war on the rocks

On Anticipating Surprise

March 5, 2018

Editor’s Note: This is the second installment in “Off Guard,” a series on surprise in war inspired by a new CSIS study. Read the first installment here.

Anticipating surprise is a real and longstanding security challenge, despite being an oxymoronic notion. “The future is not foreseeable,” former Secretary of State George Shultz once said, “however prescient we may think we are.” Yet, large institutions, including the U.S. armed forces, have to make strategies and investments based upon some conception of the future. Policymakers must make strategy despite an uncertain grasp of what tomorrow will bring. The only question is whether they do so badly or well.

The arts of strategy and forecasting are interrelated. Good strategy requires both long-term forecasting and imagination. As Colin Gray so often reminds us, defense planning is conducted to prepare for a future that is not reliably knowable. The intellectual component of strategy formulation is conducted in shadows of uncertainty, with the future masked by ambiguity and the unknown reactions of adversaries. The making of strategy can also be complicated by institutional preferences and mental frames that lock out options worth exploring. This brings to mind Thomas Schelling’s quip in the foreword to Pearl Harbor: Warning and Decision, in which he noted the tendency to confuse the unfamiliar with the improbable. The making of strategy has always required an unsparing examination of the unfamiliar and the paring away of institutional biases and personal blinders that restrict imagination. Strategic choices have to be made and they can be made intelligently despite the limitations of imperfect foresight.

As Mark Cancian’s superb study, Coping with Surprise, recognizes, there are many smart things that government leaders can do to prepare to respond to catastrophic surprise and increase resilience. But we should not miss the opportunity to evaluate the techniques of pre-crisis or “left of bang” thinking. The strategy formulation period is the true staging ground for making smart choices and for increasing post-surprise resilience.

Strategy under Uncertainty

Senior policy makers must resolve an inherent tension in formulating strategy. Hal Brands observes that strategy is beset with tensions including “between the need for foresight and the fact of uncertainty; between the steadiness and purpose that are necessary to plan ahead, and the agility that is required to adapt on the fly.” Resolving that tension, between foresight and inevitable uncertainty, is the holy grail of sound strategy. A leader that has explored the contours of the possible and considered potential hedges or branch plans is far ahead of any single-minded strategic hedgehogs bent on driving down a single path.

Government agencies tend to think about the future in linear and evolutionary steps, heavily relying upon past practices and historical contexts that may no longer be relevant. This can increase the likelihood of surprise. Bureaucratic preferences can also blind leadership to potential challenges. Surprise doesn’t always spring forth from unexpected events or collective ignorance, but rather from group denial. Most shocks were envisioned by someone, warned about, but resolutely ignored. There is little incentive for leaders and their planning staffs to confront “pink flamingos” that they have studiously avoided because they require bold changes.

Rather than engage in gaining foresight into an uncertain tomorrow, established bureaucracies seek their comfort zones. But then, stuff happens. In retrospect, after a strategic shock, it is said that key hints and signals were lost in the noise. In reality, the signals were blocked by comfortable preferences and institutional habits.

Coping with Uncertainty via Scenarios

To do strategy, one must overcome what Colin Gray insists is an incurable ignorance of the future. There are a variety of techniques used to gain foresight as the CSIS study persuasively notes. Most of the major defense institutions produce estimates of the future operating environment. The National Intelligence Council, the Joint Staff, the U.S. Army, and the United Kingdom all publish open source explorations of trends. These have value but a few are, as Ben Jensen detailed in War on the Rocks, “reminiscent of alchemy” or not well connected to strategies or modernization plans. Red teaming has been emphasized as an important tool in helping decision-makers better understand the vulnerabilities of a given course of action. Wargaming and other tools of educated futurology can be employed. But to open the aperture early in the planning process, scenario-based planning is recommended as a valuable technique.

Scenarios and “multiple futures” help policymakers foresee possible inflection points and bring uncertainty into account. Scenario-based thinking facilitates the systemic incorporation of critical drivers and trends that might fundamentally change the future environment in significant ways. By identifying key trends and drivers along plausible alternative paths, from the present into different futures, scenarios help leaders avoid the “default picture” by which tomorrow looks much like today. This helps policymakers break out of rigid mental models and open up a discourse among senior leaders about trends, assumptions, and potential shocks.

Scenarios are often used as illustrations to explain a set of global trends or future security environments. But good scenario development can do so much more. It can be an effective counter to strategy groupthink. Properly employed, scenarios can help reduce some of the critical influences of uncertainty and friction in strategy formulation. Scenarios can sharpen the diagnosis as well as shape options for tradeoffs in strategy options, particularly in the diagnosis and formulation phases of strategy development. Without scenarios, strategists may pursue bureaucratically favored solutions masked as viable strategy. With scenarios, the same strategy team may have a better feel for how their worldview contains more bias and higher risks in different worlds.

The art of strategy ultimately employs forecasting, risk management, and the testing of hypotheses. Good forecasters, including so-called super forecasters, are more scrupulous about their personal biases and are more empirical in their risk assessments. Good trend analysis and empiricism should be employed in crafting detailed scenarios. Proper use of scenarios, by senior managers not as a staff exercise, can serve to grade recommendations from the strategic planning team, or explore the boundaries of various excursions into the future. At its best, this technique helps overcome a deficiency the 9/11 Commission called “institutionalizing imagination.”

The evaluation of effective strategies, ones that can adequately respond to contingency and the real world, is a critical part of the strategy development process. Strategy testing against scenarios helps both the decision-maker and the strategy team by exploring consequences of seemingly favorable strategic plans. There is a tendency to ascribe untested benefits to preferred courses of action. As Michael Mazarr has noted in his book on risk analysis, “the most profound risk disasters in finance and national security come from insufficient attention to and awareness of the potential risky consequences of intended or favored strategies.”

Scenarios provide a less threatening way to lay out alternative futures in which the bias, preferences, and assumptions underpinning today’s strategy may no longer be true. They are not a panacea but they can help avoid groupthink. Decision-makers should temper that possibility with astute use of scenarios founded upon the critical assumptions or uncertainties that will impact their enterprise the most. I have participated or led study efforts using this technique several times to explore investment or organizational design options.

The Pentagon uses scenarios in a variety of ways, including concepts of operations for new hardware and in assessing the future combat force. These “Defense Planning Scenarios” are more geographically and military specific and are crucial to processes for assessing military capabilities and operational requirements. They are used most often in testing the sufficiency of the U.S. military force design, particularly for capacity sufficiency and the implications of simultaneous contingencies. The Defense Department generally excels at this form of scenario analysis.

Some employ scenarios for wargames or to educate readers about how potential trends may play out. These can be very educational. Examples would include Andrew Krepinevich’s 7 Deadly Scenarios and a similar book published in the United Kingdom titled 2020 World of War. These books offer scenarios about the interplay of several key trends and can amplify foresight.

Yet, the best technique focuses on identifying combinations of potential drivers and key uncertainties to explore not a linear extension of a singular future but alternative paths. This “multiple futures” approach is a technique ideal for confronting uncertainty in strategy and in anticipating surprise.NATO has used this process in the past, and their Multiple Futures reportgenerated a debate about critical drivers and forms of risk that NATO senior leaders might face. The discourse over the trends, uncertainties, and probabilities informs a board or a Pentagon leadership team and tries to make them consider the possible unknowns that more traditional processes sweep under the table as unlikely or highly improbable.

As defense planning expert Michael Fitzsimmons noted over a decade ago, an overemphasis on uncertainty and avoiding decisions does not help either. This does not reflect a support for prediction along narrow lines. Rather, it is a call for analytically sound comparisons to expose biases or parochial thinking in order to improve strategy. Fitzsimmons, quite appropriately, calls for serious analysis up front to frame options early and for maximizing flexibility for responding to what cannot be known with any reliable detail. The chief advantage of this approach is “not necessarily more precise predictions of the future, but rather greater clarity and discipline applied to the difficult judgements about the future upon which strategy depends.” One should explore the future with rigor and even skepticism to ensure balance but to write off efforts at foresight serves only to increase the shock effect of future surprises.

Focus on Critical Uncertainties

Picking the right drivers or critical uncertainties is important to sound forecasting. The critical uncertainties should be carefully identified and empirically extended into the future. The incorporation of multiple trends highlights tradeoffs and the potential magnification of seemingly unlikely events by their combination. Some potential critical drivers and uncertainties that explore tradeoffs in the current National Security Strategy might include:

  • Rate of growth of U.S. economy, will it be able to sustain 3.0 percent growth or decline and undercut both global security and social welfare needs for its aging populace?
  • Lost public consensus on U.S. international role, does the United States return to merely another major power or does it continue to seek to lead and sustain a rules-based international order?
  • Decrease in international trade and financial flows, is the end of globalization at hand given?
  • Cohesion or severe decline severe decline of the Western and Asian security alliances
  • Rate of erosion of America’s military edge
  • Proliferation of disruptive technologies to smaller states or non-state actors.

But while scenarios provide a good way of evaluating strategies, the scenarios themselves do not generate the strategy. As former Secretary of the Navy Richard Danzig once concluded, the “propagation of scenarios, however sophisticated, broad ranging or insightful, does not obviate the need for strategies for coping with uncertainty.” The scenarios are a mechanism that abet coping since they are plausible contexts a strategy may have to face, and for which responsible policymakers may have to prepare for or adjust their strategy to hedge against a possible risk. Or they may just be better prepared to make adjustments previously examined in strategy development. As Fitzsimmons has stressed in his findings:

[T]he transparency and discipline of the process of arriving at the initial strategy should heighten the decision-maker’s sensitivity towards changes in the environment that would suggest the need for adjustments to that strategy. In this way, prediction enhances rather than undermines strategic flexibility.

Leadership Discourse

Scenarios are designed principally to illuminate potential futures, and to serve as a starting point for strategic discourse. They can help to clarify the implications of trends, underscore major assumptions, and frame potential options about a world that an institution might have to adapt to. Alternative futures help decision-makers understand potential future contexts and their implications — draw out potential issues, enhance hypotheses, and lay out signposts. In short, the use of scenarios can test that hypothesis to help “future proof” the strategy that is selected.

They are best employed with interaction with the senior leadership team, instead of an elaborate staff exercise that sits in a dusty binder on a shelf. The debate or discourse a leadership team has over multiple futures enhances its decisions, clarifies strategic options (investments, divestments, and hedging), and better prepares for future adaptation. Examining the potential emergence of multiple futures is also helpful in testing the robustness and adaptability of a strategy. Using scenarios, one can test how well the strategy can adapt — and how much risk is assumed — if the assumptions change. If the risk is too high, then the strategy should be modified or contingency plans developed to mitigate the risks and make the strategy more robust. This should preclude a dominant but narrow hedgehog approach to strategy. Across many contexts, the forecaster Nate Silver shows that the best forecasts are pluralistic or multidisciplinary.

Multiple futures or scenarios are potent tools, properly designed and employed, but they are not strategy per se. They are a means to that end, a tool for Pentagon civilian leaders to ensure that tomorrow’s military is not built entirely upon yesterday’s mental models. These models, or what Lawrence Freedman calls internal scripts, should be acknowledged and as skeptically challenged as scenarios. Experts with these techniques recognize the inherent value and numerous pitfalls in their use. In my experience the greatest pitfall is the failure of senior leaders to engage each other over fundamental trends, potential discontinuities and the infamous “unknowns.” The comfortably familiar all too often becomes the enemy of imagination and magnifies surprise.


Imagination and rigorous exploration of the future are in short supply in government, despite the plethora of agencies devoted to it, as Cancian noted. But when governments are caught by surprise events, as Freedman observed in Future War: A History, it is “often not because they were unthinkable but because there had been no prior reason to push them to the top of the security agenda.” Thus, surprise and its shock were amplified, not by the adversary but by an institution’s own thinking. Policymakers need not worship at the altars of long-term defense planning nor avoid forecasts and making hard choices. Instead, they should exploiting “multiple futures” and critical uncertainties as part of what CSIS recommends as better intellectual preparation to enhance foresight and improve strategy.


Dr. Frank Hoffman is a Distinguished Research Fellow at the National Defense University where he focuses on strategy, military innovation, and future conflict. He earned his Ph.D. in War Studies from King’s College London. This article reflects his own views and not the policy or position of the U.S. government or Department of Defense.

Image: U.S. Army/Sgt. Jim Greenhill