Previous Month | RSS/XML | Current | Next Month


March 31st, 2024 (Permalink)

Cry "Havoc!"

An article in a popular science magazine from a couple of years ago had the title: "Climate change is wrecking havoc on microbial diversity"1. What about microbial equity or microbial inclusion? Where's the microbial DEI officer when you need her?

"To wreak" is an uncommon verb with the uncommon meaning "to cause something to come about violently"2, which is most commonly encountered in the phrase "to wreak havoc". Though you may sometimes see other things wreaked―such as vengeance, disaster, or devastation―havoc is almost always wreaked, whereas the others may be brought about in less violent ways.

"Wreck", in contrast, is a familiar word, especially when applied to automobiles. Whereas "wreak" is only a verb, "wreck" can be both a verb and a noun3. For instance, your car might be wrecked in an accident, after which it becomes a wreck, like the ramblin' wreck from Georgia Tech4. The pronunciations of "wreak" and "wreck" differ only by the sound of the vowel, which is short in "wreck"―pronounced like "reck"―but long in "wreak"―pronounced recklessly like "reek".

Given the unfamiliarity of the word "wreak", those who only read the phrase "wreaking havoc" may mistake the word for the more familiar and similar-looking "wreck". This might explain the PopSci article title, above, as well as other occurrences of the misspelling. After all, you can't really wreck havoc, though havoc may wreck other things.

Another possible source of confusion is that the two words have similar meanings: to wreck something is to damage or destroy it, and to wreak is to bring something about in a destructive way. It's no accident that the two words are so similar in meaning and so often confused since both are derived from the same Indo-European root5.

A less common misspelling is "reeking havoc"6. Since "wreak" is pronounced like "reek", those who have only heard the phrase may think it's spelled "reeking havoc", though that would mean bad-smelling havoc.

"Wreaking havoc" is one of those hackneyed "turns of speech" that George Orwell warned about when "prose consists less and less of words chosen for the sake of their meaning, and more and more of phrases tacked together like the sections of a prefabricated hen-house.7" It's a boilerplate idiom: when I read it, I have a vague sense of what is meant, but it certainly doesn't conjure up any strong images of either havoc or wreaking.

Though it's not really a metaphor, it would be best to consign "wreaking havoc" to the same bin as Orwell's "worn-out metaphors":

Many of these are used without knowledge of their meaning…, and incompatible metaphors are frequently mixed, a sure sign that the writer is not interested in what he is saying. Some metaphors now current have been twisted out of their original meaning without those who use them even being aware of the fact. … [A] writer who stopped to think what he was saying would avoid perverting the original phrase.7


  1. This is the title that shows up in the browser tab at the top of the article and in search results; the article itself, by Carla Delgado, has the following headline: "Diverse microbes are key to healthy soil. Climate change is threatening that.", Popular Science, 7/3/2022.
  2. "Wreak", Cambridge Dictionary, accessed: 3/30/2024.
  3. "Wreck", Cambridge Dictionary, accessed: 3/30/2024.
  4. Evan Scott Schwartz, "Team traditions: The history behind Georgia Tech's ramblin' wreck", Sports Illustrated, 11/4/2014.
  5. John Ayto, Dictionary of Word Origins (1991), under "wreck".
  6. This seems to be verified by Google's Ngram viewer, though some of these results are probably puns; see: "Wrecking havoc,reeking havoc", Google Books Ngram Viewer, accessed: 3/31/2024. Oddly, both misspellings seem to have peaked in the first decade of this millennium.
  7. George Orwell, "Politics and the English Language", The Orwell Foundation, 4/1946.

March 25th, 2024 (Permalink)

Who's Counting?

The first question to ask about any statistic is: Who counted it? That is, who is responsible for the statistic; who produced it? Statistics do not grow on trees. To produce a statistic, things must be counted, and people can miscount. So, any statistic is only as trustworthy as those who produced it.

The answer to this first question should prompt some follow-up questions: Were those who counted in a position to do so accurately? Did they have any conflicts of interest that would undermine their objectivity? What does their history show about their trustworthiness? Have they been sloppy or otherwise inaccurate in the past? Have they been caught lying or exaggerating? If those counting were in no position to do so accurately, objectively, or are simply untrustworthy, there's no reason to trust the count.

Case in point: No doubt people are dying in Gaza as a result of Israel's invasion, but how many? A current statistic is 32,3331, a number precise down to the ones place, which makes it appear to be an exact count rather than an estimate2.

The answer to the first question of who produced this statistic is: the Gaza Health Ministry (GHM). The GHM is the official health agency of Gaza and, since Gaza itself is controlled by Hamas, the GHM is also controlled by Hamas3. Hamas is a terrorist group that targets civilians, kills unarmed people of all ages4, takes hostages, and rapes5 and tortures its victims, including fellow Gazans6. In addition, Hamas members do not wear uniforms, hide among the civilian population of Gaza, use hospitals and schools as shelters, then exploit the inevitable civilian casualties that result for propaganda7. If Hamas is willing to use civilian deaths as propaganda, why should it not be willing to exaggerate the death toll in Gaza for the same reason?

Hamas has a strong motivation to exaggerate the death toll, and it's impossible to believe that it has any moral compunctions against doing so. Those defending the statistics put out by the GHM argue that past numbers have been in line with those gathered by the United Nations and other independent agencies8. However, we don't have to go back to earlier wars to question the GHM's death toll statistics. Remember the alleged Israeli airstrike on the al-Ahli hospital? It was GHM that almost immediately claimed that 500 people were killed in the explosion, though it later revised its statistic down a little. We later learned that almost everything about the initial reports of the explosion were wrong9: it was caused by an errant Palestinian rocket rather than an Israeli airstrike, and the rocket did not hit the hospital but exploded in a nearby parking lot10. Was the death toll also wrong?

Here's what a Human Rights Watch report on the incident had to say about that death toll:

The Ministry of Health in Gaza reported that 471 people were killed and 342 injured. Human Rights Watch was unable to corroborate the count, which is significantly higher than other estimates, displays an unusually high killed-to-injured ratio, and appears out of proportion with the damage visible on site.11

The "killed-to-injured ratio"―also referred to as the "wounded-to-killed ratio"―refers to the fact that the number of those wounded in war usually outnumbers the dead. This is common sense, but it's also borne out by history, as one historian writes: "the typical ratio of those wounded to those killed in conflict has historically hovered around the 3:1 mark.12" Due to advances in modern medicine, the ratio has tended to increase in recent decades as more casualties are able to survive their injuries. In contrast, the dead outnumber the injured in the statistics provided by the GHM.

The second question to ask about any statistic is: How was it counted? In the case of the hospital incident, the 500 number was released by the GHM only about an hour after the explosion. It seems unlikely that an actual count could have been done so quickly, so this was probably an estimate as is also suggested by the roundness of the number. However, the later downward revision of the number to 471 could have been the result of an actual count.

In contrast, independent agencies estimated between one and three hundred killed13. These, of course, are only estimates since such agencies don't have the access needed to do an actual count. In the absence of a reliable count of deaths, the best that we can do is rely on estimates from reliable sources.

There are other reasons to doubt the death toll put out by the GHM14, but the above considerations are enough to treat it with skepticism. Given that we cannot trust the GHM, what can we do? We'll probably have to wait until the war is over for anything approaching an objective count of the casualties. We'll also have to wait until Hamas has been removed from power to find out how much control it exercised over the statistics coming out of the GHM, and how much the death toll was inflated. In the meantime, all that we can do is guesstimate the death toll based on what we know; let's do so by estimating a maximum and a minimum.

First, the maximum: the GHM's statistics are not completely useless because we have every reason to believe that they're inflated. So, we can take the GHM's current statistic of around 32K as a maximum.

The minimum is only a little more difficult: Israel itself claims to have killed 12K members of Hamas15, so clearly the order of magnitude (OOM) will be at least tens of thousands. Therefore, both the minimum and the maximum OOM is tens-of-thousands, so that we can claim with some confidence that the death toll in Gaza is in the low tens-of-thousands, but that's about as precise as we can reasonably be.


  1. "Health Ministry In Hamas-run Gaza Says War Death Toll At 32,333", Agence France-Presse, 3/25/2024. The Gaza Health Ministry's own website appears to be inaccessible, presumably due to the war.
  2. One thing to be wary of in such statistics is over-precision: due to wartime conditions, there's no way that such a number can be completely precise even if it is a count and not an estimate. What is its range of measurement error? We're not told, which is a reason to be skeptical. See: Overprecision, 8/27/2022.
  3. Isabel DeBre, "What is Gaza’s Ministry of Health and how does it calculate the war’s death toll?", PBS News Hour, 11/7/2023.
  4. "14 kids under 10, 25 people over 80: Up-to-date breakdown of Oct 7 victims we know about", The Times of Israel, 12/4/2023.
  5. Farnaz Fassihi & Isabel Kershner, "U.N. Team Finds Grounds to Support Reports of Sexual Violence in Hamas Attack", The New York Times, 3/5/2024.
  6. "‘Strangling Necks’: Abduction, Torture and Summary Killings of Palestinians by Hamas Forces During the 2014 Gaza/Israel Conflict", Amnesty International, 5/25/2015.
  7. Ahmed Fouad Alkhatib, "The Origin of Hamas's Human Shields Strategy in Gaza | Opinion", Newsweek, 2/27/2024.
  8. For instance: Jessie Yeung, Duarte Mendonca, Abeer Salman & Eyad Kourdi, "UNICEF defends accuracy of Gaza death toll as horror unfolds in ravaged enclave", CNN, 11/9/2023.
  9. Yascha Mounk, "How the Media Got the Hospital Explosion Wrong", The Atlantic, 10/23/2023.
  10. John Leicester, "French intelligence points to Palestinian rocket, not Israeli airstrike, for Gaza hospital blast", AP, 10/20/2023.
  11. "Gaza: Findings on October 17 al-Ahli Hospital Explosion", Human Rights Watch, 11/26/2023.
  12. Tanisha M. Fazal, "Nonfatal Casualties and the Changing Costs of War", International Security, 11/2014.
  13. Jeremy Herb, "Between 100 and 300 believed killed in Gaza hospital blast, according to preliminary US intelligence assessment", CNN, 10/19/2023.
  14. For instance, see: Abraham Wyner, "How the Gaza Ministry of Health Fakes Casualty Numbers", Tablet Magazine, 3/6/2024.
  15. Emanuel Fabian, "IDF says 12,000 Hamas fighters killed in Gaza war, double the terror group’s claim", The Times of Israel, 2/20/2024.

March 18th, 2024 (Permalink)

How to Lie with Headlines

Some recent headlines: Trump says there'll be 'bloodbath' if he loses, ramps up anti-migrant rhetoric

Trump suggests "it's going to be a bloodbath" if he loses the election1

Trump says there will be a ‘bloodbath’ if he isn’t reelected2

Trump says there will be a 'bloodbath' if he loses the election3

This is just a selection; virtually all of the usual suspects in the establishment news media followed suit. Given the similarity of these headlines, it's tempting to think that a memo went out to these different news outlets and they all just followed directions. However, I think that temptation should be resisted; instead, I expect that this is the result of group think and herd behavior: some bellwether of the flock was the first to put out such a headline, and the others quickly followed without bothering to ask where they were going.

Some of the above reports provide enough context to debunk their own headlines. For example, beneath the CNN headline, we read:

Former President Donald Trump suggested Saturday that if he were to lose the 2024 election, “it's going to be a bloodbath for the country." The remark came as Trump promised a "100% tariff" on cars made outside the US. “We're going to put a 100% tariff on every single car that comes across the line, and you're not going to be able to sell those guys if I get elected," Trump said in Vandalia, Ohio. "Now, if I don't get elected…it's going to be a bloodbath for the country."4

Here's what Trump said in context:

If you look at the United Auto Workers what they've done to their people is horrible. They want to do this all-electric nonsense where the cars don't go far, they cost too much, and they're all made in China, and the head of the United Auto Workers never probably shook hands with a Republican before. … China now is building a couple of massive plants where they're going to build the cars in Mexico, and they think that they're going to sell those cars into the United States with no tax at the border. Let me tell you something to China: … those big monster car manufacturing plants that you're building in Mexico right now, and you think you're going to not hire Americans and you're going to sell the cars to us now, we're going to put a 100% tariff on every single car that comes across the line, and you're not going to be able to sell those guys. Now if I don't get elected it's going to be a bloodbath for the whole―that's going to be the least of it―it's going to be a bloodbath for the country, that'll be the least of it, but they're not going to sell those cars.5

Trump is difficult to understand because he tends to talk in one long, rambling, run-on sentence, frequently interrupting himself in the middle of a thought to go off on some digression, sometimes returning to finish the thought and sometimes not. So, there's often some excuse for reporters misunderstanding what he says, but little excuse in this case. In context, it's clear that the "bloodbath" he was talking about was to the automobile industry in the United States, and not some kind of civil war or rioting. Out of context, in the headlines, the false impression is created that Trump is predicting, or perhaps threatening, political violence if he loses.

Some of the news outlets quoted above have now edited their headlines to make them less misleading; for instance, the ABC News headline shown above, which is a screenshot of the original, now reads:

Trump, addressing auto industry, says there will be 'bloodbath' if he loses election6

This is better, but it still doesn't make it clear that the "bloodbath" would be to the industry, rather than that he was just "addressing" the industry when he said it.

Many people read only the headlines, and even those of us who make an effort to follow the news often do not read beyond the headlines. In fact, I usually read just the headlines, perusing the underlying articles only if the headline piques my interest. The headline-only reader will be misled into thinking that Trump was threatening a bloodbath if he's not re-elected. When I first saw these headlines, I thought instead that he was probably warning that his followers would be angry enough to riot if he were to lose again. I had to actually read one of the articles to find out that he was talking about a "bloodbath" to the automotive industry. So, even in the case when the article itself includes enough context to show that the headline is misleading, the headline will still mislead many readers.

The headlines from establishment news sources now resemble those of the old tabloid newspapers, which were notorious for promising more than the story delivered.


  1. Kit Maher & Alayna Treene, "Trump suggests 'it's going to be a bloodbath' if he loses the election", CNN, 3/16/2024.
  2. "Trump says there will be a ‘bloodbath’ if he isn’t reelected", Today, 3/17/2024.
  3. Emma Barnett & Jillian Frankel, "Trump says there will be a ‘bloodbath’ if he loses the election", NBC News, 3/16/2024.
  4. Ellipsis in the original; paragraphing suppressed.
  5. "Donald Trump talks about Bernie Moreno, his presidential campaign WCPO 9in Ohio", WCPO 9, 3/16/2024.
  6. Gabriella Abdul-Hakim, Libby Cathey & Fritz Farrow, "Trump, addressing auto industry, says there will be 'bloodbath' if he loses election", ABC News, 3/17/2024.

Recommended Reading: Leo Benedictus, "The media must stop using misleading headlines", Full Fact, 5/28/2021

Detail of Photograph
March 13th, 2024 (Permalink)

Seeing is Disbelieving


  1. Brian Melley, "Why the AP retracted the first official photo of the Princess of Wales since her abdominal surgery", AP, 3/11/2024
  2. "News agencies withdraw photo of UK's Princess of Wales", Reuters, 3/11/2024
  3. "Kate apologises for ‘confusion’ after digitally editing family photo", PA, 3/12/2024
  4. Bill Chappell & Fatima Al-Kassab, "What to know about the 'confusion' over Kate Middleton's edited family photo", NPR, 3/11/2024
  5. See:
  6. How to Lie with Photographs, 12/9/2023
  7. Liam Reilly, "Deadspin’s entire staff has been laid off after the sports site was sold to a startup", CNN, 3/11/2024
  8. Ahjané Forbes, "Family sues Deadspin after blackface accusation at Kansas City Chiefs game", USA Today, 2/8/2024
  9. See: Jonathan Turley, "Deadspin Defamation: Parents of Holden Armenta Move Toward Libel Action Over Black Face Allegation", 12/6/2023

March 3rd, 2024 (Permalink)

What Red Said

Three people nicknamed Goldilocks, Brownie, and Red were at the hair salon. Each had hair of a different color: one was a blonde, one a brunette, and one a redhead, but not in that order.

One of the three said: "I just noticed that none of us have natural hair color that matches our nicknames. Isn't that odd?"

Another replied: "That's true, but the really odd thing is that all of us are getting our hair dyed a color that doesn't match our natural hair color or our nicknames."

Red added: "Well, I'm not dying my hair brown."

Assuming that what each of the three said is true, what is the natural hair color of each and the color after the hair is dyed?

Recommended Reading
March 1st, 2024 (Permalink)

Heterodoxy Vs. Heresy & the Dog Ate My Data

* See: Illiberal Journalism & Tea with Terrorists, 1/1/2024

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 and rearranged the order of the excerpts in order to emphasize points.

Previous Month | RSS/XML | Current | Next Month