The crisis vocabulary, with mechanisms
Polycrisis, permacrisis, metacrisis, omnicrisis. Each word names something real, and each is used far more loosely than it deserves. This page asks of each one what it claims, what mechanism would make the claim true, and what it asks you to do.
Crisis as decision
Koselleck traces the word to the Greek krisis: judgement, decision, the turning point of an illness at which the outcome is settled. A doctor watching a fever was not watching a catastrophe. She was watching for the moment when the patient would turn one way or the other, and when what she did next would matter most.
So in its original sense a crisis is a moment that demands discernment. It has a before and an after, and it calls for someone to judge.
From moment to condition
Much of the current vocabulary has lost that meaning. Crisis has drifted from the moment of decision to a permanent condition, something we live inside rather than something we pass through. The newer words mostly describe that condition from different angles: how long it lasts, how its parts connect, what generates it, and who has the standing to deal with it.
That drift matters, because a condition does not call for judgement in the way a turning point does. It calls for endurance, or for atmosphere. The aim of this page is to get the decision back.
The family
Each word, with the mechanism that would make it true
Used loosely, each term is atmosphere. Used with its mechanism, each forces a specific question.
| Term | Source | The question it forces | Complexity mechanism | Failure when loose |
|---|---|---|---|---|
| Crisis | Koselleck, 2006 | What has to be decided now, and by whom? | A threshold, where the outcome is still open and timing decides it | Assumes a stable normal to return to |
| Creeping crisis | Boin et al., 2020 | What threat is developing slowly, in plain sight, without being addressed? | Slow variables; critical slowing down | Labelled only in hindsight |
| Transboundary crisis | Ansell et al., 2010 | Which crises cross jurisdictions, sectors and time, outrunning any single authority? | Coupling across scales | Coordination talk without mandate |
| Permacrisis | Collins Dictionary, 2022 | What changes when crisis becomes the normal condition? | Persistent regime; chronic stress | Fatalism and fatigue |
| Polycrisis | Lawrence et al., 2024 | How are crises in different systems causally entangled? | Common stresses, domino effects, inter-systemic feedbacks | Mood without mechanism |
| Metacrisis | Rowson, 2021 | What generates the entangled crises in how we perceive, value and coordinate? | Generator dynamics; reflexivity | Totalising and hard to falsify |
| Omnicrisis | Patriotta & Voronov, 2026 | Whose authority to define and repair the crisis is in question? | Tipping in institutional and epistemic orders | Relativism about expertise |
| Super wicked problem | Levin et al., 2012 | What happens when time is short and the solvers are also the cause? | Irreversibility; discounting of the future | Paralysis |
A ladder of depth
From an event to a way of seeing
Each rung locates the explanation deeper, and asks more of a response.
| Rung | Core claim | Where it locates the explanation | What it asks of a response | Main weakness |
|---|---|---|---|---|
| Crisis | A discrete disruption with a beginning and an end | Event | Restore normal function | Assumes a stable normal to return to |
| Permacrisis | Crisis is the prolonged normal condition | Time and duration | Endurance and adaptive capacity | Descriptive, with no mechanism |
| Polycrisis | Crises in different systems are causally entangled | Inter-system mechanisms: common stress, dominoes, feedbacks | Decoupling, buffers, cross-sector anticipation | Becomes a mood unless the mechanisms are specified |
| Metacrisis | Entangled crises share generative roots in how we perceive, know, value and coordinate | Worldviews, sense-making, civilisational dynamics | Change in meaning, coordination and the capacity to act | Hard to falsify, and can be totalising |
| Omnicrisis | Disruption spans institutional domains and becomes a crisis of authority and expertise | Institutions and occupational groups | Coordinated repair across expertise; building a new normal | Newest; its empirical base is still forming |
The exchange
What each side gives the other
What crisis thinking gives complexity
Stakes. Without the crisis literature, emergence can be treated as an intellectual curiosity: interesting patterns in ant colonies and traffic. Crisis thinking insists that the same dynamics are breaking things people depend on, and that the time to understand them is short.
It also brings in politics, which complexity science tends to leave out. Crises redistribute power. Every crisis frame advantages some actors over others, because whoever names the crisis also names what counts as a response (Boin et al., 2016). A model of cascades that ignores who benefits from calling something a cascade is missing half the system.
Two of the newer terms add something more. The metacrisis literature adds interiority: the worldviews and meanings that shape what people perceive and value, and so what they do. The omnicrisis literature adds authority: the question of who is entitled to define the crisis and lead the repair.
What complexity gives crisis thinking
A demand for mechanisms. A polycrisis claim should say which pathways connect which systems. Lawrence and colleagues name three: common stresses, where one underlying stress weakens several systems; domino effects, where a crisis in one system triggers crises in others; and inter-systemic feedbacks, where crises amplify each other through loops. Without that, the vocabulary becomes atmosphere.
A shift from event to regime. Complexity reads a crisis as a system operating near its thresholds, with heightened sensitivity and reduced resilience. That moves attention away from the headline event and towards early warning, slow variables and buffers. The question becomes what is eroding quietly, and how much slack is left.
And it gives back krisis. A system near its thresholds is exactly where small, well-timed actions can have large effects (Scheffer, 2009; Centola et al., 2018). So crisis is the moment of greatest leverage as well as greatest danger. That is the original meaning restored: the moment that calls for judgement.
A warning
When the word does the work
Roitman’s warning
Roitman argues that “crisis” works as a blind spot. To call something a crisis is to posit a norm from which the present has departed. That forecloses a question: whether the norm was the problem. It also authorises some actions and makes others unthinkable. Once the word is in play, returning to normal looks like the only sensible aim.
Complexity thinking supports the critique. In adaptive systems there is no stable normal to return to. Recovery means reorganisation (Gunderson & Holling, 2002), or in Patriotta and Voronov’s terms, the construction of a new normal.
What survives the critique
The vocabulary is still useful, used in a particular way. Crisis language is defensible when it is diagnostic and time-bound: it names the type of crisis, the mechanism it points to, and the decision it calls for, and it stops being used when the decision has been made. Crisis language that only creates urgency, without naming a mechanism, is atmospherics.
The test is simple to state. If you removed the word, would anyone know less about what to do? If not, the word was setting a mood and doing no analysis.
For futuring
A layered vocabulary for scanning
A layered vocabulary
Futures work gains precision by being explicit about level. Polycrisis is for cross-system mechanisms. Metacrisis is for the underlying drivers and worldviews. Omnicrisis is for institutional authority and expertise. They are complements, not competitors. A serious practice needs all three in view, and has to be clear at any moment about which one it is using.
Each step down the ladder moves the explanation deeper and widens what counts as an intervention. Polycrisis is the most tractable and the best connected to complexity science’s tools. Metacrisis is the most ambitious and the most exposed to the charge of vagueness. Omnicrisis is the most institutionally precise.
Which signals each level tells you to watch
This is where the vocabulary earns its place in a scan. In the futuring strand, signals are sorted by lens: social, technological, economic and so on. Sorting them by level as well changes what you look for.
At the polycrisis level, watch for coupling: a stress showing up in more than one system at once, or a failure in one sector appearing in another sector’s data a few months later. At the metacrisis level, watch for shifts in what people take for granted: the futures they assume, the things they stop arguing about, the meanings that change before the behaviour does. At the omnicrisis level, watch for contests over standing: professions disagreeing in public, guidance being ignored, new voices claiming the right to define the problem.
Two other words add signals of their own. A creeping crisis tells you to watch slow variables, the ones that move too gradually to make the news, and the sign that recovery from small shocks is getting slower. Permacrisis tells you to watch the people as well as the system: fatigue and fatalism are themselves forces in how a system behaves. A vocabulary that only produces dread will be ignored, so the scan has to keep asking what can be done at a human scale.
Reading
Where it comes from
Start with the first on the list.
Koselleck, R. (2006). Crisis (M. W. Richter, Trans.). Journal of the History of Ideas, 67(2), 357–400.
The history of the word, and the source of crisis as the moment of decision.
Roitman, J. (2014). Anti-crisis. Duke University Press.
The strongest case that crisis narratives hide the question of whether the old normal was the problem.
Lawrence, M., Homer-Dixon, T., Janzwood, S., Rockström, J., Renn, O., & Donges, J. F. (2024). Global polycrisis: The causal mechanisms of crisis entanglement. Global Sustainability, 7, e6. https://doi.org/10.1017/sus.2024.1
The most serious attempt to turn polycrisis from a mood into named causal pathways.
Rowson, J. (2021). Tasting the pickle: Ten flavours of meta-crisis and the appetite for a new civilisation. Perspectiva.
The most careful treatment of metacrisis, and one that anticipates the objections to it.
Patriotta, G., & Voronov, M. (2026). Crises unbound: Institutions, occupational groups, and the construction of omnicrises. Journal of Management. Advance online publication. https://doi.org/10.1177/01492063261448817
Omnicrisis as a crisis of authority as much as of events, and recovery as a new normal.
Boin, A., Ekengren, M., & Rhinard, M. (2020). Hiding in plain sight: Conceptualizing the creeping crisis. Risk, Hazards & Crisis in Public Policy, 11(2), 116–138.
Why slow-building threats go unaddressed until they are obvious.
Boin, A., ’t Hart, P., Stern, E., & Sundelius, B. (2016). The politics of crisis management: Public leadership under pressure (2nd ed.). Cambridge University Press.
Crisis as politics: who gains from a frame, and what leaders do under pressure.
Levin, K., Cashore, B., Bernstein, S., & Auld, G. (2012). Overcoming the tragedy of super wicked problems: Constraining our future selves to ameliorate global climate change. Policy Sciences, 45(2), 123–152.
Problems where time is short and the people solving them are also causing them.
Scheffer, M. (2009). Critical transitions in nature and society. Princeton University Press.
Thresholds, tipping and early-warning signals, the mechanics behind crisis as regime.
Gunderson, L. H., & Holling, C. S. (Eds.). (2002). Panarchy: Understanding transformations in human and natural systems. Island Press.
Why recovery in an adaptive system means reorganisation, not a return.
Unsettled
What I would still argue about
The questions this page does not settle.
- How to tell a creeping crisis from ordinary noise while it is still creeping. Hindsight is the usual failure mode, and I have no method that reliably beats it.
- Whether metacrisis claims can be made testable enough to guide action, or whether the term is most useful as a prompt to look at worldviews, and no more.
- How to rebuild professional authority in an omnicrisis without reverting to closed expertise. Participation is the usual answer. It is slow, and crises are not.
- How to keep crisis language time-bound inside institutions that benefit from the urgency it creates. Roitman explains why the word persists. She does not say how to retire it.