A Justice Department trial attorney repeatedly contacted Massachusetts Institute of Technology researchers asking, eventually under threat of subpoena, about research they had conducted on the 2019 Bolivian presidential election, according to emails obtained by The Intercept. Sent between October 2020 and January 2021, the emails point to the existence of the Justice Department inquiry and add new evidence to support Bolivian allegations that the United States was implicated in its 2019 coup.
The emails reveal the Justice Department’s involvement in the Bolivian coup regime’s criminal investigation into alleged voter fraud, which has not previously been reported. The inquiry targeted a pair of respected MIT researchers about their work for the Center for Economic and Policy Research, in which they broadly refuted suspicions that Bolivia’s socialist party had rigged the election.
The short-lived coup regime reached power following a clear script: In the weeks leading up to the Bolivian presidential election in October 2019, the opposition pumped endless propaganda through social media and television networks, warning that incumbent President Evo Morales would exploit widespread fraud to win reelection. Morales had become the first Indigenous president elected in Bolivia in 2005, at the head of his party Movement Toward Socialism, or MAS, and by 2019, he was running for his fourth term. He faced intense opposition, often framed in explicitly racist terms, from a Frankenstein coalition of right-wing Bolivians of European descent and supporters of former President Carlos Mesa, once a member of Bolivia’s left revolutionary party who had become hostile to Morales’s social democratic government.
As the votes were counted on election night, Morales was ahead as expected. The question was whether he would win by enough to avoid a runoff, which in Bolivia is triggered when a candidate wins by a margin of fewer than 10 points. In an unofficial tally, Morales led Mesa by 7.9 points, giving the opposition hope for a second round. But when the official count was released, Morales had won by 10.6 points. There would be no runoff.
Without evidence, the opposition immediately leveled fraud charges. It was backed up the next day by the Organization of American States, the powerful hemispheric cooperation organization based in Washington, D.C.
“The OAS Mission expresses its deep concern and surprise at the drastic and hard-to-explain change in the trend of the preliminary results revealed after the closing of the polls,” read the OAS’s incendiary statement. Protesters took to the streets; the military called for Morales to step down; and the opposition installed a new leader, Jeanine Áñez, after three weeks of unrest. Far to Mesa’s right, Áñez assumed office and swiftly attempted to eliminate the sense of enfranchisement for Indigenous people that the Morales government had brought. While 14 out of 16 members of Morales’ first Cabinet were Indigenous, Áñez did not appoint a single Indigenous person to her first Cabinet. In the two months before assuming office, she had tweeted that Morales was a “poor Indian” and implied that Indigenous people cannot wear shoes. When she reached the presidency, she declared that “the Bible has returned to the palace.”
The coup, roughly the same play President Donald Trump would attempt a year later, was complete.
But the U.S. press refused to call it that, instead accepting the allegations of fraud at face value.
“The line between coups and revolts can be blurry, even nonexistent,” wrote Max Fisher for the New York Times. He cited what political scientist Jay Ulfelder calls “Schrödinger’s coup”— those cases which “exist in a perpetual state of ambiguity, simultaneously coup and not-coup”— and dismissed the distinction as “old binaries” now considered “outdated” by scholars.
The Times did not undergo such hand-wringing over allegations that Morales’s party had rigged the election. Its October 2019 coverage reproduced the opposition’s promises for a “damning” unreleased OAS report, raising “the prospect that a victory by Mr. Morales would be regarded by the international community as illegitimate.” The Trump administration’s top diplomat for Latin America, Michael Kozak, condemned the Morales government and vowed that the U.S. “will work with the international community to hold accountable anyone who undermines Bolivia’s democratic institutions.”
Even a surface-level look at the vote-counting process suggested that the surge for Morales was utterly predictable.
But even a surface-level look at the vote-counting process suggested that the surge for Morales was utterly predictable. The bulk of the votes that were left to be counted on election night in 2019 had been cast deep in the country’s rural areas, where Indigenous miners, coca growers, and other working-class people overwhelmingly favored Morales. (The former president hails from the Chapare and was previously the head of the coca growers’ union.) It should have seemed obvious that their votes had put him over the top.
Just over a year later, in November 2020, late-counted Democratic votes put Joe Biden over Donald Trump in the U.S. presidential election, and Trump called foul. “We were winning everything, and all of a sudden it was just called off,” Trump said on election night. “We’ll be going to the U.S. Supreme Court, we want all voting to stop. We don’t want them to find any ballots at 4 o’clock in the morning and add them to the list.” The U.S. media had no difficulty explaining why the surge for Biden was legitimate. But when reporting on Bolivia, all of the American election expertise evaporated.
The OAS followed its October statement with a more in-depth analysis in November 2019, this time finding perhaps as many as a few hundred cases of apparent vote-tampering. But the data in the report did not sufficiently support the organization’s allegations of widespread fraud. In a letter to the OAS later that month, Rep. Jan Schakowsky, D-Ill., asked if the organization was “aware that this steady increase in Evo Morales’ margin was the result of precincts that were, on average, more pro-Morales reporting their results later than precincts that were, on average, less pro-Morales? Why is this apparently obvious conclusion — from the publicly available data — never mentioned in the EOM [Election Observation Mission] press statements or reports?”
The New York Times did not exercise the same scrutiny. “After the Organization of American States declared on Sunday that there was ‘clear manipulation’ of the voting in October,” the paper editorialized, “Mr. Morales was left with no choice but to resign, bitterly tweeting from an unknown location that ‘The world and patriotic Bolivians will repudiate this coup.”
In fact, it would be statisticians who repudiated the coup. Researchers at MIT, commissioned by the Center for Economic and Policy Research, took a closer look at the data and evidence behind the allegations and concluded what many other independent observers had already found: The fraud claims were bogus, according to a statistical analysis conducted by Jack R. Williams and John Curiel of MIT’s Election Data and Science Lab.
The fallout from the MIT researchers’ analysis, which was published by the Washington Post in February 2020, was considerable. In a stunning reversal, the New York Times published an article on the findings, saying that it “cast doubt on Bolivian election fraud.”
The prestigious release was a major blow to the coup regime, leading to references in many of the same major media outlets that had peddled the coup government’s election fraud narrative. The new insight sapped the coup government’s international credibility, which was further degraded as it repeatedly delayed a new election. With La Paz shut down by protesters — this time the crowds were on the side of MAS — the regime was finally forced to hold an election on October 18, 2020.
Three days before the vote, the researchers received the first of the Justice Department’s requests. Trial attorney Angela George identified herself as an attorney at the Justice Department’s Office of International Affairs, or OIA, and said she had “received a formal request from Paraguay” for assistance in an ongoing criminal investigation. Curiel told her she had the wrong researcher, as he had not worked on any Paraguayan election study, and she told him that Bolivia was the one she had meant.
George never provided details about the nature of the criminal investigation, the existence of which has not been previously reported. Attempts to reach the coup government’s minister of justice, Álvaro Coimbra, were unsuccessful, as he is in prison facing charges of sedition related to the coup.
“We have a few questions about the data report, and we would appreciate if you could let us know when you are available to speak with us via telephone before or by November 6, 2020,” George wrote to the researchers. When Williams explained that his research was based on publicly available information, she replied threatening “a subpoena being served on you and the lab” but also dialed down her demand, saying that an interview might not be necessary. “I am simply trying to find out if the report, Analysis of the 2019 Bolivia Election, that is embedded in the Washington Post article referenced below includes your research and is an authentic copy of the report that was produced … and includes the comprehensive research you and Mr. Curiel conducted,” the prosecutor wrote.
The Justice Department inquiry frightened election researchers in the academic community and may have had a chilling effect on subsequent research.
The threat of subpoena was an extraordinary move, as the Justice Department has strict protocols to protect the freedom of the press and prevent government intimidation. According to a source familiar with the investigation, who was not authorized to speak publicly, the Justice Department inquiry frightened election researchers in the academic community and may have had a chilling effect on subsequent research.
A former Department of Justice trial attorney who also worked at the OIA told The Intercept that the correspondence was unusual for several reasons. Requesting anonymity to avoid professional reprisal, they said that professional investigators trained in interview techniques usually contact subjects, and there are stiff rules governing any interactions with the media.
“Generally, OIA would enlist the FBI or other investigative agency to execute an incoming MLA request such as a voluntary witness interview or inquiry like this one. It’s unusual for an OIA attorney to handle it,” the former trial attorney explained.
They also said that interactions with the media require authorization from senior Justice Department leadership.
“There is a whole set of onerous protocols in place for trial attorneys seeking information from a media organization, and the decision to move forward would be made at high levels at the DOJ. This particular request is not your run-of-the-mill criminal investigation, so you can be fairly sure that it received very high-level exposure,” the source said.
Justice Department spokesperson Joshua Stueve declined to comment.
Earlier in 2020, the U.S. government-funded media organization Voz de América, the Spanish-language complement to Voice of America, singled out the same two researchers by name in an article. The story implied that they could be taken to court over their study.
“Bolivia roundly rejected the supposed study from the Massachusetts Institute of Technology (MIT, by its English initials), which assured that there had been no electoral fraud in Bolivia,” begins the story dated March 5, 2020, by Yuvinka Gozalvez Avilés.
Avilés writes that Karen Longaric, then Bolivia’s minister of foreign affairs, “dismissed the idea of pressing charges against the two people who published the article,” and warned that there are harsher sanctions than a judicial investigation, namely to be professionally discredited.
“Both experts belong to MIT; however the institution denied any participation or authority in said document, clarifying that both people ‘saw the project through as independent contractors of the Center for Economic Policy and Research,’” Avilés continues. The Center for Economic Policy and Research told The Intercept that they had not received any communication from Voz de América for the article, nor did they hear from the Justice Department about the investigation. MIT’s press team did not respond to The Intercept’s requests for comment.
The article also echoes a baseless allegation from Longaric that the MIT researchers’ report “is linked to people connected to the disputed president of Venezuela, Nicolás Maduro, an ally of former president Evo Morales.” Avilés does not provide any evidence for this claim but quotes Longaric saying of the researchers: “We can assert that once again those enemies of democracy tried under false pretenses to disrupt the rule of law in Bolivia and obstruct the elections.” (Trump allies also claimed that Venezuela had a hand in stealing his own election.)
Leading up to the second Election Day, the right-wing media ecosystem was once again rife with claims that the vote would be rigged, but the effort failed the second time, as MAS won in a landslide. Morales, then still in exile, did not run, but his protégé Luis Arce won 55 percent of the vote. Once again, there would be no runoff.
Áñez had dropped out of contention a month before the new election, leaving Mesa again as the leading opposition candidate. Morales has since returned to Bolivia from exile, and Áñez has been arrested, charged by the new government with terrorism, sedition, and conspiracy.