22 December 2023
CO23188 | The War Puzzle Revisited: Overconfidence and the Russia-Ukraine War
SYNOPSIS
A major paradox in international relations is the widespread fear and anxiety that underlies the security dilemma in times of peace, and the prevalence of overconfidence on the eve of war. This phenomenon is no better illustrated than in Russia’s fear for its security and its ill-fated invasion of Ukraine in 2022. While Russia and many outside observers expected a quick victory, two years on their early gains have been reversed and they have become bogged down in a long and costly war, fighting a hardened opponent with powerful allies. However, while this is a new war, it is an old story, on which human nature sheds some light.
COMMENTARY
The apparent overconfidence underlying Russia’s invasion of Ukraine is not unique. Russia also severely underestimated its opponent in the Russo-Japanese War of 1904-1905, despite Japan’s superior navy and an army that outnumbered theirs by three to one. Costly defeats on land and the annihilation of the Russian navy at the Battle of Tsushima betrayed a striking Russian overconfidence.
Of all the Great Powers, by 1914 Russia might have had the least illusions of victory. Yet Russia again severely underestimated its opponents, settling on “an extremely ambitious strategy” involving a three-pronged attack on Germany and Austria simultaneously.
But we need not single out Russia. All of the European powers showed evidence of overconfidence prior to World War I, and even that cataclysm did not upset the trend. In World War II, Britain, France and the Soviet Union underestimated Germany, Germany underestimated Britain and the Soviet Union, and Japan underestimated the United States.
From the Peloponnesian War (431-404 BCE) to the World Wars, to Vietnam, Iraq, Afghanistan, and now Russia-Ukraine, overconfidence in war is a gloomily persistent phenomenon. It recurs across time, across all types of actors and contexts, and among both defenders and attackers. Political scientist Stephen Van Evera concluded that “At least some false optimism about relative power preceded every major war since 1740”. The ubiquity and regularity of overconfidence, despite the lessons of history, suggest there must be some common – and persistent – underlying cause.
The War Puzzle
The recurrence of overconfidence on the eve of war speaks to a broader debate in international relations on the idea that war is a “puzzle”. The argument goes that if states are rational actors, they should be able to avert conflict in a pre-war bargain that reflects their relative power and avoids the costs of war. Yet, wars occur, even though both sides would be much better off finding a negotiated solution since the costs are so high – for both winners and losers alike. This leaves war a puzzle.
While explanations for war have been proposed within this rational actor paradigm (such as faulty information), my argument is that states – and their leaders – are not “rational” in the first place. States do not make rational assessments of their relative power compared to rivals. In fact, states systematically overestimate their power. But why?
The Psychology of Overconfidence
Whilst we may lament the havoc of overconfidence, psychologists would not be surprised. Cognitive psychologists, such as Nobel Laureate Daniel Kahneman, have demonstrated that all of us, as human beings, have numerous “biases” in judgment and decision-making that violate the expectations of rational choice theory, such as “anchoring”, “cognitive dissonance”, “loss aversion” and, not least, overconfidence.
Cognitive biases have helped us to understand many otherwise puzzling decisions in international relations, such as the role of analogical reasoning in escalating the Vietnam War, the role of attribution error in heightening hostility during the Cold War, and aversion to losses contributing to Japan’s decision for war in 1941.
While there are many such biases, overconfidence – a tendency to overestimate our capabilities, our control over events, and our future success – is arguably one of the most important. Kahneman said that, if he had a magic wand, overconfidence is the bias he would most like to eliminate.
Variation in Overconfidence: The Negativity Bias
Overconfidence is prevalent, but not a constant. There are in fact many sources of variation, which are important because they allow us to make predictions about when overconfidence is likely to be weaker or stronger and therefore alter the probability of war.
Here I highlight just one key source of variation, where we see not just the absence of overconfidence, but precisely the opposite. Besides the vast literature on overconfidence, psychologists have also found that people are subject to a powerful “negativity bias”, where we tend to over-weight negative information and events compared to equivalent positive ones – so much so that psychologists dubbed the phenomenon “Bad is stronger than good”. In international relations, this manifests itself in a powerful tendency for states to exaggerate the number and severity of threats they face.
How do we reconcile the fact that we seem to have both positive and negative biases? In fact, they coexist because they apply to different objects of assessment: positive biases centre on perceptions of ourselves and our capabilities, whereas negative biases centre on perceptions of other actors and the environment.
States thus exhibit two persistent biases: a negativity bias in their exaggerated fear for security, and an overconfidence in their capabilities. This mix of fear and overconfidence is a recipe for disaster because it means states think they are facing greater threats and yet are overconfident in their ability to deal with them, increasing the number of enemies and the probability of war.
Implications for the Russia-Ukraine War
How might these biases help us to understand the Russia-Ukraine War? Both the negativity bias and overconfidence seem to have played an important role in the causes of the war. Russia was motivated to act in the first place because it perceived what many see as inflated threats to its security and yet, at the same time, it believed it could resolve this problem through war.
Russia clearly overestimated its capabilities and control over events. This is illustrated in its effort to seize Kyiv at the start of the war, as well as in its broader strategy, logistics, and manpower, its failure to achieve key objectives, and subsequent withdrawals. Russia also underestimated the ability and resolve of Ukraine to resist, the breadth and resolve of international sanctions, and the level of support from NATO and Ukrainian allies. If anything, this became even more striking as time went on, with Russia doubling down in the face of a hardened Ukraine, tightening sanctions, and a growing arsenal.
What do these biases mean for the prospects for peace? Recalling the war puzzle, if one (or both) sides do not correctly perceive their relative power, they are unlikely to agree on any mutually acceptable solution. As long as each side sees a great enough threat and yet believes they can win (or at least can keep fighting), the war will continue and a negotiated solution is unlikely any time soon.
Implications for Wider International Relations
For NATO and Russia in Europe, the straightforward prediction of the negativity bias is that Russia is exaggerating the threat from NATO, and NATO underestimates the threat it poses to Russia. Meanwhile, both sides overestimate their relative capabilities (and thus their deterrent effect on the other), as well as their control over events (and the attendant risks of escalation), and their prospects of success in the future, all of which suggest the conflict with Russia will be more costly and will drag on longer than it otherwise would.
For the US and China in Asia, we reach similarly cautionary predictions. The negativity bias suggests that both sides will tend to exaggerate the threat that the other poses, especially the US as the declining power facing a growing peer competitor. Meanwhile, both sides will tend to overestimate their relative capabilities (and thus the strength of their deterrence), their control over events (heightening the risk of conflict at many potential flashpoints in the South and East China Seas), and their belief in ultimate success, making them stubborn and less willing to compromise.
People disagree about the real status and durability of Russia’s, the US’s, or China’s economic and military power. But the reality of that power does not matter as much as how decision-makers perceive it. The problem of cognitive biases is that dangers and opportunities will appear greater than whatever they are in reality, and it is the perception that shapes decisions. This is especially so in a world of changing geopolitics when there is greater ambiguity about relative power and therefore more room for perceptions to fill the gap.
Conclusion
Overconfidence has been a recurrent and powerful contributory cause of war throughout history, and we should expect it to do so again in the future. Combined with the negativity bias exaggerating the threats we face, there is often a toxic mix of fear and aggression. This yields gloomy prospects for the Russia-Ukraine war, and the danger of wars beginning or expanding in Europe and Asia may be greater than we realise. We might like to think that our rivals and allies are rational actors and will behave that way, but we should not expect them – or us – to meet that expectation, especially in times of crisis when tensions are running high and time and information are limited.
There is, however, some good news. Armed with knowledge about cognitive biases and the conditions in which they arise (in ourselves as well as in others), we can calibrate our strategies accordingly. If we want to successfully deter rivals, resolve conflicts and, when necessary, defeat opponents, then we could do more to exploit the science and history of cognitive biases and use this to our advantage. As Sun Tzu might have advised, we should use this knowledge to better “know the enemy and know yourself”.
About the Author
Dominic D. P. Johnson is Alastair Buchan Professor of International Relations at the University of Oxford, and Fellow of St. Antony’s College. He was the visiting S. Rajaratnam Professor of Strategic Studies at the S. Rajaratnam School of International Studies (RSIS), Nanyang Technological University (NTU), Singapore, in August 2023. His most recent book is Strategic Instincts: The Adaptive Advantages of Cognitive Biases in International Politics (Princeton University Press, 2020), and his previous books include Overconfidence and War: The Havoc and Glory of Positive Illusions (Harvard University Press, 2004).
SYNOPSIS
A major paradox in international relations is the widespread fear and anxiety that underlies the security dilemma in times of peace, and the prevalence of overconfidence on the eve of war. This phenomenon is no better illustrated than in Russia’s fear for its security and its ill-fated invasion of Ukraine in 2022. While Russia and many outside observers expected a quick victory, two years on their early gains have been reversed and they have become bogged down in a long and costly war, fighting a hardened opponent with powerful allies. However, while this is a new war, it is an old story, on which human nature sheds some light.
COMMENTARY
The apparent overconfidence underlying Russia’s invasion of Ukraine is not unique. Russia also severely underestimated its opponent in the Russo-Japanese War of 1904-1905, despite Japan’s superior navy and an army that outnumbered theirs by three to one. Costly defeats on land and the annihilation of the Russian navy at the Battle of Tsushima betrayed a striking Russian overconfidence.
Of all the Great Powers, by 1914 Russia might have had the least illusions of victory. Yet Russia again severely underestimated its opponents, settling on “an extremely ambitious strategy” involving a three-pronged attack on Germany and Austria simultaneously.
But we need not single out Russia. All of the European powers showed evidence of overconfidence prior to World War I, and even that cataclysm did not upset the trend. In World War II, Britain, France and the Soviet Union underestimated Germany, Germany underestimated Britain and the Soviet Union, and Japan underestimated the United States.
From the Peloponnesian War (431-404 BCE) to the World Wars, to Vietnam, Iraq, Afghanistan, and now Russia-Ukraine, overconfidence in war is a gloomily persistent phenomenon. It recurs across time, across all types of actors and contexts, and among both defenders and attackers. Political scientist Stephen Van Evera concluded that “At least some false optimism about relative power preceded every major war since 1740”. The ubiquity and regularity of overconfidence, despite the lessons of history, suggest there must be some common – and persistent – underlying cause.
The War Puzzle
The recurrence of overconfidence on the eve of war speaks to a broader debate in international relations on the idea that war is a “puzzle”. The argument goes that if states are rational actors, they should be able to avert conflict in a pre-war bargain that reflects their relative power and avoids the costs of war. Yet, wars occur, even though both sides would be much better off finding a negotiated solution since the costs are so high – for both winners and losers alike. This leaves war a puzzle.
While explanations for war have been proposed within this rational actor paradigm (such as faulty information), my argument is that states – and their leaders – are not “rational” in the first place. States do not make rational assessments of their relative power compared to rivals. In fact, states systematically overestimate their power. But why?
The Psychology of Overconfidence
Whilst we may lament the havoc of overconfidence, psychologists would not be surprised. Cognitive psychologists, such as Nobel Laureate Daniel Kahneman, have demonstrated that all of us, as human beings, have numerous “biases” in judgment and decision-making that violate the expectations of rational choice theory, such as “anchoring”, “cognitive dissonance”, “loss aversion” and, not least, overconfidence.
Cognitive biases have helped us to understand many otherwise puzzling decisions in international relations, such as the role of analogical reasoning in escalating the Vietnam War, the role of attribution error in heightening hostility during the Cold War, and aversion to losses contributing to Japan’s decision for war in 1941.
While there are many such biases, overconfidence – a tendency to overestimate our capabilities, our control over events, and our future success – is arguably one of the most important. Kahneman said that, if he had a magic wand, overconfidence is the bias he would most like to eliminate.
Variation in Overconfidence: The Negativity Bias
Overconfidence is prevalent, but not a constant. There are in fact many sources of variation, which are important because they allow us to make predictions about when overconfidence is likely to be weaker or stronger and therefore alter the probability of war.
Here I highlight just one key source of variation, where we see not just the absence of overconfidence, but precisely the opposite. Besides the vast literature on overconfidence, psychologists have also found that people are subject to a powerful “negativity bias”, where we tend to over-weight negative information and events compared to equivalent positive ones – so much so that psychologists dubbed the phenomenon “Bad is stronger than good”. In international relations, this manifests itself in a powerful tendency for states to exaggerate the number and severity of threats they face.
How do we reconcile the fact that we seem to have both positive and negative biases? In fact, they coexist because they apply to different objects of assessment: positive biases centre on perceptions of ourselves and our capabilities, whereas negative biases centre on perceptions of other actors and the environment.
States thus exhibit two persistent biases: a negativity bias in their exaggerated fear for security, and an overconfidence in their capabilities. This mix of fear and overconfidence is a recipe for disaster because it means states think they are facing greater threats and yet are overconfident in their ability to deal with them, increasing the number of enemies and the probability of war.
Implications for the Russia-Ukraine War
How might these biases help us to understand the Russia-Ukraine War? Both the negativity bias and overconfidence seem to have played an important role in the causes of the war. Russia was motivated to act in the first place because it perceived what many see as inflated threats to its security and yet, at the same time, it believed it could resolve this problem through war.
Russia clearly overestimated its capabilities and control over events. This is illustrated in its effort to seize Kyiv at the start of the war, as well as in its broader strategy, logistics, and manpower, its failure to achieve key objectives, and subsequent withdrawals. Russia also underestimated the ability and resolve of Ukraine to resist, the breadth and resolve of international sanctions, and the level of support from NATO and Ukrainian allies. If anything, this became even more striking as time went on, with Russia doubling down in the face of a hardened Ukraine, tightening sanctions, and a growing arsenal.
What do these biases mean for the prospects for peace? Recalling the war puzzle, if one (or both) sides do not correctly perceive their relative power, they are unlikely to agree on any mutually acceptable solution. As long as each side sees a great enough threat and yet believes they can win (or at least can keep fighting), the war will continue and a negotiated solution is unlikely any time soon.
Implications for Wider International Relations
For NATO and Russia in Europe, the straightforward prediction of the negativity bias is that Russia is exaggerating the threat from NATO, and NATO underestimates the threat it poses to Russia. Meanwhile, both sides overestimate their relative capabilities (and thus their deterrent effect on the other), as well as their control over events (and the attendant risks of escalation), and their prospects of success in the future, all of which suggest the conflict with Russia will be more costly and will drag on longer than it otherwise would.
For the US and China in Asia, we reach similarly cautionary predictions. The negativity bias suggests that both sides will tend to exaggerate the threat that the other poses, especially the US as the declining power facing a growing peer competitor. Meanwhile, both sides will tend to overestimate their relative capabilities (and thus the strength of their deterrence), their control over events (heightening the risk of conflict at many potential flashpoints in the South and East China Seas), and their belief in ultimate success, making them stubborn and less willing to compromise.
People disagree about the real status and durability of Russia’s, the US’s, or China’s economic and military power. But the reality of that power does not matter as much as how decision-makers perceive it. The problem of cognitive biases is that dangers and opportunities will appear greater than whatever they are in reality, and it is the perception that shapes decisions. This is especially so in a world of changing geopolitics when there is greater ambiguity about relative power and therefore more room for perceptions to fill the gap.
Conclusion
Overconfidence has been a recurrent and powerful contributory cause of war throughout history, and we should expect it to do so again in the future. Combined with the negativity bias exaggerating the threats we face, there is often a toxic mix of fear and aggression. This yields gloomy prospects for the Russia-Ukraine war, and the danger of wars beginning or expanding in Europe and Asia may be greater than we realise. We might like to think that our rivals and allies are rational actors and will behave that way, but we should not expect them – or us – to meet that expectation, especially in times of crisis when tensions are running high and time and information are limited.
There is, however, some good news. Armed with knowledge about cognitive biases and the conditions in which they arise (in ourselves as well as in others), we can calibrate our strategies accordingly. If we want to successfully deter rivals, resolve conflicts and, when necessary, defeat opponents, then we could do more to exploit the science and history of cognitive biases and use this to our advantage. As Sun Tzu might have advised, we should use this knowledge to better “know the enemy and know yourself”.
About the Author
Dominic D. P. Johnson is Alastair Buchan Professor of International Relations at the University of Oxford, and Fellow of St. Antony’s College. He was the visiting S. Rajaratnam Professor of Strategic Studies at the S. Rajaratnam School of International Studies (RSIS), Nanyang Technological University (NTU), Singapore, in August 2023. His most recent book is Strategic Instincts: The Adaptive Advantages of Cognitive Biases in International Politics (Princeton University Press, 2020), and his previous books include Overconfidence and War: The Havoc and Glory of Positive Illusions (Harvard University Press, 2004).
