A little over a year ago, the Trump administration withdrew from the Iran nuclear agreement negotiated by the Obama administration and reimposed sanctions on Iran. The U.S. goal was to impose economic pain on Iran until it agreed to a new deal that placed even more restrictions on Iran.

But Iran has not given in. In addition to making recent provocations like shooting down a U.S. drone, the Iranian government has resumed nuclear activities that were restricted by the deal — and claimed that its stockpile of enriched uranium will soon exceed the threshold imposed by the deal.

Recent U.S. actions, including new sanctions, cyberattacks and warnings of “great and overwhelming force” if Iran attacks “anything American” suggest that the Trump administration is seeking to show its resolve to bring Iran’s leaders to the bargaining table. But my research with Andrew Kydd shows that when bargaining over something that can change the balance of power — such as nuclear weapons — having too much resolve can actually backfire. In some cases, it can be helpful for leaders to communicate clearly that they will not attack if concessions are made.

What is ‘resolve’ and what are its benefits?

Researchers who study international conflict typically define “resolve” as willingness to fight. Generally, demonstrating resolve is a good thing. If adversaries believe that you are willing to fight, then they are more likely to make concessions — meaning that you can often get what you want without actually fighting.

Countries often try to “signal” their resolve through actions like public statements, military maneuvers and sanctions. Danielle Lupton, Roseanne McManus and Keren Yarhi-Milo explain that this type of signaling can often be successful at intimidating adversaries. The evidence shows that when the conditions are right, sending signals of resolve can lead to better outcomes in international disputes.

Why then have the Trump administration’s signals so far failed to persuade Iran to make concessions? The reason might be that in some circumstances, it is actually possible to have too much resolve.

Too much of a good thing?

To persuade your adversaries to make concessions, convincing them that you are willing to attack if they do not concede is an important requirement. But it’s also important to convince your adversaries that you will not attack if they do concede.

If your adversaries believe that you will attack them regardless of what they do, they have little incentive to do what you want. Among the first to articulate this insight was Thomas Schelling, who wrote, “To say, ‘One more step and I shoot,’ can be a deterrent threat only if accompanied by the implicit assurance, ‘And if you stop I won’t.’ ”

How does resolve play into this scenario? If we define resolve as willingness to fight, then high resolve will make your adversaries more likely to believe that you will attack if they do not comply with your demands. On the flip side, high resolve will also make them more likely to believe that you might still attack even if they do comply with your demands. That’s why having too much resolve can sometimes make it harder to bring adversaries to the bargaining table.

My research with Andrew Kydd shows that this dual-edged effect of resolve is most problematic when countries are bargaining over something that causes a big shift in the balance of power. Here’s why: Before making concessions that weaken its military position, a country must be confident that the rival demanding the concessions will not exploit its weaker position to attack in the future.

This means that in bargaining with Iran, the United States might actually be “too resolved” for its own good. The United States already has vastly superior military capabilities to Iran, and in asking for restrictions on Iran’s nuclear program, it is demanding concessions that would further weaken Iran’s position.

Iran is unlikely to agree to these restrictions without confidence that the agreement would lead to permanent peace. The statements of hawks within the Trump administration, such as Secretary of State Mike Pompeo and national security adviser John Bolton, are unlikely to give Iran this confidence. If Iranian leaders believe that a U.S. attack — either now or in the more distant future — remains a realistic possibility regardless of any concessions that they make, then it is logical for them to maintain their nuclear program.

What’s the way forward?

With a few exceptions, the U.S. strategy toward Iran has focused mostly on signaling resolve. But if the problem is too much resolve, then it might be helpful for the United States to adjust its strategy and also send signals of moderation and reassurance.

Our research suggests it’s possible for nations to simultaneously signal resolve and moderation, without either signal undermining the other. In the case of Iran, this would mean sending Iran the message that the United States is resolved to attack if Iran does not meet certain demands but will absolutely not attack if it does comply.

Research suggests that it is possible for public statements of moderation to be credible to foreign adversaries through the same mechanism that makes statements of resolve credible — public accountability. Jack Levy, Michael McKoy, Paul Poast and Geoffrey Wallace show that the U.S. public disapproves of leaders who say they will not use force and then attack as well as leaders who threaten force and fail to follow through.

Signaling moderation is no guarantee that Iran will come to the bargaining table. Trump’s history of going back on his statements — and the fact that his public approval changes little when he does — might make statements of moderation by him less credible. There is also the possibility that Iran has sufficiently aggressive ambitions that it would want to retain its nuclear program even if it trusted the United States completely.

Still, the current U.S. emphasis on signaling mostly resolve is not the only possible balance. Signaling moderation as well would probably make a complete breakdown in relations less likely.

Roseanne W. McManus is an associate professor at Pennsylvania State University and is the author of “Statements of Resolve: Achieving Coercive Credibility in International Conflict” (Cambridge University Press, 2017).