Some time ago, Trump made the claim that “In certain swing states, there were more votes than people who voted, and in big numbers.” Politifact ran analysis of Trump’s statement, and then I ran analysis of Politifact’s analysis. It got me thinking. I remembered hearing similar claims by people other than Trump, so I did an internet search.

Unfortunately, all I could find was one article allegedly debunking some of these claims. Yet, without hearing the original claims in their entirety, in context and with supporting evidence, and relying only on the article “debunking” these claims, I have now become convinced that the claims are almost certainly true.

The Story: Republicans in the state legislature claimed that there were about 200,000 more votes cast in Pennsylvania than there were people who actually voted. The department of state spokesman responded that it was a misunderstanding, explaining that not all counties had finished reporting into the SURE system, yet had separately certified to the secretary the number of voters, adding that it is the certification that determines the number of voters, not the balance in SURE.

Well, if certification determines the number, what is the SURE system for? Could it be that we don’t trust those doing the certifying and have publicly accessible software to check inputs in real time? Is that what SURE is? What else could it be if not that? Based on the context, we know it is in part some way of tracking the number of voters, even if it might be imperfect.

Why is it that as late as December 29 (when the story came out), the numbers were still not uploaded into SURE? One would think that adding up the number of people that vote should be very quick and easy and should be made public before ballots are counted so that poll workers cannot add more names to the system as needed to match the number of ballots already counted thereby covering up fraudulent ballots. Instead, this was weeks after the election! If they had time to count up their numbers and certify, how did they not have time to upload them to SURE? Did they even know their numbers before they certified?

That is damn suspicious. If the state had simply said the Republicans were lying, and made no further comment, I would not know who to believe until I had heard the Republicans make their case. However, by mentioning that there is indeed some system that tracks voters, but then dismissing it without giving a reason, while claiming no discrepancy in whatever system is used for certification, the state gives the GOP claims some credibility.

Furthermore, because I have some previous knowledge of such things going back many years before this particular election, I am skeptical of elections in general:

I happen to know that states do not always keep their voter rolls updated to eliminate those who have moved or died (it’s an ongoing problem) and that some states were mailing out ballots to everybody on the list, meaning some residents received ballots for every previous occupant of the home. Combined with unmanned drop-off sites and lack of ID checks, there is no way to know that some people didn’t vote twice. Given human nature, it is virtually certain that many people did vote twice (or more) – on their own – without need for a large-scale conspiracy. Did the extra votes split evenly between all candidates? Given demographic differences, possibly not.

I also know that state secretaries do not keep close track of how many people vote because the news is full of stories of poll workers finding more ballots to count days after election day. In all cases, the number is too small to make a difference and there is no reason to suspect ill intent rather than incompetence, but in all cases they are said to be surprises as if they didn’t know that some ballots must be missing.

Trust no one.

Some time ago, Trump made the claim that “In certain swing states, there were more votes than people who voted, and in big numbers.” Politifact ran analysis of Trump’s statement, ranking it as “Pants On Fire.” Their analysis was very strange. Instead of showing proof that the numbers matched, or focusing on the lack of evidence to the contrary, the author of the post feigned ignorance of what Trump meant.

The Politifact author suggests Trump might have meant that there were more votes than the number of those registered to vote and then easily disproves this. Then he suggests Trump might have meant that some voters voted more than once and claims that a system exists to prevent this.

I find the author’s lack of understanding basic English to be very disturbing. I don’t think that’s what Trump was referring to at all. Extra votes could mean double-voting, but could also mean votes matched with made-up names or simply that there were additional ballots lying around to be counted that could not be tied to any names. Trump never gave a reason for the discrepancy; he merely claimed there was one. All he was saying is that if you add up all the people who came in person, who requested an absentee ballot, and who used a drop-off kiosk (however those are counted), the total number was less than the total number of ballots checked for Biden, Trump, Jorgenson, West, and left blank. Where did these extra ballots come from?

Trump very well may have been lying, but nobody has yet demonstrated this, certainly not Politifact.

It can be difficult to know the truth when different media outlets don’t even agree on the basic foundational facts of the story – or when they don’t even cover the same story so coverage can be compared. Establishing reliability by checking the sources does little good since the sources might not be reliable either. However, sometimes one can know when they are being lied to, and count this against the credibility of the source on other things.

On 31 May 2021, Dean Obeidallah posted to CNN.com the article The most ominous part of Texas’ voter suppression move. It makes several claims:

The article claims that the bill prohibits early voting on Sundays before 1:00 pm and that this will somehow make it harder for minorities to vote. What are minorities doing Sunday afternoon that the white majority isn’t? The article never explains. It only says that “souls to the polls” campaigns exist to get minorities to vote after church, and that having to wait (assuming that church gets out long before 1:00 – not all do) will somehow discourage them from voting at all. Does waiting until 1:00 also discourage church-going whites from voting in equal amounts? The article never explains. One could easily imagine a parallel universe in which a decision to move the time earlier would result in churchgoers claiming their votes were being suppressed because lines start forming while they are still in church. It sounds like no matter what time is chosen, somebody will claim suppression. It’s obviously not a valid claim.

The article also claims that the bill prohibits drive-through voting and that this will somehow suppress minority votes. Will minorities not vote if there is no drive-through option? Will whites vote anyways? Why? What is the difference? The article mentions that this option was used in the last election mostly by minorities, but it also mentions that the option was only used in one county. If this county is predominantly inhabited by minorities, then that alone explains the discrepancy. It is not clear where the alleged suppression is.

The article also claims that the bill lowers the burden of proof of election fraud needed for judges to “overturn the will of the people.” This is a strange way to word a sentence. If there is a claim of election fraud, it means that the “will of the people” is in dispute. How else other than going to a judge can the dispute be settled? A judge might overturn the people’s will, but a judge might also overturn the will of the fraudsters on behalf of the people. It’s not a perfect system, but neither is anything else. What’s the alternative?

Reading the article carefully, it sounds like if there is clear proof of fraud, and clear proof of enough fraud to have made a win, but not necessarily proof that the fraud did in fact make a win (very hard to prove), the integrity of the election is tainted enough that it is voided and nobody wins. It is as if no election happened. How is it suppression if everyone loses equally?

This article completely fails to make its case.

It can be difficult to know the truth when different media outlets don’t even agree on the basic foundational facts of the story – or when they don’t even cover the same story so coverage can be compared. Establishing reliability by checking the sources does little good since the sources might not be reliable either. However, sometimes one can know when they are being lied to, and count this against the credibility of the source on other things.

Tucker Carlson ran a story on 27 March 2021 called US military has gone full woke, waging war on those who disagree with them, but he did a lousy job of supporting his case.

Tucker quotes Lloyd Austin warning us of extremists, but extremists are never defined. For all I know, “woke” extremists are meant to be included. Yet, Tucker finds it a problem that the word isn’t defined and uses this as evidence for the worst-case scenario, claiming that even those who merely voted for Trump would be seen as extremists, while obvious extremist groups Antifa and BLM would not be.

Tucker quotes two senators and a congresswoman speaking against white supremacy and extremism, but these are never defined either and there is no hint in these quotes that these labels are meant to be used against Trump supporters or those who would question the establishment.

Tucker quotes Ramon Colon-Lopez, who seems to make a distinction between extremists of the left and right, but the quote is too broken up and mixed with Tucker’s opinions to understand for sure what he is saying.

Tucker mentions a man fired from the DEA merely for being in DC on January sixth and a man charged with a crime by the FBI merely for witnessing some of the events of January sixth. If true, these stories certainly do make the DEA and the FBI look bad, but not necessarily the military.

Tucker links to a military PDF supporting “diversity and inclusion.” Most of it is platitudes and corporate-speak that can be interpreted any number of ways. It is not overtly “woke.”

However, this operation will be run by Richard Torres-Estrada, and this is where it begins to get a little bit interesting. According to Tucker, this man has some questionable Facebook posts, and I do not have the time to confirm this. However, assuming Tucker is telling the truth, it is disturbing that this man is in any sort of leadership position in the military.

Furthermore, Tucker quotes Lt. General Brad Webb saying that “…we are deep underway updating pilot tests and also officer candidate tests that, at its root, you know, you get a weighted score ... if you have a private pilot license. Well, that’s a socioeconomic influencer. In other words, if you’re rich enough to afford to have private pilot time, you can get a license. That ought not be weighted in such a way that you exclude, you know, various ethnic groups.” There is no mention of how pilot licenses should be weighted, and there is no overt claim that the current system excludes and has to change, but how else other than lowering standards can we be sure it won’t exclude in the future? Lowering standards just to have a more ethnically diverse military sounds awfully “woke.” It is General Webb’s statement that is the most troubling.

While Tucker does a lousy job of making his case that the military has gone “woke,” he still does a much better job than the AP did when they claimed that the military was racist. Tucker at least has dubious quotes from those in high leadership positions. The AP only had quotes from those claiming they had experienced discrimination low in the hierarchy.

It can be difficult to know the truth when different media outlets don’t even agree on the basic foundational facts of the story – or when they don’t even cover the same story so coverage can be compared. Establishing reliability by checking the sources does little good since the sources might not be reliable either. However, sometimes one can know when they are being lied to, and count this against the credibility of the source on other things.

AP ran a story on 27 May 2021 called Deep-rooted racism, discrimination permeate US military. Reading the entire text of the article, there is no evidence presented to support the headline’s use of the words “permeate” or “deep-rooted.” What exists in the article are anecdotes from a tiny number of individuals – and we don’t even know for sure that we are getting the whole truth of those incidents. In some cases, it is possible that what they experienced was not racial discrimination at all, but merely normal military discipline, or perhaps some other form of discrimination.

We need statistics, not one-sided anecdotes. The only statistics cited are that in 2020 there were 750 claims filed by service members of discrimination by race or ethnicity. This is only a tiny fraction of military personnel (about 1.4 million active status), and it does not mean any of them were valid claims. Just as racism is a real thing, so are false accusations of racism. In 2017, there were 265 from just five ships. Assuming that 2017 has similar numbers to 2020, this means that almost half of the racism is concentrated in those five ships rather than being widespread. There is a lot more to the military than five ships. Also in 2020, civilians filed 900 claims of racial discrimination and 350 claims of discrimination by skin color. Why are these claims listed separately? Do these categories overlap partially, completely, or not at all? This is not explained. Of course, it is possible that most incidents do not get reported, but the article never even tries to make that case. Also, I find it suspicious that the three numbers from 2020 (750, 900, and 350) are all round multiples of fifty. What are the chances?

The article also mentions that while there are plenty of minorities who are eligible for enlistment, there are fewer who are eligible to be officers. 73% of officers are white, while 8% are black and 8% are Hispanic. However, this should not be surprising since according to Wikipedia, 73% of the United States population is white (though that page does not count Hispanic as a race, so at least some of those whites might also be Hispanic). This means that minorities actually make up a perfectly proportionate number of officers. If the article is correct that there are a smaller percentage of minority officers than enlisted, this means that minorities actually make up a disproportionately high percentage of the enlisted! Does the military discriminate against the white majority? Even if it could be shown that whites made up a disproportionate number of officers, this would only beg the question of what the eligibility requirements are that blacks have a harder time meeting them. It might not be racial discrimination on the part of the military, but symptoms of a problem elsewhere in society.

Furthermore, the article quotes those who suggest that there is nothing wrong with minorities grouping together and self-segregating, and that anyone suggesting otherwise is the racist, yet anyone with a few years of experience in this world knows that if segregation was allowed to persist, it would be taken as evidence of racism. It puts people in a tight spot where there is no way to escape the charge. If they support segregation, they’re racist, and if they oppose segregation, they’re racist. It is impossible to take such claims seriously.

The bottom line is that the article utterly fails to support the claims imbedded in its headline.

As it turns out there are others doing my job for me by teaching rationality. Julia Galef does so on her YouTube channel, podcast, and through her organization. Subjects include exposing sphexish thinking and introduction to Bayesianism.

One thing I have always done when finding new evidence to support a hypothesis of mine is to ask if the evidence was likely to exist anyways even if the hypothesis was false. It happens so automatically, that I don’t notice it. For example, suppose I have indigestion after drinking a lot of milk. Does it mean I’ve finally become lactose-intolerant? Probably not if I normally have indigestion even when I don’t drink milk. This is the type of thing Julia teaches.

Another concept I was always aware of but had no name to put to it was “orphan beliefs.” This is what happens when you adopt a belief based on certain evidence, but then fail to update your beliefs when the old evidence is invalidated. It has happened before that I have planned on making an egg sandwich, decided that a turkey sandwich would take less effort, put the bread in the toaster, and then stupidly dropped an egg in the frying pan anyways. It seems the brain does not work like a Bayesian net. However, a habit I have had for as long as I can remember is to sit and think about things I already “know” from different perspectives, describing them from basis to conclusion and back again down multiple branches. When encountering old ideas I no longer believe, I reevaluate the entire structure. Thus, I catch these “orphan beliefs” eventually.

Logic. Love. Liberty.