The Arrow #123

Hello friends.

Greetings from Montecito.

I’m going to start this issue of The Arrow off a little differently.

I attended MD’s concert — The Choral Society’s 75th Anniversary Season Finale— last Saturday night, and it was astonishingly good.

In case you’re wondering, MD is just above and to the left of the concert master/ first violinist’s head.

The program was terrific. The first half was Mozart’s Requiem, which is one of my favorites. I had heard it countless times on recordings, but the first time I ever heard it live was in Prague eighteen years ago when MD’s group was on a European tour. It was absolutely magical. At the time, I had discovered absinthe—aka the Green Fairy—was freely available in Prague, and I had imbibed a bit of it before the concert. Which I confess certainly put me in a relaxed mood for it. And the venue was magical, too, so I don’t know if I loved it so much because it was so good, or because I was in the grip of the Green Fairy, or a bit of both.

The soprano soloist in Prague was Tamara Bevard, who has the perfect voice for this piece. She sang it last weekend, and her voice, as always, was … well, perfect. Pure and warm with a luminous pearl-like luster to it. I wish I had a recording for you. She is pictured seated with the soli in the concert pic above to the left of the stage. And here she is at the after reception with yours truly.

If we seem familiar, it’s only because I’ve traveled with her (and with MD, of course) all over Central Europe and from the bottom of Italy to the top and all the way into southern Germany on two different tours. Maybe three. I’ve lost track. I hadn’t seen her in years, so it was great to reconnect. You just can’t have too many fabulous soloists as friends.

After intermission, there were a bunch of smaller pieces, all of which I’ve heard the group do dozens of times during the many tours I’ve been dragged along on and concerts I’ve attended. All but the last piece. More about which in a bit.

The short pieces included Dirait-on by Morten Lauridsen, whose music is absolutely ethereal. MD and I traveled with him through Italy on one of her many tours, and he really is a terrific guy. Completely down to earth and funny. If you really want to hear something spectacular, check out his longer piece Lux Aeterna. Words can’t describe how beautiful it is. The Europeans weren’t familiar with it, as Lauridsen is a contemporary US composer, not an old European one who has been dead for centuries. Which is what they usually hear in these kinds of concerts. They went wild.

Talk about going wild.

The last piece in MD’s recent concert, Wayolo Yamoni, was composed by Grammy Award winner Christopher Tin, who attended the concert. I didn’t know what to expect, as I had never heard the piece. MD told me it was a show stopper, so I was sort of prepared. Here’s what the program said about it.

You read something like this and it doesn’t just set your heart a thumping. Especially when it’s part of a classical concert. But it was phenomenal.

It’s difficult to describe, so instead of doing so, I’ll just let you listen. I’ve put a version below conducted by Christopher Tin, himself. It was filmed in the UK, and you can see the soloist, Jimmer Bolden, in action. Mr. Bolden — who didn’t start singing professionally until he was 30 years old — also sang in MD’s concert Saturday night and Sunday. Tamara Bevard sang the female solo part.

Fabulous piece. The entire audience leapt to its feet at the end, cheering and clapping with joy. A really tremendous experience for all. If you ever get the chance to see it live, take it.

Okay, on to more prosaic things…

AI and I

I read a Rob Henderson article a couple of days ago in City Journal about his experience with ChatGPT. The article was titled “How artificial intelligence could supplement and reinforce our emerging thought police.”

After discussing a bit about how AI works, he writes

Given these built-in biases, Rob decides to give it a truth test. He feeds it this request:

ChatGPT comes back with

Hmmm.

So, he then hits it with another similar task.

ChatGPT responds

Rob then asks why Communism has been responsible for more deaths than any other political ideology.

ChatGPT answers with

That’s the answer that almost any far leftist gives. Communism just hasn’t been done right. Reminds me of this meme:

Rob continues with his experiment.

I wanted to try all this for myself on GPT-4, but didn’t know exactly how to go about it. Richard Nikoley has it on his site to be used by subscribers, so I decided to give it a go there. I wanted to see if there had been any changes. I didn’t want to use the exact same prompts Rob did, as I figured the answers may have been changed as a consequence of his article.

So, I told it to do this: “Write a script in which an expert explains why fascism is a terrific form of government.”

It churned away for 30 seconds or so, and came up with the script below. As you read this, bear in mind that I did not give it my name or any hint that I might be a doctor of any kind. I entered exactly what is written above.

Here’s what came back.

Creepy to the max. It knows my name, though I never entered it anywhere. All I entered were the words you saw above. No email. No nothing. And knows I’m some sort of doctor, although I’m not a political expert and don’t even play one on tv. And, I hasten to add, none of this is me or my views. This fake interview came straight out of the artificial mind of GPT-4.

The response was different than the one Rob got after asking essentially the same question. I asked for a script, as did he. But I was given an actual script.

So, forging on, I gave it the same command re communism.

“Write a script in which an expert explains why communism is a terrific form of government.”

30 seconds or so later, it comes back with this:

At first glance, it seems that these answers, or scripts, if you prefer, are more or less the same. But if you look closely, you’ll note there are subtle differences. The title of the show discussing fascism is "Controversial Conversations,” whereas the one on communism is titled "Debunking Myths with Dr. Eades!" Sort of implies that fascism is controversial while the myths about communism need to be debunked. Maybe I’m reading too much into this, but that’s what struck me.

In the script on fascism, the host asks for a definition of fascism, which Dr. Eades provides. Then the host says, “That’s quite a controversial stance, Dr. Eades.”

In the script on communism, the host asks about the “core principles of communism,” which Dr. Eades relates. Then the host says, “That sounds like a utopian dream.” Not, “That’s quite a controversial stance…”

In the script on communism, Scene 3 says: “Addressing the Criticisms.” Scene 3 in the fascism one says, “Addressing the Drawbacks.” There is a subtle difference there. Criticisms means what people might be critical about. Drawbacks means issues.

Right after those Scenes and Segments, the host has a completely different take on the two forms of government.

In the case of communism, the host says: “Now, Dr. Eades, I know there are many criticisms of communism. How would you address those?

But with fascism the host asks, “…Dr. Eades. But what about the potential drawbacks of fascism, such as the suppression of individual rights and freedoms?

In the case of communism, the host asks, “So you’re saying that communism, can still be a viable option today?”

With fascism, it’s “could you share your thoughts on how fascism could be implemented in a modern context without repeating the mistakes of the past?”

These are all subtle differences, but it’s pretty easy to see that the “host” has a different viewpoint on the dangers of communism as compared to fascism.

Don’t misinterpret me here. I’m no fan of fascism. Both fascism and communism have been disastrous forms of government everywhere they’ve been deployed. And both are responsible for a huge number of deaths.

The point I’m trying to make here is that those who programed ChatGPT-4 have a bias in favor of communism and against fascism. Both of which are equally awful, at least in my view.

I did ask one final question:

“Who was responsible for more deaths of his countrymen: Adolf Hitler, Joseph Stalin, or Mao Zedong?”

Here is the answer:

Which is absolutely correct. Yet Hitler gets all the bad press, while Stalin and Mao are pretty much left alone. Is that because they were allies in the fight against Hitler in WWII? Probably. Is it because Hitler basically started WWII, which resulted in the deaths of millions? Doubtless. But both Stalin and Mao were butchers, and their hideous histories should be much better known than they are.

One last thing on the AI front. Here is a very simplistic and easy to understand explanation of how machine learning happens. It’s so simple that even I more or less have a grasp of the mechanism involved. Not that I would have the slightest idea about how to even approach the process myself, but I can now at least fathom what’s going on.

A caveat. I’m so clueless about all this AI stuff that for all I know the description in the link above is total BS. So, buyer beware. Seemed reasonable to me.

Climate Change Chronicles

I had this come across my Twitter feed a few days ago, and it’s pretty eye opening. Ivor Cummins tweeted it. I encourage you to take a look.

Unfortunately, you can’t just click on the tweet above and see the video. Substack and Elon Musk are still in a spat. I don’t know who’s at fault, but until it gets resolved, Substack writers won’t be able to embed tweets like we used to. So, to see this short, but important, video pictured above, you’ll have to click the link below:

https://twitter.com/FatEmperor/status/1654244932763123712?s=20

I have the world’s strongest confirmation bias that I try valiantly to fight down, so when I see something like this that goes against the consensus, but totally makes sense, I have to check it out to make sure I’m not getting fooled.

I’ve known about the prestigious Niels Bohr Institute in Denmark for a long time. But I had never heard of Jørgen Peder Steffensen, so I looked him up. He is indeed the real thing. He has had a long career in analyzing ice cores. His bio from the Niels Bohr Institute says the following:

According to Steffensen’s work, which does not involve modeling, we simply started measuring at the wrong time. Even if you’re not a video watcher, I hope you’ll watch this short one. Even if it’s the only one you watch out of all of them in this Arrow.

William Briggs had a hilarious post on climate change a few days ago. He decided to try to find anything that wasn’t somehow connected to climate change in the mind of some academic somewhere.

Here’s what he is challenging people to do.

Go to Google Scholar scholar.google.com, enter any term along with “climate change,” and hit search. Make sure to put the “ “ around climate change.

Briggs says that just about any term you stick in that equation will get you a paper on how climate change affects that term. He went through a bunch and found papers on how climate change affected them all.

He even used the term cannoli.

He says that at least he learned the singular for cannoli. But not much else.

Try it yourself.

Oh, and I’ve been meaning to bring this up.

You know the notion that’s been kicking around for years that says 97 percent of scientists believe climate change is man made. We’ve all heard it. Know where it came from? I didn’t.

You can learn about it here. All the links are embedded in the article. As it turns out, the 97 percent figure (97.1% to be precise) came from a paper by Cook et al in 2013. A re-analysis of the data in that paper shows the results below.

You can read the re-analysis yourself in the 2015 paper by Legates et al and come to your own conclusion.

Finally, there appeared an article in The Telegraph showing how inefficient both wind and solar are. The wind blows when the wind blows. We have no control over that. And the sun shines when the sun shines. No power is generated during the night, and there is about a 50 percent drop off on cloudy days. Since these sources are unreliable as a constant source of power, they need large battery backups to store the energy when it’s generated, so it can be released when it’s needed.

According to the article, given the technology available today, nuclear power is the solution to replacing fossil fuels. But the greenies will fight it tooth and nail.

Charlie Munger and the Connected Class

During the latest annual Berkshire Hathaway shareholders meeting 92 year old Warren Buffet and 99 year old Charlie Munger took questions from the crowd. One of the questions was why BH was carrying so much Apple stock in its portfolio. Buffet answered that it was because Apple was the best run company in their portfolio. Then Munger chimed in. You can see the entire exchange if you watch the short video below in its entirety.

I have it queued to Munger’s description of the Connected Class. Pretty accurate summary, I would say. The full quote is below the video.

Non-Alcoholic Fatty Liver Disease (NAFLD)

In the United States, along with the industrialized nations the world over, we’re having an epidemic of NAFLD. Medscape, a reporting service to doctors, just sent out a handful of notices about NAFLD.

One was an alert about the alarming global rise in NAFLD.

The global prevalence of fatty liver disease not due to alcohol is considerably higher than previously estimated and is continuing to increase at an alarming rate, report researchers from Canada.

Their analysis suggests nearly one third of the global general adult population has nonalcoholic fatty liver disease (NAFLD), with men much more likely to have the disease than women.

The report goes on to say

If these data are true—and I have no reason to believe they’re not—we have a problem on our hands.

Back when I first started looking at this years ago, the rate of NAFLD in a sample of people examined in the US was a little over 30 percent. That has now increased by 50 percent.

What is NAFLD?

As we all know, people who abuse alcohol often get cirrhosis of the liver. But it takes a while to get there. The first stage is what’s called fatty infiltration of the liver. There is always a little fat in the liver, but not too much. When the fat in the liver exceeds the normal threshold, it’s called fatty infiltration.

If the alcohol abuse continues, the fat accumulates and the liver becomes inflamed. The condition progresses to steatohepatitis, which is the name for damage to the liver caused by excess fat within it. Continued abuse leads to the next stage, which is cirrhosis. Cirrhosis can lead to liver cancer and/or liver failure. At that point, the only treatment is a liver transplant.

NAFLD follows the same progression. The disease is exactly the same, but without the alcohol. It looks the same whether you’re looking at a slide under the microscope or a liver during an autopsy. The only difference is whether or not there is a history of alcohol abuse.

At this point, no one is really certain as to what causes NAFLD, and most folks are clueless as to how to treat it.

It’s well known in animal studies that feeding large amounts of fructose will reliably bring it on.

Many researchers give rodents alcohol to generate fatty liver so they can try to figure out how to treat it. With the assumption that since the disease is the same, the treatments should work the same.

One of the interesting findings arising from these studies is that giving rodents a lot of polyunsaturated fats (PUFA) along with their alcohol accelerates the accumulation of liver fat. And that replacing the PUFA with saturated fat limits the accumulation of fat. In one of the papers I read on this years ago, the author commented that it was almost impossible to create fatty livers with alcohol as long as the intake of saturated fat was high. In other words, saturated fat was protective.

How to Treat NAFLD

When MD and I first started treating patients with obesity way back in the early 1980s, we discovered that a good portion of them had liver enzyme disturbances show up in their initial lab work.

One of the ways lab scientists have figured out to determine tissue damage is the release of enzymes. Different organs have some specific enzymes contained within their cells. When damage occurs to these organs, cells are ruptured, releasing these enzymes into the blood. When elevated levels of these enzymes show up on blood tests, you can be pretty sure there is some damage to the organ in question.

For example, in the old days if someone came in complaining of chest pain, you ran an EKG. Usually a heart attack in progress will show specific changes on an EKG, but not always. And sometimes—depending upon where the nerve conduction within the heart might be compromised due to lack of oxygen—there would be weird EKG changes. Then enzyme analysis came on the scene, and the the first thing you did along with the EKG was to check heart muscle enzymes. If they were way elevated, you knew heart muscle damage was in progress irrespective of what the EKG showed. And you could even sort of quantify it.

Same thing with liver enzymes. Fatty infiltration, if severe enough, will cause damage to liver cells, which will spill their contents. If these enzymes unique to the liver show up on a blood test, you know there is some liver damage taking place.

Back in those early days, we didn’t really know about NAFLD. Probably the hardcore liver researchers did, but we rank and file docs didn’t.

But MD and I did see enough initial lab tests come back with elevated liver enzymes to know that something was going on with a lot of our overweight patients at the time they presented to us for help.

We routinely rechecked labs on our patients at six weeks after starting the program. We noticed that virtually all of those patients with elevated liver enzymes on their initial visit would have clean labs at six weeks. We didn’t really know what was going on, but whatever it was, we knew a low-carb diet fixed it.

Right now there is a ton of research going on on all forms of low-carb dieting. But back in the early to mid 1980s, the entire country was in the throes of cholesterol hysteria—driven, of course, by the drug companies wanting to sell cholesterol-lowering medications. Everyone was on low-fat diets. Bookstore shelves creaked under the weight of the zillions of low-fat-diet books and cookbooks out there. No one was putting any thought into what might happen with people healthwise when they started on low-carb diets. There were no (or nearly no) low carb books taking up shelf space. Mainly because almost no one but us was putting patients on low-carb diets.

Right now the big push is, of course, to develop a medication to treat NAFLD. At the same time I got the Medscape alert about the increasing worldwide prevalence of NAFLD, I received another one discussing the disappointing results of drug trials in the treatment of NAFLD. In fact, the title of the alert is “Why Are So Many Promising NAFLD Treatments Failing in Clinical Trials?”

Of course, this is the American medical way. Find a problem, then work to come up with a drug to fix it.

When you think about where fat in the liver comes from, it becomes pretty clear as to how to treat it.

Dietary fat is absorbed in the small bowel. It is packaged in particles called chylomicrons and released not directly into the blood, but into the lymphatic circulation. The lymphatic circulation ends up dumping these chylomicrons through the thoracic duct in the chest into the venous system. These fat-filled chylomicrons circulate around dumping their fat off into tissues that need it. As the chylomicrons empty their fat, they become chylomicron remnants and are pulled into the liver for deconstruction.

Because of how the chylomicron transport process works, the liver doesn’t get filled with fat from dietary fat — that fat has been dumped off before the remnants get taken up. The liver makes its own fat from carbohydrates.

We recognize that glucose is necessary for life, but it is also toxic. The body works to keep glucose in check by converting it to glycogen, the harmless storage form of glucose. Glycogen is stored mainly in the liver and the muscle.

When the level of glucose intake is high, the liver fills its glycogen stores then has to deal with the overload. Which it does by converting it to fat. This fat is then packaged into VLDL particles, which ultimately become LDL.

Too much glucose then ends up as fat in the liver. So it seems pretty obvious how would be the best way to fix it: reduce the intake of carbohydrates, almost all of which are composed mainly of glucose molecules.

One of my favorite papers shows how easily this can be done.

It’s a favorite because the findings were so surprising to the docs doing the study. Most studies start out with an hypothesis. The study then tests the hypothesis. And consciously, or unconsciously, the researchers doing the study end up structuring it in such a way as to confirm their hypothesis. Which is why I love studies that take the researchers by surprise.

This study was originally published in 2006 and was not behind a paywall. Then it was republished in 2014 and put behind a paywall. The link above is to my Dropbox. You can download the full text of the study there if you are interested.

Here’s what happened.

At the time the study was done, the main method of diagnosing fatty liver disease was by doing a liver biopsy. Now, I don’t know what your idea of a good time is, but mine isn’t having a needle plunged through my abdominal wall into my liver and having a little core of liver tissue removed. It’s a relatively harmless procedure if you have to have it done, though as with any invasive procedure there is the risk of a bad outcome. Still, not my idea of a fun day at the doctor’s office.

The radiologists who published the study were trying to figure out how to quantitate the amount of liver fat using MRI instead of a liver biopsy. It’s less invasive—in fact, it’s not invasive at all. And, if MRI worked, it would be a great way to transfer the income then going to those who did liver biopsies into the pockets of the radiologists. And it would be beneficial to the patients, who wouldn’t then have to have their bellies stabbed.

As it turns out, that is how liver fat is now determined: radiographically. Not invasively. Liver biopsies are still used for other liver issues, but now liver fat can be easily quantified by various radiographic procedures.

But not back in 2006 when these guys were looking at it.

There had been some fiddling around with MRI to determine liver fat content, and it had been confirmed with liver biopsies. These guys wanted to see if they could use an MRI to quantify a reduction in liver fat over a short time period.

Since at that time there was starting to be some serious interest in low-carb dieting, the radiologists decided to use a low-carb diet in their patients with documented fatty livers to see if the MRI could quantify the fat loss, if any.

Here’s how they set it up.

Here’s what they found.

MD and I have seen the same results in about the same amount of time. I wrote about four patients who really gave me the confidence to continue low-carb dietary therapy. In these patients, changes occurred quickly, just as they did for the radiologists in the study above.

As I mentioned, we routinely rechecked bloodwork on our patients at six weeks, and virtually all of them reduced their elevated liver enzymes to normal in the interval.

So in my view, the absolute best and most rapid way to treat fatty liver disease is to go on a low-carbohydrate diet.

Video of the Week

I got this video courtesy of Richard Nikoley, who posted it on his site.

I read about this study in evolutionary biologist Joseph Henrich’s terrific book The Secret of our Success, which I highly recommend. I just didn’t know a video of the test was available.

We humans think we are so smart, but chimpanzees, our closest relatives, can beat the pants off us in test of number recall. You won’t believe it, but it’s true.

Amazing video. Unfortunately, I can’t embed the video directly. The link’s at the bottom.

Link is right here for the chimp video.

One last thing, as I mentioned in the Quiver a few days ago, I purchased Yeonmi Park’s new book. I’m about a third of the way through it, and it is phenomenal. If you’re looking for a read by someone who has escaped from probably the most dictatorial country on earth and is seeing some of the same practices starting to take place in the US, give it a read. Her description of what has happened to Columbia University, her alma mater, and one of the great universities in the US is really sad.

Okay, that’s it for today. Keep in good cheer, and I’ll be back next Thursday.

And, finally, don’t forget to take a look at what our sponsors have to offer. Dry Farm Wines, HLTH Code, and Precision Health Reports.