Some two years since Russia’s involvement in the 2016 presidential election roiled American politics, anger has again erupted over Moscow’s potential role in the midterm elections. At a White House press briefing last week, Intelligence Community leaders averred that Russia “continues to engage in malign influence operations” and warned that Russian hackers might resume targeting America’s voter rolls and voting machines as they did in 2016. The briefing undoubtedly had a domestic political purpose, but it is safe to assume that these officials would not make such assertions, nor would the White House host their briefing, absent real evidence of Russian activity.

In testifying to the reality of the Russian threat, however, the briefing left a major question largely unaddressed: What could the Russians possibly be thinking? If Moscow is hoping to pursue the ambitious bilateral agenda that President Putin proposed in Helsinki, continued election meddling would seem tailor-made to preclude it. On the other hand, if the Kremlin is looking not to normalize relations, but rather to divide and conquer American democracy by sowing dissension, then it could not have ginned up a more counterproductive tactic. Although Americans seem deeply divided about nearly every issue in our national debate, there is a strong bipartisan consensus on Capitol Hill that the country must do something serious about Russian meddling. As Putin himself knows well, few things are more effective in uniting a fractured polity than battling a common external enemy. How then can we explain Russia’s actions?

Influence Operations, Not Sabotage

Before addressing Russian intentions, we need to examine what we know and what we might reasonably infer about what Moscow is doing. Although IC officials were clear that they regard Russia as a genuine threat to the midterm elections, they were not specific about what Russia’s continued “malign influence operations” entail. In an earlier address that described a range of threats against our nation, however, Director of National Intelligence Dan Coats stated that “we are not yet seeing the kind of electoral interference in specific states and voter databases that we experienced in 2016.” This suggests that, for the time being, Russian cyber operators have not resumed efforts to penetrate election-related computer networks in advance of the midterms, nor have they gone beyond 2016 activities to engage in sabotage of election infrastructure.

But Coats also warned that Russian influence operations, such as the social-media trolling activities carried out in 2016, have continued. “Despite public statements by the Kremlin to the contrary,” he said, “we continue to see individuals affiliated with the St. Petersburg-based Internet Research Agency creating new social-media accounts, masquerading as Americans, and then using these accounts to draw attention to divisive issues.” Whether the scale of this activity has changed since the 2016 elections is unclear. Facebook announced last week that it had detected and removed thirty-two Facebook pages and fake accounts that have been posting information on divisive social issues, although it stopped short of linking this activity specifically to Russia.

Plausible Motivations

One of the challenges of intelligence analysis is that dispositive evidence—information that is consistent with only one explanation, thus ruling out others—is rare. Most intelligence aligns simultaneously with multiple pictures of an evolving situation, and analysts must put together what they believe is the most likely explanation while leaving open the possibility that other narratives might also be accurate. Intelligence failures often flow from anchoring prematurely on a hypothesis that analysts expect to be true while neglecting or downplaying alternative hypotheses they are inclined to doubt. Exploring alternative explanations for what Russia is up to is therefore critical.

If it is true, as Coats has stated, that the Russians are not yet targeting election-related infrastructure, this casts doubt on assertions that Moscow hopes to help or hurt candidates from either party in the elections. Russia could not realistically expect a modest social-media campaign to sway midterms toward either party; anything more ambitious would require more extensive—and eminently detectable—activity. Modern elections are won through micro-targeting of like-minded voters to stimulate turnout. Any serious Russian effort to affect the vote would require either detailed knowledge of electorate demographics or access to voting machinery that would facilitate falsification. It appears from Coats’ statement that the Russians are not pursuing either avenue at present.

Why then continue a social-media influence effort that is of dubious utility in affecting American voters, but is unquestionably effective in undermining bilateral relations? The Washington consensus holds that Moscow has no genuine interest in rapprochement. Rather, it bears an ideological hatred of democracy and hopes to undermine our system of government by exploiting social divisions. This explanation is not implausible, but it ignores the reality that Russia has friendly relations with many of the world’s prominent democracies, including Israel, India and Japan. And it underplays the origins of Russian social-media trolling operations, which started inside Russia in the mid-2000s as a defensive response to “color revolutions” in ex-Soviet republics and intensified in the wake of the social-media-fueled “Arab Spring” in 2011, which the Kremlin worried was a U.S.-backed regime change campaign that might soon spread to Russia. Russians honed their trolling skills inside Russia against perceived American fifth-columnists, and gradually took that battle abroad to Ukraine and then the United States.

If Moscow views its social-media campaign as a defensive response to American meddling in Russian politics, then it might see little to gain and much to lose by giving up this activity without getting reciprocal U.S. concessions. Putin often recalls occasions when he made unilateral concessions to the United States, such as closing Russia’s intelligence collection site in Cuba and withdrawing from its naval facility in Vietnam in 2001, without gaining anything in return. As a result, many Russian nationalists view him as too eager to court Washington and too reluctant to stand up to American pressure. He may be wary of feeding those perceptions by ordering Russian trolls to stand down absent an American agreement to abstain from meddling in Russia’s domestic affairs. And, in view of the U.S. media’s near obsessive focus on Russia’s activity, he may well see the continuation of that activity as a bargaining chip too valuable to be tossed aside.

Mutual Misfires

The problem with this approach, of course, is that by failing to stop Russia’s social-media meddling, Putin has retained a bargaining chip but precluded any possibility that America might bargain. Moscow appears to have badly misjudged how much damage its cyber intrusions and influence operations have inflicted on U.S.-Russian relations and, by extension, Russia’s own security. The Kremlin’s studied ambiguity about what it has been up to has narrowed rather than widened the Trump administration’s room for maneuver on Russia policy. Dismissing U.S. concerns as a temporary wave of domestic political hysteria, for which Moscow bears no responsibility, has become an excuse for Kremlin inaction in addressing a dangerous state of affairs. At this point, it is difficult to envision a resolution without Russia’s acknowledgment of its actions.

By the same token, Washington must reckon with the reality that it has miscalibrated its approach to deterrence. For deterrence to be effective, the Russians must believe not only that we will follow through with threatened punishments, but also that we will reward good behavior. So far, America’s response to Russian meddling has been almost exclusively punitive, crafted to make it exceedingly difficult to lift sanctions in the event Russia complies with our demands. In encouraging Moscow to believe punishments will remain in effect even if it halts meddling, we have unintentionally incentivized continued Russian defiance.

Prospects that either side might change course appear dim at present. Each believes the other is solely to blame for the impasse. Each is continuing policies that increase the likelihood of confrontation. The politics on both sides rewards those who demonstrate toughness. Escaping our bilateral spiral of hostility, however, will require statesman courageous enough to balance firmness and force with diplomacy and compromise.

George Beebe is director of the Center for the National Interest’s intelligence program. He formerly served as chief of Russia analysis at the CIA, and as special advisor to Vice President Dick Cheney on Russia and the former Soviet Union.