In 2016, the Director of National Intelligence told the Senate Armed Service Committee that “unpredictable instability” is the new normal.Footnote ¹ But is this a new normal? After all, surprises have been far from rare in world politics. Mere weeks before the outbreak of the Bolshevik Revolution in February 1917, Lenin predicted that the Russian revolution would come only after his death. Unexpected peoples’ revolutions toppled regimes in Asia in the 1980s; ended the Cold War in 1989; led to the breakup of the Soviet Union in 1991; and convulsed the Middle East during the Arab Spring of 2010–12. In 2016, voters in Britain and the United States handed the incumbent parties and their neoliberal programs stingingly unexpected defeats. And we were similarly unprepared in recent years for the financial crises of 1997 and 2008; Al Qaeda’s and ISIS’s entry onto the international security landscape; tidal waves of migrants heading for developed regions’ southern borders; and the social changes brought about by radical innovations in science and technology. How do we make sense of the unexpected in world politics?

In answering this question, scholars scramble to recalculate power configurations and alignments, point to distinct forms of control, such as soft powerFootnote ² and discursive framing,Footnote ³ or simply invoke exogenous change as the source of puzzling surprise.Footnote ⁴ Steadfastly, they hold on to the assumption that the world is dominated by calculable risk. If only we could accurately map and measure all of the different components of power, we would know the probabilities of outcomes, at least in principle. Unexpected change is typically thought of as part of the diffusion of the power to control events and peoples. This is an old trope of international relations scholarship. Harvard professor and power theorist Joseph Nye restates the insights of liberals and realists like Ray Vernon and Susan Strange from decades past: power is diffusing away from states to a kaleidoscope of non-state actors.Footnote ⁵ Repeating Henry Kissinger’s arguments from the late 1960s, a former head of Policy Planning under President George W. Bush and the current President of the Council on Foreign Relations, Richard Haass, concurs: “Power is more distributed in more hands than at any time in history.”Footnote ⁶ Although the diffusion of power is often not aligned with the interests of political actors accustomed to exercising control, it is a relatively orderly and predictable process that lends itself to social scientific analysis.Footnote ⁷ Rationality points to the feasibility of controlling legible, linear history. And this model of a “general linear reality” writes Andrew Abbott, “has come to influence our actual construing of social reality.”Footnote ⁸ We put the unexpected aside at the cost of being tripped up by it time and time again.

This failing, we argue, has two roots. An exclusive focus on existing control power capabilities overlooks the actualization of potential capacities that mark what we call here protean power.Footnote ⁹ We define protean power as the effect of improvisational and innovative responses to uncertainty that arise from actors’ creativity and agility in response to uncertainty. Furthermore, the assumption that the world is governed only by risk overlooks the pervasiveness of uncertainties not amenable to probability calculations. The result is to underline the efficacy of control power and slight the importance of protean power. Unexpected changes or shocks are not exogenous to how power relations unfold, but to how our theories depict them. The actualization of potential power capacities in conditions of uncertainty always loom. Machiavelli is not alone in reminding us of the importance of chance in the affairs of states. Actors at the front lines of financial, humanitarian, energy, environmental, and other political crises routinely acknowledge the pervasive intermingling of the known and unknown, and direct our sight to potentialities in the shaping of power dynamics.Footnote ¹⁰ The fluidity of those dynamics is what prompted former President Obama to echo Thucydides by invoking “hope in the face of uncertainty.”Footnote ¹¹

Our argument embraces the usefulness of risk-based power calculations in many situations. At the same time, we must take account of the existence of uncertainty that is experienced as familiar by most international actors. The power to control thus must always be viewed in its relation to protean power, which is not a mere appendage of control power. Instead, it can pass from potentiality to actuality in a flash, changing power’s terrain, often dramatically. Effects of actions in contexts of risks, experienced as such, can be understood in terms of control power; effects of actions in contexts of uncertainty, experienced as such, in terms of protean power. The two kinds of power co-exist and co-evolve.

How, for example, was it possible for the Berlin Wall to fall? The answer to this question encapsulates our central point: the confluence of two different kinds of power. Mary Sarotte focuses on the accidental nature of the Wall’s opening. Her analysis stresses the agency of local actors and historical contingency such as the misreading of a list of government instructions that was handed to a government spokesman named Günter Schabowski during a press conference on the evening of November 9, 1989.Footnote ¹² That mistake permitted people to stream across a border that had been hermetically sealed for a generation. This constituted a heartening, though rare, event of citizens disarming peacefully a repressive regime. People power as the actualization of protean potentialities was one part of the story. Diplomatic and financial control power was the other. During the 1980s, economic power drained away from East Berlin as the GDR leadership became dependent on Western capital. Lacking sufficient productivity gains in manufacturing to serve the escalating cost of its debts, the unforeseen collapse of the price of oil in 1985 sharply reduced earnings from the GDR’s most important export product, mineral oil refined from Soviet crude.Footnote ¹³ Gorbachev’s reform program in the Soviet Union put additional pressure on the East German government. East Germany’s leadership faced only unappealing options: sharp reductions in living standards or blood on the streets. Permitting emigration in the hope of further West German loans with lenient conditions thus became the preferred policy that the government planned to adopt before the end of 1989. While the specific details of what happened on the night of November 9, 1989 were contingent, the diffusion of control power away from East Berlin was central for matters to evolve as they did. Significantly, the GDR’s financial and political straits produced consequences that Western actors did not foresee.Footnote ¹⁴

To help us better understand the unexpected in world politics, our argument in this chapter takes three steps. First, we begin the analysis by reviewing the discussion of the different faces of power, ending with the notion of power demarcating fields of political possibilities. Second, we distinguish between two kinds of power. Control power seeks to dominate; operating in a world of risk, it penetrates and diffuses. Protean power results from the improvisations and innovations of agile actors and processes of the actualization of potentialities; coping with uncertainty, it creates and circulates among actors and sites. Control power operates most clearly, and reliably, in situations marked by calculable risk that actors experience as such; protean power arises in situations of deep-seated uncertainty that actors often experience as a crisis. Because they can create room for each other, the two types of power are not mutually exclusive. As hopes of deliberately controlling outcomes diminish, protean power potentials loom large. The balance between them follows from an interaction of two dimensions affecting actor practices: the degree to which such actors experience the world to be risky or uncertain and whether it is, in fact, so. Third, in contrast to conventional international relations scholarship, we show that control and protean power analysis requires us to conceive of world politics as an open rather than a closed system.

Power

One of the many paradoxes of power is this. It is an explanatory construct practitioners and scholars of international relations cannot do without. It is also a concept that needs to be explained, rather than do the explaining. The prevailing understanding that power is a thing we “have” or “lack” in order to create a desirable effect is a starting point of our political experience and analysis.Footnote ¹⁵ In the study of international politics, for example, power is widely understood to be about capabilities typically measured by indicators such as military spending, the size of the economy, or technological advancement; articles and books proceeding in this manner fill libraries. Such capabilities are then used to explain or predict specific effects or outcomes.

Yet what remains normal in the analysis of international relations, theorists of power have dismissed as inadequate long ago. Unfortunately, their writings have had little discernible effects on the field of international relations, which treats the concept of power as a synonym for more or less narrowly construed actor capabilities. While not denying the importance of the base and means of power, theorists of power insist that power is grounded in the relationships among actors rather than in their attributes.Footnote ¹⁶ Along with David Baldwin, we thus view “the elements of national power” approach with its exclusive focus on national capability as profoundly misleading.Footnote ¹⁷

A relational view of power has been the shared premise of a vigorous and prolonged debate about the different faces of power, here understood as different forms of control. Ultimately, the debate has centered on where and how to draw a distinction between “free action and action shaped by the action of others.”Footnote ¹⁸ Generally speaking, over time scholars have broadened substantially the empirical context where we should look for the effects of power.

For Lasswell and Kaplan “political science, as an empirical discipline, is the study of the shaping and sharing of power.”Footnote ¹⁹ Building on what he called Lasswell’s seminal contribution, Robert Dahl started the modern debate with his definition of power as the ability to get others to do what they otherwise would not.Footnote ²⁰ Dahl drew a distinction between the base of an actor’s power and the means of employing the base, on the one hand, and differences in the scope of responses elicited and the number of comparable respondents, on the other. For the purpose of comparing the power of actors, Dahl insisted, we need to focus primarily not on the actions of A but on the responses of B;Footnote ²¹ power base and means, though important, do not provide us with a comparison of the power of actors.

In an important critique of Dahl, Bachrach, and Baratz broadened the context of the effects of power by drawing a different distinction between free and constrained action. They focused on political dynamics that Dahl’s analysis of bilateral power relations, revealed in concrete decisions about key issues, blended out. Two in particular: power exercised to limit the scope of the political process to safe issues; and power exercised to avoid taking a decision. Non-participation and non-decisions are effects of power that can stop a conflict from arising and from being acted upon. Unobservable processes and issues thus can be the effects of power and help to maintain the status quo in the absence of overt conflict.Footnote ²²

Steven Lukes broadened further the context where we should track free and constrained action. He pointed to a basic agreement between Dahl and Bachrach and Baratz. All three assumed that power was exercised by actors. Lukes focused also on the effects of structures that can shape the wants, needs, and desires through the impersonal workings of socio-cultural arrangements and practices.Footnote ²³ To have effects, power does not need to be intentional or active.Footnote ²⁴ Lukes argued that power should neither be reduced to its exercise nor its means, and that it operates within and upon structures.Footnote ²⁵ His theory highlighted structural features of society that make actors powerful without having to exert control directly. Yet, like Dahl, Lukes insisted that we need to study both the agents and the subjects of power. Power is about an agent’s potential capacity and specifically the scope for personal reasoning and self-definition. “Power identifies a capacity: power is a potentiality, not an actuality – indeed a potentiality that may never be actualized.”Footnote ²⁶ Lukes’ theory is thus both subject- and agent-centered.Footnote ²⁷

Building on and adapting different aspects of the writings of Michel Foucault, theorists of power, including in the field of international relations, have broadened still further the context of tracking the effects of power.Footnote ²⁸ Foucault’s analysis is subject- rather than actor-centric. Power both controls and generates through every-day mechanisms of discipline. It creates the characters of actors and streamlines, among others, their sexual, health and mental practices so that they fit existing social and political arrangements. Disciplinary power molds souls and inscribes bodies.Footnote ²⁹

Informed by Lukes and Foucault in particular, Clarissa Hayward’s subsequent analysis proves especially fruitful for our purposes. Hayward argues that power’s mechanisms are best conceived not as instruments that powerful actors use but as social boundaries. “Power defines fields of possibility.”Footnote ³⁰ Laws, rules, norms, customs, identities, and social standards are such boundaries. They enable and constrain all forms of action, including for the most powerful. Actors can change the shape and direction of power through practices that result from both structured fields of possibility and actor endowments. Conceived as social boundaries and endowments, power defines what is possible for self and other. Contrary to Dahl’s strong rejection, “action at a distance” for Hayward is an identifiable and important site for tracking power effects.Footnote ³¹ In global politics, the possible can be constrained or enabled at long distance without the existence of any discernible connection between the source and the target of power. To inquire into the workings of power we should not ask “how is power distributed” as we seek to distinguish between conditions of power and powerlessness. We should ask instead “how do power’s mechanisms define the (im)possible, the (im)probable, the natural, the normal”?Footnote ³² What matters is the mutability of asymmetries in power that define the field of what is possible.Footnote ³³

Control and Protean Power

Power is an elusive concept. Hence, no single framework can “claim to have found the essence of power.”Footnote ³⁴ Instead, each partial conceptualization can provide some important insights about key aspects of power.Footnote ³⁵ Typically, analysis focuses exclusively on the shifts in the dynamics of control power operating under conditions of risk. The concept of protean power broadens the analysis by acknowledging the existence and explanatory potential of power dynamics operating under conditions of uncertainty. Including both types of power promises more analytical breadth and a richer explication of unexpected change in world politics.Footnote ³⁶ As a first step we distinguish between two ideal typical situations. When the context and the experience of power are marked either by risk or by uncertainty control and protean power form an ideal typical distinction (Table 1.1).

Of all the theorists of power Robert Dahl has been most explicit about the close affinity between control power and risk. Probabilities of an event with and without the exercise of power is for Dahl an indispensable way of comparing the power of different actors.Footnote ³⁷ Observations of the two different conditions may be difficult but are “not inherently impossible: they don’t defy the laws of nature as we understand them.”Footnote ³⁸ Many decades after the quantum revolution in physics, Dahl’s appeal to the laws of nature remained Newtonian and was expressed in classical notions of probability. Half a century later there is no indication that conventional views of international politics have changed – even though it is time for international relations scholarship to wake up from its “deep Newtonian slumber.”Footnote ³⁹ Arguably, today quantum physics and quantum probabilities define the laws of nature “as we understand them.” They resonate with the concepts of possibility and potentiality that are central to protean power dynamics.Footnote ⁴⁰

The incalculable provides the context and experience of what we call protean power. It arises either through direct relations between actors or indirectly in the follow-on effects that reconfigure complex systems. Protean power is the effect of actors’ improvised and innovative responses to an incalculable environment or their experience of the world as equally uncertain. This type of power cannot be harnessed consciously. It is a creatively generated shift in accepted problem-solving that circulates across different sites of political life. It emerges in specific moments. It is an inextricable part of variable combinations of risk and uncertainty that encompass affirmation and refusal as well as improvisation and innovation.

Protean power has generative effects on the broader context. These can be entirely unanticipated and as such bypass all attempts to exert control. While the processes underlying the two power types may co-occur, and converge, their relation to actor experiences of the world are diametrically opposed. From the perspective of those amassing control capabilities, the effects of protean power in settings of uncertainty enhance the unpredictable and result in frustration.

In our understanding, the unexpected is an integral part of power dynamics. This means that we should add the concept of what is possible to what is probable and what is natural. The mutability of the world goes beyond the predictable effects that constitute control power. It includes convention-defying uncertainties that destabilize the world. Admittedly, in common language risk and uncertainty are often used as synonyms. The confusion between the two concepts is both perfectly understandable and intellectually damaging. The Merriam Webster dictionary, for example, defines risk in terms of uncertainty, as “the possibility that something bad or unpleasant (such as an injury or a loss) will happen.”Footnote ⁴¹ Despite this confusion, we should distinguish clearly between the concepts of risk and uncertainty. Both are relevant for an analysis of power and unexpected change.

Terminological confusion has been deepened by a questionable translation of Max Weber’s analysis into English. A widely accepted view holds that Weber’s definition of power is operating only in the world of risk – power as the likelihood of achieving one’s will while overcoming the resistance of others. The conventional view is based on a problematic and theoretically constricting translation of the capacious German concept of Chance. That term has two valid translations: one probabilistic risk (Wahrscheinlichkeit), the other possibilistic uncertainty (Möglichkeit).Footnote ⁴² Following Weber, we hold that power operates in the world of risk and uncertainty. Actors accomplish their objectives over others in dominating relations (potestas), as well as with others in enabling relations (potentia). Weber’s conceptualization of power thus invites us to look simultaneously at control power in terms of processes that connect capabilities with effects in relations that penetrate and diffuse, and at protean power in terms of agilities that create and circulate.

How do actors facing risk and uncertainty choose their practices? Risk-based models of power-as-control assume that they are playing the odds. Eager to apply statistical techniques he had learned on Wall Street to professional sports, after three disappointing seasons, the general manager of the Philadelphia 76ers basketball team, Sam Hinkie, observed ruefully in his resignation letter that “the illusion of control is an opiate … It is annoyingly necessary to get comfortable with many grades of may be.”Footnote ⁴³ Confronting uncertainty, actors can turn to prior beliefs (priors over priors in the language of economics) in order to make reasoned decisions based on implicit probabilities. Unfortunately, no plausible answers exist to the question of which prior beliefs are chosen and why. Actors can also turn to imagined futures of the possible and impossible, something international relations scholarship tends to overlook.Footnote ⁴⁴ Hence, most actors cope and muddle through, typically informed by standards of reasonableness rather than rationality. The assumption of rational decision-making may, of course, be correct for some individuals and situations, for example, American traders on Wall Street or American defense officials in the Pentagon. But what about Japanese traders in Tokyo or Japanese defense officials in the Self-Defense Forces? They do not differ from Americans because they adhere to inherently irrational beliefs. Instead, differences in institutional and intellectual settings suggest distinctive engagements with the theory and practice of arbitrage and coercion. They underline how much conceptual redefinition, extension, and ambiguity can occur in different settings.Footnote ⁴⁵ To insist that the mix of risk and uncertainty will always and everywhere yield the same probability calculation does not help us understand better power dynamics in the domain of the unexpected. It seems more sensible to let go of the notion of invariant, omnipresent, rational probability calculations and to acknowledge the existence of variable standards of reasonableness under conditions of risk and uncertainty. Control and protean power thus are brought into one analytical perspective as they make crises normal and endogenous to world politics rather than abnormal and exogenous.Footnote ⁴⁶

The theoretical shift in perspective that is needed to explain the surprises assumed away by risk-only views is that power is not only a cause of empirical patterns but also their effect. Figure 1.1 captures the connection between practices and power outcomes by depicting two dimensions: attributes of the underlying context and actor experiences.Footnote ⁴⁷ The co-existence and co-evolution of control and protean power do not occur along a simple continuum. Instead, the four cells in Figure 1.1, populated by characteristic political practices, are produced by the interaction of the two dimensions. As such, they reflect both the degree to which complete knowledge or ignorance of probabilities prevails and the degree to which actors seek it in the first place. We acknowledge that empirically, in the depth of a crisis, for example, the effect of actor experience and context attributes on political practices, and therefore power, may not be readily distinguishable. This can, however, be done in principle and certainly in retrospect. Both dimensions thus have their place in the framework.

Figure 1.1 Context, Experience, and Power

Each of the political practices captured in the four cells generates power dynamics that feed back on uncertainty and risk depicted along the two dimensions. The context- and experience-altering impact captured by the arrows in Figure 1.1 thus makes control or protean power the effects of diverse political practices. Affirmation, in the top left cell, is the recognition by actors that capabilities can be amassed and deliberately deployed to exercise power. From the perspective of those subject to such power, affirmation may take the form of acquiescence or compliance in the context of predictable risks. In the end, as the short arrows show, it enhances the utility of probability calculations concerning future outcomes; reliance on established power templates reinforces the risk-based nature of the world and is experience by actors as such. This is the domain of control power. The discipline of international relations is replete with examples of authors assuming, mistakenly, that this is the only world in which politics unfolds.

Our analysis highlights the existence of two other worlds captured by the other three cells in Figure 1.1. Depicted in the bottom right cell, for example, innovation is a response to a second, fundamentally uncertain world. It generates protean power, shifting the goal post for exercising control in the process and necessitating still more agility in the future. Protean power, then, is the effect of innovation that generates further uncertainty and at the same time underscores the futility of control power. Finally, we can also find ourselves in a third world that mixes risk and uncertainty. During an emergent crisis, actors operate in the top right cell: uncertainty has made probability calculations impossible, though actors do not realize it. This is the root of the disorienting nature of most crises. Actors assume “old ways” still apply when the ground has already shifted to make possible unexpected outcomes. When they discover that familiar solutions no longer work, they are compelled to improvise to stay afloat in increasingly unstable and uncertain contexts. Conversely, as previously earth-shattering solutions evolve into best practices and uncertainty is replaced by risk, actors’ assumptions of pervasive uncertainty may persist. They continue to make decisions affecting their immediate environment only, refusing attempts at risk-based decision-making, without any desire to control others directly. The shortcomings of control power experienced by actors generate room for surprising solutions, while success transforms protean power into control power. This is captured by the bottom left cell.

The four cells in Figure 1.1 exemplify but do not exhaust the range of practices available. In our labeling we focus on particular practices that relate actor experience and context attributes to power manifestations, and the degree to which the latter reinforce or undermine the different constellations of risk and uncertainty. Power as either cause or effect is not coterminous with political practice, a common mistake that invites the spinning of tautologies. It is instead analytically separate from practice as it affects the experience and context of risk and uncertainty. Illustrated by the two large arrows, innovation, the response to immediate experience of uncertainty in an uncertain world, generates protean power and so exacerbates further the uncertain conditions from which it arose. It is for this reason that we find it impossible to link protean power to specific attributes or capabilities and instead highlight its agile nature that jettisons any semblance of regularity.

Each cell in the figure can be populated by empirical illustrations, some of which we present in this volume. For example, Jennifer Erickson’s discussion of arms control during the Cold War in Chapter 11 approximates the situation of a risky world that actors experience as such (top left cell). The analysis of science and technology in Chapter 6 discusses the opposite case of an uncertain world that is experienced as uncertain (bottom right cell). The awareness and acknowledgment of pervasive uncertainty at all levels shapes how actors engage in innovation and how protean power is generated. In Chapter 8, Erin Lockwood and Stephen Nelson offer an analysis of a mixed case of risk and uncertainty (top right cell), evidenced particularly well by the growing instability of mismatched responses to financial crises. They show how market players, operating in the domain of both risk and uncertainty, have relied on modeling conventions and contractual clauses that illusorily seek to transform uncertainty into manageable risk. In Chapter 5, Noelle Brigden and Peter Andreas analyze protean power effects of migrant improvisations and innovations as well as the anticipated, yet unintended, escalatory dynamic between more police control and more migrant evasion. And in Chapter 7, Rawi Abdelal addresses the relations surrounding hydrocarbon flows in Europe that also mixes risk and uncertainty (bottom left cell). He offers an excellent illustration of actor experiences that generate protean power under crisis conditions. Such innovative solutions may briefly settle into control mode, leaving a landscape of (rerouted) pipelines behind. But in the interaction of the two power types, actors will find control disrupted further down the line. In short, conventionally deduced behavioral implications of different power constellations conceal important variations in the degree of uncertainty, and thus can easily mislead us. Specifically, they make us overestimate the importance of control power in world politics.

In its relationship to uncertainty and risk, control power can be compared with a game of billiards with its discrete movements. There is room for strategy, but there is no question about the rules, which are closely linked to laws of motion in physics that govern a player’s decisions and constrain their execution. By contrast, protean power resembles a game of interactive fluidity, like tennis. It is about “being in the right place, at the right time” that extends well beyond coincidence. For the world’s leading physicist of tennis, Howard Brody, there was nothing flighty about the game. Yet he would have acknowledged that individual ball control, motivation, mutual weakness recognition, and interaction with the spectators produce enough uncertainty to make the exact score unpredictable.Footnote ⁴⁸ Such is the world of protean power, moving past simplified equations of force.

Even though an actor may be too weak to exercise “power over” (understood here as actual capability) the human or non-human world, she or he may nonetheless be sufficiently empowered to have “power to” or “power with” (understood here as the capacity to actualize potentialities, without or with others) to be able to navigate in that world successfully.Footnote ⁴⁹ One way of illustrating the operation of protean power is to focus on the effects of human action without design. Under conditions of uncertainty it is not necessarily strategic actions but their emerging byproducts that create the most consequential effects.Footnote ⁵⁰ It is clear that actors want to do something in response to the uncertainty that surrounds them. What should be done, however, is typically unknown. Actors do their best, guessing and coping, uninformed by calculable probabilities and unknown determinants of success or failure. Once their actions have resulted in outcomes, ascribed power effects are linked to specific actors who are seen as having caused the outcomes. Who wins is therefore determined through traceable (ex post) but not predictable (ex ante) assessments. We thus gain a deeper understanding of the fragility and limits of control power, not a handbook of how to beat fortuna at her game. Figure 1.1 is a useful reminder that the two kinds of power are analytically separate. Drawing on the empirical case studies in this book, Chapters 2 and 13 argue that they are also deeply interrelated. Uncertainty makes control power fragile, tugs our conceptualization toward protean power dynamics, and sets the stage for the co-evolution of both power types.

Complexity and Power

How was it possible for China to transform itself within a generation, lifting hundreds of millions out of abject poverty? Nobody inside or outside China foresaw this revolutionary change in 1979 when Chairman Deng announced his reform package. Now almost everybody assumes that it happened because of one or several well-known factors, such as less state supervision, smart technocrats, unleashed entrepreneurship, better access to world markets, or more secure property rights. Reminiscent of the story of migration (Chapter 5), Yuen Yuen Ang offers a very different answer: the inherent unpredictability of the reform journey and the co-evolution of control and protean power.Footnote ⁵¹ The guiding tenet of that journey was to “cross the river by touching the stones,” toes gripping hard but with an unknown destination on the other side of the river. Central leaders were at times alarmed about the unanticipated consequences of their decisions. The reforms empowered local state and party officials and market actors to pursue adaptive development strategies that permitted improvisation and innumerable specific solutions to ever-changing problems. Chairman Deng, for example, was totally surprised at the proliferation of township and village enterprises, the centerpiece of the early reforms. “This result was not anything I or any of the other comrades had foreseen; it just came out of the blue.”Footnote ⁵²

In answering “how was it possible?” Ang turns to complexity theory.Footnote ⁵³ It highlights the adaptive character of open systems and their unpredictable, emergent properties.Footnote ⁵⁴ Complicated systems are predictable. Complex systems are not. They produce “outcomes that cannot be precisely controlled.”Footnote ⁵⁵ Sharing hidden, communal lives, trees are complex and resilient. Solitary toasters with no secrets to hide are complicated and lack resilience.Footnote ⁵⁶ Complexity demands incessant improvisation and successive approximation, innovation by recombination, local knowledge, and accumulated experience. It acknowledges the inescapability of uncertainty that control power cannot conquer.Footnote ⁵⁷ But even if it could in particular instances, Robert Jervis reminds us that “local predictability, if not simplicity, produces a high degree of complexity and unpredictability.”Footnote ⁵⁸ Often that complexity reflects a momentary indeterminacy in the cross-balancing of control and protean power.

The circulation of protean power comes into play in situations of uncertainty, fueled by the effects of improvisation and innovation. The fit between improvising solutions and particular aspects of an uncertain context matter, even though it becomes apparent only in retrospect. By contrast, control power operates in situations of calculable risk. In relatively stable and predictable environments, the effects of control power emerge directly. Implicitly, our understandings of control power tend to assume that its predicted effects occur in closed systems, such as laboratory settings, which invite partial equilibrium analysis that holds constant all variables that might confound the stipulated power effect. This is not so in open systems. Although large numbers yield predictable averages in the aggregate, individual behavior is typically unpredictable and seemingly erratic. Furthermore, in open systems the interaction of a sizable number of factors form wholes that may not be readily captured by linear models of the world.Footnote ⁵⁹ In a linear world, small things follow from large ones. In a non-linear world, large things can follow from small ones.Footnote ⁶⁰

The worlds of risk and uncertainty and control and protean power resemble the well-known difference between clocks and clouds.Footnote ⁶¹ The French astronomer and mathematician Pierre-Simon Laplace was convinced that the world is a big, complicated clock. As science developed, more and better knowledge about the clock’s inner workings would enable us to predict the future with deterministic or probabilistic equations. All that is needed is work and patience. The present state of the world is the effect of its past and the cause of its future. In the world of clocks, there is, at least in principle, no uncertainty. Like the past, the future is fully knowable to an omniscient present. Various insurance markets are clocklike in their predictability. And our experience confirms daily many of our predictions. We could not function in a world in which everything was possible. This is the risk-based world of control power.

Protean power operates in the world of clouds. Modern meteorology knows vastly more than in the past about the conditions under which clouds form, and its predictive power of general weather patterns has improved greatly. Yet it is much less confident in its ability to make specific predictions about the shape of particular clouds. Historical probabilities summarize the possible ways the future could unfold. And the curve summarizing those possibilities looks nothing like a normal bell-shaped curve that is necessary for the calculation of risks. It has fat tails that describe a much more volatile world than risk models lead us to believe. Historians are the first to understand intuitively and acknowledge explicitly that the world we experience as the only real one is the result of statistical distributions of possible worlds that emerged from once-possible worlds. So are playwrights such as Nick Payne.Footnote ⁶² “The past did not have to unfold as it did, the present did not have to be what it is, and the future is wide open.”Footnote ⁶³ The indeterminacy that inheres in the field of power points to almost infinite alternative pasts and futures – the field of protean power possibilities.

Complexity thus brings into one perspective risk and uncertainty and control and protean power. “Risk,” Mary Douglas writes, “is not a thing, it’s a way of thinking.”Footnote ⁶⁴ The same is true of uncertainty. Searching for a combination of both, Karl Popper settled for something “intermediate in character, between perfect chance and perfect determinism.”Footnote ⁶⁵ Popper’s solution points to a kind of freedom that is not mere chance. He contrasts his preferred “plastic-control” to “cast-iron control.”Footnote ⁶⁶ Following Popper, Almond and Genco argue that we are living in an open system with emergent, creative properties, regularities with a short half-life, human inventiveness, and low-probability conjunctions. Plastic control endows the exercise of power with a looseness of fit that undercuts planning.Footnote ⁶⁷ This is an apt description of a complex world marked by risk and uncertainty and the operation of protean and control power.

The evolution of the universe, biology, geological patterning, climate, hurricanes, and other processes in the natural world are often modeled as a set of complex, open systems, governed not by universal laws and equilibrium but by pervasive chaos and disequilibrium. Within and across such systems volatility sets free a “protean capacity of self-organization” … containing “the potential for creative evolution.”Footnote ⁶⁸ System trajectories can be made intelligible ex post but are not predictable ex ante. For the analysis of control and protean power this is the ontological foundation of analysis. It is at odds with the control power logic of international relations scholarship based on the assumption of closed systems. The experimental method that seeks to uncover general laws is inadequate to come to terms with the emergent properties of open systems. For practical reasons, linear causality does not capture such properties. It is, of course, entirely possible that open systems contain simple rules that we should be able to decipher. But in the social world predictive capacity is systematically limited by the the time it takes the system to run through enough iterations to watch how things map out. Stephen Wolfram calls this “computational irreducibility.”Footnote ⁶⁹

Open system analyses of control and protean power differ in how they make sense of the world. The reason is simple. Causality is understood and works differently in the domains of capability and of the capacity to actualize potentialities, of control and of protean power. When mapping causal configurations, current convention draws the causal influence in an unbroken line from actor A to actor B; when modeling two-way causation a broken arrow is typically drawn to connect actor B to actor A. This does not mean that the first arrow is in some ways stronger or more important than the second. To the contrary, for Dahl and many other theorists of power the main action is in and from B, not in and from A.Footnote ⁷⁰ In the language of contemporary discussions of causality, the first account of causality can, in principle, overcome the problem of endogeneity. Because it focuses on more than efficient causation, the second cannot. Less or more capacious views of the social scientific enterprise make scholars choose differently at this juncture. Less capacious conceptions treat world politics as a closed system that should, at least in principle, be investigated through controlled experiments. More capacious conceptions conceive of world politics as an open system, not amenable even to quasi-experiments. This book views the world as an open system.

Explication provides a method to operate at the intersection of these two conceptions of science. Explication differs from both “mere” description and “law-like” explanations.Footnote ⁷¹ This style of analysis combines “how” questions, understanding, descriptive inference, and constitutive analysis, on the one hand, with “why” questions, explanation, causal inference, and causal analysis, on the other.Footnote ⁷² Constitutive effects are productive or generative, and in practice are difficult to distinguish from causal effects.Footnote ⁷³ “Constitutive relations are causal, albeit not causal in the neopositivist sense … [Constitutive explanation] is not a rival to causal explanation, but simply an alternative to the neopositivist focus on cross-case covariation.”Footnote ⁷⁴ The analysis of control and protean power politics thus benefits from, indeed requires, a broad notion of causality and an eclectic approach that suits the analysis of open systems.Footnote ⁷⁵ On this point we thus follow Lévi-Strauss for whom a “mind in its untamed state is distinct from a mind cultivated or domesticated for the purpose of yielding a return … it is possible for the two to co-exist and interpenetrate.”Footnote ⁷⁶

Complex, open systems undercut the efficacy of using past trends and performance as a predictor of future outcomes. If control power worked in the past, it is often assumed that it must do so also in the future. Disregarding fluctuations in interactive, entangled, competitive, complementary, parallel, or nested co-evolving factors that mark complex systems can easily lead astray any analysis of power dynamics. Although the underlying uncertainty can be the result of exogenous shocks, it typically arises endogenously through a combination of inefficacious control power and an amplification of uncertainty through the circulation of protean power. Complexity thus necessitates an “inherently humble approach that is conscious of the limitations to predictability and control.”Footnote ⁷⁷

This is a point that resonates deeply with the writings of both Friedrich Hayek and Elinor Ostrom. Their arguments for decentralization and polycentrism rest on the existence of control frameworks of ideology and institutions in which market exchanges can occur and subsidiary choices can be exercised. But despite the existence of constitutive and regulatory opportunities and constraints such frameworks provide, Hayek and Ostrom both stress the importance of the unpredictable and the advantages of decentralization.

Hayek’s analysis emphasizes spontaneity, although in a market system that he sees as existing largely in isolation from rather than closely related to and imbricated in other self-organizing and open systems. Hayek directs our attention to the market context in which control and protean power are interrelated. The complexity and unpredictability inhering in social and economic life means that all hierarchical orders, important as they are in guaranteeing property rights, have distinct limits. Scholars must accept that actors need to “adapt to the unforeseeable.”Footnote ⁷⁸

Hayek alerts us to the tensions and contradictions between the desirability of utilizing all actors’ dispersed knowledge and attempts to improve underlying orders through direct commands.Footnote ⁷⁹ The division of knowledge stands at the center of socio-economic and political life.Footnote ⁸⁰ Knowledge revolves around an infinitely complex and profoundly political process of communication. Imperfect communication produces distorting rather than self-organizing knowledge systems. The errors of centralized control are rooted in the pretentious ignorance and utter disregard of pre-scientific knowledge on which most of the theories of social scientists and political engineers rest.Footnote ⁸¹ The price system does not connect demand and supply in the abstract. It connects innumerable actors, situated in distinctive locales, acting at specific times, and with unique understandings of themselves and the expectations of others. Hayek deploys a vocabulary that applies well to the analysis of protean power and allows us “to capture the continuous reproduction and fluidity of economic processes.”Footnote ⁸²

Abstract orders, of course, require institutions that make social and economic life possible in the first place and a vibrant ideology that supports those institutions.Footnote ⁸³ Rational designs of allocative institutions that are focused on top-down control, however, are suboptimal. They are beholden to inaccurate abstractions that fail to engage the uncertainties of practical life.Footnote ⁸⁴ And when uncertainty engulfs actors, they rely on micro-level repertoires of knowledge and action to get by. In doing so, such actors add new factors to an already complex environment and exacerbate both normal, operational, and radical, crisis-induced uncertainty.

Hayek views markets as devices that coordinate activities without an omniscient center exercising control.Footnote ⁸⁵ Markets disperse knowledge and thus power.Footnote ⁸⁶ Except for their self-perpetuation, they are instances of social orders that evolve without predetermined ends. Although the power effects they produce are clearly identifiable with hindsight, these practices are inherently unpredictable, a distinct characteristic of protean power.

Stressing, like Hayek, the virtues of decentralization Elinor Ostrom’s probing treatment of environmental resource management captures the need for linking all types of speech, knowledge, and practices, working through “mechanisms of mutual monitoring, learning and adaptation of better strategies over time.”Footnote ⁸⁷ Actors dealing with a profoundly complex environment are faced with the challenge of seeking to improve, without being able to fully control. They respond with innovative solutions and continuous adjustments that thrive in decentralized, polycentric systems and create conditions for the emergence of protean power.

In her Nobel Prize acceptance lecture, Ostrom challenged the presumption that governments and centralized authority-wielding organizations more broadly do a better job than other actors who are more immersed in local contexts.Footnote ⁸⁸ Actors are placed in networks and wield power by virtue of defining and redefining webs of connection rather than by claims to their official positions. Ostrom questions the belief that we cannot do without abundant external resources to govern effectively. Rather, actors traditionally viewed as weak have a unique ability to produce governance systems. Too often we underestimate their efficacy.Footnote ⁸⁹ Reaching beyond the obstacles of collective action, she suggests that problems that span multiple levels (of action and analysis) should be addressed at the appropriate scale.Footnote ⁹⁰ Citing the original definition of polycentric systems,Footnote ⁹¹ Ostrom highlights the role of formally independent decision centers in producing often innovative and effective policy solutions in legal contexts that can operate beyond the local level.

One-size-fits-all approaches that are externally imposed and dominance-backed do not work well in polycentric systems.Footnote ⁹² For example, we may be facing the consequences of our collective failure to respond to climate change. Although, or because, the specific manifestations of the resulting environmental pressures vary, actors and societies can be compared on their ability to deal with such unknown and, in their specifics, unknowable challenges. Ostrom, with her co-authors, labels this quality “adaptedness.”Footnote ⁹³ The reflexivity in such socio-ecological systems underpins the fundamental uncertainty about what collective action outcomes will follow and what the cumulative effect of innovative steps at all scales will be. At its base, Ostrom’s account is about turning threats into opportunities,Footnote ⁹⁴ recognizing that the most pressing threats are rarely, if ever, visible looking down from the top.

Hayek and Ostrom alert us to the fact that the simplifications of scholars tend to reduce complexity to complication. And in that process of simplification, research can easily lose sight of crucial aspects of protean power dynamics. What distinguishes protean power from control power is the unknown outcomes it produces. Protean power operates in networks that are extensive, loosely coupled and self-directed rather than intensive, tightly-coupled, and authoritative.Footnote ⁹⁵ Although protean power is not readily aggregated, its effects are real and unfold in uncertain conditions that often evoke refusal or resistance and derive from improvisation or innovation. Viewed as agility in response to uncertainty, in a world that often defies control, actors cannot know what exact effects it will produce. They generate protean power through their creativity and local awareness and the creation of future potentialities as a result of new actualities, without claiming to seek or to cause specific outcomes.

Expressing widely shared sentiments, Randall Schweller writes that “we are entering a jumbled world run by and for no one, in which the nature of power itself is changing, an ungovernable place … a chaotic realm of unknowable complexity.”Footnote ⁹⁶ Yet a complex world is not necessarily chaotic and is not necessarily slipping out of control. Two compelling advocates of complexity theory, Axelrod and Cohen, for example, argue that “while complex systems may be hard to predict, they may also have a good deal of structure and permit improvement by thoughtful intervention.”Footnote ⁹⁷ In politics “governments not only ‘power’ … they also puzzle.”Footnote ⁹⁸ Forecasting is more than a statistically informed extension of past trends into the future.Footnote ⁹⁹ It requires a mind that is open to both intuition and science. Good forecasters are Isaiah Berlin’s foxes who embrace the complexity of the world, not hedgehogs whetted on one big idea or trend.Footnote ¹⁰⁰ Tetlock’s research has established that, beyond the frame of three to five years, the accuracy of the predictions of the average expert is no better than random.Footnote ¹⁰¹ And it takes skill and hard work to be a successful forecaster of possible scenarios.Footnote ¹⁰² While some uncertainties are altogether unknowable, others are not, at least in principle. This does not mean denying the importance of control power and risk. It does mean, however, that we must incorporate protean power at the micro-level that can yield unanticipated consequences.Footnote ¹⁰³

Scholars and policymakers occasionally compare international politics to a game of chess.Footnote ¹⁰⁴ That game has fixed rules and calculates probability in a complex environment. Yet it also illustrates the limits of control. The current world chess champion is a young Norwegian, Magnus Carlsen, the most highly ranked champion in the game’s history.Footnote ¹⁰⁵ In one of the most lop-sided matches in recent decades, he dethroned the defending world champion Viswanathan Anand in November 2013. This changing of the guard illustrated a broader trend. A handful of Russian grandmasters no longer dominate the sport; today more than 1,200 grandmasters of chess play the game, compared with eighty-eight in 1972. The collapsing chess order shows a dialectical relation between high levels of conformity instilled by risk-based, computerized chess training manuals and the continued relevance of improvisation and innovation. Carlsen’s genius lies in his unorthodox and surprising strategies that rely on his prodigious memory rather than the conventions of computer chess. Carlsen has an aptitude for playing many different styles of chess, adapting readily rather than searching like a scientist for the best solution to a given problem.Footnote ¹⁰⁶ His playing style confirms Adam Smith’s insight: “in the great chess-board of human society, every single piece has a principle of motion of its own.”Footnote ¹⁰⁷ In the terminology of this book, Carlsen’s huge success shows that chess is a game where risk and uncertainty and control and protean power meet.

Conclusion

Uncertainty breeds protean power and protean power intensifies uncertainty. The world is well stocked with low-probability events such as the sudden appearance of terrorist organizations operating on a global scale and waves of large-scale human migration. Typically, these events are available for risk-based political analysis only after they have happened. We have argued here that an adequate understanding of disruptive events and processes requires going beyond an analysis that focuses only on direction by and diffusion of control. It must incorporate also the analysis of the creation and circulation of protean power.

As an analytical construct and policy tool, control power operates in “normal” situations where calculable probabilities of outcomes make it, at least in principle, measurable and deployable. Protean power, by contrast, emerges typically in situations of uncertainty. This form of power thrives on actors’ agility. They can be innovative in reinterpreting the meaning of rules, and they can play without rules, relying on identity and other mechanisms for managing uncertainty.Footnote ¹⁰⁸ As such, protean power creates political dynamics that alert us to the presence of endogenous uncertainty rather than merely responding to it as an exogenous force.Footnote ¹⁰⁹ It allows actors to position themselves to derive relative advantage from unexpected challenges, while adding to the overall uncertainty everyone faces. The concept of protean power invites us to analyze refusal from the perspective of the targets of control power and to inquire into creative practices furthering mobility, ambiguity, and disorder and the improvisations and innovations that come in their wake – all markers of the circulation of protean power in contemporary world politics.Footnote ¹¹⁰

In a disorderly and at times chaotic world predictive accuracy is unobtainable. This is old news. It recapitulates for our times a long-standing connection between two types of power embedded in the known and unknown. A widespread view holds that control power is diffusing and that regional and global orders are being undermined as the world is heading from predictable order to randomness. In a world where risk and uncertainty overlap and intermingle the case studies in this book point to a more complex world. To focus exclusively on risk and control power overlooks the fact that explanations of crises and far-reaching surprises require the analytical lens of protean power thriving in uncertainty.

The political world is more unfathomable than notions of control power permit us to recognize. It is filled with more potential for improvisation and innovation than false convictions and traditional practices concede.Footnote ¹¹¹ Protean power can be creative – as in the case of Silicon Valley and innovative start-ups. And it can be destructive – as in some of the novel products and practices that made the financial industry fall off the cliff in 2008 and in the surge in terrorist violence in recent years. Smart forecasts, prudence, and resilience offer some measure of protection in a world open to a statistically staggering range of possibilities that the human mind meets with a psychological craving for often unobtainable predictability. That craving leaves many political actors and scholars of international relations, in the words of legendary investor Charlie Munger, in the position of “a one-legged man in an ass-kicking contest.”Footnote ¹¹² A broader concept of power provides needed protection and improved vision. The 9/11 attack on the United States and what some have called the “assault” of America by tens of thousands of children migrating illegally in the summer of 2014 serve as two simple reminders of one basic fact. Until we stop focusing only on control power and begin to recognize also the role of protean power, unfolding events in world politics will continue to outpace our ability to understand and cope with them.