The common assumptions-ends-ways-means-risk (AEWMR) military strategy model instructs its users to make reasonable assumptions, balance ends, ways and means, and control for risks. But it does not offer much practical guidance on how this should be done, or how to choose between relevant balanced strategies. To address this problem, recent scholarship has proposed the concept of theory of success as an analytical tool to enhance the ability of strategists to assess, formulate, and validate military strategies. This article seeks to take this body of scholarship to the next level by operationalizing the theory of success concept into a simple easy-to-use practical guide that helps its users to identify the assumptions and causal hypotheses underpinning any military strategy, and to validate it logically and empirically. The article is based on the assumption that greater analytical clarity and rigour is a prerequisite for better military strategy. It expects conceptually clear, logically consistent and empirically validated military strategies to stand a better chance of success than vague, inconsistent, and poorly validated ones. Use of the analytical tool proposed here will of course not guarantee successful outcomes. Politics, biases and flawed intelligence may still result in the adoption of flawed assumptions and formulation of ill-suited strategies. Military strategy from may also continue to fail in the execution phase as a result of clever enemy responses and unforeseen developments.
The purpose of this article is to provide students of military strategy and civilian and military practitioners with a set of easy-to-use analytical tools that will enable them to analyse, formulate, and validate military strategy in a structured way with greater analytical rigour and clarity than is generally the case today. The article uses the common assumptions (what you take for granted in your analysis), ends (what you want to achieve), ways (how you want to achieve it), means (the resources you employ to achieve it) and risk (AEWMR) strategy model as its point of departure. This is a logical starting point as this model forms the basis of contemporary Western strategic thinking and practice (Heffington, Oler and Tretler, 2019; Joint Chiefs of Staff, 2019; Marine Corps War College, 2021; NATO, 2017; UK, 2014).
The hegemony of the AEWMR model is easy to understand. It is a very powerful analytical tool enabling its users to make sense of complex situations and to formulate strategies involving the military instrument to counter threats and seize opportunities. The problem with the AEWMR model is that it only advises strategists to assess threats and logically connect ends, ways, and means in a balanced manner (Lykke, 1989; NATO, 2017, pp. 3–2; NATO 2019, pp. 3–3; UK Ministry of Defence 2014, p. 8; Yarger, 2006, pp. 14–15, 66, 68; Yarger, 2012, p. 50). The model itself provides little practical guidance on how to achieve this balance and coherence. In most accounts, striking the right balance comes down to intuition, professional experience, wisdom, and some measure of luck given the high degree of complexity and uncertainty that military strategy-making involves. Echevarria II captures this line of thinking well:
No scientific method exists for determining how much military power is enough, or when balance (in a strategy) is achieved. The answer depends largely on the professional judgement of military commanders, and on what domestic conditions will allow in terms of expenditure of fiscal resources and political capital. In truth, balance, like beauty is in the eye of the beholder. (Echevarria II, 2017, p. 6)
Former U.S. Secretary of Defence James Mattis (2008, pp. 18–19), concurs; for him, “all operating environments are dynamic with an infinite number of variables; therefore, it is not scientifically possible to accurately predict the outcome of [a military] action. To suggest otherwise runs contrary to historical experience and the nature of war.”
No student or practitioner of military strategy would disagree with these observations. There is more to strategy-formulation than Jomini-style scientific calculation. Selecting and combining ways and means to attain desirable ends is a creative process more akin to composing songs or writing fiction (that is, art) than doing mathematical calculations (that is, science). The (song) writer and the strategist are both trying to produce successful new products by using their professional technical skills to link the component parts of songs, books, and military strategies in original ways (Eikmeier, 2015). The process is also highly politicised and conflictual, meaning that the formulation and balancing of ends, ways, and means result from political and bureaucratic compromises (Allison, 1969; Watts, 2012). Yet this does not mean that the process cannot be underpinned by scientific methods. Eliot A. Cohen has defined strategy in a way that nicely combines the two elements:
Strategy is the art of choice that binds means with objectives. It is the highest level of thinking about war, and it involves priorities (we will devote resources here, even if that means starving operations there), sequencing (we will do this first, then that) and a theory of victory (we will succeed for the following reasons). (Cohen, 2009)
According to Cohen, the art of choice shaped by intuition, experience, politics, and creativity must be underpinned by a theory of victory – a persuasive argument that the chosen combination of ways and means is likely to produce the desired ends without excessive costs and risks. A theory of victory, or success, as most prefer to call it (victory is too narrow a term), is based on a causal hypothesis: if we use force (ways and means) in a certain manner, then we are likely to achieve our end without running unacceptable risks, because we expect this cause and effect relationship to apply in this particular situation for the following reasons: A, B and C.
A theory of success is not used to explain how a specific strategy came about. It is not a claim about how strategy is made in a particular situation but a validation tool that can be used by strategists to assess the likelihood that execution of the chosen ways and means will bring about the desired ends. It can be used to test the validity of an existing strategy, or a proposed strategy yet to be executed.
The theory of success concept has in recent years been promoted by a growing number of scholars in order to enhance the analytical clarity and rigour of the strategy-making process (Biddle, 2015; Hill and Gerras, 2018; Hoffman, 2020; Meiser, 2016–17; Meiser and Sitara, 2018). While they have done a good job laying out its causal logic and suggesting how it can be tested, they have not provided detailed practical guidance showing how theories of success should be formulated and tested. This article will fill this gap by demonstrating how students and practitioners of military strategy can formulate and validate military strategies based on the causal theory of success logic. In addition, this article supplements the concept of the theory of success with the concept of a theory of threat, demonstrating that the same logic can also be used to perform the threat (or opportunity) assessments on which military strategy rests.
The article has four parts. The first presents the AEWMR model and the strategy formulation process; the second introduces the article’s understanding of causal theory and demonstrates how it can help practitioners to assess, formulate and validate military strategies; the third discusses the limitations of causal theories in the social world and advises military strategists to hedge their bets by considering multiple theories in order to assess threats/opportunities and to select the optimal combination ways and means in a given situation. The concluding part sums up the main findings and highlights the strengths and weaknesses of the proposed analytical tools.
The purpose of (military) strategy and the AEWMR model is to create a better future. The name of the game is the correct assessment of threats (the risk of enemy attack) or opportunities (chance to break through enemy lines) in your strategic environment, the formulation of desirable ends on the basis of this assessment and then the selection of the combination of ways (how to use force –counter-insurgency, deterrence, or show of force) and means (materiel, money, men) most likely to achieve them. The selection of ways and means should be informed by a risk assessment ensuring that their execution does not incur unacceptable risks. In the execution phase, the chosen ways and means are implemented in order to achieve the desired ends. Evaluation of the results and enemy reactions will typically create a need for adjustments or fundamental revisions of the entire strategy (Lykke, 1989; Yarger, 2006; 2012).
Strategists using the AEWMR model to structure their strategy formulation process must answer the following four questions:
Questions 1 and 4 are difficult to answer because they involve assessments of future developments: assessments of developments in the strategic environment and assessments of the likely effects of implementing the chosen combination of ways and means. To give an example, it was not hard for NATO to agree that Russia constituted a potential military threat following its use of force in Ukraine in 2014. But how big was the threat to the alliance, what form did it take, and what was the best way to address it in a cost-effective manner without incurring unacceptable risks? The threat assessment depended to a large extent on the perception of Russian intentions. Was the Russian leadership revisionist and driven by a desire to expand territorial control westwards to create a larger geographical buffer between Russia and NATO? Or was the Kremlin status quo-oriented, driven by fear of further NATO expansion and trying to prevent its geographical buffer from becoming ever smaller (Götz, 2016; Lanoszka and Hunzeker, 2019)? If Russia wanted to expand its territory to the west, the military threat to NATO was high – NATO weakness invited aggression. If Russia, on the other hand, wanted to maintain the status quo by preventing further NATO expansion, the risk was much smaller. If NATO expansion stops, no or very little military threat exists. But how should NATO strategists determine the correct threat assessment on which to base their strategy formulation? The AEWMR model itself did not help them to answer this question.
The same applies with respect to question 4 posed by the AEWMR model: what combination of ways and means was most likely to achieve the desired ends with acceptable risk? NATO responded to the threat posed by Russia to the Baltic countries and Poland by deploying enhanced Forward Presence (eFP). Yet the alliance has been heavily criticized for choosing this particular combination of a way (deterrence by punishment in the form of a tripwire force) and military means (5,000 personnel). A much-debated RAND study concluded that Russia could overrun the Baltics in less than 60 hours, and that seven brigades (some 35,000 personnel) were needed to stop a large-scale Russian surprise attack at the border (Shlapak and Johnson, 2016). This study consequently advocated a different way (deterrence by denial) and much larger military means.
Which of the two combinations of ways and means outlined above are most effective with respect to keep the Baltic states safe from Russian aggression? Which of them are most likely to achieve NATO’s desired end with acceptable risks? Did NATO chose the best combination of ways and means when opting for deterrence by punishment in the form of a tripwire force? The AEWMR model cannot answer these questions by itself.
These examples demonstrate that military strategists face the same analytical challenge when they chose among different threat/opportunity assessments and different combinations of available ways and means. Both choices are based on assumptions and assessments or hypotheses about future events: that the adversary is most likely to act in a certain manner because of Z; that execution of selected ways and means will bring about the end desired, because of Z. Such hypotheses are based on causal theory. It follows that strategists can use the same analytical techniques as social scientists to make hypotheses/assessments of future events.
How strategists can do this is the question we turn to now.
A causal theory, in its most basic form, can be expressed as: if we do X, then Y will happen, because of Z. It has four components: assumptions, a hypothesis that X causes Y, a logical “because of Z” argument linking X and Y, and evidence validating this argument (see Christensen and Raynor, 2003, p. 3). Causal theory serves as a simplifying device that enables its users to determine what is important and what is not. A theory functions as a pair of binoculars highlighting some causal relationships at the expense of others. It indicates that some causal factors are more important than others and specifies causal relations among them (Waltz, 1979, p. 8).
Assumptions are things that the theory takes for granted and hold constant, permitting the specification of the conditions under which the causal hypothesis – the cause and effect relationship – is expected to hold. Military strategy is also based on assumptions (Hammes, 2010; Lombardi, 2011; Marine Corps War College, 2021, pp. 101–109). An assumption provides a supposition about the current situation or future course of events, assumed to be true in the absence of facts (ATTP 5–0.1, 2011, pp. 4–9). Strategists have to rely on assumptions to plug knowledge gaps to assess what the enemy is most likely to do in the future and to devise a course of action that will enable them to achieve their desired end. In the NATO example above, it is necessary for NATO to assume that the Russian leadership will act rationally and base a future decision whether or not to attack the Baltic countries on a cost-benefit calculation. If Russia does not calculate its costs and benefits in the way that NATO assumes, then the deployed tripwire force may not prevent a Russian attack.
A causal hypothesis predicts that doing X will cause Y to happen – in our NATO example that the deployment of a tripwire force will deter Russia from attacking the Baltic states. The associated “Because of Z” argument explains why this assessment can be expected to hold. But it must be validated logically and empirically to increase confidence that the assessment is sound, and that NATO should risk basing its choice of ways and means on it.
Thus, assessment based on causal theory constitutes the core of military strategy formulation. The task of the strategist is to assess what the enemy is most likely to do, and on that basis formulate a desirable end, and assess the combination of available ways and mean most likely to attain it with acceptable risk.
Let us illustrate the logic returning once again to NATO’s 2016 decision to deploy four multinational battlegroups to the Baltic countries and Poland. The strategy-making process was initiated by a major change in the Alliance’s strategic environment – Russia’s annexation of the Crimean Peninsula and its resort to force in Eastern Ukraine in 2014. Again, beginning by defining the problem, NATO strategists now had to answer the four questions posed by the AEWMR model:
To answer question 1 with the help of causal theory, NATO strategists must identify and validate the most relevant theories of threat, each with their own set of assumptions, causal if-then-because hypotheses linking cause and effect, and supporting evidence. The procedure should be repeated with respect to question 4 in order to validate the chosen combination of ways and means. The procedure is demonstrated in Tables 1 and 2 below, where two competing theories of threat and success and their policy recommendations are presented. They were chosen because they dominated the NATO debate about how to deter a Russian attack in the Baltic region (for an overview of the debate, see Lanoszka and Hunzeker, 2019).
|THEORIES||OFFENSIVE REALISM (Mearsheimer, 2001)||DEFENSIVE REALISM (Taliaferro, 2000–01)|
|Theory of threat||Decision making assumption||Rational actor basing decisions on cost-benefit calculations||Rational actor basing decisions on cost-benefit calculations|
|Intentions assumption||Actors are expansionist||Actors are status quo oriented|
|If-then-because hypothesis||IF opportunity for cheap expansion, THEN Russian attack, BECAUSE Russia wants to maximize security and will expand territory until costs > gains||IF NATO expansion stops, THEN no risk of Russian attack, BECAUSE Russia wants to prevent loss of security and primarily uses force to protect its sphere of influence|
|Internal validity – conceptual clarity and logic||
|External validity – empirical testing and comparison with similar historical cases||
|⟹ Threat assessment||High risk of conventional attack||Low risk of conventional attack|
|NATO end||Prevent conventional attack||Prevent conventional attack|
|THEORIES||DETERRENCE BY DENIAL (Mazarr 2020)||DETERRENCE BY PUNISHMENT (Mazarr 2020)|
|Theory of success||Decision making assumption||Rational actor||Rational actor|
|Intentions assumption||Actors are expansionist||Actors are status quo oriented|
|⟹ Threat assessment||High risk of conventional Russian attack||Low risk of conventional Russian attack|
|NATO end||Prevent conventional attack||Prevent conventional attack|
|If-then-because hypothesis||IF NATO demonstrates capacity to deny cheap gain (fait accompli), THEN Russia will not attack, BECAUSE the benefits of attacking only exceed the costs if cheap fait accompli/quick success is likely||IF NATO signals willingness to fight if attacked and ceases from further eastern expansion, THEN Russia will not attack, BECAUSE Russia only attacks to prevent loss of territorial buffer/influence|
|Internal validity – conceptual clarity and logic||
|External validity – empirical testing and comparison with similar historical cases||
|⟹ Best ways and means||Deployment of 7 brigades in the Baltic region||Deployment of tripwire force in Baltic region|
|Risk||Moscow fear of NATO surprise attack triggers Russian pre-emption||Russian conventional attack overrunning tripwire force creating a quick and cheap fait accompli.|
Table 1 demonstrates how causal theory can be used to make relevant threat (or opportunity) assessments (theories of opportunity/threat) and validate them to ensure that the best threat/opportunity assessment is used as the basis for the subsequent formulation of ends, ways and means (see Table 2).
Threat assessments are generally based on a simple theory of threat stating that if the relative military balance of capabilities is favourable and offensive intentions are present, then an attack is likely to occur, because the potential attacker will assess the prospects of military success as high (for the original formulation, see Singer, 1958). This theory is based on a rational actor assumption and the probability of an attack will depend on a host of contextual factors such as adversary intentions (which is where the two theories in Table 1 differ fundamentally), intensity of the interests at stake (existential versus non-existential), relative balance of power, offence versus defence dominance, geography, the way the actors view war (strategic culture), and whether the adversaries have nuclear weapons and a situation of mutually assured destruction (MAD) applies (see Angström and Widen, 2015, p. 36; Heffington, Oler and Tretler, 2019, p. 31).
The assessment of adversary capabilities is anything but easy, as the first three months of the Russia-Ukraine 2022 war illustrate. Most analysts expected a relatively swift Russian military victory because of its vast quantitative superiority in men and materiel. Yet the Ukrainian defence forces defeated the Russian attempt to encircle the capital city of Kiev and prevented Russian forces from conquering the Donbas province. Dogged and competent Ukrainian resistance turned the conflict into a war of attrition, the outcome of which is difficult to predict as of June 2022. Capabilities cannot merely be assessed by counting men and equipment; terrain, morale, training, doctrine, preparation, and third-party support also matter, and their impact on actual fighting power is hard to estimate in advance.
Intentions are even harder to decipher for two reasons. First, aggressors have a strong incentive to conceal their intentions in order to preserve an element of surprise. Second, systems that appear offensive to outside observers may have been built and deployed for defensive reasons and vice versa (Tang, 2009). To give an example, the Anti-Access/Area-Denial (A2/AD) systems deployed by Russia in the Baltic region have been interpreted by some analysts as evidence of Moscow’s offensive intentions, because they make it difficult for NATO to reinforce the Baltic states in the case of war. Others view them as defensive and intended to compensate for Russia’s general military inferiority and inability to spend as much on defence as NATO and the United States. The problem for NATO strategists is that the Russian A2/AD systems can support both offensive and defensive strategies. It is for this reason that strategists tend to treat intentions as assumptions – as unknowns that cannot be proven in advance but need to be accepted as given to allow for strategy formulation and planning. There is consequently a tendency in military planning to apply a worst-case logic and assume that states will be as offensive as their capabilities and the relative balance of power allow them to be.
It is here use of causal theory becomes indispensable as an analytical tool to assess the Russian threat. Table 1 demonstrates how the two theories of threat dominating the debate over NATO’s military strategy in the Baltics draw very different conclusions about the character of the Russian threat. It shows how their assumptions, if-then-because causal hypotheses, and their validation produce markedly different threat assessments. The advantage of conceptualizing strategy-making as causal theory should also be obvious. It forces strategists to state their assumptions upfront and consider their implications, to be explicit about their hypotheses, to explain their assessments and recommendations why a given development constitutes a threat requiring a military response, and to validate logically and empirically. Explicitly stating each constituent part of a threat assessment makes it easier to detect logical and empirical weaknesses in the because argument presented to justify it. All four components of a threat assessment should be subjected to critical scrutiny.
Table 2 displays the two competing theories of success dominating the NATO debate on how the Russian threat to the Baltic states might be deterred. The first four rows in Table 2’s theory of success have been copied from Table 1 to show how the two competing threat assessments, and especially the assumptions on which they are based, play a key role in shaping the two theories of success and their recommended combinations of ways and means most likely to achieve NATOs desired end – deterrence of a conventional Russian attack.
The Offensive Realism assumption that states are expansionist produces the threat assessment that Russia is looking for an opportunity to attack (Table 1), which in turn produces the theory of success suggesting that NATO need to deny Russia a cheap fait accompli in the Baltics to deter an attack (Table 2). Hence the recommendation to deploy seven brigades in the region (deterrence by denial) (Mazarr, 2020). The Defensive Realism assumption that states are status quo-oriented produces the threat assessment that the risk of a conventional Russian attack is low, if NATO stops its eastern expansion (Table 1). This assessment gives rise to the theory of success suggesting that it is enough to signal a willingness to fight to deter a Russian attack (Table 2). This leads to the tripwire force recommendation. A tripwire force is enough because the risk of subsequent conventional and nuclear escalation is sufficient to prevent a Russian attack (deterrence by punishment) (Mazarr, 2020).
As was the case with respect to the threat assessments (Table 1), the different recommendations regarding the combination of ways and means most likely to attain the desired NATO end must be subjected to critical scrutiny (Table 2). Internal validity: is the causal logic clear and internally consistent, are the ends and means balanced? External validity: has Russia acted in accordance with the theoretical expectations on previous occasions? How has use of the proposed combination of ways and means worked in similar historical situations?
To consolidate the validity of threat assessments and choices of ways and means, military strategists should use the same validation techniques as social scientists to ensure that their analytical choices have the best possible logical and empirical foundation. Scientists validate their theories by examining their internal logical consistency and clarity (internal validity) and their empirical support – whether or not the expected causal relationship between X and Y has been documented in real world events (external validity; see Mearsheimer and Walt, 2013, p. 434; Meiser and Nath, 2018).
Military strategists can validate the internal consistency of proposed threat assessments and combinations of ways and means by examining whether the components – assumptions, if X-then-Y hypotheses and because of Z arguments –follow logically from one another. The theories exemplified in Tables 1 and 2 pass this test. Their concepts are clear, their hypotheses follow logically from the assumptions, and their because explanations are plausible.
External validity is examined by testing whether the chosen assumptions and the if-then-because hypotheses enjoy empirical support. For instance, has Russia acted as a status quo-oriented actor on previous occasions and refrained from expanding territorially when opportunities to do so arose? With respect to the choice of a tripwire force, it should be validated by examining whether deployment of tripwire forces deterred status quo-oriented powers from attacking in comparable historical situations. If this expectation can be confirmed, then confidence in deterrence by punishment is increased. Arguably, the successful deployment of a tripwire force by the Allied powers in West Berlin during the Cold War in order to deter a Soviet attack suggests that the NATO tripwire force is likely to succeed in the Baltics as well (for a counterargument, see Reiter and Poast, 2021).
Different techniques, quantitative and qualitative, exist for validating assumptions and causal hypotheses, and the best technique will obviously depend on the theory being tested and the character and quality of the available data. This said, strategists should, at a minimum, ask the following seven basic questions in order to validate the assumptions and the cause-effect hypotheses that their threat assessments and proposed ways and means rest on. The questions below have been culled from standard guides or “checklists” used by social scientists to test the validity of causal claims (Andersen, Lausten, and Cecchini, 2020, pp. 107–109; Brady, 2011, pp. 30–31; Hill, 1965; Marini and Singer, 1988, pp. 366–379).
Direction: Is X affecting Y as expected, or it is the other way around? This is the classic chicken and egg problem: has NATO expansion since 1989 been a cause of stability as claimed by NATO, or a cause of conflict increasing the risk of war, as claimed by Russia? Since NATO expansion and increased Russian hostility occur simultaneously, determining whether NATO expansion caused Russian hostility or vice versa is difficult. In such situations, solutions may involve asking more precise questions and adding more explanatory factors to obtain a more detailed understanding of the issue at hand.1 The answer as to whether NATO or Russia, or both are to blame for the growing tensions since 2014 depends on the assumptions, if-then-because hypotheses, levels of analysis, and the timeframes adopted (for an overview of the debate see Goldgeier and Shifrinson, 2020). The main positions in the NATO expansion debate can be plotted into a table similar to Table 1 allowing strategists to take the theories producing the different arguments apart and subject their constituent parts to critical scrutiny.
Frequency: Does the expected relationship between X and Y occur often? If the expected relationship can be found in other comparable situations, confidence in the hypothesis is strengthened.
Mechanisms: What causal mechanisms link X and Y? In addition to examining whether a change in X is followed by a change in Y, one can also examine the causal links or mechanisms linking X to Y. This is known as process tracing (Beach and Pedersen, 2011). The advantage of this method is that it allows you to break the expected causal link into more observable effects that logically have to exist if X is to cause the expected change in Y. This logic is illustrated in Figure 1 which deduces three logical observable implications from the causal hypothesis formulated by Mearsheimer (2014) that NATO expansion caused the Russian use of force in Ukraine in 2014.
If the following three statements can be corroborated, then confidence in Mearsheimer’s hypothesis is increased:
Whereas the two first statements can be validated empirically, the third cannot. Ukrainian NATO membership was not imminent in 2014, and the crisis that culminated in the Russian annexation of the Crimea was triggered by a disagreement related to an EU association agreement – not NATO membership. NATO membership was a distant prospect for Ukraine in 2014 and no written or verbal evidence suggests that fear of NATO membership triggered the Russian use of force. While NATO expansion and Russian fear that Ukraine might become a member of the alliance sometime in the future is likely to have played a role in Russian decision-making, it does not appear to have been a determining factor in the decision to intervene in Ukraine in 2014. While Putin justified his takeover of the Crimea with this fear after the fact, it was never expressed by him or other Russian decision makers prior to the intervention (McFaul, 2020, p. 119). Moreover, Ukrainian NATO membership had not been on NATO’s agenda since 2008, and the new Ukrainian leadership taking over from pro-Russian president Yanukovych reassured the Kremlin repeatedly that it would not seek NATO membership or deny Russia assess to its large naval base in Crimea (Sestanovich, 2014). It further weakens confidence in Mearsheimer’s causal explanation that Russia had not exhausted or tried all the diplomatic and economic means at its disposal before resorting to the use of force. Interestingly, Mearsheimer’s argument may have more explanatory power with respect to explaining the Russian attack in February 2022. EU and NATO countries increased their economic and military support to Ukraine markedly after 2014, and NATO granted it Enhanced Opportunity Partner status in June of 2020 (Mills, 2022). This rapprochement between Ukraine and EU and NATO increased the Russian incentive to launch a pre-emptive war to prevent Ukrainian membership of NATO before it was too late.
Plausibility: is the posited cause-effect relationship between X and Y plausible? Is it plausible to claim that NATO expansion caused the Russian use of force in Ukraine in 2014? According to Mearsheimer (2014), the answer is yes. Drawing on Realist Balance of Power Theory, he argues that the Russian use of force was a logical and predictable reaction to the Western offer of NATO membership to Ukraine in 2008, which threatened to remove Ukraine from the Russian sphere of influence. Others have disputed this view (Goldgeier and Shifrinson, 2020), but Mearsheimer is able to support his argument by presenting a coherent theoretical argument consisting of assumptions, if-then-because hypothesis, and supporting evidence (see Table 3).
|Theory of threat||Decision making assumption||Rational actors basing decisions on cost-benefit calculations|
|Intentions assumption||Actors balance military threats to their national security by all means necessary2|
|If-then-because hypothesis||IF Russia perceives Ukrainian membership of NATO as likely, THEN Russia will use force to prevent it, BECAUSE Russia views NATO expansion as a threat to national security and wants to maintain/maximize its power and influence|
|Internal validity – conceptual clarity and logic||
|External validity – empirical testing and comparison with similar historical cases||
|⟹ Threat assessment||High risk of Ukrainian NATO membership|
|Russian end||Prevent Ukrainian NATO membership and loss of sphere of influence|
|⟹ Best ways and means||Quick use of force (fait accompli) to prevent Ukrainian membership of NATO|
Omitted Variable Bias: Is the relationship between X and Y caused by one or more third variables influencing both X and Y and causing them to co-vary? For instance, critics of NATO expansion argue that NATO has not created stability in Central and Eastern Europe by preventing a power vacuum from developing between Russia and Western Europe. They contend that the stability created in Central and Eastern Europe was caused by the economic and political reforms set in motion by EU expansion (Goldgeier and Shifrinson, 2020). In view of the critics, the apparent causal relationship between NATO expansion and the increased stability within and among NATO Central and Eastern European member states is caused by the accession of these countries to the EU. An obvious way to test these competing hypotheses would be to compare the level of internal stability experienced by European countries that joined NATO after 1989 with the level in countries that only joined the EU, stayed out of both organisations, or joined both of them. This would give some indication as to whether NATO membership had an independent effect on the level of stability in the countries joining the alliance after 1989.
Strength: Is X necessary for Y to happen? In the social sciences, several independent (X) factors are likely to influence Y and cause it to happen. The purpose of theory is to identify the independent (X) factors with the strongest impact on Y, and looking for necessary conditions is one way to do this. A necessary factor has to be present for Y to happen. Usually a number of necessary factors have to be present to cause Y to happen. Another way to examine the strength of a theory is to compare it to other theories. Do other theories claiming to explain/predict the outcome of interest support or contradict the explanation proposed? Support will increase confidence in the explanation, whereas contradictions will invite scepticism. Finally, confidence in a causal explanation will be enhanced if it is difficult to find disconfirming evidence. It is possible to confirm or verify almost any plausible hypothesis by scouring the history of warfare from Thucydides’s account of Peloponnesian War to the present day. Strategists should therefore look for evidence that contradicts or falsifies their expectations, and examine whether these contradictions create a need for revising or rejecting their hypotheses. To give an example, strategists interested in examining whether Russia under President Putin is driven by expansionist ambitions, should not only look for evidence supporting this hypothesis. They should also look for evidence that Putin has refrained from exploiting opportunities for increasing the territorial buffer between Russia proper and NATO (see, for instance, Götz, 2016). Such examples would question the validity of the revisionist hypothesis put forward by Offensive Realism.
Time: Is X before Y in time? Logically X has to be prior to Y to cause it. In addition, distance in time is also an important consideration. The closer the distance in time between X and Y, the greater the confidence that X caused Y. The hypothesis that the Russian recognition of the independence of Abkhazia and South Ossetia in 2008 was caused by the Western recognition of the independence of Kosovo earlier that year is more credible than the hypothesis that NATO expansion in the 1990s caused the Russian use of force in 2014, because the distance between the latter events is so much greater.
While the understanding and use of causal theory presented above can enhance the quality of threat/opportunity assessments and the selection of ways and means, there is no escape from the “fog of theory” generated by the existence of competing theories, their limited predictive powers and their inconclusive empirical support (Mueller, 2001, p. 1). Causation is more complicated than the simple if-then-because statements in Tables 1 and 2 imply. Causation in the social world is not deterministic. If X is done, then Y does not automatically follow. Causal relationships are at best probabilistic: if X is done, then Y is likely to occur, and more likely than Z (Hill and Gerras, 2018). Since causation in the social world is not deterministic and accurate prediction is impossible, one should never attempt to go beyond the first order effects of a given military action and attempt to estimate what the second or third order effects are likely to be. This is where Effects-Based Operations (EBO) theory in its early versions went too far. While early EBO theory, like the theories of threat and theory of success proposed here, provided Western military planners with a useful analytical tool basing operational planning and targeting on a clear causal logic, it went too far in its expectation that it would be possible to predict the “full range of direct, indirect, and cascading effects—effects that may, with different degrees of probability, be achieved by the application of military, diplomatic, psychological, and economic instruments” (Davis, 2001, p. xiii). This is now generally accepted as mission impossible, and the current Effects-Based Approach to Operations (EBAO) employed by the U.S. Air Force is far less ambitious and explicit about the impossibility of prediction (AFDP 3–0, 2016). Yet it continues to use EBAO because of the analytical advantages that its causal logic offers to mission planning and targeting. It is with this experience in mind that understanding of causality presented in this article does not go beyond first-order effects and come with the clear warning that prediction is impossible (for the EBO debate, see Mattis, 2008; Ruby, 2008; Vego, 2006).
A further complication is created by the fact that causal relations in the social world vary across time and space because they are partly constructed by human actions. As Clausewitz (1989, p. 593) pointed out, “every age has its own kind of war, its own limiting conditions, and its own peculiar preconceptions.” It follows that assumptions and theories of war that worked in the past cannot automatically be expected to do so in the future. While the high level of economic interdependence among the European great powers failed to prevent the outbreak of World War I, this does not mean that a high degree of economic interdependence will not help to prevent a future great power war between the United States and China. MAD, a high degree of economic interdependence and a more negative view of great power war may be sufficient to prevent China and the United States from going to war in the future and falling into the so-called Thucydides’s trap that Allison (2017) worries about.
To reduce the risk of being wrong, strategists should always consider more than one theory (Marine Corps War College, 2021, pp. 76–77). As is clear from Tables 1 and 2, it is difficult to assess the character of threat posed by Russia and select the best combination of ways and means, because the two competing theories can be validated both internally and externally. In such a situation, a hybrid strategy might be called for in order to reduce the risk of being wrong. This was essentially what NATO opted for. The alliance supplemented the 5,000-strong tripwire force recommended by defensive realism with the establishment of the 20,000-strong quick-reaction Very High Readiness Joint Task Force (VJTF) deployable within a few days and the Four-Thirties Readiness Initiative putting 30 mechanised battalions, 30 air squadrons and 30 combat vessels on 30 days-standby (Belkin, 2020). While these additional measures cannot stop a large-scale Russian surprise attack on the border as recommended by Offensive Realism Theory, they do increase the risks and potential costs of launching a large-scale attack on the Baltic countries. In addition, NATO has also taken steps to counter Russia’s use of hybrid warfare. These steps have been reinforced by complementary EU measures (Jakobsen, 2019). In response to the Russian attack on Ukraine in February 2022, NATO executed its hybrid strategy of deterrence by deploying 40,000 troops in the eastern part of the Alliance along with significant air and naval assets (Euronews, 2022).
The theories of success in Table 2 address immediate threats; they are fire-fighting measures doing little to alter NATO’s deteriorating relationship with Russia in the short term. Yet, as Liddell Hart (1961, p. 372) reminds us, the purpose of military strategy is to create a better peace. The long-term end for NATO strategy must be the creation of a situation of peaceful co-existence, where the Baltic tripwire is no longer needed, and the Russian use of hybrid warfare ceases. NATO’s existing measures should therefore be complemented with ways and means that can achieve this long-term end. Theories of success proposing how this could be done are presented in Table 4.
|THEORIES||SPHERES OF INFLUENCE THEORY (Allison 2020)||REASSURANCE THEORY(Stein 1991)||POSITIVE ENGAGEMENT THEORY NINCIC (2010)|
|Theory of success||Decision making assumption||Rational actor||Rational actor||Rational actor|
|Intentions assumption||Russia seeks to maintain a sphere of influence close to home||Russia seeks to maintain a sphere of influence close to home||Russia seeks great power recognition and partnership with NATO on equal terms|
|NATO long-term end||Peaceful co-existence||Peaceful co-existence; reduce risk of miscalculation and inadvertent escalation||Transform Russia from enemy to friend|
|If-then-because hypotheses||IF NATO refrains from expanding further East and from interfering militarily in the Russian sphere, THEN Russia will cease its use of (hybrid) warfare, BECAUSE it is motivated by fear of NATO expansion/aggression||IF NATO signals restraint and lack of hostile intent, THEN Russia will reciprocate thus paving the way for diplomatic negotiations, BECAUSE Russian hostility is motivated by fear of NATO expansion/aggression||IF NATO offers Russia inducements for cooperation, THEN Russia will reciprocate and initiate a virtuous spiral of cooperation, BECAUSE the gains involved will have a transformative impact on Russia-NATO relations in the longer term|
|Internal validity – conceptual clarity and logic||
|External validity – empirical testing and comparison with similar historical cases||Mutual recognition of Western and Soviet spheres of influence stabilized Western Europe after 1945||Western reassurance helped to stabilize Soviet-US relations during the Cold War and prevent inadvertent escalation||Western use of inducements improved relations between Russia and the West in the 1990s|
|⟹ Best ways and means||Stop of further NATO expansion to the East||
||Restraint may be mistaken for weakness||
Tables 2 and 4 suggest that NATO should formulate a strategy composed of the combination of ways and means most likely to deter a conventional attack in the short term as well as the combination of ways and means most likely to pave the way for improved Russia-NATO relations in the longer term so that the eFP deployment becomes unnecessary. Tables 1, 2, 3, 4 demonstrate that military strategists can draw inspiration from a large array of theories when assessing threats/opportunities and selecting ways and means. This is why it is important to use the techniques presented above to reduce the number of relevant theories to a manageable number. The key to success is to keep the number of theories as low as possible.
In practice, the choice of strategy will never be determined by causal logic alone. While it should play a key role in rooting out theories of threat and success that cannot be logically and empirically validated, the choice of strategy will ultimately be determined by politics. Alliance, domestic, financial, ethical, and moral considerations will determine which validated options are deemed most feasible, suitable, and acceptable by the political and military decision-makers in charge (Yarger, 2012, p. 7). The recommendation provided by sphere of influence theory in Table 4, the cessation of NATO expansion, is politically unacceptable for NATO, which welcomed Finnish and Swedish membership applications triggered by the Russian attack on Ukraine in 2022. Similarly, the NATO choice of a hybrid strategy to deter Russia in the Baltics reflected a political compromise between the 5,000-strong tripwire force proposed by defence realism/deterrence of punishment theory, and the 35,000-strong force proposed by offensive realism/deterrence by denial theory. While the Baltic countries and Poland wanted a large deterrence by denial force, the vast majority of NATO countries did not want to pay the costs associated with fielding and sustaining it (Jakobsen and Ringsmose, 2018, p. 42). Hence, faced with a choice between two competing strategies that appeared equally valid, logically and empirically, the NATO members settled for a political compromise consisting of the trip-wire force backed up by reaction forces deployable at short notice.
The common AEWMR model does not help strategists to choose among competing relevant threat/opportunity assessments or to choose among competing relevant combinations of ways and means. Conceptualizing military strategy as causal theory provides strategists with specific tools for doing so. The article has consequently presented an easy-to-understand definition of causal theory (if X, then Y because of Z) and shown how strategists can use it to determine whether their threat/opportunity assessments (theory of threat/opportunity) and choices of ways and means (theory of success) are logically sound (internal validity) and supported by historical evidence (external validity). It has also provided an introduction to the basic techniques that social scientists use to validate causal theories, and shown how they can be employed to validate military strategies as well.
Conceptualizing strategy-making as causal theory has the advantage of forcing strategists to state their assumptions up-front and consider their implications, to spell out why a given development constitutes a threat/opportunity requiring a military response and why a proposed combination of ways and means is most likely to achieve the desired end with acceptable risks. It also forces strategists to validate their threat/opportunity assessments and choices of ways and means, both logically and empirically.
The approach proposed here does not make strategy making easier, nor does it increase the likelihood of strategic success per se. Success ultimately depends on the ability of strategists to make the right choices in a context of uncertainty and incomplete information and to react quickly when strategy execution creates unexpected results. This said, training practitioners – civilian and military – in the techniques presented here will make it easier for them to understand and formulate military strategy with higher internal and external validity. It takes practice to use the AEWMR model and causal theory properly, and it requires practitioners to familiarize themselves with theories from several disciplines. Strategy-making is a multidisciplinary team sport and causal theory is an essential tool for success (Cavanaugh, 2014, pp. 15–16). Hopefully, this article has shown in a user-friendly way how theories of threat/opportunity and success and causal theory can enhance the analytical quality of the military strategy making process.
1For a good short video illustrating how additional explanatory factors or causal paths may improve explanatory power, see: https://www.youtube.com/watch?v=38RlXdr4Np0.
I would like to thank the other contributors to this special issue, my Master’s students and colleagues at the Royal Danish Defence College, my colleagues at the Center for War Studies at University of Southern Denmark as well as Matt Cavanaugh, Frank Hoffman, Beatrice Heuser, Jeffrey W. Meiser and the two anonymous reviewers for constructive criticism and good advice. Their feedback increased the quality of the final product significantly.
The author has no competing interests to declare.
AFDP 3–0. (2016). The effects-based approach to operations (EBAO). U.S. Air Force. https://www.doctrine.af.mil/Portals/61/documents/AFDP_3-0/3-0-D06-OPS-EBAO.pdf.
Allison, G. (1969). Conceptual models and the Cuban missile crisis. The American Political Science Review, 63(3), 689–718. DOI: https://doi.org/10.2307/1954423
Angstrom, J., & Widen, J. J. (2015). Contemporary military theory: The dynamics of war. Routledge. DOI: https://doi.org/10.4324/9780203080726
ATTP 5–0.1. (2011). Commander and Staff Officer Guide. Washington DC: Headquarters, Department of the Army. https://www.globalsecurity.org/military/library/policy/army/attp/attp5-0-1.pdf.
Beach, D., & Pedersen, R. B. (2011). What is process-tracing actually tracing? The three variants of process tracing methods and their uses and limitations. APSA 2011 Annual Meeting Paper. https://ssrn.com/abstract=1902082.
Brady, H. E. (2011). Causation and explanation in social science. In R. E. Goodin (Ed.), The Oxford handbook of political science (pp. 1–62). Oxford University Press. DOI: https://doi.org/10.1093/oxfordhb/9780199604456.013.0049
Brands, H. (2020). Don’t let great powers carve up the world. Foreign Affairs. https://www.foreignaffairs.com/articles/china/2020-04-20/dont-let-great-powers-carve-world.
Cavanaugh, M. (2014). On strategic understanding teaching strategy from the ground up. Military Review, 94(4), 13–18. https://www.armyupress.army.mil/Portals/7/military-review/Archives/English/MilitaryReview_20140831_art006.pdf.
Christensen, C. M., & Raynor, M. E.. (2003). Why hard-nosed executives should care about management theory. Harvard Business Review, 81(9). https://hbr.org/2003/09/why-hard-nosed-executives-should-care-about-management-theory.
Echevarria II, A. J. (2017). Military strategy: A very short introduction. Oxford University Press. DOI: https://doi.org/10.1093/actrade/9780199340132.001.0001
Eikmeier, D. (2015). Operational art, design and the center of gravity [Video]. https://www.youtube.com/watch?v=nBStKk3fE4E.
Euronews. (2022). NATO now has 40,000 soldiers on Europe’s border with Russia, 18 May. https://www.euronews.com/my-europe/2022/05/18/nato-now-has-40-000-soldiers-on-europe-s-border-with-russia.
Götz, E. (2016). Russia, the West, and the Ukraine crisis: three contending perspectives. Contemporary Politics, 22(3), 249–266. DOI: https://doi.org/10.1080/13569775.2016.1201313
Hammes, T. X. (2010). Assumptions – A fatal oversight. Infinity Journal, 1(1), 4–6. https://www.militarystrategymagazine.com/article/assumptions-a-fatal-oversight/.
Hill, A. A., & Gerras, S. (2018). Stuff happens: Understanding causation in policy and strategy. Parameters, 48(2), 13–25. DOI: https://doi.org/10.55540/0031-1723.2941
Hill, A. B. (1965). The environment and disease: association or causation? Proceedings of the Royal Society of Medicine, 58, 295–300. DOI: https://doi.org/10.1177/003591576505800503
Goldgeier, J., & Itzkowitz Shifrinson, J. R. (2020). Evaluating NATO enlargement: scholarly debates, policy implications, and roads not taken. International Politics, 57(3), 291–321. DOI: https://doi.org/10.1057/s41311-020-00243-7
Jakobsen, P. V. (2019). New threats to European security. In D. J. Galbreath, J. Mawdsley and L. Chappell (Eds.), Contemporary European Security (pp. 153–172). Routledge. DOI: https://doi.org/10.4324/9781351235624-10
Jakobsen, P. V., & Ringsmose, J. (2018). Victim of its own success: how NATO’s difficulties are caused by the absence of a unifying existential threat. Journal of Transatlantic Studies, 16(1), 38–58. DOI: https://doi.org/10.1080/14794012.2018.1423603
Joint Chiefs of Staff. (2019). Joint doctrin note 2–19: Strategy. Department of Defense Headquarters, 10 December. https://www.jcs.mil/Portals/36/Documents/Doctrine/jdn_jg/jdn2_19.pdf?ver=2020-01-15-171758-647.
Lombardi, B. (2011). Assumptions and grand Strategy. Parameters, 41(1), p. 29–40. https://press.armywarcollege.edu/parameters/vol41/iss1/10. DOI: https://doi.org/10.55540/0031-1723.2572
Lykke, A. F., Jr. (1989). Defining military strategy = E + W + M. Military Review, 69(5), 2–8. https://cgsc.contentdm.oclc.org/digital/api/collection/p124201coll1/id/504/download.
Marini, M. M., & Singer, B. (1988). Causality in the social sciences. Sociological Methodology, 18, 347–409. DOI: https://doi.org/10.2307/271053
Mattis, J. N. (2008). USJFCOM Commander’s guidance for effects-based operations. Parameters, 38(3), 18–25. DOI: https://doi.org/10.55540/0031-1723.2437
Mazarr, M. (2020). Understanding deterrence. In F. Osinga and T. Sweijs (Eds.), NL ARMS Netherlands annual review of military studies 2020: Deterrence in the 21st century: insights from theory and practice (pp. 13–28). Springer Nature Switzerland AG. DOI: https://doi.org/10.1007/978-94-6265-419-8_2
McFaul, M. (2020). Putin, Putinism, and the domestic determinants of Russian foreign policy. International Security, 45(2), 95–139. DOI: https://doi.org/10.1162/isec_a_00390
Mearsheimer, J. J., & Walt, S. M. (2013). Leaving theory behind: Why simplistic hypothesis testing is bad for International Relations. European Journal of International Relations. 19(3), 427–457. DOI: https://doi.org/10.1177/1354066113494320
Meiser, J. W. (2016–17). Ends + ways + means = (bad) strategy. Parameters, 46(4), 81–91. https://publications.armywarcollege.edu/pubs/3298.pdf. DOI: https://doi.org/10.55540/0031-1723.3000
Meiser, J. W., & Sitara, N. (2018, 9 August). The strategy delusion. Strategy Bridge. https://thestrategybridge.org/the-bridge/2018/8/9/the-strategy-delusion.
Mills, C. (2022, 4 March). Military assistance to Ukraine 2014–2021. Commons Library Research Briefing, No. 7135. https://researchbriefings.files.parliament.uk/documents/SN07135/SN07135.pdf.
Mueller, K. (2001). The essence of coercive air power: A primer for military strategists. Air & Space Power Journal—Chronicles Online Journal. https://www.airuniversity.af.edu/Portals/10/ASPJ/journals/Chronicles/mueller.pdf.
NATO. (2019). AJP-5 – Allied joint doctrine for the planning of operations. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/837082/dcdc_doctrine_nato_planning_of_ops_ajp_5.pdf.
Nincic, M. (2010). Getting what you want: Positive inducements in international relations. International Security, 35(1), 138–183. DOI: https://doi.org/10.1162/ISEC_a_00006
Reiter, D., & Poast, P. (2021). The truth about tripwires: Why small force deployments do not deter aggression. Texas National Security Review, 4(3), 33–53. DOI: https://doi.org/10.26153/tsw/13989
Ruby, T. Z. (2008). Effects-based operations: More important than ever. Parameters, 3(38), 26–35. DOI: https://doi.org/10.55540/0031-1723.2441
Shlapak, D. A., & Johnson, M. W. (2016). Reinforcing deterrence on NATO’s eastern flank: Wargaming the defense of the Baltics. RAND Corporation, RR-1253-A. DOI: https://doi.org/10.7249/RR1253
Singer, J. D. (1958). Threat-perception and the armament-tension dilemma. Journal of Conflict Resolution, 2(1), 90–105. DOI: https://doi.org/10.1177/002200275800200110
Stein, J. G. (1991). Reassurance in international conflict management. Political Science Quarterly, 106(3), 431–451. DOI: https://doi.org/10.2307/2151741
Taliaferro, J. W. (2000–01). Security seeking under anarchy: Defensive realism revisited. International Security, 25(3), 128–161. DOI: https://doi.org/10.1162/016228800560543
Tang, S. (2009). The security dilemma: A conceptual analysis. Security Studies, 18(3), 587–623. DOI: https://doi.org/10.1080/09636410903133050
Yarger, H. R. (2006). Strategic theory for the 21st century: the little book on big strategy. U.S. Army War College. https://ssi.armywarcollege.edu/strategic-theory-for-the-21st-century-the-little-book-on-big-strategy/.
Yarger, H. R. (2012). Toward a theory of strategy: Art Lykke and the U.S. Army War College strategy model. In U.S. Army War College guide to national security issues (5th ed.), Vol. I: Theory of war and strategy. Strategic Studies Institute, U.S. Army War College (pp. 45–51).