Seth's Blog

My interest in fermented food started in January, at the Fancy Food Show in San Francisco, where I had a theoretical idea: The pleasure we get from sour, umami, and complex flavors had the effect,  when it evolved, of increasing bacteria intake. This suggests we need to consume plenty of bacteria to be healthy. Three things happened at that convention that supported these ideas: (a) Someone trying to make a high-end non-alcoholic drink said he found it impossible to get enough complexity without fermentation. (b) I remembered that after a trip to Japan, I had started eating lots of miso soup. Miso (fermented soy beans) is an unusually effective flavoring agent. (c) A Stonyfield Farms employee told me that her health improved a lot when she started eating yogurt every day two years ago. (Stonyfield Farms makes yogurt.)

Recently I learned more about the health improvement. She started eating more yogurt about two years ago because she changed jobs — from an architecture firm in Boston to Stonyfield, in New Hampshire, where the employee kitchen has a refrigerator full of free yogurt. In Boston, she ate yogurt about once/week; at Stonyfield, she eats it once/day (for breakfast).

When she moved to New Hampshire, she also changed her diet in other ways. She now eats more foods that are “natural and organic” and less fast food. She doesn’t eat anything with aspartame any more; she also avoids caffeine. She eats more fruits and vegetables. Maybe the biggest change is that she eats three good meals every day instead of one meal on the run. Other changes in her life include less stress, a different atmosphere, and more exposure to nature.

In Boston, she had lots of colds and sinus infections, maybe 3-4/year. When she got sick it took a long time — 2 weeks — to get better. She also felt sick to her stomach a lot. In Boston she got mononucleosis; it took six months to completely recover. In New Hampshire, she’s had only 1 cold in the past year and it only lasted 3-4 days. No other illnesses. Another change she’s happy about is that she gained weight. In Boston she weighed about 90 pounds; now she weighs about 110. (She’s 5′ 4″ and 30 years old.)

She’s noticed that Stonyfield employees are healthier than other places she’s worked (as this study suggests). Fewer people are sick and when they’re sick they aren’t sick as long. Everyone eats the free yogurt, except the lactose-intolerant. Stonyfield yogurt contains less than half the lactose of milk; for some lactose-intolerant people that’s low enough, for others it isn’t low enough. (Stonyfield makes a soy yogurt without lactose.)

This fascinating article by Daniel Kish, a blind psychologist, describes how he navigates via tongue clicks. The echos tell him about his surroundings. I was struck by the similarities with self-experimentation:

1. Don’t wait for experts. A blind person could wait for a sighted person (“At the time I went to school, blind kids either waited for people to take us around or we taught ourselves to strike out on our own”). Just as I could have waited for a sleep expert to figure out why I was waking up too early. But I didn’t: I struck out on my own via self-experimentation.
2. Many little probes. Kish guided himself by clicking his tongue many times. Likewise, effective self-experimentation, in my experience, involves many little experiments.
3. Free. Kish can go where he wants when he wants. It costs nothing. Likewise, my self-experimentation needs no grant, and allows me to study whatever I want and reach any conclusion.
4. Learning by doing. An experiment, like sonar, involves doing something, getting feedback, and moving forward based on interpretation of the feedback.
5. Active better than passive. “Passive sonar that relies on incidental noises such as footsteps produces relatively vague images. Active sonar, in which a noise such as a tongue click is produced specifically to generate echoes, is much more precise,” writes Kish. Likewise, I’ve learned more from active experimentation than from measuring something day after day, which relies on natural variation.
6. Ancient. “The readiness with which people learn sonar suggests to me it may be an inbuilt skill,” writes Kish. Self-experimentation is a form of trial and error, which predates humans.
7. Verification in other ways. “Ultimately, students verify what they hear by touching,” writes Kish. The solutions I come up with via self-experimentation I verify by using them. Do they work? Another kind of verification is with experiments involving others.

The broad similarity is that self-experimentation, at least mine, is a way of navigating a world with plenty of important cause-effect relationships I don’t know about (e.g., what makes my sleep better or worse). Rather than continually bumping into them.

Americans spend more on health care than people in 29 other rich countries but our health is near the bottom of the list. Shouldn’t more money buy better health? This is the American health paradox. What causes it?

In the latest issue of The New Yorker, Atul Gawande, in an excellent article, tries to find out how the money is wasted. He visits a small Texas town where he finds an entrepreneurial attitude among doctors — a tendency to order more tests and do more procedures because doing so will generate more revenue. (A weakness that my own surgeon may have succumbed to.) Gawande does his best to figure out how things could be better but comes up short. He finds better systems of care — but they seem to be losing rather than winning. I think Gawande is too close to the problem he is writing about to see the really large forces at work.

In The Economy of Cities, Jane Jacobs pointed out that Marx got it wrong: The fundamental conflict in society isn’t between owners and workers, it’s between those who benefit from the status quo and those who benefit from change. There are plenty of owners and workers on both sides. The balance — or rather imbalance — of power determines what happens. The more powerful the status quo, the less change. Lack of change means lack of innovation; lack of innovation means that problems build up unsolved.

If the status quo is powerful enough, the problems get worse and worse, remaining unsolved — until the whole thing collapses. (This is what Jared Diamond failed to understand in Collapse.)  A city economy relies heavily on a single product; the resources to make that product run out (Jacobs often pointed out that nothing lasts forever), often suddenly; and the whole city dies. Manchester (cloth) and Detroit (cars) are modern examples. Was the current financial crisis due to reckless lending? Not really. That was an opportunistic infection. It was due to a problem building up unsolved: lack of affordable housing, which was due to lack of innovation in the housing industry. Lack of real solutions made room for a phony solution that, funny coincidence, benefited the powerful: rip off poor people by lending them too much money. (A new form of predatory lending that took advantage of the human tendency toward speculative bubbles.) Just like resource depletion, the phony solution worked and worked and worked, until, all of a sudden, it stopped working and the whole giant structure fell down, hurting the poor and powerful alike.

The cause of the American health paradox is American inequality. America is more unequal than other countries. Everywhere, in every country, the powerful prefer the status quo but in America the rich and elite are especially powerful relative to the poor, so the status quo is especially entrenched and innovation especially well-squelched. America has a lot of health problems building up unsolved. Perhaps the most obvious is obesity, which affects the poor far more than the rich. The further the rich from the poor — that is, the more inequality — the more the rich can ignore it. And they have: The healthcare establishment’s record on prevention and treatment of obesity is terrible. Staggeringly bad. In one tiny example, when I proposed a rat experiment to test an idea behind the Shangri-La Diet, I was denied permission by the UC Berkeley Animal Care and Use Committee: My idea couldn’t possibly be true, I was told. Had there been plenty of poor people on the committee, instead of none, I think the outcome would have been different. Problems such as depression, allergies, autoimmune disorders, and autism are likewise building up with no real progress being made. An example of a real solution is home glucose monitoring for diabetes. This came from outside the healthcare establishment — from Richard Bernstein, an engineer with diabetes.

Although The Economy of Cities was published in 1969, it has not received the attention it deserves. Lots of well-read people dislike inequality, and the connection between inequality and poor health has been documented many times, especially by Richard Wilkinson, but the Jacobian point that more inequality means less innovation means problems stacking up unsolved is not widely appreciated. In a whole book about the badness of inequality (Inequality Matters, 2005), I didn’t see this point made even once. In his New Yorker article, Gawande fails to understand Jacobs’s point that farmers didn’t invent tractors; the big improvements to American (and world) health are not going to come from doctors or anyone now powerful in healthcare. They are too wedded to the status quo. (Notice that this recent innovation in affordable housing, the nano home, comes from a car company — an Indian one.) Gawande, being a doctor, surrounded by the powerful at Harvard (where he teaches), is in a poor position to figure this out. Where will the big improvements in health actually arise? From people who benefit from change. A reasonable healthcare policy would try to empower them.

James Fallows wonders if the decline of newspapers is one effect of a much larger trend: the decline of advertising. He quotes a reader:

The real problem is, advertising is dying. It’s just pulling down newspapers along the way. Next up: TV, radio, and Google.

Advertising isn’t cost-effective, the reader says. This is becoming increasingly clear. Companies can no longer justify the expense.

I would bet a lot of money this is wrong. Advertising isn’t dying; it is moving to a more differentiated personalized form, as has happened in dozens of industries. Jane Jacobs wrote about this in The Economy of Cities: the historical flow is from artisanal production to mass production to differentiated production. An example is software. Long ago, programs were written by individuals: artisanal production. Then came software produced by large companies, such as Microsoft: mass production. Now we are entering the age of highly individualized software. The usual term is open source but open source software is enormously customizable. For example, some Tsinghua students made a version of Firefox specifically for Tsinghua students. Internet Explorer will never be as easily customized as Firefox. Which means, according to history, IE is doomed.

Fallows’s reader is wrong for another reason: The central role of advertising in human evolution. Language was the first advertising. Single words served to say (a) you had something to trade and (b) you wanted something. This is how and why language began — it facilitated trade. Language was so successful as advertising that lots of other uses evolved on top of that use, just as newspapers and magazines do a lot besides carry advertisements. Human evolution, in my view, is the story of how we became occupational specialists; by increasing trade, advertising was central to that. In the form of language, it’s been a huge force pushing evolution for the last 100,000-odd years. Given that longevity, the probability it will disappear in the next 100 years is very low.

The language evolution theory makes a prediction. Words can easily be used (a) to announce you have something (“toothpaste!”) and (b) to ask for something (“toothpaste?”). The first is push advertising; the second is pull advertising. We don’t hear much about pull advertising. But the current imbalance — huge amounts spent on one, almost nothing on the other — doesn’t make sense. Historically, both work. We use language both ways, including a lot of pull advertising. Surely most people say what they want (“I’m hungry”) more often than they say what they have to trade for it. (In China, some peddlers, such as the father of a friend of mine, do spend their day saying what they are selling.)

Based on history, I predict the imbalance will be corrected; pull advertising will become much more important. Not a brilliant prediction because it is already happening. Searching online for something you want, e.g. via Google, is a form of pull advertising. Guru.com, where you post a job you want done and wait for bids, is another example. An example that doesn’t yet exist is a free concierge-by-phone service. You call them, they help you buy something.

Here is an interesting article about the danger of microwaved food:

Comparing the blood chemistry of people after eating food cooked in conventional and microwave ovens, a dismayed Hertel explained that “blood cholesterol levels are less influenced by cholesterol content of the food than by stress factors.” . . .

So was the blood chemistry of consumers. These abrupt measurable changes included a decrease in high-density lipoprotein (good cholesterol) and a sharp rise in low-density lipoprotein (bad cholesterol) levels following the consumption of microwaved food.

The two researchers also discovered marked declines in the number of red blood cells that carry oxygen to the tissues and collect carbon dioxide, as well as in white blood cells that fight infections.

The researchers say these bad effects happen because microwave heating makes cells “easy prey for viruses, fungi and other micro-organisms.” The author adds, “bad bugs are everywhere.”

Whereas I believe the opposite: The problem with microwaved food, when there is one, is that it is too sterile. The article later reports an experiment in which E. coli. grew much faster on microwaved milk than conventionally heated milk. I interpret that to mean the microwaved milk was more sterile: less competition for the E. coli.

Note I don’t mean to say don’t use your microwave. I use mine all the time to heat water and defrost stuff. It’s the experimental data and their interpretation that interested me