A guide to key concepts for policy analysis and public management in the Anthropocene (longer read)


I doubt anything like a formal manual would be possible or useful for complex policy analysis and management, today and ahead. Instead, this is intended as a guide to key concepts and a framework for undertaking analysis and management in the Anthropocene.

The most likely readers are those who understand that the Anthropocene requires different ways of thinking through and analyzing big policy and management issues. No more chop-logic about starting with risks, see what the tradeoffs are, and then establish priorities. No more about too-little too-late, there is no alternative but [fill in the blank], and anyway next is worse. Even where that holds, it holds only so far, and these chop-logics certainly do not go far enough. The Anthropocene is too complex for that.

Indeed, it is that complexity which enables recasting what are popularly called intractable problems. To telegraph ahead, a complex policy or management issue labelled “intractable” is one that has yet to be recast more tractably without simplifying the complexity.

Some policymakers, policy analysts and public managers already know this and are acting accordingly. Others will do so in the future, along with more social critics, policy and management academics, social scientists, and even policy wonks and media specialists. This guide is for those who are searching for key concepts and a framework for their application.


The guide’s framework has four pillars that differ from conventional analytical and management approaches:

–The guide is for major policy and management issues that are ineradicably complex. That complexity is on the rise because the number of components (elements) in major issues, the functions (roles) each element has, and the interconnections between elements and functions are increasing in ways that challenge measurement, monitoring and comprehension.

–The increased complexity does not mean intractability. The more complex, the more opportunities to recast the issue tractably. For this guide, the enemy of purportedly intractable policy and management is their complexity.

–This means policy analysts and managers can advise—more frequently and realistically than might be supposed—decisionmakers, “Even if what you say holds, you can go further. Here’s how and still be policy relevant. . .” None of this is easy or guaranteed.

–Difficulties, inexperience and not-knowing are always encountered in recasting and pushing further. Fortunately, the more complex the issue, the greater chances in usefully distinguishing between managing, controlling and coping ahead with respect to the complexity. Setbacks are expected but are more likely to be positive.

Together, the four pillars argue against thinking the Anthropocene can be universalized or reified or abstracted when in reality it is highly differentiated for the purposes of discrete policy and management. Here too many people already know this. Indeed, it has always been a complex, uncertain, interrupted and conflicted Anthropocene for much of the world.

Key concepts


In ordinary language and even though different, policy and management are considered intractable when complex, uncertain, unfinished and conflicted. Climate change is one obvious example.

To expand, a policy or management issue is uncertain when causal knowledge about it is found wanting by decisionmakers. Complex when the issue’s components or elements are more numerous, varied and interconnected. Incomplete, when efforts to address the issue are interrupted or left unfinished. And conflicted, when individuals take very different positions on the issue because of its uncertainty, complexity and incompleteness. To stop there would however be to end in the exaggeration of wicked problems.

The argument here is that problems in analysis and management of these issues arise when those relying on ordinary language do not differentiate terms, like uncertainty or intractability, as contexts change or already differ. Let me state the implication formally. When you qualify uncertain or complex or unfinished or conflicted by asking “with respect to what,” risk ends up being differentiated from uncertainty, uncertainty with respect to consequences is not the same as uncertainty over the likelihoods of those consequences, and unknown-unknowns are altogether another matter. These differences are important for better policy and management.

In the same way, “highly complex” and “intractable,” which are easily conflated in ordinary language, must be differentiated from the get-go. Complex, let alone complicated, does not mean intractable. In fact, the opposite holds when real people with real problems are operating in real time.

Since readers might consider “complexity is the enemy of intractable” to be counter-intuitive, let’s be clearer about the formal definition of complexity going forward.

Policy and management complexity

This guide adopts what is arguably the best-known definition, that of po­litical scientist, Todd R. La Porte and his construct of organized system complexity: “The de­gree of complexity of organized social systems (Q) is a function of the number of system components (Ci), the relative differentiation or variety of the components (Di), and the de­gree of interdependence among these components (Ik). Then, by definition, the greater Ci, Di, and Ik, the greater the complexity of the organized system (Q)”.

This definition has the merit of highlighting five features left ambiguous in ordinary language about complexity and by extension policy and management complexity in the Anthropocene. Discussion of the very important fifth feature—the increased opportunities to recast a complex issue—is left to the section immediately following.

The first feature that ensues from the definition: Complex is a comparative property of systems; that is, a system is more or less com­plex than another system in terms of the respective number of components, the differ­entiation of components, and their interrelatedness. While it is common to say, “this or that is complex (full stop),” such statements beg the question of more or less complex with respect to what: Just what is the baseline for establishing “complex”? Of course, the discussion of complexity has unavoidable ambiguities—just what is a “system” that it is more complex?—while still being less ambiguous than many ordinary language discussions.

Second, this definition of complexity illustrates how, say, the Earth can be the most complex ecosys­tem among ecosystems here (i.e., the Earth entails all its ecosystems, all their differing ecosystem functions and services, all their interconnections). The methodological point is not, however, that you “scale up to complexity,” but rather the system of interest becomes more (or less) complex by way of comparison.

Third, the definition illustrates how difficult it is to quantify complexity beyond numbers of compo­nents and functions attributed to each component. There is no broadly accepted quantitative measure of interconnectivity (a better term for our purposes, as some connections are unidirectional and not bidirectionally interdependent). The same could be said for identifying inter-related “functions.” Some ordinary language terms, nevertheless, such as “in­creasing resource scarcity,” capture a sense of the interconnectivity at the global scale.

Fourth, to distinguish a system’s components from each other, the different functions or roles each component has, and the interconnections between and among the functions and compo­nents is its own methodological imperative: First, differentiate! The more you differentiate the case at hand, the more unlikely you are to find reduced-form crisis narratives such as the Global Financial Crisis (the most salient feature of the 2008 financial crisis was that it was not global) or the Tragedy of the Commons (its premise of a homogenous pasture open to like herders is exactly what can not be assumed).

Recasting the intractable

The fifth feature following from the guide’s definition of policy and management complexity is by far the major one. For the chief feature of this complexity is surprise, and surely the greatest surprise is how many recastings are possible for issues of many components, multiple differentiations, and high interconnectivity. The recastings of interest here are those that keep the complexity for tractability purposes rather than conspire to reduce it in the name of Keep It Simple! To be clear, recast here does not mean simplify; its synonyms are: reframe, redescribe, recalibrate, revise.

To begin to see this kind of complexity and its import means you can start analysis anywhere. You see the forest; I see a mountain of poison against insects. I witness the birth of the family’s first child; you see the first child give birth to a family. I ask, when is biotechnology bestiality? You ask, are gardens zoos without the cruelty? Is burglary a kind of architectural criticism? Does burning down a lumber yard mean forgone structures have been destroyed?

Doesn’t our continuing inability to safely store nuclear weapons waste reveal the Cold War to be the first war in modern times where the continental US took direct hits because of an enemy? What does the US look like when it is a country where more men might be raped than women? (Think: prison male populations). What if those lengthy studies to model and validate the lifecycles of threatened species become weapons of their mass destruction?

Or consider recastings already familiar to some: General Motors—a pension system producing cars. McDonalds—a real estate multinational selling hamburgers. Uber—the world’s largest taxi company owning no vehicles. Facebook—the world’s most popular media owner creating none of its content. Alibaba—the most valuable retailer with no inventory. Airbnb—the planet’s largest provider of accommodation owning no real estate. The US government—a massive insurance conglomerate with an empire’s army.

Consider also less familiar recastings. Your bad policy mess: It’s said today some 790 million people remain without access to electricity and 2.6 billion people depend on polluting fuels for cooking. At one point, three to four billion people—up to two-thirds of the world’s population—lived in regions without adequate water supplies or sanitation. More recently, it’s been estimated that 2.2 billion people on Earth live without safe drinking water. My good policy mess: Truly those are very, very large numbers, right? In fact, even today the distribution of people worldwide without adequate water supplies and energy is so large that many of them must be doing better than others. That means there are tens of millions—hundreds of millions?—of people who do not see themselves as victims and who have helpful things to say about how to better survive without adequate water to those millions more who do see themselves as victimized because of lack of adequate water. But where then are the campaigns, e.g., in the World Bank or the IMF, to do just that?

His bad policy mess: It has been said that one out of every two young African-American men in major US urban areas is enmeshed in the criminal justice system. But that too is a very large number. Her good policy mess: Why are we not interviewing the other 50 percent of young urban African-American males outside the criminal justice system to find out what they are doing, and what the rest of us could learn from them?

Their bad policy mess: A reported 11 million people are in the US illegally. Our good policy mess: If those numbers are anywhere close to accurate, then there must be thousands—hundreds of thousands? many more? —who are already acting as if they were good US citizens. Or consider this: It’s estimated that of 280 million-plus migrants worldwide, some 82 million have been forcibly displaced. If that isn’t bad enough, what happens next? We look first to international and national organizations, and if not them, then philosophers and ethicists, to come up with answers that work better than others!

Examples are easily extended, but the point remains: There is no one way, let alone one single right way, to see and interpret conditions already complex. This is not news. The world is not one way only because the world’s complexity—again, its many components, each component with multiple functions (e.g., my simultaneous roles as husband, father, author…interacting with others), and the many interconnections between and among components and functions (what ordinary language calls part of “the wider context”)—enable all manner of complex ways of seeing.

“Even if what you say is true as far as it goes, it needs to go further. . .”

Since recastings of complex issues—including “wickedly” intractable ones—are not only possible but to be expected, the most policy-relevant thing we can say to decisionmakers, analysts and managers is: “yes, but” or “yes, and.” This way we index the complexity out of which recastings emerge. Here, the addition of “and yet. . .” has the performative function of confirming it’s complexity from which we draw recasting(s).

“Yes, it’s complex; but you need to push the matter further…” The part that is “yes” is affirmation that taking a decision does matter; the part that is “but” or “and” is the insistence that follow-ons also matter. “Yes, your recommendation holds, but it’s usefully amended in this way. . . “ To be able to say that means one must be less vague by first asking of themselves: What am I missing that is right in front of me? How can I better recast the issue without losing complexity’s seriousness and timeliness? (Should it need saying, there are policy analysts, managers and decisionmakers working this way already.)

Note the point of “yes, but” or “yes, and” is not to stymy action but to turn complexity to its singular advantage. Yes, obviously, Planet Earth is a very complex, approximately closed system, but equally closed with respect to everything? The mess we’re in—and it’s a good mess—is that the global climate crisis can’t be about the planet and science, all the way down. All the way down takes us quickly to all manner of “yes, but. . .” So too for other major issues: The differences matter, and matter in real time. For if we can’t manage better than we are right now with all this complexity, why ever would we believe promises to manage the better later on in an even more complex Anthropocene?

Here is how taking complexity seriously matters so. We know that policy and management are contin­gent on all manner of factors—societal, political, economic, historical, cultural, legal, sci­entific, geographical, philosophical, governmental, psychological, neurological, techno­logical, religious, and what-not. In fact, why close off analysis of policy and management complexity at “what-not,” when we can further understanding with “yes, but” or “yes, and”? In contrast to those glued to the rhetoric about the right person with the right policy at the right time, complexity’s barrel is so full of fish it’s difficult to track anything like one at a time.

The “right” policy? And yet. . . policy analysis and management are socially-constructed. Of course, the core concepts of risk and uncertainty, like others, are historicized. (Not only is your risk not mine; 19th century uncertainty looks very different from 21st century versions.) But to stop there is to end in exaggeration. Acknowledgement of the historical, social, cultural, economic…roots of policy analysis and concepts for management (including those in this guide) has rarely been pushed far enough in the view here.

For there is a major corollary to the social construction: Humans know only that which they create. (Such is the insight of Augustine for philosophy, Giambattista Vico for history, Roy Bhaskar for science….) Humans know mathematics in a way they cannot know the galaxy, because the former is a human creation about which more and more can be made to know. The latter’s uncertainties are socially constructed in a way that, for lack of a better word, “unknowledge” about the galaxy and beyond is not. This corollary means that to accept that “risk and uncertainty are socially constructed concepts easily historicized” is the start of analysis, not a conversation stopper.

What needs to be pushed further are the details of the interconnections among risk, uncertainty and associated terms that we make and the meanings we draw out for these connections, often under conditions of surprise and case-by-case. (It’s more complex than: Just recast!) Our creations frequently surprise us and this guide takes to heart that humans are often apt to explain the surprises by means of analogies, methods and counternarratives that extend the range of what they—we—call knowledge. (Of course, other humans are more than apt to revert to the same-old stereotypes, the same-old analogies or worse.)

It should not need repeating that terms like “system,” “with respect to,” and more are also rooted in time and place. But to stop there, again, stops short of the wider push ahead for the Anthropocene: That which we have created by way of risk, uncertainty and the like—and continue to create—has become very complex indeed. In fact: so complex as to continually provoke more differentiations for useful knowledge. (Again useful for whom? Again: for those whose actions demonstrate they take complexity seriously. Again: Yes, we know that’s not everybody.) One particularly useful set of differentiations is that related to “control” and its limits in the Anthropocene.

Control, manage, or cope ahead

In ordinary language, it is common to conflate “manage” and “control.” That will not do for policy and management complexity. Here, we need a richer set of terms and definitions. For when it comes to that complexity, people manage because they can’t control, and they try to cope better even when they can’t manage.

To start formally, think of a system in terms of inputs, outputs, and the processes to convert inputs into outputs. Inputs, outputs, and processes are, moreover, differently variable rather than uniformly alike. Control is when the system’s input variance, process variance and output variance are rendered low and stable. Think of the nuclear reactor plant: guns, guards and gates are used to ensure outside inputs are controlled; processes within the nuclear facility are highly regulated to ensure few or no mistakes are made (operations and procedures that have not been analyzed beforehand are not permissible); and the output of the plant – its electricity – is kept constant, with regulated low variance (nuclear power is often considered “baseload,” on top of which are added other types of electricity generation).

The problem, again stated formally, is that the number of critical systems having low input variance/low process variance/low output variance are fewer and fewer because of increasing political, economic, environmental and social unpredictabilities in the Anthropocene. By way of example, electric generation sources—and very important ones—now face high and higher input variability. Think of climate change, unrest, regulatory failures and other external impacts on the inputs to energy production (such as solar and wind). Such pose the challenge of managing what can no longer be controlled. In response, operational processes inside power plants have had to become more varied (this reflecting the so-called law of requisite variety), with more options and strategies to process and produce what still must be low-variance output: namely, electricity at a regulated frequency and voltage.

Coping in critical systems embraces cases where process variance can no longer be managed to match input variance and/or where output variance is no longer low and stable. That in fact is what makes earthquakes and fires catastrophic. The best to be done in these situations is to cope better, though attempts to command, control or manage will continue as well. But this isn’t to be just coping. Not only are we expected to be resilient as regards better absorbing the shocks, we are at the same time expected to try to better plan the next steps ahead. The coping here is coping ahead in the face of real-time unknown-unknowns and involves planning above-and-beyond the reactive.

The centrality of inevitable difficulties, inexperience, not-knowing and setbacks

Coping because we cannot manage, managing because we cannot control, and trying to control that which proves uncontrollable or unmanageable signal a policy and management world full of difficulty and setbacks, where inexperience and not-knowing move front and center. Some people take this to be proof of the intractable. For this guide, difficulty and inexperience and not-knowing are the persistent goad to recast—reframe, revise, redescribe, recalibrate—the issue complexity.

The more experience we have with not-knowing and with complexity, the more we must resist behaving as if our inexperience and its difficulties are also decreasing. Things getting easier is a complacency we can’t afford in the Anthropocene. What matters the most in pushing and pulling us to recast policy and management problems is the continual experience that inexperience is always center in analysis and management and that, for wont of something better, inexperience is the best proxy we have for not-knowing. Our track record with not-knowing and inexperience is central in the Anthropocene.

Why does this matter? Setbacks—unanticipated, unwanted, and often sudden interruptions and checks on performance—are commonly treated as negative in policymaking, implementation, and operations. What you hoped to be a temporary setback looks to become a permanent letdown. Less discussed, however, are both those setbacks that prove positive and our track record in pulling the positive out. A long known example is when a complex organization transitions from one stage of its life cycle to another by overcoming obstacles at the stage in which the organization finds itself.

Other positive setbacks serve as a test bed for developing better practices (but no guarantees). Different setbacks are better thought of as design probes for whether that organization is “on track,” or if not, what track it could/should be on. In yet other circumstances, a common enough observation has been that setbacks serve to point operators and managers in the direction of things about which they had been unaware but which do matter.

In these ways, positive setbacks end up as optics for rethinking major points of departure. Our track record in doing so–that is, coming to understand we didn’t know what we thought we did as well as finding out we knew more than we initially thought–becomes an absolutely central point of focus in the Anthropocene. Track records of development practitioners surmounting different setbacks, for example, look a good deal more useful when compared to the irreproducibility of research findings in relevant peer-reviewed publications.

What then are “wicked problems”?

By this point the guide’s answer should be clear: Wicked policy problems are complex problems that have yet to be recast through their very complexity. This is not optimism; it’s realism.

As with much in contemporary policy and management, wicked problems have ended up as exaggerations: Even where that may be true as far as it goes, the truth of the matter needs to be pushed further.

In fact, the litmus test that an issue is overly complexified or overly simplified is whether or not it can be recast in ways that open up fresh options for intervention without gainsaying its complexity. If—and yes it is a big “if”—a simplification can be recast as complex in ways that new interventions are now plausible or if the issue thought to be so complex no further action is possible can be recast to show otherwise, then the matter has been pushed and pulled beyond the current exaggerations. To recognize that your account can and should go further is, in profound contrast, no exaggeration.

Problems aren’t wicked when they are hard problems that profit from being left open. Further, declaring something a wicked problem can create The Ultimate One-Sided Problem—it’s, well, intractable—for humans who are everything but one-sided. In effect, the one-siders of intractability, Anthropocene or not, have taken the generous notion of intractably human and scalped it.

Last Point

Notably missing from the above discussion is the chief alternative to this guide’s call for recasting complex policy and management problems, namely: Learn and manage adaptively! Why recast, if we can learn and manage our way through adaptively? Of course, micro-learning and adaptation are critical; without them, we’d be dead. But for the purposes of this guide, adaptive learning and management, writ larger, are not an option.

Basically, learning from the past is difficult for the same reasons predicting in the future is under Anthropocene conditions (both require stable objectives, institutional memory, positive redundancy and low environmental uncertainty, among others). Recasting becomes a very necessary focus in these circumstances.

Principal sources

My thanks to Arjen Boin, Rob Hoppe, Janne Hukkinen and Ian Scoones on earlier drafts. Remaining defects testify to my stubbornness. This entry is part of a longer guide that will provide applications, examples and extensions to special topics.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s