How Mueller detailed the Trump-Russia deal that stole the 2016 election. How Bill Barr buried it.
By J.J. Goldberg
On August 2, 2016, the chairman of the Donald Trump for President campaign met secretly in New York City with a suspected cutout for Russian intelligence to discuss things they could do for each other. It was not their first such meeting, but it appears to have been a decisive one. It came at a strategic moment in the campaign, 12 days after Trump was coronated at the Republican national convention and three months before Election Day.
The Russian, political consultant Konstantin Kilimnik, described by the FBI and others as actively “associated” with Russian intelligence, was hoping that Trump, if he won, would help Moscow regain partial control of Ukraine, its former satrapy. The American, Trump campaign chief Paul Manafort, discussed what looks very much like a plan for Russian internet trolls to help Trump win. He singled out four problematic swing states: Michigan, Wisconsin, Pennsylvania and Minnesota.
In the end, both sides got pretty much what they wished for from each other that evening.
A third item on their agenda was a tangled, multimillion-dollar financial spat between Manafort and a former client who was now Kilimnik’s patron, billionaire Russian aluminum magnate Oleg Deripaska, one of a small handful of Vladimir Putin’s most trusted oligarch-advisers.
It might sound like a cheap spy novel or a deranged conspiracy theory, but the August 2 meeting is documented fact. It’s described in detail in the official U.S. Department of Justice “Report On The Investigation Into Russian Interference In The 2016 Presidential Election,” better known as the Mueller Report. The account of the meeting begins on page 138 of the report’s Volume I, which examines evidence of Russian interference activity. (Volume II explores evidence of coverup and obstruction.)
According to Mueller, the two men met over a late dinner at the Grand Havana Room, a swank gentlemen’s cigar club situated atop Jared Kushner’s cash-hemorrhaging property at 666 Fifth Avenue, three blocks from Trump Tower. Also present was Manafort’s longtime deputy Rick Gates. Afterward they took care to leave by separate exits to cover their tracks.
Once seated, Kilimnik presented Manafort with a purported peace plan to end the deadly, long-running Russia-Ukraine conflict, essentially on Moscow’s terms. The plan would, according to Mueller’s report, “create an autonomous region with its own prime minister” in the war-torn Donbass region of eastern Ukraine, bordering Russia. It would be formally under Ukrainian sovereignty but dominated by Moscow. Manafort told Mueller’s investigators that the plan “was a ‘backdoor’ way for Russia to control part of eastern Ukraine.” Manafort and Kilimnik agreed in later emails that the plan would require backing from a prospective Trump White House.
In turn, Manafort described to Kilimnik, a longtime friend and business associate, the current state of the presidential race, dwelling specifically, the report says, on the importance to Trump of winning a handful of closely contested “battleground” states. As noted, Manafort named four states of particular concern: Michigan, Wisconsin, Pennsylvania and Minnesota.
Afterward, Manafort asked Gates to give Kilimnik a 75-page printout of internal Trump campaign polling data prepared earlier that day. Campaign polling is designed to identify persuadable voters, craft persuasive messages for them and get them out to the ballot box.
As it turned out, three of the four states Manafort named, Michigan, Wisconsin and Pennsylvania, were the ones Trump won November 8 by infinitesimal, utterly unforeseen margins, giving him the Electoral College and the presidency despite his multimillion-ballot loss in the popular vote.
Strangely, the American public seems utterly unaware of what happened that evening. This is an injustice to the Mueller investigation and to American democracy. The meeting’s vital importance is a matter of public record. Mueller’s deputy, Andrew Weissman, told a federal court in February 2019 that the incident “goes, I think, very much to the heart of what the Special Counsel’s Office is investigating.”
And yet, just six weeks after that court hearing, on March 22, when the freshly completed Mueller Report was delivered to the Justice Department, the Manafort-Kilimnik meeting had been downgraded from “the heart” of the investigation to an inconclusive incident, one among many that the special counsel examined but could not prove indicative of any Trump-Russia collusion. By early May 2019 the meeting would be effectively erased from America’s public consciousness before it could even register — buried, it appears, by some well-timed flimflam and outright lying from the top of the Trump Justice Department.
Mueller’s account of the August 2 meeting is drawn from testimony by all three participants. Manafort and Gates were formally interviewed in 2018 by the special counsel’s office. Kilimnik discussed it with an American associate, Sam Patten, who then described it to Mueller’s team. Their testimony is bolstered by considerable documentary and digital evidence gathered by Mueller’s investigators.
The handover of polling data to Kilimnik at the meeting does not appear in the Mueller Report. Rather, it emerges piecemeal in other documents, including a federal court transcript and related courtroom exhibits. However, Mueller does describe a series of other such polling data transfers from Manafort to Kilimnik during the campaign, beginning in early May and continuing beyond August 2. The understanding between them, seen in intercepted email exchanges, was that Kilimnik would pass the data along to three pro-Putin Ukrainian oligarchs as well as to his boss, Deripaska.
Reading Mueller’s overall Grand Havana Room narrative, it seems evident that he suspected the data somehow found its way from Kilimnik to Putin crony Deripaska and from there to fellow Putin crony Yevgeny Prigozhin, a catering mogul popularly known as “Putin’s chef.” Prigozhin was the founder and main funder of the Internet Research Agency or IRA, the private social-media troll farm indicted in 2018 as a central player in Russia’s election meddling.
It should be noted that both Kilimnik and Deripaska have denied participating in any U.S. election scheme.
The IRA social-media effort should not be confused with the better-known but separate operation by Russia’s GRU military intelligence to hack massive data from Democratic-linked computers and pump embarrassing bits back into America via WikiLeaks.
Prigozhin’s IRA trolls presumably would have used Manafort’s polling data to identify persuadable voters in Manafort’s battleground states and custom-tailor effective messaging. But the special counsel was unable to prove it.
Instead, he was fed distractions. Gates told the investigators “that he did not know why Manafort wanted him to send polling information, but Gates thought it was a way to showcase Manafort’s work, and Manafort wanted to open doors to jobs after the Trump Campaign ended.”
At the same time, intercepted email messages between Manafort and Kilimnik indicate that Manafort was hoping the data would somehow help resolve his huge debt — “to get whole,” as he put it in one email — with Deripaska.
Manafort denied to Mueller’s investigators that he ever instructed Gates to send polling data to Kilimnik, until he was confronted with hard evidence that he had. He also denied having any further involvement in the Ukraine scheme after August 2, until he was confronted with documentary evidence in the form of emails, airline passenger records and more. What the evidence showed was that Manafort and Kilimnik continued discussing the plan through 2017 and into early 2018.
Mueller plainly didn’t believe either explanation for the data dumps, neither the job search nor the debt resolution effort. The report flatly states (page 130), after relating both versions, that the special counsel’s office “could not reliably determine Manafort’s purpose in sharing internal polling data with Kilimnik during the campaign period.”
Nor did Mueller believe that the data trail ended with Deripaska or the Ukrainian oligarchs: “Because of questions about Manafort’s credibility and our limited ability to gather evidence on what happened to the polling data after it was sent to Kilimnik, the Office could not assess what Kilimnik (or others he may have given it to) did with it.” That is, he couldn’t find the missing piece of evidence, but he didn’t consider the case closed.
What’s more, the report states, “The Office has not uncovered evidence that Manafort brought the Ukraine peace plan to the attention of the Trump Campaign or the Trump Administration.”
Manafort’s known lies about the Grand Havana dinner, the polling data and the Ukraine plan were the main basis for Mueller’s 2019 request to a federal court that it vacate Manafort’s plea agreement and send him to prison. That was the subject of the February 4 hearing where Mueller’s deputy Weissman described the Grand Havana repast as going “to the heart of what the Special Counsel’s Office is investigating.” He was explaining to Judge Amy Berman Jackson why the lies stemming from the Grand Havana were sufficiently material to a criminal investigation to warrant prison time for Manafort.
Still, the fact remains that Mueller could not prove Manafort’s polling data actually reached Russia’s social-media trolls and thus influenced the election. Accordingly, the report does not say that the Grand Havana exchange constituted collusion (or as he terms it, “conspiracy or coordination”) between the Trump campaign and the Russian government. Ever the straight-arrow prosecutor, Mueller will not allege a crime unless he believes he can prove it.
In fact, the report states flatly at the very beginning, on page 2 of the 448-page document, and repeats periodically in various phrasings, that “the investigation did not establish that members of the Trump Campaign conspired or coordinated with the Russian government in its election interference activities.”
That statement might lead the ordinary reader to conclude that the 22-month, $25 million Mueller probe came up empty and cleared the Trump campaign of any suspected collusion. That would be dead wrong. Just a few sentences later on the same page 2, Mueller explains that the word “establish” is used in the report in a narrow, specialized, lawbook meaning: to indicate that “substantial, credible evidence enabled the Office to reach a conclusion with confidence.” Evidence, as explained further on page 8, that “would probably be sufficient to obtain and sustain a conviction.” Sufficient, that is, to prove the case beyond a reasonable doubt.
That is followed immediately by this blunt caveat: “A statement that the investigation did not establish particular facts does not mean there was no evidence of those facts.”
Mueller further stipulates on page 14, in the final words of his executive summary, before jumping into the main body of the report, that because of multiple instances of lying, claims of privilege and lack of means to keep digging, “the Office cannot rule out the possibility that the unavailable information would shed additional light on (or cast in a new light) the events described in the report.”
As we’ve seen, the report goes on to specify precisely which pieces of evidence are missing: how the polling data might have gotten from Deripaska to Prigozhin and how, if at all, the Trump White House was told of its expected role in the Ukraine plan. Because he couldn’t establish a direct chain of evidence leading from the Grand Havana to the suspected end-users in either direction, Mueller could not allege a Manafort-Kilimnik conspiracy. He could only hint at it.
Looking back a year after the report’s release, it appears that few Americans read as far as page 2, thanks to some deft verbal sleight of hand by Attorney General William Barr that made the Grand Havana dinner effectively disappear, together with its leadup and fallout. If the public had read it, the words “Michigan,” “Wisconsin” and “Pennsylvania” should have jumped out like flashing neon lights. Instead, the incident lies all but forgotten.
The attorney general, newly sworn in just weeks earlier, gave Americans their first peek at the Mueller Report on Sunday morning, March 24, 2019, two days after he received it from Mueller, in a four-page letter he sent to Congress and then released to the public. Barr’s letter claimed to “describe the report and to summarize the principal conclusions.” To this end the letter offers just three sentence fragments quoted directly from the 448-page report. The rest consists of paraphrasing, broad interpretation and Barr’s own conclusions.
One of those three direct quotes had a decisive impact on everything that followed, namely Mueller’s statement that the “investigation did not establish” Trump-Russia conspiracy or coordination. Barr quotes it at the end of a long paragraph in which he reminds readers that Mueller’s mission was to investigate Russian election interference, and that within that quest, the question of Trump campaign conspiracy-or-coordination was “a primary consideration.” Then Barr offers Mueller’s conclusion — that he “did not establish” such collaboration — but only after he first rephrases it and almost invisibly reframes it. The Mueller investigation, Barr writes, “did not find” such collaboration. Then he quotes Mueller’s original “did not establish.”
Barr repeated his “did not find” formula twice more in the next two paragraphs. At no point in the four-page letter did he explain Mueller’s definition of “establish.” The net effect was to suggest that the Mueller investigation was a wild goose chase.
If Barr’s paraphrasing was intended to mislead, it worked brilliantly. “Did not find” was the universal takeaway, understood nationwide to mean that there was nothing to be found. The public would have to wait a month before reading — if they still bothered — what Mueller actually wrote: that he found evidence, but not yet enough to take to court.
The ploy succeeded instantly. Within minutes after Barr’s congressional letter went public on Sunday afternoon, veteran NBC News correspondent Andrea Mitchell tweeted out what became the standard storyline: There was “no collusion.”
AG Barr says Trump campaign did not engage in conspiracy with Russia despite Russian efforts to reach out. In other words, no collusion by Trump or his campaign to tacitly or expressly coordinate with Russia.
Shortly afterward, The New York Times published its own more detailed account, leading with this:
WASHINGTON — The investigation led by Robert S. Mueller III found no evidence that President Trump or any of his aides coordinated with the Russian government’s 2016 election interference, according to a summary of the special counsel’s key findings made public on Sunday by Attorney General William P. Barr.
Mueller complained in a letter to Barr the following Thursday, March 27, that Barr’s letter to Congress “did not fully capture the context, nature, and substance of this Office’s work and conclusions.” Mueller continued: “We communicated that concern to the Department on the morning of March 25. There is now public confusion about critical aspects of the results of our investigation. This threatens to undermine a central purpose for which the Department appointed the Special Counsel.”
Mueller was not explicit about what Barr’s letter “did not fully capture.” News reports indicated that the complaints centered on the obstruction question. But Mueller’s letter was kept hidden from the public and the media for another five weeks, until the House released it on April 30.
In the interim, Barr was busy. After his March 24 letter he kept the report under wraps for another four weeks, releasing it to Congress and the public only on April 18. Those four weeks provided enough time to redact sensitive, classified or grand jury-related material. Enough time, too, for the misleading message in Barr’s congressional letter to sink in: Mueller didn’t find any collusion. After a month with nothing but Barr’s letter to go on, the public, politicians and the press seem to have concluded by April 18 that there was nothing worth reading in Volume I. The nation’s attention had turned instead to Volume II and the case for obstruction.
Barr now allowed himself to up his game from misleading to outright lying. At a 23-minute press conference on April 18, moments before releasing the redacted report, the attorney general moved beyond the deceptive paraphrasing of “did not find” to a flat untruth: that the special counsel “found no evidence” of Trump campaign conspiracy or coordination with Russia.
Two weeks after that, Barr expanded his assault on the truth even further, this time in sworn testimony May 1 before the Senate Judiciary Committee. First, in his prepared opening remarks, he reiterated his fiction that “the Special Counsel found no evidence that any Americans … conspired or coordinated with the Russian government.” Then, in the question and answer period, came this telling exchange with the committee chairman, Republican Senator Lindsey Graham of South Carolina (seen here live on YouTube at 1:07:32):
GRAHAM: Do you think the President’s campaign in 2016 was thoroughly looked at in terms of whether or not they colluded with the Russians?
BARR: Yes.
GRAHAM: And the answer is no, according to Bob Mueller.
BARR: That’s right.
None of that was true. Mueller found plenty of evidence pointing to collusion. The evidence just wasn’t enough to “establish” collusion beyond a reasonable doubt. Nowhere did he state that it didn’t happen.
In fact, he states just the opposite, again on page 2: “The report describes actions and events that the Special Counsel’s Office found to be supported by the evidence collected in our investigation.” That is, if it’s in here, it means there’s evidence of it.
Indeed, of the 199 pages of Volume I, probing Russian interference, no fewer than 107 pages are spent examining evidence that suggests possible conspiracy or coordination between the Trump team and Russians who might help tamper with the election. (Another 55 pages detail Russia’s own interference activity, much of it familiar from earlier Mueller indictments. The rest of Volume I covers methodology, charging decisions and a 10-page executive summary.)
Of the dozens of cases whose evidence fills those 107 pages, the longest and most damning section, at 15 pages, covers the Grand Havana meeting and its context. The dinner is further illuminated by other information squirreled away like a trail of breadcrumbs elsewhere in Volume I.
Restating Mueller’s narrative in chronological order, the story begins in 2014, when Yevgeny Prigozhin’s Internet Research Agency began focusing its trolling operations on disrupting American politics, spreading chaos, division and mistrust (pages 24-25). In early 2016 the agency began to focus on the upcoming U.S. presidential election, aiming to discredit Hillary Clinton and boost Donald Trump.
Paul Manafort was hired by the Donald Trump campaign as convention manager on March 29, 2016 (page 134). He immediately instructed Rick Gates, his deputy, to prepare memos touting his new status to Oleg Deripaska, his former client and now angry creditor, along with three pro-Putin Ukrainian oligarchs. The next day Manafort sent the memos and a press release to Kilimnik to translate and disseminate. On April 11 he checked back with Kilimnik to ensure Deripaska had seen the material. Kilimnik assured him he had (page 135).
On April 19, the report says (page 25), the “first known IRA advertisement explicitly endorsing the Trump Campaign was purchased.” This apparently marked the opening shot in a full-bore, covert Russian social media campaign to elect Trump.
Eighteen days later, on May 7, Kilimnik met with Manafort in New York. Over an early breakfast he briefed Manafort on the state of his Ukraine peace plan. Manafort, in turn, briefed Kilimnik on the state of the Trump campaign. Coincidentally, Manafort was promoted to overall Trump campaign chairman 12 days after that, on May 19.
Following the May 7 Manafort-Kilimnik breakfast, Gates recalled (page 136n), Manafort instructed him to begin sending internal campaign polling data via WhatsApp to Kilimnik, scrubbing the app’s record after each dispatch. Kilimnik was to pass it along to Ukrainian oligarchs. Beyond that, the report says, “Gates understood that the information would also be shared with Deripaska.” Gates continued sending Kilimnik batches of the sensitive data regularly up to and beyond the August 2 meeting until August 19, when infighting and reports of his past Russia ties forced Manafort to quit the campaign. Gates stayed on and continued sending data to Kilimnik, but without Manafort’s access to confidential polling, the transmissions were reduced mostly to publicly available material. They soon petered out.
Manafort and Kilimnik continued discussing the Russian’s Ukraine plan through the presidential transition and well into the Trump presidency, including at least one face to face meeting in January 2017 in Virginia and another the next month in Madrid, as well as in numerous email exchanges. As late as 2018 Manafort was preparing to poll Ukrainian opinion on elements of Kilimnik’s plan.
Both sides ended up getting much or all of what they wished for at the Grand Havana. Russia got a president in the Oval Office who proceeded, almost from the moment he took office, to repeatedly undermine and humiliate Ukraine’s leaders. Trump’s treatment visibly undercut Ukraine’s image of U.S. support, thus weakening its bargaining position, even as Russia deepened its proxies’ hold on the disputed zone.
As for Trump, he got Michigan, Wisconsin, Pennsylvania and the White House.
Trump’s presidential jabs at Ukraine were individually reported in the press, but rarely tied together as a pattern. The administration put a brief but controversial hold on a shipment of Javelin missiles in December 2017, nearly two years before the notorious August 2019 missile hold — and long before Joe Biden entered the presidential race. Trump repeatedly pushed for Russia’s readmission to the G-8, despite its continuing occupation of Ukrainian Crimea, the reason Moscow was expelled in the first place. He mounted a controversially flaccid response to Russian attacks in 2018 on Ukrainian shipping in the Black Sea. He denied the newly elected Ukrainian President Volodymyr Zelensky the prestige of an Oval Office visit, a customary gesture to any incoming allied leader.
Finally, having conspicuously shorn Ukraine of its undivided U.S. backing, Trump pushed a weakened Zelensky into December 2019 peace talks with Putin. Zelensky managed to hold the Russian to a draw, leaving the situation on the ground no better but not much worse.
None of these actions reversed Washington’s basic support for Ukraine, which is mandated by Congress. Both Javelin missile holds ended in deliveries. Russia denied interest in rejoining the G-8. Trump’s actions were mostly temporary and technically within congressional mandate. Their only lasting impact was to raise doubts about the strength of U.S. backing for Ukraine. But that was their apparent purpose.
Manafort and the Trump campaign, of course, gained something of historic magnitude: the presidency.
It could be pure coincidence that Trump’s surprise victory came from tiny margins in precisely the states that Manafort described to Kilimnik as critical on August 2. Given the end results, however, coincidence seems far-fetched.
Exit polls showed that Trump’s victory followed a sudden, sizeable and still unexplained shift of undecided voters in Trump’s direction in those three states in the last seven days before the election. A shift was seen in some other states as well, though less pronounced or decisive. Undecideds choosing sides in the final week came to 20 percent of the vote in Wisconsin and Michigan and 15 percent in Pennsylvania. Trump won those last-minute undecideds in the three states by margins of 11, 29 and 17 percent respectively, for a combined total of nearly 1.3 million last-minute votes.
It was just enough. Trump won those three states by a combined total of 77,744 votes. This won him the Electoral College despite losing the popular vote nationwide by a convincing 2.9 million votes, or about 2.1 percent.
The result made Trump the third American president to win the White House after losing the popular vote by multiple percentage points. The other two were John Quincy Adams in 1824 and Rutherford B. Hayes in 1876. Both of them were put across the finish line, historians say, by post-election backroom deal-making. Two other presidents, Benjamin Harrison in 1888 and George W. Bush in 2000, took office after losing the popular vote by a fraction of a percent (of the votes that were counted, that is).
Democrats have been arguing ever since Election Day 2016 over what it was that cost Hillary Clinton the presidency. Theories have ranged from misguided campaign strategy to faulty messaging, to a mistaken assumption that the old Obama coalition would rally, to Clinton’s failure to visit Wisconsin and finally to Clinton’s own personality or demeanor.
Clinton herself argued as early as September 2017, in an interview with The New Yorker magazine’s David Remnick, that “there had to be some information provided to the Russians by someone as to how best to weaponize the information that they stole.” Her intuition was at least partly correct: Information had been provided. It’s possible that the stolen, weaponized information she was referring to — the data hacked from Democratic-linked computers by Russia’s GRU military intelligence and passed to WikiLeaks, separate from the IRA social-media campaign — could have hurt her among the 130 million-odd citizens in the nationwide popular vote.
But Clinton did not lose the popular vote. She won it by a respectable margin of nearly 3 million votes. It was not Trump’s 62.9 million voters who somehow defeated Clinton’s 65.8 million. On the contrary, Trump was elected by a total of 77,744 voters, surgically extracted at the last minute in three Great Lakes states, who tipped the Electoral College. Put differently, the president of the United States was chosen by an electorate small enough to be seated together in the University of Michigan football stadium, with room left over for social distancing.
The lessons for Americans are stark. First, it should be clear by now that anything, literally, can happen. Regardless of what the polls show, there are endless ways Trump and his allies might conjure up a victory out of certain defeat. Warnings are surfacing already from across a broad political spectrum that the president might refuse to leave the White House after losing the vote. He might, for example, resort to some dramatic public measure such as declaring an emergency or insisting the vote count was fraudulent, as he has already begun to do in his attacks on mail ballots.
Republican officials are taking blatant steps to suppress Democratic votes by massively purging voter rolls, reducing the number of polling places in minority neighborhoods, discrediting mail-in ballots and much more. Game-planning is already underway in some circles to address the threat, though it will be an uphill climb, and one shouldn’t underestimate the powers of incumbency, particularly a Trump incumbency.
Still, threats like those leave democracy with at least one advantage: their visibility. They can be seen unfolding in real time, allowing a functioning democracy to mount defensive countermeasures, however long the odds. On the other hand, events like the Grand Havana scheme — if that’s what it was — are invisible.
Even the GRU-WikiLeaks hacking operation left enough traces to be caught by the FBI months before the election, though partisan action by GOP Senate leader Mitch McConnell hobbled a government response. Besides, its impact on the election results hasn’t been and probably couldn’t be measured.
By contrast, what happened at the Grand Havana Room wasn’t discovered until 2018. As decisive as its impact appears in hindsight, the operation left no visible traces. Investigators couldn’t look for the trail, because they didn’t know it existed. The technology, and the shamelessness of it, were so new that such a scheme could hardly be imagined. Even today, with all the evidence, the story ultimately remains a set of hard facts connected by conjecture, as this article shows. Mueller and his team might shed further light, but they have consistently refused to expand beyond the report’s text.
The upcoming presidential election won’t likely experience a repeat of the Grand Havana incident. The trick has been exposed, so investigators know what to look for. The next scheme out of Russia — or China, Iran or Venezuela — will be something we can’t even imagine today.
J.J. Goldberg 