Editor’s Note: Part of the job of intelligence officials is to give bad news to policymakers. But should officials at times soften their assessments or otherwise pick their battles in order to maintain the access to policymakers that is vital for intelligence to be relevant? Joshua Rovner of Southern Methodist University explores the Iran nuclear deal in this context, warning of the temptation of what he calls the “soft politicization” of intelligence and the risks for both intelligence and policymakers of doing so.
Secret intelligence is playing a public role in the ongoing debate over the Iran nuclear deal. If the deal is finalized in June, Iran will sacrifice much of its existing uranium enrichment capabilities in return for lifting some economic sanctions and will have to accept an intrusive inspections regime to verify its compliance for more than a decade. The presence of inspectors will create new opportunities for intelligence collection: Not only will intelligence agencies benefit from inspection reports dealing with Iran’s nuclear program, but they will be free to explore other areas of Iran’s scientific and associated industrial infrastructure.
Obama administration officials have expressed confidence that the intelligence community will be able to watch Iran closely, and the intelligence community has returned the compliment. In a recent public appearance, CIA Director John Brennan expressed satisfaction that Iran had made so many concessions and applauded U.S. diplomats for securing a deal that was “as solid as you can get.” There is no reason to doubt Brennan’s sincerity. Indeed, while relations between U.S. policymakers and intelligence leaders are sometimes fractious, the two sides are on the same page when it comes to Iran. Declassified U.S. estimates are broadly consistent with administration statements on Iran’s nuclear progress. Since 2007, the intelligence community has assessed that Iran does not have an active nuclear weapons program, but that it is committed to maintaining its enrichment capability. While some hawks have criticized these estimates, there is nothing to suggest that they were disputed by President Bush or President Obama. And if intelligence on Iran is as good policymakers believe, then there is no reason the intelligence community would worry about its own ability to monitor Iranian compliance.
But this synergy between intelligence and policy may not last forever. What will happen, for instance, if the intelligence community discovers that Iran is cheating? Having staked itself to the nuclear deal, the administration may be reluctant or unable to accept this kind of bad news. Worse yet, the intelligence community will be under pressure to report on Iranian activities in public, given that policymakers used intelligence as a major selling point in the U.S. ability to verify compliance. Instead of keeping intelligence under wraps, policymakers will be tempted to politicize it by pressuring officials to report findings that are aligned with their own views.
At this point intelligence officials will face a difficult choice. On the one hand, they can push back and resist pressure to change the tone or substance of their estimates. At the same time, they might worry about spending all their political capital in one shot, leaving them isolated and pushed out of the policy process. As in any advisory relationship, intelligence-policy relations are iterative. However much intelligence officials would like to speak truth to power today, they fear the price will be losing influence in the future. Iran is not the only game in town, of course, and intelligence officials presumably want a seat at the table in discussions of issues ranging from counterterrorism to great power politics. For this reason they may engage in “soft politicization” by toning down their conclusions or making it appear as if there are a range of equally plausible interpretations from the same underlying evidence. If this occurs, policymakers will be able to claim that they made decisions on the basis of the best available information.
Intelligence officials have often yielded to the temptation to soft-peddle estimates that cut against policy views. It is not hard to understand why. Confronting policymakers with bad news seems like a recipe for losing access and influence, especially when intelligence findings challenge the logic of policy decisions. “Outright pandering clearly crosses the line,” Columbia University’s Richard Betts writes. “But what about a decision simply not to poke a policymaker in the eye, to avoid confrontation, to get a better hearing for a negative view by softening its presentation when a no-compromise argument would be certain to provoke anger and rejection?” Moreover, some issues are more important than others, so it makes little sense to alienate policymakers on relatively minor issues. If it is true that intelligence chiefs have a finite stockpile of political capital, they should spend it wisely.
Engaging in this kind of soft politicization might seem like common sense. Unfortunately, there is little evidence that it actually works.
Consider the case of Director of Central Intelligence (DCI) Richard Helms, who on more than one occasion bent to policy pressure in order to keep intelligence from becoming irrelevant to the policy process. In 1969, for instance, the Nixon administration leaned on Helms to exaggerate the capabilities of a new Soviet missile while it was trying to secure Senate approval of a new missile-defense system. Even though many analysts were skeptical of Soviet capabilities, Helms intervened to ensure that published estimates were consistent with the administration’s public position. He also appeared before Congress alongside the Secretary of Defense as a show of support. As he explained later, “I was not prepared to stake the Agency’s entire position on this one issue… I was convinced we would have lost the argument with the Nixon administration, and that in the process the Agency would have been permanently damaged.” Unfortunately for Helms, his action did little to improve his position or the standing of the intelligence community, which was increasingly excluded from high-level discussions. Helms was unceremoniously fired in 1972, and the Board of National Estimates, which had once been the focal point for intelligence community analysis, was dissolved in 1973.
The case of DCI George Tenet is a more recent cautionary tale. Following the perceived intelligence failure that led to the September 11 attacks, Tenet worked hard to restore the Bush administration’s confidence in the intelligence community. At the outset this meant moving aggressively against Al Qaeda and the Taliban in Afghanistan, but Tenet also came under pressure to help the administration make the public case against Iraq in 2002-2003. Although he was well aware of analysts’ doubts about the Iraqi threat, he chose to declassify intelligence that was consistent with the president’s claim that Iraq was a “grave and gathering danger.” These efforts to mollify the administration proved to be futile. In the aftermath of the March 2003 invasion, when it became clear that Iraq possessed no real weapons of mass destruction, relations between the White House and the intelligence community broke down completely. Policymakers claimed that shoddy intelligence analyses were responsible for their false claims about Iraqi capabilities. Intelligence officials accused the administration of bullying them into exaggerating the threat. And Tenet himself accused administration officials of feeding stories to credulous reporters to shift the blame to the intelligence community. He resigned one year later.
What should we draw from these cases? For intelligence officials, the main lesson is to stop trying to curry favor by softening estimates. In the last decade the U.S. intelligence community has weathered repeated controversies, not because it has played politics but because it has performed well. If it detects Iranian cheating in the aftermath of the nuclear deal, it should say so bluntly, even if this means upsetting policymakers who invested so much in the effort. The long-term consequences of soft politicization far outweigh the short-term discomfort of being honest.
For policymakers, the lesson is to take intelligence out of the spotlight. Using intelligence to sell the Iran deal will reinforce the expectation that future assessments will also be public. This will create a temptation to pressure intelligence agencies to make sure their findings are consistent with administration statements, and intelligence officials might tailor their findings so they are inoffensive. The result will be mushy conclusions useful to no one. Intelligence on Iran is very solid today, and intelligence-policy relations are healthy. The best thing the administration can do to preserve this happy status quo is to remove secret intelligence from public view.
Joshua Rovner is the John Goodwin Tower Distinguished Chair in International Politics and National Security at Southern Methodist University.