Coming Down from Olympus: A Call for Normative, Descriptive, and Phenomenological Distinctions in Strategic Theory

Photojay | Dreamstime.com – Soldiers Wait For Helicopter In Iraq Photo

Defense analysts often wish for better strategies and tactics. This lament has multiple legitimate sources, from the gap between expectations of strategic excellence and dismal realities to the seeming absence of coherent decision-making frameworks altogether. The cruel paradox, perhaps, is that we live in an era in which we have never been more aware of the importance of strategic and tactical knowledge, but nonetheless seem to make nothing but poor choices.

I argue that this is an inevitable problem of failing to distinguish between different types of theories about how strategic decisions are made. Not all strategic theories are created equal; some are rooted in what strategists should do, others explain what science expects strategists to do, and the most traditional simply set out a framework for what strategists should know. While these separate bodies of knowledge may have enormous overlap, they should not be thought of as functionally identical.

First, I examine confusion in modern day defense analysis between strategy as we want it to be and strategy as it unfortunately often is. Next, I review a cross-section of challenges to both qualitative and quantitative theories and methods used to model strategic decision-making in conflict. Lastly, I argue for the importance of distinguishing between the purposes of different types of strategic theory. By keeping all of these theories distinct, those who care about strategy may select the right intellectual framework for the job.

The View from Mount Olympus

It may be questioned whether or not such parsing and distinctions are really justified in the first place. But consider, however, this recent Anthony Cordesman missive expressing frustration with American strategy:

In the real world, however, a government does not create a strategy by issuing wish lists and empty lists of its desires. A real strategy has to have a tangible plan, it has to have a clear program to implement that plan, and it has to have the budget and resources to make it work. This means making difficult trade-offs and setting clear priorities. It means establishing accountability and having measures of effectiveness. It also means justifying the choices with a clear analysis of the risks and costs involved.[i]

Right away, Cordesman makes a strong descriptive claim: in the real world, governments do not make strategies by issuing wish lists and empty lists of their desires. However, it is empirically false that in “the real world,” governments and other organizations avoid strategy-as-wish-list. It is, sadly enough, often the case.[ii] Next, Cordesman argues that a “real strategy” has to have a clear program to implement a tangible plan, with budget and resources allocated accordingly, and priorities, tradeoffs, and other important aspects specified upfront. This is also empirically suspect. Organizations have successfully “muddled through” without any of these things.[iii] Others have developed loose schemes for managing change, often incorporating scenario thinking.[iv]

There are also strong reasons to doubt that the kind of pre-formalized, rational design approach Cordesman envisions is feasible for many organizations and their problems.[v] In general, Cordesman’s view of strategy is one that, as noted by organization theorists, takes an “Olympian” view of organizational strategic rationality.[vi] Like a mighty Greek god haughtily perched high up on the mythical home of Zeus and Athena, Mt. Olympus, Cordesman’s strategist seems to be one that suffers from none of the frailties of mortal men and women.

Yet the biggest problem is that Cordesman’s analysis mixes descriptive and normative claims about strategy. In an ideal world, a “real strategy” would constitute everything he discusses. But in the real world, real strategies do not resemble his vision for a variety of reasons. It would be a mistake, however, to peg this as just one analyst’s confusion. The question of what strategy is “in the real world” – and the consequences of the answers for practical decisionmaking and achievement of desired goals – has dominated strategic debates for decades.

A consistent theme of the last few decades in particular has been a split between those that feel that, on the one hand, formalized and detailed theories of strategy are a useful tool and those that feel that strategy and strategic theorists merely rationalize the un-rationalizable. I will offer a short overview of different critiques of strategy in a variety of academic fields that either directly deal with strategic theory or overlap with it. While individual critiques of deviations from the notional “Strategic Man” can be combatted, collectively they are far more problematic.[vii]

I will also argue that the difficulties in settling these debates suggest that different kinds of strategic theory may serve differing needs and functions, and these distinctions should be respected when evaluating the desirability of any one strategic theory.

The Ideal and the Real in Strategic Theory

Challenges to strategic thinking can be broken up into a variety of categories, from individual disputes over aspects of strategy such as instrumental reason to doubt whether strategy can survive the collective observed human impediments to sound strategic action.

First, the instrumental character of strategy itself has been challenged on multiple grounds. Anthony Burke, for example, argues that the strategic catastrophe that the Iraq War represents is a formidable empirical and normative strike against the idea of strategy as a process that bridges ends, ways, and means.[viii] Martin Shaw declares that strategic theory is too often a fig leaf for “slaughter,” or at the very minimum downplays the connection between strategic thinking and transgressive activities such as genocide and mutual nuclear annihilation.[ix] As Hedley Bull has noted, much of this stems from anxiety over the perceived notion that strategy is not only inaccurate and fantastical in nature, but also immoral in character.[x]

Of course, much of this ignores that not all strategists cast war and conflict simply as an instrumental mapping of objectives to actions. The notion of the Clausewitzian “wondrous trinity,” for example, suggests that attempts to instrument violence to policy is just one of the several guiding influences that act on war in general.[xi] Others acknowledge a central place for passions and cognitive-affective notions in general within strategic thought.[xii] Moreover, studies of civil war suggest that even fairly brutal and seemingly illogical forms of political violence can be accommodated by strategic explanations.[xiii] The fact that dynamics of revolt, oppression, and retaliation may be generated as emerging products of decentralized interactions between myopic agents with simple strategies should suggest some foundation for strategy as an explanation, even when it suggests uncomfortable things about war and conflict.[xiv]

Other perspectives have criticized particular notions of strategic decision-making for purported indifference to distorting cultural biases. Ken Booth has argued that ethnocentrism acts as a distorting influence on strategic decision-making.[xv] Of course, this is not necessarily a strike against strategic theory, as Colin Gray and others have persuasively argued that identity and strategic culture alone is a weak explanation and is best nested within strategic theory overall.[xvi] Other similar challenges to strategic theory take issue with the idea of a single decision-maker, noting that decisions may be imperfect aggregates of groups, institutions, or other collective entities.[xvii] But this, at best, suggests institutional constraints on strategic decision, something that many strategic theorists acknowledge.[xviii]

Decision theory, game theory, and rational choice theory, the most prominent mechanisms for mathematically modeling strategic decision, have legions of critics in the various sciences. Criticisms range from cognitive implausibility and mismatches with data and experiments to uncertainty over what kind of “game” decisionmakers believe they are playing in the first place.[xix] Others have focused increasingly on both structural and cognitive-affective explanations that might explain deviations from strategic rationality.[xx] Some also argue that more qualitative ideas of strategy disregard varieties of strategic reasoning and competencies that are often combined in practical strategic work.[xxi] Finally, others have argued that complexity theory has invalidated traditional notions of strategy and conflict.[xxii]

It is difficult to address these criticisms collectively, but while they pose challenges they also have problems of their own. First, while game theory and other mathematically rooted models of strategic interaction can mislead, critics have had far more success criticizing these models than proposing alternative mechanisms that are both realistic and may function as a replacement for “unrealistic” notions of decision. Second, while varieties of strategic reasoning and competencies exist and ought to be taken seriously, isolating particular individual competencies and types of problems is an inherently fraught enterprise. It is true, for example, that logistical and administrative competencies concern tasks of accounting, estimation, and organization that may be treated as optimization problems. But the manner in which people solve them may not conform to stereotypes of bean counters.[xxiii]

Finally, treatments of complexity theory in military and policy settings are often developed with a studied indifference towards the mathematical and computational methods used to actually do complexity research in the social and natural sciences. This is not just misleading, but also actively pernicious, as the benefit of such research lies in making complex system interactions and foundations explicit. Without such explicitness (either in equations or running computer code), complexity functions as a pseudo-scientific, New Age-like explanation that analysts may twist at will.

If all of these criticisms have problems examined individually, they are more persuasive collectively. Richard K. Betts tallies up an enormous amount of theoretical and methodological problems with the way that strategic studies students think about their discipline, asking “is strategy an illusion?”[xxiv] Betts tries to answer his own question by asserting that it is not; strategy may be difficult but it is by no means impossible. This perspective is mirrored in Lawrence Freedman’s recent survey of strategy and a Center for Strategic and Budgetary Assessments monograph, both of which imply that as haphazard as human strategic reasoning may be, strategic behavior happens by default and necessity. Our strategies may be uncertain, flawed, and bounded, but we strategize anyway because we must.[xxv]

The problem with these studies is that some succeed at exposing all of the human and organizational frailties that prevent strategic competence, without making convincing cases as to why we can achieve good strategic behavior at all. For example, it is certainly plausible that one’s own strategy need not be perfect, but merely better than that of an opponent. But what, then, makes it better than the opponent’s strategy, given all of the barriers to good strategic decision making? And how do we reconcile the imperfect and haphazard nature of strategy as practiced with the often lofty ideals of strategic theory?

Coming Down From Mount Olympus

The answer may be in realizing that a “one size fits all” approach to strategic theory has poorly served both scholars and practitioners. Without a distinction between the ideal, the real, and other types of theories, scholars will find themselves unable to answer in any real way the question of whether or not strategy is an illusion. The most obvious retort to such a question is “what kind of strategy?” Some strategic ideas may be illusions, but useful ones. Others may be harmful if they lead to confusion and mismatched expectations.

Are we talking about, say, the unbounded decision-maker envisioned by Daniel Bernoulli in 1738 or the flawed human that makes decisions bounded by cognitive limits as well as heuristics and biases?[xxvi] Is our strategist an individual that skillfully connects ends, ways, and means, or a collective groping its way through “dialogue and negotiation” to a consensus?[xxvii] When it comes to using strategy as “theory for practice,” the notional strategist envisioned by theory need not be all of those things at once.[xxviii] There exist a variety of strategic theories, usable for a variety of purposes.

Some theories presume an “Olympian” level of competence and capacity in the decision-maker and thus do not explain how we make decisions.[xxix] However, just as the Greeks viewed gods and goddesses as models for mere mortals to emulate, we may use these unrealistic ideas as normative goals to strive towards in how we make our own decisions.[xxx] Checking actual decision-making against normative theories of strategy can be helpful in improving strategic performance, even if it is impossible for anyone but Zeus to attain the standard of Olympian capacity that these theories posit. As long as we are willing to settle for as close as imperfect humanity may get to the normative ideal and we do not conflate the ideal with the real, there is no harm in using normative theory.

Other theories are descriptive in that they describe how we actually make decisions, and may be rooted in observational data or experiments.[xxxi] While these theories may describe, at times, heuristics and biases that impede strategic decision-making, they also may suggest ways to exploit observed decision-making characteristics for better decisions (or at the very minimum be aware of common pitfalls).[xxxii] It is important that decision makers do not also conflate descriptive features of empirically observed decision and strategy for desired ones. Just because individuals and groups have a variety of heuristics and biases does not necessarily make these shortcuts desirable or useful.

A counterpoint to this is a popular line of cognitive science research maintains that simple heuristics actually outperform more elaborate strategies, but as with all science there are certain limitations and qualifications for these findings.[xxxiii] Certainly it would be unwise to view all descriptive knowledge about decision making and strategy as uniformly negative narratives of bias, delusion, and ill-chosen shortcuts. However, a distinction must be made between the use of descriptive theories as a way of detecting error or having realistic expectations about strategy and a normative approach of valorizing certain decision making processes and theories as desirable in and of themselves.

All of this would still preserve a prominent role for the most traditional form of strategic theory; phenomenological depictions of the nature of armed conflict that aim to serve as a body of objective knowledge for decision-makers to utilize as “theory for practice.” It is easy to forget, but strategy is far more than just theories of how people make decisions in adversarial environments. Antulio Echevarria has argued for the notion of Clausewitzian theory as an attempt to provide future strategists with a body of correct knowledge about the nature and dynamics of war and armed conflict, a kind of mental toolbox that could compensate for lack of direct experience in war usable in a pinch by the wise leader.[xxxiv] Such an approach may seem old-fashioned, but it actually mirrors the thrust of recent research in artificial intelligence (AI) on strategy in both war and other areas.

Clausewitz’s notion of strategic wisdom as a body of correct knowledge has allowed military officers and computer scientists to develop a Clausewitz AI that uses an ontology of war based on the American military doctrinal interpretation of Clausewitzian theory. While the interpretation itself is debatable, the manner in which the AI uses its knowledge base about the nature of conflict has some striking similarities with the manner in which Echevarria presents the task of Clausewitzian theory.[xxxv] More broadly, a large-scale study of strategy across knowledge domains has shown that strategy may be seen as a “shared relational structure” that generalizes from individual cases to classes of different similar cause and effect relationships about complex interactions with other human beings.[xxxvi] This is certainly plausible, as decisionmakers – for better or worse – draw connections, use analogies and cases, and otherwise query their mental databases for clues as to what kinds of decisions to make.[xxxvii]

Hence dismissing old-school strategists that often mirror Clausewitz in their analytical approach, as useless old fuddy-duddies is an enormous mistake. This has unfortunately been the approach taken by many young strategic theorists who have sought to tear down older antecedents without appreciating the methodology and purpose of these older approaches, and why they have remained useful to soldiers, analysts, and academics for so long.[xxxviii] They may be useful as long as the underlying purpose that motivated their work is appreciated, and as long as their concepts and approaches are also not regarded as useful by default simply because they are old and venerable. It may be trivially demonstrated that every strategic theorist anticipated X or Y situation without it being useful or meaningful.

One may also observe that, as a matter of both theory and practice, normative, descriptive, and phenomenological ideas about strategy may be combined. While this may seem difficult to imagine given the differences between these types of theories, it can be done. For example, though not by design or temperament, a social scientist in the manner of modern quantitative political science, Clausewitzian theory has informed the latest descriptive theories of decision-making in war in quantitative international relations.[xxxix]

Gods, Demigods, and Mere Mortals

Strategy is hard, and decision-makers need a variety of tools. Existing strategic analysis had not clearly differentiated between the nature, function, and optimal use of intellectual tools, hence the confusion of many observers pondering the future of American strategy. It is fine to look up to Mt. Olympus for inspiration and guidance as long as strategists do not themselves believe that they are or could be noble gods throwing thunderbolts from the sky. It is fine to use strategic theory as a way of understanding the nature, dynamics, and experience of human conflict so that when decisions are made correct knowledge may be utilized, as long as the purpose of theory is understood in these terms. It is acceptable to use social and behavioral science to predict what kinds of choices decision makers will make based on recurring observed trends, but not if doing so leads to conflation of expected actions with desired ones.

As difficult as strategy may be, it is not so difficult that a little clarity cannot go a long way. One does not have to be a god among men or women to understand what intellectual tools in the vast array of strategic literature is best for the job at hand.