This is a little bit of a trick question. It really implies three questions: What are the circumstances that make unrest in a population possible or likely? What circumstances need to occur in order to precipitate expressions of unrest in particular places? And what circumstances are conducive to spreading (or damping out) these local expressions?
First, how might we define the concept of "unrest"? To my ear the concept involves grievance and activism. Grievance involves the situation where individuals and groups feel that they have been badly treated by someone. Activism implies a disposition to act visibly and politically to protest or alleviate this mistreatment. Grievance can exist without activism, and there are instances of activism that stem from emotions other than grievance. But when these emotional and behavioral states come together we can refer to the resulting stew of behavior as "unrest".
So let's start with the causes of grievance. Power relations create the emotions of grievance: excessive conscription or taxation, insufficient attention to the interests of an ethnic minority, abusive and disrespectful treatment by the police. When individuals and groups believe they are being treated in ways that unfairly harm their interests or reduce their dignity, they are likely to feel aggrieved.
Grievance is a propositional emotion; it involves a subject, a harm, and a perpetrator. And this means that grievance is not simply a reaponse to suffering. Take a population that is experiencing dearth in the early stages of famine. Individuals and sub-groups may differ in their experience of grievance; they may hold different social actors responsible for their suffering (landlords, lenders, city people, the military, or the state, for example). And these differences have implications for how and when these groups may be aroused to protest and action.
What about "activism"? What kind of psycho-political state is this? It is a propensity to make the transition from political emotion to action -- to go from resentment of the state's behavior to the choice of joining a street demonstration; to go from anger about conscription to joining an anti-draft organization; to go from frustration about the landlord's unwillingness to restore the heat to joining with others in a rent strike. "Activism" appears to be a complex characteristic of individuals and groups. For one thing, it seems to have a substantial component of culture and tradition baked into it. Cultures seem to differ in their responses to mistreatment; some communities seem to have resources for activist mobilization that others lack. Second, there appears to be a substantial degree of social learning through imitation involved in becoming "activist." So it is likely that there is a degree of positive feedback involved in the spread of activist psychology.
So back to the original question: what causes the spread of unrest? There needs to be an issue that creates a grievance in a significant number of people. Something needs to happen to make this issue salient relative to other concerns. There needs to be a critical mass of people who share the grievance and possess the components of the social emotions of activism. And there needs to be a "spark" that allows activists to mobilize others.
Consider a hypothetical example -- a company with dozens of factories in different parts of the country that is imposing a unilateral change in its contributions to worker retirement accounts. Suppose each factory has several thousand workers; and suppose that there is a range of responses to the retirement changes in the various factories along these lines:
- Quiescence and grudging acceptance
- Widespread grousing but no organized action
- Wildcat strikes
Several possible explanations might be considered:
- The presence/absence of effective rank-and-file leaders
- The presence/absence of effective local managers' countermeasures (persuasion, cooptation, threats)
- Strong/weak traditions of activism in different locations
- Alternative narratives about what the changes mean ("inevitable in this business climate", "better this than a lot of layoffs", "higher management is taking this opportunity to stiff us", ....)
- High/low impact of the management changes on the interests of workers in each location
- Strong/weak channels of communication among workers in different factories
We might speculate, then, that unrest is most likely to occur and spread when there is an abuse that affects a large number of people; there is a generally shared understanding of the nature of the abuse; there are effective local activists capable of arousing the indignation of the rank-and-file; there are accessible communication vehicles permitting the spreading of messages of dissent; the population has a tradition of activism; and the state's managers are ineffectual in damping down the occurrences of protest. These conditions appear most favorable for the dissemination of unrest.