Not only did Joe Biden declare that he would appoint “the first African-American woman to the US Senate“–when one of the two to have already been elected, Sen. Kamala Harris (D-CA), was campaigning against him for the Democratic nomination–but Senators are not appointed by the President! His own has been rife with gaffs, including the claim that “150,000,000 Americans have died from gun violence since 2007–which would be half the US population of around 330,000,000–but that he himself was a “candidate for the US Senate” and that he worked on the 2016 Paris Climate Accord with Chinese leaders Deng Xiaoping, who died more than 20 years ago!
Joe Biden appears to be severely cognitively impaired from a pair of brain aneurysms he suffered, which he acknowledged during a campaign event in New Castle, NH, last year after his eye filled with blood. Even the late endorsement of Jim Clyburn, the most influential black political leader in South Carolina, a few days before the primary cannot possibly account for his stunning 30-point victory over Bernie Sanders. Something else had to be involved here, what is commonly referred to in social science research as an intervening variable. In this instance, provided you have the proper background knowledge, it’s not that difficult to figure out what happened. Here’s the outcome:
As those who have been following the campaign for the Democratic Party nomination are aware–which, however, is only a small proportion of the American electorate–the primary was held on 29 February 2020 (because 2020 is a Leap Year and February has 29 days), but only a few days earlier, news reports (that appear to be accurate), showed that, on Face the Nation (23 February 2020), Biden’s lead over Sanders was around 5 points at 28% to 23%, with Tom Steyer in third place at 18%. Those who have been paying attention, therefore, properly anticipated an outcome (less than a week later) that would closely parallel these proportions of the polling when the primary would be held:
Moreover, Face the Nation also reported that the first choice for President among Black Democratic voters was Biden at 35%, Steyer (not Sanders) at 24% and Bernie at 23%. The difference across the state, of course, is because, although South Carolina has around 60% black voters, the other 40% had differing preferences, while yielded the 28/23/18% percentages across the entire Democratic electorate, where Steyer, in particular, was polling stronger with non-Blacks than with Blacks:
Perhaps most significantly, however, CBS News also reported the downward trajectory of Biden support, where he had enjoyed massive preference at 54% in November, but which had dropped to 35% in late March (days before the primary), while Steyer’s support had risen during the very same interval from just 2% to 24% and Bernie remaining constant at 23%, which is unsurprising, given Biden’s dismal performance in Iowa (finishing 4th with 15.8%), in New Hampshire (finishing 5th with 8.4%) and in Nevada (finishing 2nd with 20%).
Speaking generally, electorates are a lot like the Titanic: they are massive and slow-moving, where a stunning increase from a 5% lead on 23 February 2020 to a (nearly) 30% lead less than a week later defies rational belief (absent some dramatic development with obvious impact upon the election, which was not the case with Jim Clyburn’s endorsement, which had long been widely expected). Something else appears to have been at work here, the “intervening variable”, which is missing because it is non-public but necessary to understand what happened here. (Any statistician can confirm my observations.) This extremely improbable outcome, however, appears to be explicable, if you have been paying very close attention.
On 16 July 2019, Robert Epstein, Ph.D., a research psychologist who earned his doctorate at Harvard and has been the Editor-in-Chief of PSYCHOLOGY TODAY, testified to Congress about his research into the use of algorithms by Google in a way that shifted at least 2.6 million votes in favor of Hillary Clinton in 2016, where the total may have been as much as 10.4 million. What he presents–based upon massive research–seems to explain why, although Hillary was drowning in minuscule turnouts compared to Donald Trump, she claimed to have won the popular vote, in spite of which Trump won the Electoral College Vote, because of which many Democrats now want to eliminate the Electoral College from the Constitution.
Epstein explained that, although he regards himself as “center-left” or even further left on the political spectrum and supported Hillary Clinton in 2016, his dedication to America and the electoral process has motivated him to share his research results with Congress, in the hope that it might act to preserve the integrity of democracy in the United States for the sake of the Constitution and the American people. He also reported that Google had made a major effort in 2018, which undoubtedly contributed to the Democrats regaining control of the House and Trump not picking up a couple more Senate seats, as I myself was predicting on the basis of contemporary polling, which simply baffled me at the time.
Having been caught with its pants down, Google of course dismissed his findings as “a poorly constructed conspiracy theory” (which by itself virtually confirms it) and Wikipedia included an obscure professor from Wellesley claiming “we know that this [manipulation by Google] did not happen”. But how could he possibly know? A key aspect of Dr. Epstein’s testimony to Congress is that Google’s manipulations are “invisible” and virtually impossible track, because they can be done at the push of a button (depression of a key on a keyboard) to send “Get Out and Vote!” reminders to a targeted audience, which would have been Democrats in selected states in the past during the 2016 and 2018 elections.
And that appears to be exactly what happened here. Consider the two hypotheses: (h1) this was a normal election with no outside interference by a hidden variable (in this case, Google); versus (h2) this was another instance (like 2016 and 2018) where Google manipulated the outcome behind the scene. The more likely hypothesis is the one that confers the higher probability upon the available evidence, in this case, including Biden’s dismal performance in Iowa, New Hampshire and Nevada and downward trajectory in South Carolina. Since Clyburn’s endorsement was expected, the probability of (h1) on the evidence is very low, but on (h2), it is very high.
Evaluating the alternative on the basis of the available evidence–which should include Biden’s gaffs, which suggest that he is seriously cognitively impaired and would be a disaster on the campaign trail, should the Democrats actually nominate him–supports the conclusion that Google manipulated the outcome to favor Biden. This should come as no surprise, since Google executives have declared that they are going to do everything they can do to defeat Trump in 2020. Many Democrats think Bernie would lose the election and want to stop him. Boosting Biden would help. Google had the motive, the means and the opportunity to influence the outcome. It would be irrational not to conclude that they have done it again.
Jim Fetzer, a former Marine Corps officer, is McKnight Professor Emeritus on the Duluth Campus of the University of Minnesota and co-founder of moonrockpress.com.