Introduction

Deterrence is the act of making someone decide not to do something, it is about preventing a particular act or behaviour from happening. Deterrence is a strategic effect.¹ In political terms, deterrence means: ‘developing … military power so that other countries will not attack your country’.² The principal question then becomes what comprises sufficient military power to ­deter a potential aggressor? In classical military theory, a generally accepted rule-of-thumb states that a numerical superiority ratio of three-to-one is required for winning an offensive with acceptable probability and level of risk. Consequently, a certain ratio might be neces­sary for maintaining reassuring deterrence. However, it is possible to consider such classical theory both classical, as in possibly obsolete, and merely theory, as in not necessarily relevant in practice. In other words, there might be more to air power, deterrence, and capability, than relative numbers from a theoretical perspective. Therefore it seems relevant to ask, is the classical theory relevant when discussing deterrence?

Such philosophical reflections about theory and practice are necessary as deterrence and capability are abstract concepts requiring theory for discussing their composition, from any perspective. Furthermore, the theoretical perspective may be more relevant from a relative position of strength than from a small nation’s perspective, as large power allows enforcing practice to align with theory. From the perspective of a relatively small military power, the distinction between theory and practice is therefore particularly interesting.

The Swedish Air Force Chief of Staff highlights three criteria for success in employing air power: numerical, tactical, and technical superiority; he recognizes that numerical superiority cannot be the Swedish norm, and concludes that focus has to be on tactical and technical superiority.³ Arguably, ‘tactical superiority’ denotes the ability ‘in practice’ to make relevant use of ‘technical superiority’, while ‘numerical superiority’ aligns with the theoretical perspective of being super­ior in a statistical sense. Regardless of this argument’s validity, the assumption that ‘tactical superiority’ is mainly an effect ‘in practice’ is fundamental to the present reasoning, which thereby concerns the difference between a theoretical and a practical perspective on military capability more than deter­rence specifically. Nevertheless, the classical theory is a good starting point for discussion.

Quantity vs Quality, is a Numerical Rule-of-Thumb Relevant?

The theoretical comparison of entity counts assumes implicitly a one-to-one relationship between cap­ability qualities. With modern technologies, this is not necessarily the case. One entity might be cap­able of evading several others solely based on different / superior performance characteristics. However, in the times of classical warfare, the supposed origin of the rule-of-thumb, differences between entities and vary­ing situational conditions were practically irrelevant because with large numbers, on average, individual performances tend to balance. Conversely, comparison of forces with fewer entities increases the signi­ficance of specific individual capabilities as well as the operational impact of local conditions. Today, numbers are usually small, thereby making every entity and situation count. Hence, as statistically significant relations do not account for specific cases, the theory of numerical superiority may be irrelevant in practice today.

Furthermore, modern capabilities, particularly air power capabilities, tend to build upon advanced technologies and system of systems designs. Such ­designs render the otherwise straightforward counting of entities difficult to collect because system integration at various levels of abstraction blurs component boundaries and intertwines capabilities. Some entities contribute non-crucial aspects to a range of capabilities, while others individual entities provide complete capabilities. Simply put, system components count differently depending on purpose and viewpoint. Hence, the theory of numerical superiority is obsolete when considering the relative quality of involved entities.

However, it is undeniable that quantity is a quality in itself, although with a complex relation well captured by the assertion: ‘quality is better than quantity, especially if deployed in large numbers!’⁴ The problem ­becomes identifying what matters more in a specific situa­tion, quantity or quality. It is a judgment that requires comparing two substantially different phenomena. Quantity is a measure, ideally objective and context independent, representing a real world ­aspect, predominantly some kind of physical property. Quality, on the other hand, is essentially a subjective value, context dependent, a psychological and socially situated phenomenon. Obviously, both aspects are significant, in some sense representing theory and practice respectively. Therefore, the question remains: What is more important, theory or practice?

Disowning Practice, Undesired Consequences from Focusing on Predictions and Theory

Concrete measurements of physical aspects are easy to interpret as objective facts, especially when re­presented by ‘the pure language of science, mathe­matics’.⁵ Consequently, statistically significant predictions are easy to interpret as objective facts as well. However, the problem of collecting relevant entity counts shows that even straightforward measurements may be ambiguous. In fact, all measurements and mathematical calculations are associated with subjective assumptions about their meaning, which means that they also require contextual interpretations. Moreover, statistical predictions require theoretical models, but ‘all models are wrong’.^6,7 Over­simplification by assuming that entity capabilities are equal on average may prove a model irrelevant. Models are, in fact, always simplified and limited descriptions of reality, capable only of depicting general relations in principle. Consequently, models can only predict principle effects in general, if assumed conditions return repeatedly. Models say virtually nothing about what actually will happen in any specific real world case, only what will happen in theory.

From the perspective of the individual, the actual ­result is always more important than any general prin­ciple regardless of its statistical validity. Quality in general, is relevant only in advance, as grounds for decisions, or afterwards, as data for analysis. In practice, actual quality is what matters. Subjective phenomena, such as quality, are fundamentally unpredictable precisely because they are subjective and therefore both local and specific. They depend on interpretations and judgments not yet made, which in turn depend on contexts and conditions not yet existing because they are the result of future actions and decisions. Furthermore, people tend to understand better what they also know practically than what they only know theoretically, and for people to know things practically they must be practically involved. For relevant judgments of subjective phenomena, people require the kind of understanding of options and consequences that comes from personal involvement. This makes subjective phenomena have a self-generating character that implies circular dependencies perhaps impossible to capture in mathematical models with sufficient relevance.⁸ The theoretical perspective simply is qualitatively different from the practical perspective. Ironically, theory is nonetheless required for also understanding the practical perspective.

Unfortunately, theories and models are also treacherous. They are powerful tools for understanding and thus powerful tools for persuasion and self-deception as well. Statistical predictions, for example, inherently disown practice because they predict general future effects from principles, thereby disregarding the impact of deliberate actions in specific cases. Furthermore, our apparent preoccupation with modelled ­aspects implies a theoretical focus in which the practic­ally oriented human nature often becomes a problem. This shows, for example, in the contemporary tendency to consider automated systems categorically safer than manual ones, which arguably comes from mistaking predictability for safety.⁹ While, in fact, enforced predictability of system behaviour among unforeseen hazards may cause accidents.¹⁰ However, with an understanding limited to theoretical aspects, the mistake is indistinguishable. In the stereotypical world of theoretical principles, predictable behaviour is perfectly safe and logically relevant, and the success of automation a self-fulfilling prophecy. That is, if we believe that models depict and predict the truth, then there is no room for human agency and creativity, and no need for responsible decisions.

Tactical Superiority Requires Capability in Practice

For the military, enforced predictability and stereo­typical behaviour is clearly undesirable, unless you are the opponent. That is, exaggerated focus on theo­retical aspects of capability and a disowning of practice implies reduced relevance of ‘tactical superiority’, which by the Swedish Air Force Chief of Staff was recognized as essential for air power and deterrence from a small nation’s perspective.¹¹ Tactical superiority was then argued to require the capability in practice to make relevant use of technology, but relevance is a soci­ally situated and unpredictable phenomenon. The ability to assess relevance of effects requires therefore the ability to interpret local contexts, which is a human-oriented activity in practice.

The Swedish military pilot education system is renowned for producing creative and responsible officers, which presumably is the result of a general focus on screening and coaching instead of large intakes and extensive dismissals. Coaching allows for a supporting atmosphere in which people are encouraged to strive for insight and comprehensive understanding, thereby facilitating self-governing and implicitly a non-stereo­typical behaviour. The competitive environment that follows from a dismissal approach, on the other hand, puts focus on formal aspects, thereby fostering a model compliant behaviour. The different perspectives are distinguishable in different education approaches and the Swedish system appears to have the former, practical perspective, in mind.

However, in the contemporary era of computerization, automation, and formal evaluation according to modelled parameters, the theoretical perspective appears continuously gaining ground, also within the Swedish system. While some aspects of this development might be beneficial, the consequences of a lost practical perspective may be a hindrance. New know­ledge and theory that makes the practical perspective and its implications comprehensible is therefore required. The following conceptual framework is a candidate theory that includes the practical perspective.

Practical Capability, a Conceptual Framework Honouring Practice

The Swedish Military Strategic Doctrine¹² defines cap­ability in terms of three pillars: physics, concepts, and morality (Figure 1, left side). The definition distinguishes between what ‘can be done’ and what ‘will be done’ because the three pillars extend the view of capability from being merely a calculable consequence of physic­al conditions to include psychological and social aspects. However, the definition is still rather theoretical, focusing on defining the capability. It fails to capture the practical perspective appropriately, and how that capability comes to be.

Figure 1 illustrates a conceptual framework for cap­ability explicitly honouring practice and human agency by complementing the theoretical pillars with a corresponding set of practical aspects. On the one hand, there are theoretical aspects describing and defining the capability. These aspects aim to be general and context independent principles, thus they are rather static. Practical aspects, on the other hand, describe effects deriving from potential capabilities. These aspects regard specifics and strive for context dependency, thus they are concerned with the dynamics of reality and human decision-making. The framework helps distinguishing between theoretical and prac­tical aspects of capability.¹³ Moreover, by depicting aspects in layers, the framework also supports distinguishing between levels of abstraction.

At the bottom level, concrete physical aspects make up hard conditions for capability, such as technical ­systems, personnel, infrastructures, performance prop­erties and availability. Physics is normally indisputable, thereby determining effectively what ‘can be done’. With­out physical means to fly, there is simply no flying capability. Physics translates into practical possibilities. With physical means to fly, there is a potential to use the flying capability so that flying occurs.

At the middle level, conceptual aspects make up abstract rules for ‘what should be done’ and ‘how’. Rules govern how people should think and act, regulated by doctrines and policies. Without rules for using a cap­ability, potential effects are essentially unknown. ­Concepts translate into practical skills. With skills governed by rules, for example through training, there is a potential to use the flying capability such that flying occurs in a structured manner.

At the top level, moral aspects make up reasons for ‘why things should be done’. Morality governs how people want to think and act, affected by leadership, grounding values, ethics, motivation, etc. Without purposes for using the capability, potential effects will likely be irrelevant. Morality and purposes translate into practical incentives. With an incentive to use the flying capability, there is a potential to apply skills and use the flying capability to generate relevant effects.

Summary

Having relevant military capability is undoubtedly a necessary condition for deterrence. The question is how to define what is relevant? Theoretical models help us understand the world, and we use models to predict principle effects in general and assess relevance. However, statistical predictions disregard effectively what happens in practice. An over-emphasis on theoretical aspects thereby disowns human agency, which arguably is essential for tactical superiority. In fact, disregarding the practical perspective turns predictions into self-fulfilling prophecies because without knowledge about the practical aspect of values, predicted behaviour is successful by definition. However, predictability from stereotypical behaviour is clearly undesired from a military perspective, unless you are the opponent. Therefore it is crucially important to have a view of capability that honours the practical perspective, a view in which the importance of tactical superiority becomes comprehensible. For the prosperity of such a view, theory honouring the practical dimension is required.

From the perspective of a small nation unable to enforce predictability on operations, the practical perspective is particularly important. Capability in practice is distinctive for the Swedish Air Force’s view on air power. The presented framework is currently applied in analyses and development efforts aiming to improve the balance between technical possibilities, practical skills, and relevant incentives, in order to increase the practical capability further. Military power is fundamentally about people, and desired effects such as deterrence are mainly social phenomena. Air power operates essentially within systems of social systems that exhibit complex dynamics. Practical capability is in this environment much about understanding meanings and values comprehensible only by sufficiently involved human beings, crucial for the ability to affect an established social structure with relevant use of effective technological properties. If timely and skilfully applied in a dynamic situation, a stable structure might be knocked over with a feather.

Gray, Colin S. (1998), ‘Explorations in Strategy’, Westport, Conn.; London: Praeger, ch. 3.

http://www.merriam-webster.com/dictionary/deterrence.

Bydén, Micael (2014), in ‘European Air Power: Challenges and Opportunities’. Edited by John Andreas Olsen. Lincoln, NB: Potomac Books.

Prof. Dr. phil. Holger H. Mey, Swedish Defence University Air Power Seminar, Karlberg Castle, Stockholm, 24 Feb. 2015.

Allegedly, a statement of Galileo, ‘the father of modern science’ (Whitehouse, David (2009), ‘Renaissance Genius: Galileo Galilei & His Legacy to Modern Science’). Combined with the view that science defines truth, the Galilean legacy implies ‘what cannot be described mathematically cannot be true’. Arguably, there is a related tendency to believe also, ‘what is described mathematically is true’.

Sterman, John D. (2002), ‘All Models Are Wrong: Reflections on Becoming a Systems Scientist’, System Dynamics Review, vol. 18, no. 4: pp. 501 – 31.

Box, George E. P. (1976), ‘Science and Statistics’, Journal of the American Statistical Association, vol. 71, no. 356 (Dec.): pp. 791 – 799.

Taleb, N. N. (2010), ‘The Black Swan, The Impact of the Highly Improbable’, Second Edition, New York: Random House.

Reason, James (2000), ‘Safety Paradoxes and Safety Culture’, Injury Control & Safety Promotion vol. 7, no. 1: pp. 3 – 14b.

Perrow, Charles (1999), ’Normal Accidents: Living with High-Risk Technologies’, New Jersey, Princeton University Press.

Ibid. 3.

Försvarsmakten (2012), ’Militärstrategisk doktrin (MSD 12)’, M7739-354023, 09 833:60820.

For an analogous distinction between theoretical (calculative) and practical (situated) aspects of usefulness, cf. Stensson (2014), ’The Quest for Edge Awareness, Lessons not Yet Learned – PhD thesis on Practical and Situated Usefulness of Advanced Technological Systems among Inescapable Uncertainties and Competing Interests in a World of Dynamic Changes’, Uppsala University.