Join War on the Rocks and gain access to content trusted by policymakers, military leaders, and strategic thinkers worldwide.
Iran launched hundreds of drones and missiles at Israel last April — and the attack might have been designed to fail.
The strike was a show of force built for spectacle rather than destruction, revealing a new rung on the escalation ladder where some states can act like they’re at war without paying the usual price. This is performative aggression — a game of bumper cars in a world long defined by the more familiar and high-stakes game of chicken. Characterized by limited-damage strikes that capitalize on a defender’s high-end air defenses and designed to evade escalation, this new kind of confrontation presents challenges to traditional strategies of deterrence. Fighting performatively, under the cover of effective air defenses, is a unique dilemma for a more powerful state against a weaker opponent.
To determine how policymakers should respond to this new rung of escalation, we should first define the challenge and distinguish performative aggression from other kinds of attack. Iran’s varied strikes on Israel over the past two years illustrate these differences in practice. We should then confront the limits of traditional deterrence — whether by denial or punishment — within this emerging paradigm. Crucially, understanding whether performative aggression is a one-off event or a repeated game reshapes how we evaluate the tactic’s effectiveness, and design defensive policies. There is no straightforward “solution” to performative aggression, just as no deterrence problem has a fixed answer. But knowing which game you are in is essential for crafting a strategy that can endure.
Two Kinds of Games
Performative strikes are de-escalatory in intent, but they are qualitatively different than the more recognizable “escalate to de-escalate” model. The distinction is captured in the difference between a game of chicken and a game of bumper cars. In chicken, both players understand they are at risk, and the winner is the player with the superior risk appetite for catastrophe. There is high potential gain — prestige, intra-war deterrence, escalation dominance, control of the conflict tempo — but high downside: war and runaway escalation. In bumper cars, neither player expects any significant damage, and thus at least one may be inclined to test limits. Bumper cars has lower potential gain — symbolic wins, face-saving off-ramps — but lower downside: less risk of provocation, assuming defenses hold. We can look to Iran as a case study to differentiate these two models in practice.
On April 13, 2024, Iran launched a large, heterogenous drone and missile strike against Israel. The strike began with a salvo of one-way attack drones, whose slow speed of advance required hours of lag before the launch of faster cruise and ballistic missiles.
Fourteen months later, Iran launched a much smaller salvo of ballistic missiles against Al Udeid air base in Qatar, the U.S. Air Force’s main forward hub. In both cases, nearly all the projectiles were intercepted. Neither attack caused significant damage to military or key civilian infrastructure. No fatalities resulted from either strike. Both attacks were telegraphed, either through advanced notifications or through the employment of slow first salvos.
Contrast those two cases with two other incidents. On October 1, 2024, Iran launched around 200 ballistic missiles at Israel. The strikes resulted in little damage to military sites, but two fatalities and tens of millions of dollars in infrastructure damage denote a difference in results.
Similarly, successive Iranian salvos over the course of the 12-day war killed more than two dozen civilians and hospitalized thousands. Critical infrastructure such as a power station and an oil refinery were among the targets.
In short: same combatants, similar weaponry, different results. Why? The answer is the emergence of a novel rung on the escalation ladder, one that sits above the “standard” crisis but below conventional war: performative, de-escalatory strikes. Unlike other kinds of performative attacks, which capitalize on post-operation media blitzes to amplify effects, this new model is predicated on almost the opposite — a lack of destruction.
Performative aggression relies on several factors: the quality of defenders’ integrated air and missile defenses, clear indications or warning of attack knowable to the defender, and often the slow pace of one-way attack drones to capture the signaling value of an offensive strike while minimizing the downside of provoking an escalated response. They do, however, share some similarity to other attacks recently characterized as performative — managing audience perception is important to success.
Iran’s strikes on Israel in both October 2024 and the 12-day War were games of chicken. Conversely, the strikes in April 2024 and against Al Udeid in June 2025 were bumper cars. The challenge, of course, is that intentions are not always communicated clearly, making it difficult to know which game is being played. Thus, bumper cars can sometimes masquerade as something akin to “staged chicken”— a hybrid in which the attacker publicly embraces risk taking while privately hedging toward low probability of harm through reliance on the defender’s air defenses. Information asymmetry is an essential part of and risk factor in the model: The attacker knows they are playing bumper cars, but the defender may not.
Ambiguous Deterrence
When it comes to risk, there is always the chance for miscalculation, that a weapon leaks through defenses and creates damage at a scale that demands a larger response than the attacker anticipated. Or that a defender fails to realize the nature of the game — or has a higher risk tolerance for escalation. Yet there is also a subtler question related to deterrence dynamics that this new rung imposes: How might the United States and its allies be strategically and politically vulnerable to coercion from performative signaling?
That question, at its core, may come down to whether such engagements are repeated or one-shot games. In the latter, Iran (in this case) secures a boost with its domestic and proxy audiences from low cost, high signal strikes. Israel faces a choice. It can take the victory, retaliate at limited scale, or escalate. If Israel takes either of the first two options — as Iran hopes — then the game is finished and the score settled. If Israel takes option three, Iran loses the game. In a repeated game, however, the winner or loser is less clear.
On the one hand, Iran may hold an advantage in an iterative dynamic. Tehran can launch weapons against sophisticated defenders and have them intercepted without losing face that undermines the regime. External expectations on performance were mild, the regime controls internal narratives, and they can blame the West or Arab monarchies for coming to Israel’s defense. The United States or its allies, meanwhile, may not derive the same signaling value out of their own intercepted strikes: performance expectations would be much higher, there is less narrative control, and no third party can be blamed for failure. That would imply an asymmetry by which the emergence of slow, cheap one-way attack drones paired with high performance air defenses enable adversary signaling at the expense of the United States or high-end allies like NATO or Japan. This would be particularly problematic for countries pursuing deterrence by denial strategies, which may inadvertently create permissive conditions for low-scale performative aggression. When defense overperforms, it can reduce the deterrent prospect of strong punishment because the retribution would seem out of proportion, thereby inviting attacks optimized for the attacker’s target audiences — domestic constituencies or third parties — rather than for material effects.
On the other hand, too many performative strikes risk convincing the defender that the adversary is playing bumper cars because they cannot credibly play chicken. Eventually, this might lead to a moral hazard, whereby the adversary invites escalation through the perceived incompetence of their systems, while the attacker achieves marginal returns in later iterations. Analyst and political reactions after the two events characterized in this essay as performative — Iran’s first big salvo and its later strike on the U.S. air base — were far from unanimous in interpreting these strikes as designed to fail. This lack of clarity as to what game is being played may, over the long run, weaken the attacker’s hand and embolden those agitating for stronger responses.
Implications
The quality of air defenses — and the slow speed of one-way attack drones — give some actors more latitude to use offensive tools in predominantly theatrical, non-lethal gestures. Modern defenses and cheap precision systems interact to create an intermediary escalation rung where states can gesture at war without paying the usual costs. How this tactical dynamic translates into changes in U.S. strategy and policy is an important but understudied feature of drone proliferation.
The phenomenon of performative aggression blurs the line between coercion and theater. Managing it will require deterrence strategies that punish intent, not just impact. Yet how we action that imperative is beset by paradoxes.
One option would be to impose some declaratory policy on performative aggression to push risk back onto the adversary, increasing the expected cost regardless of the strike’s outcome. Such an option could include telegraphed, pre-scripted escalation triggers to impose punishments through automatic escalation (i.e., deterrence by punishment). But prescriptive approaches work, by definition, by proactively limiting a leader’s decision space (or risk a credibility crisis for defaulting on the declaration). Prescriptive responses may also simply broadcast the thresholds under which adversaries can operate to avoid triggering a costly reaction.
An alternative option would therefore be to pursue a more ambiguous response policy to buy space for offramps and impose denial through doubt. Of course, here too there is risk, in this case that the adversary believes they can operate below the unspecified reaction threshold. These are fundamental questions of deterrence theory operating on a new rung on the escalatory ladder.
Not all wars and not all engagements will fit the pattern of performative aggression. Even in the case of Iran, there are two models in play in the last two years: games of chicken and games of bumper cars. Yet the risk of future low-impact, high signal strikes seems unlikely to abate as cheap drones and missiles intermingle with sophisticated air defenses. In these circumstances, the United States and its allies may have to learn to fight symbolically under the cover of strong defenses, with implications for the kinds of responses policymakers deem proportionate to the provocation. Ideally, those policymakers should also find a successful middle path to deterring the opportunism that capitalizes on an asymmetry of offense and defense, averting the paradox of modern missile defenses making confrontation more tempting because of their perceived futility. The task ahead for strategists and policymakers is therefore not only to prevent damage from drone and missile attacks, but to deny adversaries the ability to weaponize restraint itself in the aftermath of performative aggression.
Joshua Tallis, Ph.D., is a principal research scientist at the Center for Naval Analyses and the author of The War for Muddy Waters: Pirates, Terrorists, Traffickers, and Maritime Insecurity.
Image: Wikimedia Commons
