Dynamics of Contention

Review by Andy Blunden of “Dynamics of Contention” by Doug McAdam, Sidney Tarrow and Charles Tilly, Cambridge University Press, 2001.

This book outlines a general approach to the analysis of social movements understood in the broadest possible sense of the term. The book was co-authored by three writers who were each at the time of writing already recognised authorities in the discipline, and in the preface, 132 individual scholars and institutions were acknowledged and thanked for their collaboration over a period of 5 years. So this is clearly an authoritative work to which anyone claiming to innovate in the theory of ‘contentious’ social change needs to respond.

The first move the authors make is to reject the project of a theory of ‘social movements’, understood in the narrow sense as a kind of social phenomenon distinct from, for instance, revolutions, democratisation and nationalism, proposing instead a broad domain of ‘contention’ in which processes may move from one kind of process to another, such that a process under study could not be examined without taking into consideration its relation to and transformation to and from other kinds of contention, and by relying on processes observed in other kinds of contention. Although the material examined in the book is limited to contentious processes in the political domain, with national scope, and focuses on what they call ‘transgressive’ contention, they expect (correctly I believe) that their results would be equally as applicable to contentious activity that is not political as such (e.g. within other kinds of institution), may be quite localized in scope, and may be carried out by existing actors within the bounds of the kind of activity in which they are expected to engage ('contained’ contention, i.e., not transgressive). The analysis is organised around nine case studies which are so chosen as to include classic instances and cover a wide range of historical and socio-cultural settings. In other words, the authors proffer a general theory of social contention.

Secondly, the authors also make it clear that they do not propose, and indeed they oppose, any attempt to formulate covering laws supposed to govern the emergence and course of contentious political episodes, including standard models and trajectories. In this also I think they are correct. As Paul Ricœur (1984) has argued so forcefully, the point of science in this domain is to make history intelligible not to discover the ‘laws of history’. This does not exclude, however, the idea of a logic of contention, a possibility that the authors do not explicitly examine. Further they emphasise the ubiquity of contingency and while making the trajectory of some episode intelligible they emphasise that in another instance the same process could unfold in the opposite trajectory – the theory is explanatory, but not predictive. The authors prefer to see episodes of contention (and it is impossible to talk of more than ‘episodes’ if one rejects ‘laws of history’) as conceivable under the concept of causality. But every cause is, of course, also an effect in the infinite chain of causality.

Thirdly, the authors distinguish their approach from Structuralism (and I have to agree with, for example, Anthony Giddens (1984), that “there is no such entity as a distinctive type of ‘structural explanation’ in the social sciences”), Rationalism (that is: “explaining social processes in terms of rational decisions made by individuals in the light of previously defined interests, resources and situational constraints,” p. 21) and phenomenological approaches.

However, the authors include under phenomenological approaches what they call Culturalism which “attributes causal power to norms, values, beliefs and symbols.” Insofar as they mean to reject mentalist explanations of social phenomenon, I must agree, but I am at a loss to understand how phenomena of social change can be understood without recognising the power of norms and symbols. Further, I think it untenable to draw a line between what can be counted as “beliefs and symbols” on one hand, and social and material resources on the other. I believe that the authors wish to distance themselves from the focus on mental or textual explanations of social behaviour, but norms and symbols constitute the very fabric of social formations of all kinds.

In rejecting both structuralist and mentalist approaches, the authors definitively focus on the meso level of human life, processes which are beyond the scope of actions of a single individual or group of associates, but not so extended as quasi-objective processes, historical products such as nation-states, markets, languages, etc. which are beyond the scope of individual actors to create or destroy. The meso level includes projects which an individual actor can launch but much more usually joins, and which have a beginning, a middle and an end – a life-cycle – and in that sense can be described as ‘episodes’ even if they extend for decades and pass through multiple phases. This way of delineating the scope of their discipline makes sense, in opposition to efforts to draw lines between different kinds of meso-level processes or subsume meso-level processes under mental processes on one hand or structural processes on the other. However, I question whether such an approach is feasible so long as it is blind to the processes known to psychology and a broader historical perspective, and incomplete so long as it relies on other sciences for these.

Although the authors are at pains to point out that social actors are mutable, unstable and overlapping entities ? which appear from seemingly nowhere, transform one into another, shift their object, upshift and downshift in scale, and change their identity ? they never question that social actors are groups of people. What the authors present as a virtue of their approach (that they allow that social actors are fluid and changeable) is more correctly seen as a symptom of the difficulty their approach creates for itself with their conception of social actors as collections of individual persons. I suggest instead that the actors in social life are projects (or practices or activities) understood as aggregates of individual actions, not aggregates of individual people.

What then, is to be the content of a ‘dynamics of contention’ as opposed to the description and explanation of a specific historical episode?

What the authors offer is a range of different species of mechanisms and processes which are to be found in appropriately modified forms in political contention in almost any social, cultural or historical setting – a kind of vocabulary of social mechanisms. They name and describe a range of such mechanisms and their varieties which would be familiar to most of us who have studied or have wide experience in contentious activity, but may not have identified as more or less universal mechanisms of social change as opposed to sui generis episodes. All of the mechanisms are ever subject to contingency.

The mechanisms include:

Constitution of new political actors and identity formation

Actor constitution (a.k.a. political identity formation) is not a mechanism but rather a fact which needs to be explained by the various mechanisms. The most important mechanisms of actor constitution are category formation, certification / decertification and collective attribution of threats / opportunities.

For these authors, actors which did not exist one moment may be there the next and the real interest is not so much what the actors do once constituted in new social arrangements, but how they are brought into being as new actors.

Category formation (via invention, borrowing or encounter)

My claim is that the actors in society are activities, i.e. aggregates of actions, not categories of persons, so category formation is a challenge. Category formation is clearly a reality; as a scientist I may see that social actors are activities (or projects), but the logic inculcated into the minds of the ordinary citizens, whether by tribalism or bureaucratism, is that the world is made up of categories of people and it is these classes which are the actors in history and social life. Nonetheless, it is activities which create categories of persons, and the authors define three methods by which this is done.

Invention is the authoritative drawing of a boundary on some basis and the proscription of relations across that boundary; borrowing entails the adoption of a boundary practice from elsewhere; and encounter occurs when two internally coherent practices come into contact with one another and through some kind of competition generate a boundary between them. All of these means of category formation are of course activities not groups, so the idea that the social actors produced are groups of people is a reification of exactly the kind being described. Categories of persons are in fact the individual actors implicated in specific actions – always a fluid and ambiguous entity with fuzzy boundaries.

Category formation is itself a mechanism to be explained by other mechanisms.

Certification and Decertification (by an external authority)

An actor can be brought into being by being certified by an existing external authority or social actor; a new nation is born by certification by the UN or through recognition by other nations. Certification may be the recognition given to a movement by a prominent social figure or institution, but these actions are more about the ‘enhancement’ of a political identity, or the conversion of a non-political identity into a political identity, or implicit identity into an explicit one, rather than creation of an identity. Decertification can also bring a political identity into existence where none existed previously. For example, shooting wildlife for sport is transformed from a pastime implying no kind of identity at all until it is made illegal, upon which shooting may become the marker of being an outlaw and could generate a powerful collective agency and identity.

However, I have to make two qualifications about this idea. (1) In every case there is already a movement or social practice which is in fact at least part of a social identity, and what certification does is give recognition, reality if you will, to what was previously only implicit. Of course, it is no small thing if one contingent attribute of a person (such as their religion) is suddenly picked out and made the marker of an illicit or valued identity. (2) Certification and decertification are actions, but if there is no action in response to them, then no change in political identity takes place. For example, the banning of hunting might generate a Shooters Party or some such thing, or it might not -shooters and/or park rangers could simply ignore the law, or the shooters could passively accept the new law. Whether or not the (de)certification results in the formation of a political actor is a question of activity, practices, not persons or groups of persons.

Collective attribution of Threats and Opportunities (and ‘spirals’)

Essential to the constitution of some practice is the collective recognition of an object of activity. Generally such objects are a threat to be countered or an opportunity to be seized, and it is the collective pursuit of that object which constitutes the activity as a social actor.

The authors correctly emphasise that attribution is the essential element, and not the existence of the threat or opportunity in itself. Indeed, there might be no objective change in the situation of a group of individuals at all, but some action may trigger the relevant attribution and thereby the associated activity.

This mechanism is one of the most frequently referenced in the book, and that is as expected, because it is the object of activity which is the distinctive feature of any activity or project. On the other hand, the idea of the ‘object of a group’ is actually quite incoherent.

Suddenly-imposed grievance

The authors give special mention to the sub-type of this mechanism: ‘suddenly imposed grievances’, because although all the same qualifications may apply, the sudden imposition of some measure which is attributable as a threat or attack can be particularly effective in triggering subject-formation.


While under certain conditions a repressive move can prod a new social actor into life, successfully applied, repression can also suppress or extinguish an actor.

Scale Shift (either upward or downward)

An important mechanism for the creation of a new political actor is upshift in the scale of an activity. For instance, if participants in a neighbourhood argument call in their respective religious, family, ethnic or other leaders to take their side, an argument with no political significance is thereby transformed into a wide-scale political dispute between parties whose identity was previously implicit and of little significance. Again, it is not the upscale of the number of people, but of the scale of activity which is actually at issue.

Contrariwise, if protagonists, for example, at the national level, decide to devolve their activity down to the local level where matters can be sorted out more concretely, the result can be the dissolution of the national-level actor.


Mobilization is the initiation and maintenance of some form of practice, particularly some energetic collective practice, the actualisation of a practice which constitutes the bringing-into-being of a social actor.

Mobilization is another one of those facts which need to be explained by the mechanisms, though it is of course a mechanism in itself.

The chief means of mobilization, according to the authors, are (1) collective attribution of threats and opportunities, often resulting in ‘spirals'; (2) Social appropriation; (3) Innovative action; and (4) Framing, though the authors de-emphasise framing because they see it a part of the established social movement theory. Framing is easily subsumed under the other three mechanisms.

Social appropriation and Innovative action

Generally speaking, an activity (or practice or activity) is characterised by the object of the activity and the means of its pursuit. Often it is not the object pursued but the means of pursuing it which is novel. The authors mention two kinds of novelty in the formation of a new activity: the appropriation of an existing vehicle (organisation, institution or practice) for a new purpose, and the invention of an entirely new mode of action.


Second and alongside attribution of threat or opportunity as a means of mobilization is the form of social appropriation the authors call brokerage, that is, the use of a mediating activity (conceived of as a social stratum) in order to mobilise actors otherwise out of reach of the practice. This includes ‘gatekeepers’ but, more importantly, militant or trusted social strata who are responsive to the object and can communicate it to a wider public who would otherwise be indifferent to the project or unaware of it.

Co-optation of intermediaries

Brokerage is especially significant when leading layers of a successful movement are co-opted into the government bureaucracy, as what is normally called a social movement transforms itself into a mode of regulation and administration as a branch of government. This the process frequently decried by activists who conflate the relative success of their movement with its demobilisation as a ‘social movement’ in the narrow sense of the term.

Escalation & Political Competition

Escalation is the process whereby resistance to a practice energises a practice and what may have been a barely self-conscious practice transforms itself into a self-conscious movement in its effort to overcome the resistance.

A special case of escalation is where there is competition between rival projects in some domain and one or all protagonists successively escalate their efforts in order to succeed in colonising a given space, generating self-conscious, mutually hostile social actors.

Radicalization, Polarization & Convergence (radical flank effect)

In particular, escalation can take the form, not only of energising activity, but of radicalising the objectives and means of action to become more challenging and transgressive.

A particular case of radicalization is where two practices in competition with one another (such as rival political parties) polarise in their mutual efforts to overcome one another.

The radical flank effect is the reaction to polarization where the opening up of a political space in the middle leads to convergence.

Identity Shift & Object Shift

In the course of pursuing its object and possibly forming alliances and encountering protagonists, a social actor unfolds the inner logic of the object itself: in the beginning they were pursuing this or that opportunity or countering this or that threat. However, as the practice unfolds the perception of the object changes. These changes are rooted in the conception and the contradictions inherent in the concept of the object. As the practice reorients and changes its object, its self-consciousness changes and this constitutes an identity shift. Object shift is just the other side of identity shift – sometimes the reorientation is not registered as a change in the self-consciousness of the actor but is attributed to a change in their collective object.

(Cross-class) Coalition Formation

It is well-known that every social movement expresses the interests of some social class. However, so long as the movement expresses the interests of only one social class it has little chance of success. This is why Marx defined the proletariat as a ‘universal class’, because its liberation meant the emancipation of all classes from the oppression of the ‘great stumbling block’, the big capitalists. Consequently, an important mechanism in the development of any movement is its escape from a narrow class base and its constitution as a cross-class coalition. Coalitions between any two social movements is an important development irrespective of the respective social bases.

Elite Defection & Infringement of Elite interest

The most dramatic variant of cross-class coalition formation is elite defection. Even the defection of a single individual from the elite (c.f. certification) can be the mechanism for dramatic escalation or energising of a movement.

On the other hand, when a movement ‘crosses a line’ where members of the elite suddenly perceive an infringement of their interests, elite support which had hitherto been at least implicit, can be a suddenly withdrawn (for example, support for a leader whose shortcomings had hitherto been tolerated) or a sudden crackdown on government opponents.

Emulation & Diffusion

Social movements frequently puzzle over the problem of how to spread their movement to new pastures, recruit more adherents and convince more people of the merits of their object. Generally speaking it is unwise to simply rely on ‘contagion’, reliant on attribution of similarity, that is, that others will see themselves as being similar or as being in the same situation and join the movement. Deliberate attempts to recruit others by means of brokerage are more likely to be successful.

Emulation refers to the deliberate decision of others to emulate a form of practice pioneered elsewhere. It is important to recognize that it is not enough to see or hear about a practice, for example in the media. Real work is required to institute a new form of practice and will only be successful if the soil is fertile and the seed is well planted. It is one thing to hear about an idea, and agree with it, another to implement as a practice. You have to learn how to do it.

Dissolution of Patron-client & Trust networks

The authors also mention changes in the social environment which can generate changes in social movements, changes brought about by the progress of modernity and the advance of bourgeois society. First is the dissolution of patron-client networks which are the mainstay of traditional and feudal societies – once these dissolve under the pressure of commerce and foreign influence, individuals are free to adopt new practices. Secondly, the authors mention alteration of trust networks which can undermine former institutions of cooperation and democratic decision making, opening the way for new practices to development and new social actors to make their appearance.


The aim of the authors to create a theory of meso-level social processes located exclusively in the meso-level by identifying a range of mechanisms which anyone with experience in social action will be able to understand more or less viscerally, and from there reconstruct an intelligible explanation of any given episode of contentious politics.

This proved to entail the unfortunate fact that which mechanism will be active in a given situation, and which outcome will result, is left at the mercy of stimuli which come from outside the domain which the theory is able to conceptualise.

Since the authors are at pains to eschew mentalist explanations they generally exclude explanation from the actions of individuals (other than members of the elite or institutional representatives able to certify or command movements). This is a mistake because the mechanisms are activated if not actually constituted by actions, which are necessarily the actions of individuals, since collectives can act only in and through the actions of individuals. As remarked above, social actors are activities (a.k.a. practices or projects) which are aggregates of actions, not aggregates of persons. So there is nothing mentalist (phenomenological, rationalist or ‘culturalist’) in referring to actions. For example, why did political competition in one instance lead to escalation and in another to convergence. The attempt to answer this vital question takes the researcher beyond the bounds of the theory outlined, at just the point when explanation is needed, rather than description.

On the other side, while rejecting Structuralism, the authors blind themselves to obvious sources of contentious politics in the broad processes of world history, processes which cannot be reduced to ‘mechanisms’ without losing the value of units of analysis confined to the meso-level of social action which are viscerally understandable as components of social actors, rather than features of an objective social environment.

However, the processes active on the stage of world history are not of a distinct ontological status as is implicit in Structuralism. They are activities (projects or practices) which have become institutionalised, and if we can break from this conception of the social world being composed of groups of people, then we can further the authors’ laudable project of removing from the discipline of ‘social movement theory’ boundaries which only bar the way to understanding the processes of interest.

The task then becomes explaining the mechanisms which the authors have identified and the conditions under which they are realized, and this will entail drawing on a conception of processes in the meso-level of social action on the understanding that all social processes and actors are aggregates of actions.


Ricœur, P. (1984). Time and Narrative. Vol. I. Chicago: University of Chicago Press.

Giddens, A. (1984). The Constitution of Society. The University of California Press.

Andy Blunden
7th March 2020