At War on the Rocks, Mieke Eoyong intervenes in the Sanders-Clinton foreign-policy debate. Although the case made for Sanders’ foreign policy by those she critiques—including Sean Kay—is much broader, she focuses on three arguments: that “Sanders has superior judgment because he opposed the Iraq War and Clinton didn’t; Sanders would exercise restraint in intervention, where Clinton is on record supporting U.S. intervention in a number of cases; [and] Sanders would restrain defense spending.”

I’m going to respond to the first two. I do so as a recovering liberal hawk. In the 1990s, my views on foreign policy were profoundly shaped by the pages of The New Republic. But over the last fifteen years, I’ve moved further and further away from liberal interventionism. Don’t get me wrong: I’m still more of a ‘strong defense’ type than most people on the left. But the problems that I see with Eoyong’s case reflect the reasons for my own evolution.

Indeed, Eoyong’s first argument is that the real test of judgment is learning from mistakes. As she writes:

A candidate’s ability to admit he or she has made a mistake and take corrective action is far more important in the world where imperfect information and changed circumstance may render initial judgments as poor decisions. No one gets it right all the time. How do candidates cope when they get it wrong? What lessons do they learn? What steps do they take to address the problems?

Fair enough. And this is one reason why I don’t worry about a possible Clinton presidency the way that many on the Democratic left do (indeed, if and when Clinton wins the nomination my pocketbook will open up to her campaign and I will do everything I can to support it). But I think it telling that Eoyong has nothing to say about the actual lessons learned by liberal interventionists from Iraq.

What are some of these lessons? Let’s look at what Clinton said when it was clear that Iraq was turning into a quagmire. In 2004, Clinton told Larry King that she believed that she made the right choice in voting for the Iraq War. Her defense was similar to Kerry’s: the decision to authorize force was correct, the Bush administration didn’t exhaust their other options, and so forth. She also said that “The failure to plan is the most hard — of all the things is the hardest for me to understand. I mean, how could they have been so poorly prepared for the aftermath of the toppling of Saddam Hussein?” Since then, Clinton was been clearer that the Iraq War was a mistake. And I don’t for a minute think that, had she been elected in 2000, the United States would have invaded Iraq.

What’s striking about all of this, however, is that at the CNN South Carolina town hall, Clinton defended the decision to intervene in Libya using, beat for beat, the same language deployed by the Bush administration in defending the war in Iraq.

• The country has oil wealth and will be in good shape once it overcomes its internal problems? Check.
• Democracy is hard and we need to give it more time? Check.
• The country had a free and fair election before it slid into chaos? Check.
• Draw a comparison to occupied Germany and Japan? Check.
• The dictator was brutal and killing his people? Check.
• We had a broad coalition? Check.

And, of course, once critical problem with the decision to engage in militarized regime change in Libya was “the failure to plan… for the aftermath….” Recall that the United States had three options in Libya: to not use force, to defend Benghazi—which is what the Russians and Chinese thought they were agreeing to, despite the expansive language that the US got them to sign up for—and to entirely destroy the regime. The coalition did the last.

But, more broadly, the idea of the relatively clean intervention in which, to borrow a phrase from Sanders’ speech against the Iraq War, we assume that “what comes next” is relatively benign, resides at the heart of the seduction of hawkish liberalism.

Consider Eoyong’s second point:

While the costs of intervention seem clear in the wake of the Iraq War, there are serious costs to non-intervention as well. The crises in Rwanda, Sudan, and Syria all demonstrate the stark humanitarian costs of U.S. non-intervention. Terrorist groups organizing in the chaos in Afghanistan/Pakistan, Yemen, and perhaps even Libya pose direct security challenges to the United States. Russian incursions into Georgia in 2008 and Ukraine in 2014 undermine the sovereignty principles on which the modern international order is founded, and run the risk of further invasion. Sanders should clearly articulate when, as commander-in-chief, he would be willing to send American troops into danger to confront a threat, and what principles he would use to make that decision — not just when he would say no.

Putting aside the fact that neither Clinton nor Sanders have articulated a really clear test of when to use force—and for good reasons—there’s the question of what we mean by “intervention.”

One place that Sanders has put distance between himself and the foreign-policy left is drone policy; he’s made clear that he’s ready to use military tools to achieve American objectives. Indeed, there’s a certain vagueness surrounding the notion of “intervention” here that makes it difficult to respond to Eoyong. The United States enjoys a wide variety of instruments at its disposal—some military, some non-military, some involving direct US actions, others involving proxies—to intervene in conflicts. Unless Eoyong is suggesting massive escalation in the use of force with respect to terrorists groups, then many of these cases aren’t points of salient disagreement.

But then consider a few of of Eoyong’s examples.

First, Russian incursions into Georgia in 2008 and into Ukraine in 2014. Is Eoyong suggesting that the Bush administration did not go far enough in its response to Moscow’s intervention in Georgia? The US decided to send a military supply ship—the Dallas— to Georgia as a calibrated response; it did so in conjunction with a full-court diplomatic press that ultimately led to Russia withdrawing from Georgia. Should we have buzzed Russian warplanes with F-22s? Should the US have sent ground troops? Should it, in short, have publicly threatened Russia with war?

Obama’s policies in Ukraine have, if anything, been much more appropriately calibrated. Russia’s intervention in Ukraine is indefensible—whatever we think about ultimate culpability for the state of US-Russian relations and Russian-Ukrainian relations. But is Eoyong suggesting that the US should intervene militarily to push Russia back? This is a much more hawkish position than that pursued by the Obama administration. It is also dangerous. If these are examples of places where more intervention is better—or, more important, where a Clinton administration would see the need for military intervention—than this makes a clear case for Sanders.

Second, Rwanda. The Rwandan genocide was one of the great horrors of the late 20th century. International inaction—and the fact that it occurred under the Clinton administration—made Rwanda a defining moment in the evolution of liberal interventionism in the 1990s and 2000s. Samantha Power, both in her Atlantic article and in her Pulitzer Prize winning book, cemented Rwanda in the popular imagination as a case where an easy intervention could have reaped enormous rewards.  If only, the story goes, the US had been less worried about getting trapped in a civil war in central Africa,  we could have averted a holocaust.

This fable, however, is almost certainly wrong. Alan Kuperman made this case in 2001. But what if, like me, you’re skeptical of some of Kuperman’s overarching moral-hazard argument? Well, we now have the careful and painstaking work of Christian Davenport and Allan Stam. It conclusively shows that a ‘clean’ intervention was highly unlikely; the arguments underlying it depend on a misunderstanding of the dynamics of violence, the aims and objectives of the Rwandan Patriotic Front, and the likely implications of inserting foreign troops into a civil war. Could the international community have done more? Almost certainly. But that’s not an argument for direct American military intervention.

In the end, the likely variation between a hypothetical Sanders and Clinton administration, at least when it comes to the use of military force, might not prove that great. They will face the same pressures and the same contingencies. If Sanders secures the nomination, the Democratic foreign-policy establishment will offer its help—after all, the differences between Sanders and Clinton are much less stark than any of the GOP candidates.

But the lessons of Iraq are not “don’t go to war in Iraq.” Iraq should have punctured the worldview that undergirds liberal interventionism: the illusion of clean conflicts followed by rosy aftermaths. The hundreds of billions spent, many people killed, and regional instability that followed ought to have made clear that, when it comes to protecting American interests, restraint, strength, and security go together.