During that August discussion in the back room of the sports bar in Pittsburgh, someone asked me, “Mike, in this Trump-intelligence thing, who drew first blood?” It was a great question. And like most great questions, it was tough to answer.
Intelligence is always challenged to establish a relationship with the chief executive. That’s really nobody’s fault. It’s far more structural than personal (although personalities do matter). The goal, of course, is for intelligence to get into the head of the president, to meaningfully contribute to his deliberations. But intelligence and the president come at this from different perspectives. Metaphorically, they enter the Oval Office through different doors.
The intelligence door is labeled “facts”—the kind of data that intelligence steals, elicits, or otherwise acquires to inform decision making. The president’s door—any president’s—is labeled “vision,” specifically the one that people voted for in the first place.
Intelligence is fixated on the world as it is. The president and his policy team dream of the world as they want it to be. Intelligence is inherently inductive, swimming in a sea of data and attempting to draw generalized conclusions. Policymakers are inherently deductive, trying to apply their first principles, the ones you voted for, to specific situations. And intelligence analysts trend pessimistic. It comes with the turf. Bob Gates, director of central intelligence well before he became secretary of defense, says that when a CIA analyst stops to smell the flowers, she looks around for the hearse. Policymakers have to be optimistic; otherwise they never would have pursued the job.
So there is always some relationship building when a new president comes on board. The intelligence folks have to get into the head of the policymaker without crossing a line and breaking the tether to their fact-based, inductive, pessimistic worldview. After all, that’s their only legitimate reason to be in the room. The president (whether he knows it or not) needs keen and impartial observers, not cheerleaders.
The intelligence community also needs to know the character of the first customer and especially how he learns. President Bush was a voracious reader, but it seemed to me that he best learned in the dialogue, which was always lively and intense. Present Obama appeared to be more reflective, preferring to read, and learning in the private moment.
There was going to be another adjustment in 2016 either way, although adjusting to a President Hillary Clinton would probably have been a light lift. She had gotten the President’s Daily Brief for four years as secretary of state. The briefers would have known her and been familiar with her worldview, and she in turn would have understood the intelligence baseline on which any day’s material was being presented. It’s not too far off to say that she might have introduced her first briefing by simply asking, “Now, where were we?”
A President Trump was a different matter, of course. Not only was he less familiar with government, the structure of intelligence, and the global situation, but the creator had also given him a few extra doses of those deductive, world-as-we-would-like-it-to-be, vision-based attributes described above. He seemed purely instinctive, spontaneous, even impulsive, and although he had little background on the substance or processes of international affairs, he also had little patience with written or even verbal presentations. He seemed to have an eerie confidence in his own a priori narrative of how the world worked.
He also seemed disinclined to learn more, even at first pushing back on the very concept of a daily intelligence briefing, saying that he was a very smart person and did not need to be told the same things over and over again every day, itself a hideous mischaracterization of the PDB.
Mr. Trump puts a great deal of stock in personal relationships. No surprise there. I suspect we all do. But he does so to an extraordinary degree, weighing loyalty to him as an absolutely essential virtue, and giving greater weight to a conclusion based on who is presenting it to him rather than any evidentiary trail supporting it.
And if all this wasn’t daunting enough, the first time American intelligence really had to try to force its way through Mr. Trump’s bubble of skepticism, they had to engage him on a matter that was being used by other Americans to challenge his very legitimacy as president of the United States: the Russian interference in the U.S. election. That was a tragedy for the IC and it created an almost perfect storm of hostility between a rejectionist president and an intelligence community that was there to tell him the truth.
So I didn’t have a good answer for my Pittsburgh friend’s question about who drew first blood. And maybe it just doesn’t matter. It all ended up with a lot of blood on the floor.
John McLaughlin was one of the best analysts ever to work at CIA. After heading up all analysis at the Agency and then serving as deputy director, John ended his thirty-plus-year career as the acting director of central intelligence following George Tenet’s departure in 2004.
John stays plugged in, and in his calm rational way, he says that Mr. Trump’s relationship with the intelligence community has evolved over four distinct phases. The first was characterized by the candidate’s and the president-elect’s ignorance of the community and its structure, its ethos, even its purpose. There was, for example, an inordinate focus on CIA as if it comprised all (or all that was important) in the IC.
As the transition wore on and the Russia interference story moved to center stage, the president-elect slipped into an attitude of overt hostility toward the community. Hence the tweet barrage comparing intelligence professionals to “Nazis” and such.
But intelligence is essential to good governance, so as inauguration day came and the needs of governing set in, the new president seemed to more and more appreciate that he had to rely on the intelligence community. Briefings became more frequent as the realities of North Korean ICBMs and Syrian chemical weapons became unavoidable.
The first three stages blended together, of course, and even today powerful elements of each (ignorance, hostility, unavoidability) seem to weirdly coexist in the president’s worldview. That evolution is an important and fascinating story. I will work to describe it in this chapter and the next.
Of course, John said there were four stages in the Trump–intelligence community story. Stage four, John predicts, will begin when Bob Mueller reports out on his investigation of potential collusion by the president’s campaign with Russians interfering in the U.S. election. No matter what Mueller recommends, the impact of that will be immense.
Presidential transitions display both the strengths and the weaknesses of American democracy. The strengths are obvious: every four or eight years we conduct a peaceful transfer of power for the most powerful office on the planet. The weaknesses largely deal with the effects of massive changes—changes in senior personnel certainly and, very often, in policy as well—and the resulting tensions.
The incoming team is focused on change. After all, they won and believe they have a mandate.
The government they will inherit is focused on stability. After all, they believe they know how things are best done.
That puts a great burden on the transition team—that body of folks loyal to the incoming administration, but wise and experienced in the ways of Washington—to build bridges to the outgoing administration as well as to those officials in the departments and agencies who aren’t going anywhere.
One veteran of the Trump transition drew a sketch on the back of a breakfast napkin to demonstrate to me the always large information gap between an incoming political/policy team and permanent institutions like the intelligence community. In normal circumstances, he said, there is enough trust (or at least a sufficient lack of hostility) for the two sides to get together, close the information gap, and in the process form the relationships needed to eventually assemble a national security team—always an amalgam of political/policy types and career professionals.
Barack Obama pretty much campaigned in 2008 as not being George Bush, but his team gave the Bush folks high marks for how they handed off the reins of government. And the incoming Obama team was populated by a legion of seasoned Washington hands who had been waiting eight years in the wilderness for this moment. It was a pretty smooth transition.
Not so much in 2016. During the Republican primaries, Florida governor Jeb Bush predicted that Donald Trump would be a chaos president.1 It was certainly a chaos transition.
First, there was good evidence that team Trump really didn’t expect to have to come to work the day after the election.
They certainly hadn’t prepared to govern. Developing policy is usually done before the primaries kick off in Iowa. There is some time then to think a bit, to put some purpose to the campaign, even to identify key intellectual influencers. None of that happened with the Trump campaign.
There is traditionally a second shot at this between the conventions and election day. There we normally see massive policy teams turning out papers on a variety of issues. That didn’t happen with the Trump campaign, either. Nada. Nothing.
More than any other candidate in the modern era, Donald Trump ran on attitude, not policy. One intelligence senior said the incoming team was as weakly anchored on facts and on the artifacts of governance as any he had ever seen. While quick to point out that he wasn’t talking about everybody, he added that he saw some of the wackiest stuff he had ever seen during any transition. Another senior observed that the new team was so ignorant of the institutions and processes of government, it seemed like “the Wild West.”
That meant that in November 2016 the demands of manning a government (the usual transition task) were compounded with the demands to decide with some specificity what the new government actually intended to do and how they intended to do it.
The chaos was aggravated by the de facto excommunication of 172 experienced foreign policy specialists who had signed those letters condemning Trump in March and August 2016, plus some 75 retired diplomats who had signed a similar letter in September describing Trump as “entirely unqualified to serve as President and Commander-in-Chief.”2 All that left a pretty thin bench to call on.
Then there was the chasm of distrust that team Trump brought with it to the transition. Trump had run, after all, on a promise to drain the Washington swamp, and it was the swamp dwellers who were putting together the briefing books and trying to schedule meetings and to identify their counterparts on the incoming team. One senior intelligence official told me that those closest to the candidate seemed to believe that “inherently governmental” structures were second-tier compared to the private sector. Hence the lack of interest in the structure and processes of the intelligence community.
Mike Flynn could have been a bridge, but the putative national security adviser seemed more and more alienated from the community to which he once belonged. His views on al-Qaeda and ISIS were increasingly radicalized, his words on the outgoing administration increasingly condemnatory, and his zeal to restructure the entire intelligence community, with a special emphasis on CIA, increasingly obvious.
At the policy level, the far right news and opinion website Breitbart seemed to enjoy an outsized influence. Its former executive chair, Steve Bannon, was now chief strategist and senior counselor, and he would later surprise no one when he promised to battle tirelessly for the “deconstruction of the administrative state.”
Closer to the intelligence community, Bannon acolyte Sebastian Gorka, who had been editor for national security affairs at Breitbart, met with Trump as early as 2015 and provided the soon-to-be candidate with a variety of position papers on Islam and terrorism.3 One intelligence insider confided to me that Gorka was viewed as weak on analysis, someone who had made his mark as a grenade thrower, purposefully provocative, committed to ideologically driven disruption and in the vanguard of the no-peace-with-Islam movement.
All of this had the air of a hostile corporate takeover.4 I have also heard it compared within the intelligence community to explorers landing in a suspicious and hostile environment, looking to make alliances with some tribes in order to subdue all the others. Resistance to the ways of the incoming team was quickly identified as evidence of the “deep state,” a phrase previously used to describe murky military and security power centers that secretly work to thwart the democratic will in countries like Turkey.
Trump supporters like Breitbart News and Fox News’ Sean Hannity made frequent references to it. Newt Gingrich, the former House Speaker, said, “Of course, the deep state exists. . . . They create a lie, spread a lie, fail to check the lie, and then deny that they were behind the lie.”5
The Russians were happy to push a version of this idea, too. Julia Ioffe, the Russian-born American journalist, told a crowd of security specialists at Aspen that Russia liked Trump and wanted to circumvent the mostly hostile or at least suspicious American national security establishment. She colorfully described the Russian view as Trump being Gulliver tied down by an army of American establishment Lilliputians.6
The issue created really odd bedfellows. Self-described advocacy journalist Glenn Greenwald joined in against “the neoliberal and neoconservative guardians” of the current consensus, “with their sprawling network of agencies, think tanks . . . and media outlets,” and then railed against “the unelected agenda of [National Security Adviser Lieutenant General H. R.] McMaster and [Homeland Security Secretary and later White House chief of staff Marine general John] Kelly.”7
I have worked in intelligence for over three decades. I know what antidemocratic forces look like. I have seen them in multiple foreign countries. There is no “deep state” in the American Republic. There is merely “the state,” or, as I characterize it, career professionals doing their best within the rule of law.
Not that they always play nice or quietly sit in their cubicles waiting to be called on. Many of the leaks that have plagued the Trump administration have come from internal White House factions vying for advantage, but there is no doubt that some have come from career professionals. My journalist friends admit to me in a generalized sort of way that a lot of folks are certainly more willing to talk to them.
That can be bad, but painting this as a dystopian government universe inhabited by secret malevolent forces is simply inaccurate. Trump routinely defines himself against caricatured enemies. Mexicans are murderers and rapists; intelligence professionals are Nazis; immigrants are deeply unfair; refugees are dangerous; and Muslims hate us. Now the organs of the government that he was about to inherit were the secret, antidemocratic, all-powerful, conspiratorial “deep state.” It is not a particularly useful description, frankly one that enhanced rather than dampened internal opposition, and one especially ill-suited to effective governance.
On several occasions, when I have made such remarks, folks recommended that I read Michael J. Glennon. Glennon laid out a kinder, gentler “deep state” theory in a 2014 article, “National Security and Double Government,” explicitly dismissing secret collusion in some dark plot, but nonetheless expressing deep concerns about the antidemocratic effects of America’s national security establishment.
Glennon bases his theory on the writings of nineteenth-century British journalist Walter Bagehot, widely credited as editor of the Economist with pushing that magazine to political prominence. Glennon cites Bagehot’s reflections on British government where he describes dual institutions, one public and the other concealed, evolving side by side to maximize both legitimacy and efficiency. “Dignified” institutions like the monarchy and the House of Lords gave legitimacy to “efficient” institutions like the House of Commons and the cabinet that actually did the real work of government.8
Glennon transfers this duality to the modern American government. He labels our “dignified” institutions as Madisonian—the courts, Congress, and the executive that draw their powers from constitutional processes. These bodies are dignified and public, fulfilling necessary forms, endorsing decisions, and granting authority and legitimacy to governing.
But in our foreign and security policy, he says, we are more and more actually governed by the “Trumanite” entities that comprise our national security structure. Trumanite, of course, refers to the internationalist consensus that has governed American policy since 1947. Glennon sees these bodies as opaque but efficient, making decisions anchored in expertise and focusing on substance.
Glennon believes that “the Trumanite network survives by living in the Madisonian institutions’ glow,” but he also sees effective Madisonian control as fading, largely a product of the growing complexity of security issues as well as an increasingly ill-informed electorate and political class.9 He supports his argument by demonstrating the continuity in counterterrorism, surveillance, and other policies between presidents as different as George Bush and Barack Obama and the strong carryover of personnel from one administration to the next. The Trumanites still determine things even as the Madisonians swap out, he reasons.
Of course, the apparent continuity could simply reflect limited options in the face of hard international realities, and personnel staying in place the admirable depoliticization of key jobs like chairman of the Joint Chiefs of Staff or director of the FBI.
Glennon also overachieves in his description of Trumanite control and initiative. His account of expanding electronic surveillance after 9/11, for example, is inaccurate in what was done and in who first pushed it.10
Still, Glennon raises serious questions, does not feel compelled to create villains or ill intent where none exists, and makes an effort to prove his case. This sets him apart from the reflexive conspiracy-mongering crowd that identifies opposition to any Trump initiative as proof of malevolent intent by a unified cabal committed to overturning the results of the election.
There is no doubt that large bureaucracies are set in their ways, and I can aver from personal experience that it is hard to get them to change course. But the “deep state” calumny is neither accurate nor useful. And all it did was to harden the lines that presidential transitions are designed to soften and eventually merge.
To make matters worse, the pure mechanics of this transition were awful.
Take communication. The transition would happen in Washington, but decision making was still tightly held at Trump Tower in New York. Even after a formal intelligence transition team was stood up in Washington, it was the view of the intelligence community that New York was still a powerful, controlling, distant, and not particularly communicative lord.
A lot of intelligence veterans were encouraged that former congressman Mike Rogers had the national security portfolio under New Jersey governor Chris Christie’s overall transition effort. Rogers had been head of the House Intelligence Committee, was a known and trusted quantity, and got high marks for his knowledgeable and bipartisan leadership of the committee with Maryland Democrat Dutch Ruppersberger.
I cornered Rogers a week after the election during a party he was hosting at his house for America Abroad Media, an organization that promotes American values in popular culture overseas and that night was hosting the cast and writers of Homeland, Showtime’s hit CIA drama. I began to tell Rogers how much we in the community were counting on him as a bridge and a voice of reason when he interrupted to tell me that he was no longer with the campaign.
That wasn’t very comforting. Rogers didn’t volunteer a reason, but he had been attached to Christie, who himself was being shoved off the island, erasing what little preparation for governing that the Trump team had done. Rogers wasn’t really part of any of the evolving administration power centers, and Republican hard-liners (who were in the ascendency at that moment) had never forgiven him for what the intelligence community had viewed as his committee’s reasonable report on Benghazi.
Retired Army lieutenant general Ron Burgess, Mike Flynn’s predecessor at the Defense Intelligence Agency, had been quietly assembling an intelligence transition team tucked under Rogers’s larger national security effort. Burgess had retired to a comfortable position at his alma mater, Auburn, but had been asked by his friend Alabama senator Jeff Sessions (Trump establishment Republican supporter number 00001) to lend his effort to the transition. More to prevent harm than to advance any agenda, Ron agreed.
Burgess put together an impressive team of (non-letter-signing) IC veterans, bided his time until the election, and then orchestrated a twenty-page paper on the things that the incoming administration should look at in the intelligence community. Ron told me that it was what any of us would have told any incoming administration: what was going well; where challenges remained; even prosaic questions like the right number of senior intelligence executives.
With Mike Rogers out at the top, the buzz at Washington transition headquarters was that Burgess could step up to head the national security portfolio, something that he did not seek and made clear that he would not accept. No matter. Within a few days Mike Flynn descended on the Washington office from the New York campaign headquarters, set up his own team and meeting schedule, and effectively froze Burgess out of the process. Burgess, already wary of Flynn, quietly informed Sessions and Vice President Elect Pence that he would be leaving and returned home to Alabama, unsure if anyone had actually read the paper his team had prepared.
All transitions tend chaotic. Two weeks after the election, this one was chaos on steroids.
On Monday of Thanksgiving week, for example, the president-elect appeared in a video announcing his day one plans in office. The video was clearly designed to sustain some political momentum and to fill the news void over the long holiday weekend. As part of his list, Trump promised to rip up the Trans-Pacific Partnership trade agreement, which he later did. The president-elect also said that he would direct the Defense Department and the chairman of the Joint Chiefs of Staff “to develop a comprehensive plan to protect America’s vital infrastructure from cyber attacks and all other form of attacks.”
No one currently or formerly in government had any idea what this meant. It bypassed the Department of Homeland Security’s statutory responsibility to protect critical infrastructure and put America’s armed forces in an unusual and likely illegal domestic defense role. A transition senior later told me that somebody just said that “we gotta say cyber and it’s gotta sound tough.” The team went into what he described as campaign mode and drafted the statement based on their gut. Thankfully the idea was never heard of again. I suspect that the Department of Defense and the Joint Staff made sure that it died a quick and inglorious death.
There was one important and notable exception to the transition chaos: the nomination of Mike Pompeo to be the director of CIA and the process that secured Pompeo’s confirmation from the Senate. This was especially impressive since three days after the president-elect deconstructed his transition team, with nothing in place to replace it, he announced that the Kansas Tea Party Republican would be his man at Langley.
One senses the hand of Vice President Elect Pence in this surprisingly early pick. Pence had had confidence in the now departed Ron Burgess (former DIA chief) and had actually urged him to stay on despite the reshuffling of the security team. Now he seemed uncomfortable with Flynn’s sudden ascendency, a condition that did not improve with time. Indeed, within three months Flynn would be fired from his national security adviser post for lying to Pence.
Pompeo could act as a counterweight, or at least deny the CIA post to a Flynn acolyte, a choice that would almost certainly have turned the Agency and much of the IC into a war zone. Pence knew Pompeo from their together time in the House, and Pence, who once said that he was “a Christian, a conservative, and a Republican, in that order,”11 found in Pompeo a kindred spirit who also fit that description.
As a member of the House Intelligence Committee, Pompeo had taken an uncompromising and politically high-profile position condemning Hillary Clinton and the Obama administration for the 2012 terrorist attack in Benghazi. It was a bone he had locked his jaws around and would not let go of. He was also reliably and publicly hawkish on hot-button issues like surveillance, detentions, and interrogations.
But while he was filling the public political role of a tough Tea Party Republican, Pompeo earned high marks from intelligence professionals for his serious study of their work. He visited multiple stations, asked good questions, and stayed late.
Shortly after his nomination had been announced, Pompeo began contacting his predecessors in the CIA post. I met with him privately in his congressional office for over an hour and was impressed with his genuine enthusiasm for the job. He seemed to understand the challenges of the task, but he wasn’t intimidated by them either.
Since the director-designate was a West Point graduate and had spent six years on active duty as a cavalry officer, I asked him if his wife intended to be active at the Agency. Although it is never required (at least not in the modern age), military spouses often sacrifice a great deal of their time and energy on behalf of the command. Once he confirmed that she did indeed want to be involved at the Agency, I offered that my wife had also played that role and volunteered that she would be happy to talk with her about it. I was impressed that Susan Pompeo was in my kitchen that afternoon for an extended conversation with my wife, Jeanine.
Pompeo’s confirmation team was led by two Bush administration veterans, Ben Powell, former associate White House counsel and general counsel for the director of national intelligence, and Juan Zarate, former deputy national security adviser for combating terrorism and first-ever assistant secretary of the treasury for terrorist financing and financial crimes. Neither man could be viewed as a Trump partisan, but neither had signed any “never Trump” letters; neither had ambitions to serve in the Trump administration. Their team was focused, pragmatic, and largely autonomous within the larger transition kerfuffle. They also intentionally held back, letting the Agency tell its own story to the DCIA-designate, in effect letting the Agency adopt him as their own.
They coached the nominee smoothly through the Senate confirmation process. Even though candidate Trump had talked about bringing back waterboarding and more and Pompeo as a congressman had advocated tougher interrogations, at his hearing he wisely noted current law, said he would abide by it, and promised to come back to Congress for changes if he felt that his current authorities were insufficient to keep America safe. He even committed to disobeying any order to resume waterboarding. This was very different from other aspects of the transition. Simple. Nonideological. Pragmatic. Successful.
The Pompeo confirmation team made some other important contributions. They counseled Pompeo to get out of the car alone when he arrived at Langley. In other words, don’t bring congressional staff with you. Depend on the team you are inheriting. I don’t think Pompeo needed much coaching on that. He had already developed a healthy respect for the Agency.
Powell, Zarate, and team also insisted that Pompeo pick a deputy from within the organization, almost a given since Pompeo had no experience at Langley. They then recommended a specific Agency veteran, Gina Haspel. Gina was an inspired choice. The Agency exhaled when Pompeo was nominated. When Gina was announced, they exulted. I was certainly heartened when informally approached by the transition team for my views on the choice.
The Agency’s bland press release noted Gina’s thirty years of service, work as chief of station, and a stint working counterterrorism. Actually, her duties had taken her into some of the most controversial programs of the Agency’s recent past. Her choice would provoke some on the outside, which is why the press release was accompanied by enthusiastic endorsements from five Agency veterans (including me), a DNI (Jim Clapper), and a member of Congress (former House intelligence chair Mike Rogers).
On a larger message, Gina’s choice was pitch perfect: CIA intended to neither repeat nor repudiate its past.
With Pompeo and Haspel at the top, the Agency seemed in good hands. On this question, at least, many at Langley believed they had dodged a bullet.
Vice President Pence also had a hand in the selection of fellow Indianan and former ambassador to Germany Senator Dan Coats to be the new director of national intelligence. Coats was viewed by the community as a solid choice: sober, experienced, reluctant—the last viewed as an important virtue since there were rumors swirling that the Trump team was enlisting the billionaire head of Cerberus Capital Management (and friend of Steve Bannon), Steve Feinberg, to redo the intelligence community and the office of the DNI. Many in the IC suspected that was an overt attempt to impose more administration loyalty on the community.
I told the New York Times at the time that I was having trouble wrapping my head around “the idea of a D.N.I. nominee in the confirmation process while others consider retooling the position.”12 I suspect that Senator Coats thought so, too, and there were press accounts that he pushed back hard against anything like this being done except under his auspices . . . and only once he had been confirmed.13 He seems to have carried the day. If a Feinberg review was ever the plan, the administration quietly backed away from it.
Coats’s nomination came nearly two months after Pompeo’s name had been announced, even though the law says that the DNI should nominate the head of CIA to the president. The timing of the nominations and confirmations gave Mike Pompeo a leg up on establishing Oval Office relationships. It also likely reflected the president’s and his team’s focus on CIA and perhaps a bias on the part of the incoming team toward action (which only CIA was authorized to do) versus intelligence advice (which was the statutory responsibility of the DNI). The timing, the relevant personalities, and the statutory limits on his office promised significant challenges for the incoming DNI.
More broadly in the overall intelligence community transition, retired admiral Paul Becker had come on board the transition team when Flynn descended from New York. Becker was a Flynn associate, but also a well-regarded intelligence veteran in his own right and an island of relative calm in all of this. He worked hard to limit damage and reduce the impression that this was a hostile takeover. His team was the outward face of the transition to the IC, and they dutifully took their formal informational briefings from all the big agencies.
But these were largely “listen and learn sessions,” unconnected to the all-powerful New York transition office and whatever decisions were being made there. They weren’t really preparation sessions for an incoming team, and each meeting seemed to have its own logic rather than being part of a whole. As one senior intelligence official said to me, “We waited for the knock on the door, but it never came.” One despondent official described it as “all small ball.”
Absent too were the substantive briefings, usually rather extensive deep dives, given to the incoming team on key topics like Iran or Syria or North Korea. There were two dozen such formal sessions in 2008, hardly any in 2016. There wasn’t much interest, and there weren’t many named incoming officials either.
One exception was the new secretary of defense, Marine general Jim Mattis, always an avid consumer of intelligence, who was described as gracious and appreciative throughout. But overall, the closing of that information gap between the outgoing and incoming team never really happened and so the relationship of trust between the new governors and their future staffs was never built.
The day after the inaugural, the same day as the infamous Trump speech in front of the wall honoring the fallen at CIA, a DNI officer entered the Situation Room to preview what one official described to me as “IC 101.” Never really satisfied with their day-to-day linkages with the now departed Obama team, intelligence seniors wanted the new guys to understand the nuts and bolts of the intelligence community and its support for the incoming national security team.
The briefer was greeted by Ezra Cohen-Watnick, the thirty-year-old newly appointed NSC senior director for intelligence programs, whose qualifications for the job appeared to be his devotion to his former boss at the Defense Intelligence Agency, now national security adviser, Mike Flynn. Cohen-Watnick’s position had traditionally been filled by a senior CIA officer or, in one case, a future CIA director (George Tenet), so his appointment was an unambiguous slap at the Agency. On this day, Cohen-Watnick did not disappoint, filling the space with what was described to me as a rant against CIA, and promising, among other things, to take covert action authority away from the Agency.
Hostility at a personal level was a bad sign. Things would be harder than they had to be. More important, at the institutional level, the transition was truly opportunity lost. Concerns were high. Confusion persisted. Building teamwork was deferred.
And then there was that deteriorating personal relationship with the new president.
John Helgerson, CIA’s inspector general during my time at the Agency, has written the definitive history on presidential transitions. In it, he concludes that although all transitions have their own character, “overall it has proved easier to help the new president become well informed than to establish an enduring relationship” with him. There have always been these two tasks: informing the president-elect and forming a positive relationship with him. This transition was tough on both counts.14
Donald Trump had been a successful businessman; he clearly understood the ground rules that enabled success in real estate. Since it couldn’t have all been about bluster, bluff, bullying, energy, and ambition, he must have had to rely on some form of “intelligence.” But this was a new world for him and no one should have expected Mr. Trump to have a solid base of knowledge about international affairs and security questions. These issues were not really part of his life experience as a real estate developer, and he frankly showed little interest in substance even as he campaigned. When asked in the summer of 2016 if he read much, Trump replied, “I never have. I’m always busy doing a lot.”15
He was stumped by the meaning of the word “triad” when asked about nuclear forces by Hugh Hewitt during a debate. Earlier, in a phone-in interview with the same conservative talk show host, he admitted to not knowing the names Nasrallah, Zawahiri, or Baghdadi, the leaders of Hezbollah, al-Qaeda, and ISIS, nor knowing the distinctions between their murderous organizations. He later accused Hewitt of throwing a bunch of “Arab” names at him. Unfamiliar with the Quds Force under Iran’s ruthless general Qassem Suleimani, he confused it with America’s friends in the region, the Kurds.
All of which presented a problem for intelligence. Where do you begin if you want to “well inform” the incoming president? Intelligence briefings can be a bit like teaching or journalism, a journey from what the client knows to what he does not yet know. Where should that journey begin with a President-Elect Trump? Probably a little too basic to launch with, “Let’s start with Iraq today, Mr. President-Elect. It’s the one on the left.” But what should you assume he knows about the Sunni-Shia split and the struggle over succession to the Prophet in 632, which actually is not a trivial element in understanding today’s Middle East?
And what constitutes success? Since this isn’t a client who favors objective analysis, history, or explanation, the bar may be very different here: can you use the venue to get one level deeper with him, to get him to do better and reflect more than he otherwise would?
Although it’s not an intelligence matter, I often wonder what happened after that session with Piers Morgan where the president created in his own mind a global ice pack expanding at record levels. He had been president a year and would have been accompanied by the usual entourage of handlers and advisers. Did anyone approach him afterward to tell him of his mistake? If someone did, how did he respond? Did he welcome it? Did he even care? Did the new data matter?
Director of National Intelligence Coats suggested some broad success in this regard when he said at the 2017 Aspen Security Conference that the president has ended some sessions with, “I got it. It’s not what I thought. It’s not what I heard. But I got it.” That is essentially the challenge, since intelligence is less about predicting than it is about understanding, about providing the context within which wise policy can be made. As one PDB veteran told me about President Trump, “How do you connect for him things that he does not connect himself?” And that comes less out of any personal ignorance about a particular topic (although that is important) and more from the personality of the first client, whom former White House speechwriter Michael Gerson once described as living in the “eternal now,” a personal world without history or consequence—the antithesis of the intelligence task of delivering context.16
In his PDB sessions, Mr. Trump fit the mold of a populist leader that A. C. Grayling describes as someone who favors “simple slogans . . . instead of attention span, the immediate instead of the long term, the local and obvious instead of the larger picture and all . . . in the form of attitudes rather than worked-out ideas.”17
Briefing sessions for the president-elect were described to me as simple, visual, and chaotic: a few sentences from the briefer, a question, then a tangent with frequent interventions about intelligence, policy, and current press all competing for attention. One familiar with the sessions told me that a literal transcript would read like a James Joyce novel: a series of thoughts that appear as they come to mind.18
Early sessions often went long, in one sense a measure of the president’s interest, in another sense a reflection of the difficulty of staying on topic. It was not unheard of for the president to tweet during sessions (probably not about the intelligence content) or to get unhelpful nods of agreements from some in the room when he made controversial comments (like “We should have kept the oil in Iraq!”).
A fairly common question that intelligence people ask themselves after an important briefing session is: did they adequately explain and clearly distinguish what they knew, what they thought, and what they did not know? To that was now added, “Do you think he got that?” It’s not that we have never asked that question before, it’s just now more routine.
The incoming president was impatient with detail and demonstrably prone to action, so there was the added burden of not creating circumstances where decisions were made before adequate data was available, before complete context had been discovered and presented.
That can be challenging. There were reports that National Security Adviser H. R. McMaster irritated the president because he lectured Trump and didn’t allow him to ask questions.19 But how far can you tailor a message before you can be fairly charged with “leading the witness”? Clearly you always have to “know” your client, but how much should you have to cater to a unique personality? At what point does packaging, simplifying, or shortening become self-defeating? Or just dishonest?
Then there was the larger question of whether objective reality, even well presented, was decisive. The veteran PDB briefer cited above agreed with my suggestion that the president-elect has a verbal tic—phrases like “people are telling me,” “people are saying,” “many people are saying”—that allows whatever statement that follows to function as if it were therefore true for him.
Later, during the administration, CBS News’ John Dickerson was interviewing the president and pressing him about his charge that Barack Obama had wiretapped Trump Tower in New York. He asked him if he stood by the charge.
“I don’t stand by anything,” Trump replied. “I just—you can take it the way you want. I think our side’s been proven very strongly. And everybody’s talking about it.”20
We saw the same thing in an interview with ABC News about alleged voter fraud. Anchor David Muir asked the president, “Do you think that talking about millions of illegal votes is dangerous to this country, without presenting the evidence?”
“No, not at all, because many people feel the same way that I do. . . . Millions of people agree with me when I say that.”21
Many people feel the same way I do. Everybody’s talking about it. QED. Thus it has been demonstrated. You have your proof. Whether or not it actually happened, it is still demonstrably true.
The veteran briefer ruminated with me over whether or not the president made a distinction in his own mind between true and untrue. He raised the controversial speech the president had given in the summer of 2017 to a Boy Scout jamboree in Virginia. The speech was overly political for an audience like the Boy Scouts and occasionally tasteless to boot.
In the face of some pretty sharp criticism, the president said that Scout leadership had called him later to say that it was “the greatest speech that was ever made to them.”22 Of course, no such call ever occurred. The former briefer then asked me if I thought that the president knew that, if I thought that his mind actually made the distinction between the past that actually happened and the past that he needed to claim at that moment. His point was that you could sometimes convince a liar that he was wrong. But what do you do with someone who does not distinguish between truth and untruth? We have had obstinate first customers in the past. We have had argumentative customers. He cited President Nixon as someone who disparaged the Agency’s views, and then added that analysts often welcomed such challenges. But we never had a first customer for whom ground truth really didn’t matter or who simply might not embrace the concept that there is objective truth.
One junior observer was more direct: she said they were going to have to rechisel that quote from Saint John in the Agency lobby: And ye shall know the alternative facts, and they shall make you free.
It will take an insider’s true memoir for us to learn the actual effect that intelligence had on the president-elect. He quickly discarded his brief foray into abandoning the one-China policy that has dominated American diplomacy since Nixon, for example, but it’s unclear that intelligence had any role in that recanting. It’s also unclear that he even understood the policy he had challenged or that he had actually challenged it.
If intelligence raised the downsides of cutting loose from the Trans-Pacific Partnership—a strategic arrangement masquerading as a trade deal—it certainly didn’t stick. The United States withdrew from the treaty four days into the administration, in my eyes one of the most destructive steps taken by the administration so far. It was also the most stunning example of a red-meat, emotional campaign promise being enacted without considering the broader context and seemingly ignorant of or indifferent to the implications of that step, not just for American prosperity, but also for American security. During the campaign, it appears that Mr. Trump actually believed that TPP enabled China, rather than isolating it: “It’s a deal that was designed for China to come in, as they always do, through the back door and totally take advantage of everyone.”23 There is no record that intelligence or any other part of the government, for that matter, was consulted before this dramatic decision was made.
There was also a lot of tough talk during the transition by the president-elect that we were now going to settle things in the Middle East by defeating ISIS quickly and ruthlessly. One wonders if any analyst ever tried to respond with, “Yes sir, and we’ll do that, but you need to know that we believe strongly that you can’t kill your way out of this mess,” or was he or she too cowed, intimidated, or polite to say so?
So I have my doubts about how “well informed” the president-elect was by inauguration day.
There is no doubt, however, about where the intelligence community stood on the second task that Helgerson had laid out: establishing “an enduring relationship” with him. That one was in the toilet.
Director of National Intelligence Jim Clapper and Secretary of Homeland Security Jeh Johnson had announced in early October that Russia was interfering in the U.S. election, and doing so under the direction of the senior leadership of the Russian Federation. FBI director Jim Comey had been scheduled to be there for the release of the report, but opted out because of his ongoing investigation. He feared the optic.
Clapper’s and Johnson’s short announcement was a bombshell.
For thirty minutes.
Then the Washington Post published a video of Trump speaking in explicit and vulgar terms about groping and kissing women, bragging about what sounded like sexual assault.
Thirty minutes after that, WikiLeaks seemed to toss a lifeline to the Trump campaign by dumping emails hacked from Clinton campaign chairman John Podesta. Eventually, twenty thousand pages of emails were dribbled out, backfooting the Clinton campaign on a variety of issues through election day. Trump actually referenced WikiLeaks 164 times during the last month of the election.
With two major scandals already burying what they viewed as a major intelligence discovery, Clapper and Johnson went back to the relative obscurity of the intelligence community where President Obama had directed that a detailed ICA (Intelligence Community Assessment) be prepared on the Russian interference. The rationale here was simple: they wanted a single agreed-upon document for the next administration and Congress to refer to. The assumption was that it would be a Clinton administration, so the work took on added urgency (and drama) with Donald Trump’s election victory.
It is hard to overestimate the importance of DNI Jim Clapper to this effort. To be sure, CIA director John Brennan had become equally alarmed in the summer when it became clear that this was not like previous Russian efforts. But it was Clapper as the head of the community who made things happen. One senior told me that they would not have gotten the Russian report out without the director of national intelligence marshaling resources and making sure that everyone had full access to arguments, sources, and methods.
And everyone in this context means the Federal Bureau of Investigation (counterintelligence), the Central Intelligence Agency (human intelligence and analysis), the National Security Agency (intercepted communications), and the DNI (overall leadership).
Later President Trump would downplay the report, saying, “Let me just start off by saying I heard it was 17 agencies, I said, ‘Boy, that’s a lot.’ Do we even have that many intelligence agencies, right? Let’s check it. And we did some very heavy research. It turned out to be three or four—it wasn’t 17.”24
There is so much misleading in that verbal smokescreen. The report bore the title “ICA—Intelligence Community Assessment.” It was prepared by the three relevant agencies listed above, but it represented the community’s view (all seventeen agencies), and if the Drug Enforcement Agency or the intelligence arm at the Department of Energy or anyone else had an objection or had something to contribute, they would have been included in the authorship as well.
In early December the Washington Post began reporting that the draft ICA was concluding that the Russians had intervened in the election to try to help Trump win the presidency, not “just” to undermine Americans’ confidence in their electoral systems.
Trump predictably responded via Twitter and (equally predictably) tried to delegitimize the report’s authors rather than argue its content: “These are the same people that said Saddam Hussein had weapons of mass destruction. The election ended a long time ago in one of the biggest Electoral College victories in history. It’s now time to move on and ‘Make America Great Again.’”25
Two days later on Fox News Sunday, the president-elect continued, “They have no idea if it’s Russia or China or somebody. It could be somebody sitting in a bed someplace. I mean, they have no idea.”
This would all have to be hammered out at a meeting with the president-elect and his team that was scheduled for Friday, January 6, at Trump Tower. Even that became Twitter fuel; the president-elect tweeted on Tuesday of that week, “The ‘Intelligence’ briefing on so-called ‘Russian hacking’ was delayed until Friday, perhaps more time needed to build a case. Very strange!”
Since the briefing had always been scheduled for Friday, this sounded a little like pregame taunting to get into the head of an opponent. Which is not how the intelligence leadership viewed itself.
The president-elect was back at it the next day with some more digital trash talk, this time preemptively undercutting what he was likely to hear: “Julian Assange said ‘a 14 year old could have hacked Podesta’—why was DNC so careless? Also said Russians did not give him the info!”
To underscore the obvious: this is not normal. American intelligence had never before seen such behavior in an important client, and certainly never in a president. January 6 was going to be a big day.
That day began for the intelligence leadership (minus Comey, who was already in New York) with a briefing in Washington to the congressional Gang of Eight: the leadership of the two intelligence committees and the senior members of both chambers. The members were told that the intelligence community had high confidence that the Russians, operating under the direction of Vladimir Putin, had hacked several email accounts in the United States, including those of the Democratic National Committee, and had weaponized that information by pushing it back through platforms like DCLeaks and WikiLeaks into the American information space to sow confusion here. The effort was aided by overt Russian information platforms like RT and Sputnik as well as by a covert army of trolls who treated the data in such a way as to make it appear trending to the algorithms of a variety of search engines. (We later learned that the Russians had a sophisticated use of social media sites like Facebook and Twitter in their repertoire, too.)
This covert influence campaign appeared to be designed, at first, simply to mess with our heads, but as time went on and more information became available, it became clear that Putin wanted to punish Hillary Clinton, whom he hated, and weaken the legitimacy of what he expected ultimately to be her presidency. Eventually, however, there was more and more evidence that the Russian campaign was working to push votes in the direction of Donald Trump.
The report made no effort to judge the effect the Russian effort had on the election and specifically said that it had found no evidence of vote tampering—which is a far cry from the later claim by some Trump supporters and some members of his administration that the IC had said that the Russians did not affect the election.
The congressional Democrats in the room for the briefing seemed happy enough with the presentation. Senate intelligence chair Richard Burr asked some good questions. Speaker of the House Paul Ryan was largely quiet, but attentive. House intelligence chair Devin Nunes and Senate majority leader Mitch McConnell seemed less pleased, but said nothing.
The intelligence team then headed for Trump Tower via the airport at Newark, where they were met by a large police escort that guaranteed they would not arrive in Manhattan unobserved. They admit that they were fairly tense as they had no idea how the president-elect would respond to their message.
At Trump Tower, Jim Comey (FBI), Jim Clapper (DNI), Mike Rogers (NSA), and John Brennan (CIA) filed in and took their seats at a small conference table. Also in the room back-benching from the president-elect’s team were Mike Flynn and K. T. McFarland (national security adviser and deputy), Tom Bossert (homeland security adviser), Sean Spicer (communications), Reince Priebus (chief of staff), Mike Pompeo (CIA director designate), and Vice President Elect Pence.
For about an hour the intelligence officials laid out their case. The briefing went well. The president was affable and sociable. He and his team were in a listening mood. The intel folks had to beat back only one tangent, a short discussion about how the Republican National Committee had protected its data. Given all the buildup, the actual session felt a little anticlimactic.
The intelligence directors also explained a bit of why they were so confident in their conclusions, each of them relying on their own collection discipline to make the case. The identity of the hackers, the tools employed, the infrastructure used, and the targets selected all pointed to Moscow. To that was added technical and human intelligence that went “behind the screen” (so to speak) to give additional evidence for the high-confidence findings. And then there was their deep knowledge of how things worked in Russia, knowing the kinds of things that can only happen with Putin’s approval, for example.
There were some limits on how detailed the sourcing could go during the briefing. This was actually a fairly large group and the Russians were already in max counterintelligence mode, launching their CI apparatus in an effort to discover how we knew so much.
The president-elect disparaged HUMINT a bit, saying it relied on what he called “sellouts,” but overall there was no real pushback.
The IC script then called for Jim Comey, the FBI director, to stay behind for one more item. Priebus asked the president-elect if he wanted anyone else to stay, but Trump declined.
Comey was there to brief Trump on the infamous “Steele dossier,” a compendium of opposition research that Chris Steele, a former British MI6 case officer, had compiled on Trump’s alleged ties with Russia. There were no precedents to guide the leadership on this; it was just one more “off the chart” episode in the 2016 transition. Jim Clapper later said that he believed it rightly fell under the rubric of a “responsibility to warn,” and they felt that the best, low-key way to share this with the president-elect was a Comey stay-behind.
I did not know Steele, but he was described to me as a competent and honest officer. Still, I was cautioned to be careful and not to automatically overrate him or his report. I was also reminded that writing a commercial intelligence document for a commercial client was different from intelligence service reporting. The client was usually less knowledgeable, less aware of the importance of vetting sources, and perhaps more needful of a “useful” document. When I finally read Steele’s report it did have the feel of a genuine intelligence product; the language and syntax and pace were all very familiar. “This is actually how we write this stuff,” I reflected. The dossier runs the gamut from the obviously true (Russia was trying to influence the American political process), to the possibly true (suspicious contacts between members of team Trump and team Putin), to the hard-to-believe-it’s-true (lurid sexual encounters in a Moscow hotel).
But far from being definitive, it was most useful as an indicator of things that needed to be further investigated. If this had actually been an American intelligence product, it would have been boldly labeled: THIS IS RAW, UNEVALUATED INFORMATION. THIS IS NOT A FINISHED INTELLIGENCE PRODUCT. We then would have gone through it line by line asking, for each assertion: Who is the source? Would the source logically have had access to this kind of information? Has this source reported reliably in the past? Do we have other information that would tend to confirm or deny this particular tidbit?
Comey, of course, didn’t address any of that. The Steele dossier had not been used for the Intelligence Community Assessment on Russian interference. The community had no views on its findings. The community just wanted the president-elect to know that it was out there.
The Trump team’s statement later that day avoided any comment on the Steele dossier, of course. The statement actually began with dutiful praise for the intelligence community, then cataloged the ongoing and multifaceted cyber threat to the nation (which was true enough, but was not the focus of the briefing), and then claimed—without any reference to the sustained Russian efforts—that “there was absolutely no effect on the outcome of the election.” That last part . . . that was actually so misleading and so untethered to the briefing that Trump had just received that it warranted a description I learned in the third grade. That was a lie. Without publicly challenging the intelligence community findings, team Trump attempted to mischaracterize and then bury them.
The president has never really accepted them. Most egregiously, in a foreign country, standing next to a foreign leader, he said in midsummer 2017, “I think it could very well have been Russia but I think it could very well have been other countries, and I won’t be specific. But I think a lot of people interfere.”26 He later tweeted in September 2017 that the “Russian hoax continues.”27
So the president never accepted the findings of his intelligence community that January afternoon. Even in the face of arguments that Russian actions are not just past but present, he remains in a different place, neither inspiring nor directing his government to decisive action. There has rarely been as much distance between an urgent intelligence community judgment and a government’s response.
The Steele dossier, by the way, didn’t stay secret very long. When the Internet media site BuzzFeed published the dossier in its entirety four days after the Trump Tower meeting, the president-elect immediately tweeted, “Intelligence agencies should never have allowed this fake news to ‘leak’ into the public. One last shot at me. Are we living in Nazi Germany?”28
Actually, the dossier was more or less common knowledge in D.C. circles throughout the summer and fall of 2016. I had multiple newsmen buy me breakfast in the hope I might know something about the mysterious report. (I didn’t.) It is a matter of public record that Senator John McCain had the dossier, which he forwarded to the FBI in early December 2016.29
So Trump was wrong to implicate the intelligence community in this one, and everyone in the IC, especially Jim Clapper, was incensed at the Nazi reference. As Clapper put it to me, the Nazi tweet made him see red, and so he put a call in to Trump. Jim was surprised that he took it. Jim told me that he tried to appeal to Trump’s higher instincts, reminding the president-elect that he was inheriting a national treasure that intended to support and enlighten him. You should treat it accordingly, Jim advised.
Mr. Trump listened attentively and then asked the director of national intelligence to publicly rebut the Steele dossier. Clapper simply said that was not his job and thanked the president-elect for his time.
Donald Trump would be president in about a week, and wise heads in the transition office like Ben Powell (who had honchoed Pompeo’s confirmation) urged him to go to CIA, meet the workforce, and reboot his relationship with the IC.
So the president went to Langley the day after the inauguration.
And conducted the worst presidential visit to an intelligence agency in the history of the American Republic.
The original concept had been that this would be a swearing-in event for the new CIA director, Mike Pompeo, but the Senate had not yet acted to confirm him. Still, the president had a productive meeting with senior Agency leadership and then went to the concourse armed with talking points designed to calm the waters between him and the intelligence community.
That never happened. Perhaps sensing the cameras that were broadcasting the event live, the president shifted into full political riff mode. In a disjointed, nearly stream-of-consciousness speech in front of the wall of 117 stars honoring the Agency’s fallen, the president never quite managed to recognize their sacrifice while finding time to resurrect his bromide about keeping Iraq’s oil and dragging Director-Designate Pompeo into the discussion: “But if you think about it, Mike, if we kept the oil you probably wouldn’t have ISIS . . .”
He also claimed that “almost everybody in this room voted for me,” that he had caught the media in a beauty of a lie about inaugural crowd size, and that “they’re going to pay a big price,” and then announced that he had been on the cover of Time magazine more than anyone, even Tom Brady.
The president also announced that the media “are among the most dishonest human beings on earth. And they sort of made it sound like I had a feud with the intelligence community. And I just want to let you know, the reason you’re the number one stop is exactly the opposite—exactly.”
That to an institution of fact-based analysts.
To be fair, there was some whooping and hollering in the CIA concourse as the president spoke. CIA reflects America, and no doubt there were Trump voters in the Agency. And this was a self-selected crowd. An Agency-all email had gone out the day before asking people to come in on a Saturday to be part of the event.
But CIA seniors sat stone-faced in the front row. At the Farm, the Agency’s clandestine training site that had the ceremony piped in via direct video link, many officers simply walked out. It’s hard to imagine a deeper hole from which to begin an administration.