Recommended Reading

Previous Month | RSS/XML | Current | Next Month


May 31st, 2023 (Permalink)

Freedom of the Press & the Two Minutes Hate

The first recommended reading this month is a preface by George Orwell to his novel Animal Farm that, for some reason, was not included when the book was published. Instead, the preface was lost for decades and only rediscovered in 19711. Animal Farm was finished in early 1944, during World War Two, and Orwell spent the next year and a half trying to get it published. The following preface explains why he had so much difficulty finding a publisher. It's worth reading today because, though the details are different, the general situation is so familiar to us. Change the examples and it could have been written yesterday.


  1. Bernard Crick, "How the essay came to be written", The New York Times, 10/8/1972.
  2. According to Dorian Lynskey, "[Peter] Smollett was almost certainly the man who advised Jonathan Cape [the publisher] to drop Animal Farm when he was head of Soviet relations at the Ministry of Information." In the last year of his life, Orwell informed the British government's Information Research Department that Smollett was "some kind of Russian agent". In 1980, after his death, it was revealed that Smollett had indeed been a Soviet agent. See: The Ministry of Truth: The Biography of George Orwell's 1984 (2019), p. 204. See also: Peter Foges, "My Spy", Lapham's Quarterly, 1/14/2016.
  3. Orwell's ellipsis.
  4. Is this still true?
  5. George Orwell, Nineteen Eighty-Four (1949), chapter 1.
  6. Erica Byfield & Checkey Beckford, "Bellevue Worker in Citi Bike Fight Video Has Receipts Showing She Rented It: Lawyer", NBC New York, 5/19/2023.
  7. See: "Get to know Citi Bike", Citibike, accessed: 5/30/2023.
  8. See: Medice, Cura Te Ipsum, 6/1/2019.

Disclaimer: I don't necessarily agree with everything in these articles, but I think they're worth reading as a whole. In abridging them, I have sometimes changed the paragraphing.

May 28th, 2023 (Permalink)

Psychic Reading?

Prophecy is the most gratuitous form of error.1

On the last day of last year, in an entry on predictions for that year attributed to Nostradamus, I wrote:

A fellow named Mario Reading published a book on Nostradamus in 2006 which supposedly predicted that Queen Elizabeth II would die "circa 2022"―I write "supposedly" because I do not have this book and so am in no position to check it. Not that I doubt that he did make such a prediction, because it was a pretty safe bet given her advanced age. Moreover, the use of the word "circa" added a degree of vagueness to the prediction: if he had meant to predict that she would die this year, and only this year, then he would have written "in 2022". As it is, if she had died last year, or next year, he still could have claimed to have gotten the prediction right, as those years are "circa" 2022. So, at the very least, the prediction covered the years 2021-2023. While that's not quite a sure thing, if the prediction had failed we would simply have heard nothing about it. Sadly, Reading died in 2017, so he can't enjoy his fifteen minutes or the money from the spike in sales of his book.2

I have now obtained a copy of a later edition of Reading's book that was published in 20153. After a brief biographical note on its subject, the book is arranged chronologically in chapters ranging from 2001 to, believe it or not, 7074, when the world is supposed to end. Clearly, every chapter prior to 2015 is written with the benefit of hindsight, so that only with 2016 do the predictions get interesting.

Here's what Reading has to say about Queen Elizabeth II in the chapter on 2022: "The future King Charles III and his Princess Consort, the former Camilla Parker Bowles, will find themselves faced with a constitutional crisis on the death of Charles's mother, Queen Elizabeth II.4" In a later passage, he adds: "…Queen Elizabeth II will die, circa 2022, at the age of around ninety-six….5" So that is where the "circa" comes from, though the chapter heading would lead one to expect the event last year. As I noted in the comment quoted above, the success of this prediction was aided by the inexactness of "circa", and either 95 or 97 years of age would surely have counted as "around" 96.

As I mention in my entry on prophecy6, the more predictions a would-be prophet makes the greater the likelihood of a lucky hit. What's never mentioned in tabloid reporting is that Queen Elizabeth's death is just one of many predictions in the book for the years 2016-2022. How many of these did Reading/Nostradamus get right? That you haven't heard of any of them should tell you the answer: none.

Here's a list of Reading's failed predictions for the last seven years:

Date Prediction Page
2016 A "massive" flood in the Aegean Sea is so "catastrophic" that "no one on dry land will be safe". 195
2017 The Pope causes a scandal by forcing out his presumed successor. 203
The Pope defrocks a number of "schismatics". 210
Al-Qaeda attacks a U.S. naval base which is then "restored to Arab control". 213
2020 A new Pope. 224
2021 A biological terrorist attack on the French town of Agde. 228
A new volcanic island rises from the sea. 232
2022 The Church of England is disestablished. 235
King Charles III abdicates and Prince William becomes King 240

In addition to predicting many events that never happened in the years since the book was published, Reading/Nostradamus failed to predict the most important events that did happen. For example, where is the prediction of a worldwide pandemic in the years 2020-2022? One of the most important events, if not the most important, of those years is missing. Also, where is the invasion of Ukraine by Russia? Where is the election of Donald Trump in 2016 and of Joe Biden in 2020? Where is the return of inflation after a forty year absence?

Reading is likely to become the Jeane Dixon7 of the 21st century. Dixon, who was the best-known "psychic" of the last century, became famous because of one lucky hit―the assassination of President Kennedy―out of many hundreds of failed predictions.


  1. George Eliot, Middlemarch, chapter 10.
  2. When Prophecy Fails: 2022 Edition, 12/31/2022.
  3. Mario Reading, Nostradamus: The Complete Prophecies for the Future (2015). All page citations are to an electronic copy of this book.
  4. Reading, p. 234.
  5. Reading, p. 238.
  6. See Rule 5 in: How to be a Prophet for Fun and Profit, 6/26/2022.
  7. For more on Dixon, see: How to be a Prophet for Fun and Profit, 6/26/2022.

May 26th, 2023 (Permalink)


Q: I have a suggestion to add to your collection of fallacies. I'm not sure whether it's not completely contained in other fallacies but I didn't find a similar one. I call it the Pharisee Fallacy or Meta-Fallacy. It's kind of a reverse Fallacy Fallacy: Believing you are correct just because you haven't made any of the logical fallacies (or cognitive biases). This of course could be because we might well never have a complete list of all fallacies, or because our sources are just simply incorrect or because of some other factors.

The point of this fallacy is that knowing and following rules (in this case avoiding the logical fallacies) doesn't make you "good" (in this case correct). The rules are there to point out a flawed functioning of the human consciousness. But there is no need for rules if one's consciousness works well.

I believe the way to infallibility is not through keeping a tab on each and every fallacy but through being more conscious: not lying to ourselves; letting our inner, deepest, unconscious thoughts register so that we see the intention of our more surface level, conscious thoughts.

It might seem like I'm against collecting fallacies, but I hope it's clear that I don't believe being conscious in such a deep level is easy to do and therefore I think that having a list of fallacies is a good way to remind us of the level of consciousness we should strive for again and again.―Márton Kenessey

A: You're correct that, as a general rule, it does not follow from the fact that an argument does not commit one of the named fallacies that it is, thereby, a "good" argument. To be good, an argument needs to start with true premisses and use cogent reasoning to reach a conclusion. In other words, there are two main ways that an argument can go wrong: either the reasoning is uncogent or at least one premiss is false. If you start from a false premiss, even the best reasoning may lead you astray.

Logical fallacies are concerned primarily with the cogency of reasoning, and cogency isn't everything: good arguments are also sound. If an argument does not commit any fallacy then it may be cogent, but if it has even one false premiss then it is unsound. The truth or falsity of premisses is not generally a question for logic, but for other sciences. So, the fact that an argument does not commit a fallacy may suggest that it is cogent, but it would still be a bad argument if it has one or more false premisses.

For the above reasons, it would be a mistake to claim that an argument must be good because it doesn't commit any of the known fallacies. However, it doesn't seem to be a common claim and, therefore, not the sort of bad argument worthy of a named fallacy. At least, I don't recall ever seeing an example of this type of argument―if anyone knows of one, please let me know. In any case, you are correct that there is no such fallacy in the Fallacy Files Taxonomy, and I would hesitate to add it until I see some evidence that it is common.

As to your claims about consciousness, psychological research shows that it's normal for people to make certain types of error. These errors include traditional logical fallacies and the more recently identified cognitive biases. So, even perfect consciousness of the workings of one's own mind would not be enough to avoid bad reasoning.

How do you know whether your consciousness works well if you don't know what mistakes it might make? I agree that being aware of one's own thinking and honest with oneself is a necessary part of good if not infallible reasoning, but it's not sufficient. You also need to know what biases to look out for and what pitfalls to avoid, and the only way to do that is to learn how to recognize them.

Infallibility is, I think, an unattainable goal, but what is attainable is improvement, and knowing about logical fallacies is one way to attain it.

May 18th, 2023 (Permalink)

Junk Statistics

In a recent episode of her show, Megyn Kelly claimed:

There are millions of [pedophiles] in the United States. Millions. There are about a million in custody, and something like ten to forty percent of pedophiles get caught; the vast majority do not get caught.1

I'm sorry to be the bearer of good news, but there are less than two million people incarcerated in the U. S.2 If about a million pedophiles were in custody, that would mean that more than half of those inmates were incarcerated for sex crimes against children, which is highly unlikely.

So, how many pedophiles are incarcerated in the United States? It's not easy to find this statistic, especially since people in jail and prison are not categorized as "pedophiles" as far as I know, and pedophilia is not itself a crime. Moreover, sexual crimes against children are often lumped in with adult rape and other crimes under labels such as "sex crimes". That said, I did find a statistical break down in terms of type of offense which indicates that 12% of the federal prison population is imprisoned for "sex offenses"3. I think it's safe to assume that the percentage of inmates incarcerated for sex-related crimes in state prisons and local jails is similar. Of course, since that percentage includes sex crimes against adults, at best it sets an upper limit on the percentage of those imprisoned for pedophilia-related offenses.

Let's be generous to Kelly and assume that 12% of the people incarcerated in the U. S. are there for sex crimes against children: that means that the number of pedophiles in custody is around 200K, not a million. Let's also be generous by assuming that only 10%, instead of 40%, of pedophiles are incarcerated; that would mean that there are about two million total. I suppose that means there are millions of pedophiles in America, as Kelly claimed, but barely. However, to get up to two million we had to assume that all sex offenses involved children, and we know that's not true. We also assumed that only ten percent of pedophiles are in custody, which was the lowest percentage given by Kelly―where did she get that?

So, even if we bend over backwards to be fair to Kelly, it seems unlikely that there are "millions" of pedophiles "prowling schools and kids' clubs and kids' charities right now" as she goes on to say1. Of course, even a million pedophiles is worrisome, but let's not get hysterical and scare parents into withdrawing their children from schools or clubs for fear of prowling child molesters.

Where did Kelly get the statistic that a million pedophiles are in custody? I don't know the answer, but she probably got it from an activist group. I suspect that it's a "Goldilocks number", that is, one that is big enough to frighten people but not so big that it attracts skeptical scrutiny4.

We survived a previous moral panic about pedophiles, namely, the so-called "satanic panic" of the 1980s5. However, hundreds of innocent people were incarcerated for nonexistent crimes, and thousands had their reputations permanently destroyed by false accusations. There are signs of a return of this panic6, and journalists such as Kelly should not contribute to it. Fighting the sexual abuse of children is a worthy cause, but let's fight it with the truth and not noble lies.


  1. "Megyn Kelly Fires Back After Charlize Theron Drag Queen Comments Backlash, and Reality of 'Grooming'", YouTube, 5/16/2023.
  2. "U.S. Jail Population Increased While Prison Population Decreased in 2021", Bureau of Justice Statistics, 12/20/2022.
  3. "Offenses", Bureau of Prisons, 5/6/2023.
  4. See: The Goldilocks Number, 1/23/2022.
  5. Alan Yuhas, "It’s Time to Revisit the Satanic Panic", The New York Times, 3/31/2021.
  6. See: The Return of a Moral Panic, 1/31/2022.

May 5th, 2023 (Permalink)

How to Solve a Problem*: Think Backwards

To get the most out of this entry, give the following puzzle a shot. Don't give up too easily, but don't worry if you can't solve it as I'll show you one way to do it later.

Puzzle: What Stays in Las Vegas

One hot summer evening, Larry decided to try his luck on the strip in Las Vegas. He doubled his money playing blackjack at The Alexandria casino, but then lost $50 on roulette. Feeling like his luck had run out there, he went to the casino next door, Carnival Carnaval. This time, Larry doubled his money playing craps, but lost $50 on a slot machine. Finally, Larry went to a casino down the street, Circus Maximus, where he doubled his money playing baccarat, but lost $50 on keno. Checking his pockets, he discovered that he was tapped out! He hadn't so much as a penny left. How much money did Larry start the evening with?

You might be able to solve this puzzle by trial and error, but it would be a long and tedious procedure. Alternatively, it could be solved with algebra, so don't complain that you never have occasion to use your high school algebra. However, if you don't remember your algebra very well, it can also be solved without algebra by solving it backwards, that is, by starting at the end and working back to the beginning.

Notice that what we're asked to discover is how much money Larry had at the beginning of the evening―that's our unknown―but we know that he ended up with nothing. A good rule of thumb for problem solving is to start with what you know; in this case, that's what Larry ended up with: $0. Therefore, it's a good idea to start at the end and work backwards. Let's give it a try.

The same thing happened to Larry in each of the three casinos he visited: he doubled the money he came in with then lost $50. In mathematical terms, if we let x represent the number of dollars he entered a casino with, then what he left with was 2x - 50 dollars. So, if he left the casino with y dollars, then 2x - 50 = y. Thus, if we know y, which is what Larry left a casino with, then we can figure out what he must have had when he entered.

We're told that Larry's visit to the last casino, Circus Maximus, cleaned him out. So, if losing $50 cleaned him out, then doubling his money must have left him with $50, which means that he entered the casino with half that, $25. That is, if the amount that he entered the casino with is x, then what he left with is 2x - 50 = 0. So, x = 25.

That takes care of the last step in the puzzle. What we do now is continue this process with the preceding steps. We just figured out that he had $25 when he left the second casino, Carnival Carnaval. So, the amount that resulted from doubling his money in that casino was $75, which means that he entered with half of that: $37.50.

As we've just seen, Larry exited the first casino, The Alexandria, with $37.50 in his pockets after losing $50, which means that he doubled his money to get $87.50. Thus, the amount that he entered the first casino with was half of that: $43.75, which is the solution to the puzzle.

Wasn't that easy? Once you know how to approach the puzzle―namely, from the end―it's just a matter of working step-by-step until you reach the solution.

Now that we have another tool in our problem-solving chest, here's another puzzle to practice on:

Puzzle: Tickets to Ride

Smith, Jones, and Robbins wanted to ride the new rollercoaster together, but tickets to the amusement park cost $24. The three friends each checked their pockets for cash but some came up short. So, Smith gave Jones and Robbins as many dollars as each had, but some of the friends were still short. Then, Jones gave Smith and Robbins as much as they already had, but some still didn't have enough for tickets. Finally, Robbins gave Smith and Jones as much as they already had, and each friend now had exactly enough for a ticket. How much money did each one originally have?

* For the previous entry in this series, see: How to Solve a Problem: Contraction, 4/6/2023

May 3rd, 2023 (Permalink)

Rain, Rein, Reign

Here's another homophonic trio, that is, three words that are pronounced the same but have pronouncedly different meanings. An article from last year about the history of the Ku Klux Klan contained the following sentence: "As explained by an 1884 book written by a founding member of the Klan, this meeting bound the 'isolated dens together' with 'unity of purpose and concert of action' to supposedly reign in rogue Klansmen that had turned violent toward black people just a year after the group's founding.1"

To "reign" is to rule2, whereas a "rein" is a lead for controlling an animal such as a horse and, as a verb, "to rein" means to use a rein or reins3. Literally, the phrase "rein in" means using reins to slow a horse or other animal, or to bring it to a halt; figuratively, "rein in" can be used of any exercise of control, especially to halt or diminish unwanted behavior. While "rain" is pronounced the same, I've never noticed it mistaken for either of the other two, perhaps because it's such a familiar word. So, obviously, the authors of the article meant that the meeting was supposed to rein in the rogue Klansmen, not "reign" them in, which makes no sense.

Surprisingly, no spelling and grammar checking programs that I tried flagged the word "reign" as a mistake in the example sentence, though one checker did complain about the split infinitive "to supposedly reign". While I'm no fan of split infinitives4, there's at least an argument for this one, whereas there's no argument for a superfluous "g" in "rein". Another checker suggest that the first "by" should be replaced by "in", which would be a stylistic improvement but "by" is not a grammatical error. Despite the checkers' silence, there is one actual grammatical error in the sentence―see if you can find it5.

The main point of this series of entries on frequently misspelled words is that you shouldn't rely entirely on computer programs to edit your copy. Such programs are no substitute for a human editor, and they won't be until they have human-level artificial intelligence which, if it's even possible, is still decades away. This doesn't mean that such programs are useless, but that they should be used with real intelligence. Current spell-checking programs can only be relied on to catch misspellings that do not produce an English word, such as transpositions of letters due to overly-rapid typing. This frees up the human editor to concentrate on those misspellings that won't be caught, such as substituting "reign" for "rein".

So, don't allow your spell-checker to reign over you, and rein in any tendency you have to insert "g"s where they don't belong.

Update (5/4/2023): After posting this entry, I came across the following headline in a book I just happened to be reading:

U. N. Peacekeepers Land in Liberia to Reign in Violence6

Some peacekeepers!


  1. Anna Agresti & James D. Agresti, "Fact Checkers Cover for Democratic Party’s Sordid History With the Ku Klux Klan", Just Facts Daily, 7/29/2022
  2. "Reign", Cambridge Dictionary, accessed: 5/3/2023
  3. "Rein", Cambridge Dictionary, accessed: 5/3/2023
  4. See: Think Twice Before Splitting an Infinitive, 8/11/2021.
  5. "That" should be "who" since, no matter how bad they might be, rogue Klansmen are still people.
  6. Richard Lederer, The Revenge of Anguished English: More Accidental Assaults Upon Our Language (2005), p. 98.

Previous Month | RSS/XML | Current | Next Month