state. Anarchists and utopians have often taken the relative democracy of some crafts, shops, and local communities as warrants for the feasibility of stateless democracy on a large scale. The historical record, however, suggests another conclusion: where states collapse, other predators spring up. In the absence of effective state power, people who control substantial concentrations of capital, coercion, or commitment generally use them to forward their own ends, thus creating new forms of oppression and inequality. If high state capacity does not define democracy, it looks like a nearly-necessary condition for democracy on a large scale.

We cannot, however, draw from such an observation the comforting inverse conclusion: that expansion of state capacity reliably fosters democracy. In fact, expanding state capacity promotes tyranny more often than it causes democracy to flower. In the abstract calculation that sums over all governmental experiences, the relationship between state capacity and democracy is no doubt asymmetrically curvilinear: more frequent democracy from medium to medium high state capacity, but beyond that threshold substantial cramping of democratic possibilities as state agents come to control a very wide range of activities and resources.

Citizenship, in this view, only forms on the higher slopes of the five continua. Only where state capacity is relatively extensive, polity membership involves some significant share of a government’s subject population, some equality of access to government exists among persons who belong to polity members, consultation of those persons makes a difference to governmental performance, and persons belonging to polity members enjoy some protection from arbitrary action can we reasonably begin to speak of mutual rights and obligations directly binding state agents to whole categories of persons defined by their relation to the government in question—that is, of citizenship. Although citizenship of a sort bound elite members of Greek city-states to their governments and elite members of many medieval European cities to their municipalities, on the whole citizenship at a national scale only became a strong, continuous presence during the nineteenth century. Figure 3 sums up the five dimensions, showing the locations of anarchy, democracy, and citizenship.

Figure 3: A Five-Dimensional Taxonomy of Regimes

Democracy builds on citizenship, but does not exhaust it. Indeed, most western states created some forms of citizenship after 1800, but over most of that period the citizenship in question was too narrow, too unequal, too non-consultative and/or too unprotective to qualify their regimes as democratic. The regimes we loosely call “totalitarian,” for example, typically combined high state capacity with relatively broad and equal citizenship, but afforded neither binding consultation nor extensive protection from arbitrary action by agents. Some monarchies maintained narrow, unequal citizenship while consulting the happy few who enjoyed citizenship and protecting them from arbitrary action by state agents; those regimes thereby qualified as oligarchies. In searching for democratic regimes, we can take relatively high state capacity for granted because it is a necessary condition for strong consultation and protection. We will recognize a high-capacity regime as democratic when it installs not only citizenship in general, but broad citizenship, relatively equal citizenship, strong consultation of citizens, and significant protection of citizens from arbitrary action by governmental agents.

Both consultation and protection require further stipulations. Although many rulers have claimed to embody their people’s will, only states that have created concrete preference-communicating institutions have also installed binding, effective consultation. In the West, representative assemblies, contested elections, referenda, petitions, courts, and public meetings of the empowered figure most prominently among such institutions; whether polls, discussions in mass media, or special-interest networks qualify in fact or in principle remains highly controversial.

On the side of protection, democracies typically guarantee zones of toleration for speech, belief, assembly, association, and public identity, despite generally imposing some cultural standards for participation in the polity; a regime that prescribes certain forms of speech, belief, assembly, association, and public identity while banning all other forms may maintain broad, equal citizenship and a degree of consultation, but it slides away from democracy toward populist authoritarianism as it qualifies protection. At the edge of the five-dimensional space that contains democratic regimes, furthermore, previous historical experience has laid down a set of models, understandings, and practices concerning such matters as how to conduct a contested election. This political culture of democracy limits options for newcomers both because it offers templates for the construction of new regimes and because it affects the likelihood that existing power-holders - democratic or not - will recognize a new regime as democratic.

Over the long run of human history, the vast majority of regimes have been undemocratic; democratic regimes are rare, contingent, recent creations. Partial democracies have, it is true, formed intermittently at a local scale, for example in villages ruled by councils incorporating most heads of household. At the scale of a city-state, a warlord’s domain, or a regional federation, forms of government have run from dynastic hegemony to oligarchy, with narrow, unequal citizenship or none at all, little or no binding consultation, and uncertain protection from arbitrary governmental action. Before the nineteenth century, large states and empires generally managed by means of indirect rule: systems in which the central power received tribute, cooperation, and guarantees of compliance on the part of subject populations from regional power-holders who enjoyed great autonomy within their own domains. Seen from the bottom, such systems often imposed tyranny on ordinary people. Seen from the top, however, they lacked capacity; the intermediaries supplied resources, but they also set stringent limits to rulers’ ability to govern or transform the world within their presumed jurisdictions.

Only the nineteenth century brought widespread adoption of direct rule, creation of structures extending governmental communication and control continuously from central institutions to individual localities or even to households, and back again. Even then, direct rule ranged from the unitary hierarchies of centralized monarchy to the segmentation of federalism. On a large scale, direct rule made substantial citizenship, and therefore democracy, possible. Possible, but not likely, much less inevitable: instruments of direct rule have sustained many oligarchies, some autocracies, a number of party- and army-controlled states, and a few fascist tyrannies. Even in the era of direct rule most polities have remained far from democratic.

Mention of direct and indirect rule calls attention to features of regimes omitted by the five-dimensional scheme that surely affect the form, locus, trajectory, and transformation of contentious politics. In addition to the distinction between direct and indirect rule, they include:

Sheer size of the polity, as measured by numbers of persons, numbers of actors, geographic range, or the total time-cost of communication among all its participants
Directness and completeness of central control over peripheral locations, ranging from centralized field administration through incorporation of selected leaders and their followings to highly fragmented indirect rule
Multiplicity of governments within the polity, for example contrasts between federal and unitary systems
Steepness of gradient and sharpness of boundary between polity members and challengers with respect to access to government-controlled resources, services, and protections, as where non-citizens face close surveillance and high risk of persecution or expulsion in contrast to liberal treatment of citizens
Uniformity of governmental administration from one jurisdiction to the next, ranging from strict hierarchy and unit-to-unit replication to extensive particularism
Historically-accumulated criteria for polity membership or exclusion, as in distinctions between full-fledged corporatism and individual-by-individual citizenship
Current structure and composition of polity members, for example domination by merchants, coalitions between churchmen and landlords, major splits among dynastic factions, direct representation of peasants
Historically-accumulated governmental criteria for prescribed, tolerated, and forbidden forms of collective claim-making, for example promotion or suppression of electoral campaigns as sites of criticism or support for public policies and personalities

Experts in comparative politics or particular national political histories will surely augment the list with still more regime-based sources of variation in contentious politics. To take all these variations into account simultaneously, however, would doubly defeat my purpose: first by filling every formulation with so many qualifications that it would fall of its own weight, second by deflecting the search away from selective causal analogies toward putatively general theories of regime change and contention.

Let us retain our grip on the problem by concentrating on two sorts of regime variation: from undemocratic to democratic regimes, and from low-capacity to high-capacity states. We concentrate on these two aspects of regime variation for several reasons: 1) because they have attracted more theoretical and empirical attention from students of popular politics than have such aspects as uniformity of governmental administration or multiplicity of governmental units; 2) because within recent centuries they have made very large differences to the character, trajectories, and dynamics of contentious politics, 3) because even over the longer run the position of a regime with respect to capacity and democracy has (as any good Aristotelian would expect) profound effects on the quality of its contentious politics.

Let us return to the democratic pentagon: capacity, breadth, equality, consultation, and protection. I will spell out a line of reasoning about regime variation in contentious politics as a dimension-by-dimension set of arguments—call them conjectures, hypotheses, or speculations. The arguments rest on knowledge limited mainly to recent western experience. They therefore invite refutation from specialists who know better.

Why and how should we expect variation in state capacity to affect contention? Most generally because higher capacity means a) state agents have the incentive and means to intervene in a wider range of social interactions within the state’s zone of action, b) state actions, for whatever ends undertaken, affect a wider range of actors and interactions, hence stimulate the interested parties to make offensive, defensive, or deflecting claims of their own, c) whatever projects contenders and third parties undertake, state agents, state-controlled resources, and likely state reactions become more crucial to those projects. Conversely, in the presence of weak state capacity, most contentious politics occurs with little or no state involvement, and a high proportion of state intervention meets concerted resistance. A number of empirical inferences follow from these arguments, for example:

The greater state capacity, the larger share of all resources and activities within a polity affected by state action, hence the more likely claims directed at state agents.
The less state capacity, the higher the proportion of all claim-making consisting of violent competition between similar groups.
The less state capacity, the more direct action against renegades, moral reprobates, and agents of central authority.
The less state capacity, the more clandestine retaliatory damage, the more concerted resistance to outside threats, the more localized action, the closer ties of claim-making to embedded (rather than disjoined) identities, and the more variation in claim-making’s cultural content.
The less state capacity, the higher the proportion of state interventions that consist of violent predation and/or exemplary punishment, hence the greater probability of violent resistance.
Beyond some threshold, state capacity correlates with directness of rule, hence with the likelihood that claim-makers and objects of claims will be state agents rather than empowered intermediaries or essentially autonomous powerholders.
Higher state capacity, on average, depends on greater extraction of resources from the subject population, hence produces a greater frequency of contests over extraction of resources that subjects have committed to non-state enterprises.

All these hypotheses lead to concrete comparisons among regimes and forms of contentious politics. They have the advantage of straightforward research implications, but the disadvantage of focusing on static high/low comparisons.

What about breadth of polity membership? At the narrowest, no one who is subject to the authority of a given state enjoys any rights or mutual obligations binding them to state agents and state agents to them. At the broadest, everyone who is subject to that authority enjoys citizenship. Categorical citizenship is then either identical to or highly correlated with polity membership. With that understanding, we might expect to find a strong difference in means of contentious claim-making between narrow and broad polities, with a) claim-makers (especially non-members of the polity) in narrow polities tending to approach state power indirectly and/or covertly: through informal networks, through corruption of state agents, through external powerholders, through terror, or through subversion and b) challengers in broad polities frequently adopting means similar to those employed by polity members—although just different enough to call attention to their distinctness and disruptive potential.

Here are some more specific hypotheses that follow from this line of reasoning:

Broadening polity membership incites alliance-formation and claims of recognition, satisfaction, and membership by still-excluded actors.
Narrowing polity membership incites anticipatory resistance and alliance-formation by threatened polity members.
The narrower is polity membership, the more frequently subjects will approach state power indirectly and/or covertly through informal networks, through corruption of state agents, through external power-holders, through terror, or through subversion.
The narrower is polity membership, the higher the share of all open contention that directly defies authorities, hence occurs at a distance from the forms of claim-making prescribed or rewarded by authorities.
The broader is polity membership, the higher the share of all open contention that occurs at the immediate edges of prescribed political forms, for example as social movements or diversion of authorized public ceremonies.
A curvilinear relationship exists between the breadth of polity membership and the frequency with which dissident polity members bid for support of non-members by promoting their inclusion: rarely in the case of extremely narrow or extremely broad polity membership, more frequent in between.
The greater a split within a polity, the more frequent such coalitions.

Thus a dynamic of inclusion, exclusion, and contention begins to emerge. Once again, the hypotheses lead to fairly crisp static comparisons, but fall short of specifying dynamic cause-effect relations.

And equality of polity membership? Perfect equality of polity membership does not require equality of wealth, power, or well-being, but absolutely identical relations of all to state agents. Absolute inequality of polity membership does not require deep inequality of life condition, but person-to-person and group-to-group differentiation of relations to state agents. (It is nevertheless probably true, as Aristotle suggested, that great inequality of material condition promotes inequality of polity membership because affluent actors use their means to influence the political process and the performances of state agents, thus increasing inequality of polity membership itself.) No state has ever extended perfect equality of polity membership, if only because all exclude certain segments of the subject population—notably children, felons, and certified incompetents—from full benefits of state power. Even very democratic states with extensive rights of citizenship differentiate benefits and obligations of citizenship by gender, age, military service, penal status, and office-holding.

These arguments have strong implications for contention-by-regime maps. The more equal polity membership is, for example, the more the polity will respond to challengers’ effective displays of WUNC: worthiness, unity, numbers, and commitment. (This should be the case because WUNC signals a contender’s capacity to intervene effectively in routine consultation and to attract support of other contenders in doing so.) The more unequal polity membership, on the other hand, the greater the differences among channels by which distinct segments of the population make claims, hence the greater the variability in conditions for effectiveness of a given actor’s claims. (“Channels” means not only the course of claim-making itself but also coalition-formation, characteristic interactions with authorities, centripetal vs. centrifugal orientations, and repertoires.) Other related hypotheses include these:

The more equal polity membership, the greater the frequency with which losers in binding consultation accept the outcome, hence the rarer contentious outcomes to such consultations, including violence.
The more equal polity membership, the greater the resemblance among the claim-making repertoires of different contenders. (This despite incessant efforts at marginal innovation differentiating one claimant or claim from the next: variety within an extremely limited compass.)
Equality of polity membership, net of other effects, bears a curvilinear relationship to size of polity: greater for intermediate sizes than for very large and very small polities.
The more extensive exploitation and opportunity hoarding (hence categorical inequality) in the base population, the greater the inequality of polity membership (see Tilly 1998, chapter 7).

Thus, according to this line of argument, both equality and equalization have strong impacts on the character of contentious politics. To move into dynamic territory, however, we would have to look much more closely at actual processes that alter patterns of inequality.

Binding consultation? Democratic theorists often focus on elections as the critical institutions. Popular elections have, indeed, served as a crucial technology for consultation—binding or otherwise. But note that even in strongly electoral regimes an interplay typically occurs among electoral campaigns as such and a) displays of potential electoral strength by collective actors outside of electoral campaigns, b) legislative performance, c) candidate-selection processes, including payment for campaign costs, d) payoffs to supporters. In any case, some degree of binding consultation also occurs in various sorts of regimes through operation of patron-client networks, virtual representation, plebiscites, recall, referendum, consultative assemblies, polls, petitions, lobbying, payoffs, public rituals, and weapons of the weak. Let us concentrate here relatively public, transparent, and institutionalized forms of binding consultation.

This reasoning suggests strong interactions between contentious politics and binding consultation. The more extensive and binding is consultation of polity members, for example, the more shared interpretations arise from public discussion. Conversely, the less extensive and binding is consultation, the more shared interpretations emerge from unofficial, underground conversations and bifurcate between a) subversive indirect discourse of the sort that James Scott calls “weapons of the weak” and b) public dramaturgy drawing on unmistakable references to widely-known symbols, legends, events, dates, and persons. More detailed hypotheses follow:

Predominant forms of consultation (e.g. elections vs. audiences at court) strongly affect the location and forms of contentious politics, especially in the presence of democracy and extensive state capacity. Parapolitical and contentious claim-making shadow routine politics.
The more extensive and binding the consultation of polity members, the greater the clustering of contention around perimeters of institutionalized consultation.
The more extensive and binding is consultation of polity members, the greater the prominence of detached (rather than embedded) identities in collective claim-making.
Presence of civil liberties—freedom of speech, assembly, association, and belief as well as due process with respect to government agents’ seizure of persons and property—enhances consultation and channels contention toward perimeters of institutionalized consultation.
Extensive binding consultation promotes adoption of claim-making forms that depend on extensive organization and preparation rather than springing from non-contentious daily routines such as marketing, working, drinking, or attending religious services.
Extensive binding consultation promotes forms of claim-making that broadcast capacity, threat, and/or intentions to act—both individual and collective—rather than immediately engaging the actions in question. Such forms dramatize the worthiness, unity, numbers both of direct participants and of populations they claim to represent.
Extensive binding consultation promotes targeting of regional or national power-holders, including state agents.
Extensive binding consultation promotes activation of disjoined collective identities: identities broader than or separate from those that inform routine social relations, e.g. workers in general rather than machinists in this particular shop.
The more uniform is consultation across an entire population (obviously a function of breadth and equality of polity membership) the more similar are claim-making repertoires across that population.
Claim-making increases with social, temporal, and geographic proximity to major consultations.
Mobilized contenders excluded from major consultations commonly act to disrupt, counter, or intervene in those consultations.
The less binding consultation, the more sensitive the response of contenders to fluctuations in opportunity and threat on two fronts: change in their relations to the current regime, change in relations between the regime and outside actors.

What about protection of polity members against arbitrary action of state agents? Here we enter a conceptual and theoretical thicket, for two reasons: first, because “arbitrary” implies a standard of even-handed due process that is extremely difficult to state generally and a priori; second, because even more so than binding consultation, protection involves incessant negotiation of particular arrangements with state agents, as when demonstrators clear their planned marches with police or welfare administrators bend their rules to mitigate hardship. Nevertheless, we can persevere by thinking of a rough scale including positive elements such as publicity of state claims on citizens, routine availability of review and redress, and uniformity of agents’ practice across social categories. We can also consider negative elements such as absence of government-protected paramilitary forces, secret forms and loci of detention, or extensive domestic espionage. This general approach suggests strong hypotheses concerning interconnections between protection and contentious politics, for example:

The more protection, the greater the clustering of contention around perimeters of institutionalized politics.
The less protection, the higher the proportion of claim-making directed to seizures of state power, fragmentation of state power, or establishment of autonomy from state power.
The less protection, the greater the centrality of patron-client relations in contention.
The less protection, the greater the propensity of all contenders to acquire their own coercive force.
The less protection, the higher the proportion of claim-making events involving violence.
The less protection, the greater the reliance of claim-making challengers on protected social locations and on identities grounded in everyday social relations—that is, embedded identities.
The more differentiated protection by social category, the greater the differentiation of contentious repertoires.

This long string of hypotheses is, of course, no more than that: a set of reasoned conjectures about what we might expect close examination of regime variation in contentious politics to show us. It therefore constitutes an agenda for inquiry, not a set of firm conclusions. My inquiry, furthermore, does not aim at empirical generalizations linking types of contentious politics to types of regime, much less general laws from which such empirical generalizations might follow. Instead, I am trying to 1) establish rough empirical regularities specifying what sorts of variation valid theories of contentious politics must explain, 2) formulate partial but powerful causal analogies that cross boundaries of regimes and contentious political forms, 3) use the map of variation to promote study of contentious episodes differing significantly in setting and form, thereby demanding analytical finesse and requiring robust analogies, 4) use it again to specify scope conditions for robust analogies when they appear.

To bring some of these scattered arguments together and confirm the utility of concentrating on state capacity and democracy/undemocracy, let us explore implications of the scheme for a crucial problem in contentious politics: similarity in repertoires among different forms of contentious and non-contentious political interaction. By contentious repertoires I mean collective claim-making routines that characterize any pair of politically constituted actors. The theatrical metaphor conveys the sense in which such claim-making generally consists not of bureaucratic form-filing but of improvisatory and contingent performances, based on previous experience, drawing on existing understandings, social relations, and known practices. Contentious repertoires always include limited numbers of such performances, far fewer and far narrower than the interactions of which the parties would be technically capable.

Let us generalize the idea of repertoire to designate all the claim-making performances commonly employed within a given regime. In general, we should expect high-capacity states to feature more uniform means of claim-making (whether contentious or otherwise) than low-capacity states. That for several reasons: because high-capacity states connect dispersed actors, including challengers, more effectively with each other, thus promoting their mutual learning and collaboration in the formulation of claims; because obtrusive high-capacity states themselves generate higher proportions of all contention, hence imprint their own rhythms and structures on claim-making routines; because such states also tend to create uniform administrative organization throughout their territories as compared with the regional particularism of low-capacity states, a circumstance that increases the similarity of situations stimulating and channeling claim-making in different segments of the population under a high-capacity state’s control. For these reasons, modular repertoires—bundles of performances easily transferred from one locality, population, issue, or organization to another—should prevail in high-capacity states.

What about differences in repertoires between democratic and undemocratic regimes? Figure 4 schematizes a crude first cut. It argues that both state capacity and democracy affect overlaps among prescribed, tolerated, forbidden, and contentious public political performances. How? First, democratic regimes absolutely prescribe relatively few such performances, but they tolerate quite a range; while military conscription, tax payments, and replies to censuses come close to being compulsory for affected parties in democracies, even registering to vote remains voluntary in most democratic regimes. High-capacity non-democratic regimes, on the other hand, commonly prescribe a wide range of public political performances while tolerating few others. They also forbid a much wider variety of claim-making performances.

Figure 4: Configurations of Political Interaction under Different Types of Regime

Second, democratic regimes draw contentious claim-making toward their prescribed and tolerated forms of expression because access to power and recognition regularly pass through effective uses of those forms; thus electoral campaigns and sessions of legislative assemblies become foci of claim-making, even on the part of contenders that currently exercise little or no power. High-capacity non-democratic regimes, in contrast, typically exclude contentious issues and actors from prescribed and tolerated forms of claim-making, with the consequence that dissidents make their claims either by covert use of tolerated performances such as public ceremonies or by deliberate adoption of forbidden performances such as armed attacks.

But state capacity matters as well. According to the arguments embedded in Figure 4, low-capacity undemocratic regimes tolerate a relatively wide range of contentious claim-making, for three reasons: 1) they lack the means to prescribe many performances, and therefore settle for tribute, ritual obeisance, and a few other services from subjects; 2) they also lack the means to police small-scale contentious claim-making throughout their nominal jurisdictions; 3) their efforts to impose cultural and organizational uniformity throughout their jurisdictions remain weak and ineffectual, with the consequence that actions, emergence processes, and trajectories of contentious politics vary greatly from region to region and sector to sector.

On the democratic side, similar arguments apply. Low-capacity democratic regimes have rarely formed in history, and even more rarely survived; most have taken no more than a local scale. When they have existed, however, they have typically prescribed few performances, tolerated a great many, and passed a great deal of their public life in contention among conflicting claims, factions, and forms of action. From long Mediterranean experience with city-states, Aristotle recognized the vulnerability of low-capacity democratic regimes to takeover by factions and to external conquest. They also appear to fragment easily into polities organized around rival—or at least distinct—governments. Since many of today’s emerging democracies build on relatively low-capacity states, any leads we can find to the operation of low-capacity democratic regimes should illuminate struggles going on in the contemporary world.

These conjectures about variability of repertoires require refinement and empirical verification. They nevertheless fit recent western history well enough to encourage us in thinking that regimes varying along the two major axes—state capacity and undemocracy/democracy—generate significantly different qualities of contentious politics. State capacity and democratization therefore get much more attention than other aspects of regimes. But we break down the analysis of democratization into four dimensions: breadth of polity membership, equality of polity membership, strength of consultation, and protection. At our most general, then, we are asking how a regime’s position within the five-dimensional space interacts with the character, trajectory, and dynamics of contentious politics within that regime.

Here is the first question that emerges from such an agenda: How does the character of a regime affect a) the forms of contentious politics that occur within its perimeters, b) the dynamics of contentious politics within its perimeters? Our second question follows: How do changes in a regime’s character affect changes in forms and dynamics of contention? Translation: how do

changes in 1) state capacity 2) breadth of polity membership, 3) equality of polity membership, 4) strength of collective consultation, 5) protection of polity members from arbitrary action by state agents

affect changes in

6) repertoires of contention, 7) paths of claim-making, 8) parties to claims?

Which leads effortlessly to the third question: How do changes in repertoires of contention, paths of claim-making, and parties to claims affect characteristics and trajectories of regimes? More particularly, we are searching for partial causal analogies in these respects that cut across considerable ranges of regimes and contention. Those sought-for causal analogies three main clusters of phenomena: actors, actions, and identities in contentious politics; the emergence of contention; and trajectories of contentious struggle.

Most of the answers I have proposed here cling to comparative statics: they say what sorts of political contention we might expect to find at different positions along the five continua, or at best what sorts of changes we might expect to see as a regime moved along the continua. That happens partly because taxonomic reasoning invites comparative statics, partly because the causal arguments in and behind these conjectures remain gross or poorly articulated.

Nevertheless, reflection on regime variation and contention opens a promising program for research. The comparative program locates different contentious processes along the five dimensions of regimes and the trajectories of change they imply; those static relations deserve closer empirical attention. In a partly separate enterprise, we should be examining such change processes within well-defined historical settings where we can identify available models of political practice as well as current international constraints on regimes and contentions, for example in the turbulence of Eastern Europe’s post-communist political change. A third somewhat different research line follows particular mechanisms such as brokerage and identity formation across different regimes and varieties of contention, for example by looking for causal analogies between their operation in nationalism, ethnic conflict, and non-ethnic social movements. The agenda will keep students of comparative politics and political contention busy for quite a while.

Note

Let me explain what’s going on here. Doug McAdam, Sidney Tarrow, and I began our still-incomplete book Dynamics of Contention with the idea of relating processes and instances of contentious politics to variation in political regimes. Early versions of this paper took shape as chapter two of that book. The farther we got, however, the more we moved away from regime questions and toward the search for causal analogies across different sorts of regimes and of contention. We eventually expelled the chapter from our book, and heaved a collective sigh of relief. Paradise lost was space regained. Meanwhile, I was beginning a (likewise still-incomplete) book on contention and democracy in Europe. There, regime change figured crucially and inevitably. To clear my mind for that book and to entice other scholars into discussions of interactions between regimes and contention, I have reshaped my fallen angel into a free-flying bird. (Whether eagle, albatross, or dodo is for readers to decide.) In similar processes, I have also produced three other unpublished papers that overlap and complement this one: “Armed Force, Regimes, Contention, and Democratization in Europe since 1650,” “Processes and Mechanisms of Democratization,” and “Contentious Conversation.” I will be happy to inflict those papers on readers who agree to criticize them.

References

Barnes, Jonathan. 1984. ed., The Complete Works of Aristotle. Princeton: Princeton University Press. 2 vols.

Dahl, Robert A. 1975. “Governments and Political Oppositions,” in Fred I. Greenstein and Nelson W. Polsby, eds., Handbook of Political Science. Volume 3. Macropolitical Theory. Reading, Massachusetts: Addison-Wesley.

Finer, S.E. 1997. The History of Government from the Earliest Times. Oxford: Oxford University Press. 3 vols.

Lichbach, Mark Irving & Alan S. Zuckerman. 1997. Comparative Politics. Rationality, Culture, and Structure. Cambridge: Cambridge University Press.

Lindblom, Charles E. 1977. Politics and Markets. The World’s Political-Economic Systems. New York: Basic Books.

Scott, James C. 1985. Weapons of the Weak. Everyday Forms of Peasant Resistance. New Haven: Yale University Press. _____ 1990. Domination and the Arts of Resistance. Hidden Transcripts. New Haven: Yale University Press. _____ 1998. Seeing Like A State. New Haven: Yale University Press.

Tilly, Charles. 1998. Durable Inequality. Berkeley: University of California Press.