It’s a no brainer. Just play safe
It is a common argument in many different contexts. In its strongest form, the claim is that the choice being argued for is unambiguously right, eliminates the possibility of a bad outcome at no cost. More plausibly, the claim is that one can trade the risk of something very bad for a certainty of something only a little bad. By agreeing to pay the insurance company a hundred dollars a year now you can make sure that if your house burns down you will have the money to replace it.
Doing that sometimes is possible but, in an uncertain world, often not; you do not, cannot, know all the consequences of what you are doing. You may be exchanging the known risk of one bad outcome for the unknown risk of another.
Some examples:
Erythritol
Erythritol was the best of the sugar alcohols, substitutes tolerably well for sugar in cooking, has almost zero calories or glycemic load. For anyone worried about diabetes or obesity, using it instead of sugar is an obvious win. Diabetes and obesity are dangerous, sometimes life threatening.
Just play safe.
I did. Until research came out offering evidence that it was not the best sugar alcohol but the worst:
People with the highest erythritol levels (top 25%) were about twice as likely to have cardiovascular events over three years of follow-up as those with the lowest (bottom 25%). (Erythritol and cardiovascular events, NIH)
A single article might turn out to be wrong, of course; to be confident that erythritol is dangerous requires more research. But a single article was enough to tell me that using erythritol was not playing safe. I threw out the erythritol I had then discovered that all the brands of “keto ice cream” — I was on a low glycemic diet and foods low in carbohydrates are also low in glycemic load — used erythritol as their sugar substitute.
Frozen bananas, put through a food processor or super blender along with a couple of ice cubes and some milk, cream, or yogurt, make a pretty good ice cream substitute.1 Or eat ice cream and keep down your weight or glycemic load by eating less of something else.
It’s safer.
Lethal Caution: The Butter/Margarine Story
For quite a long time the standard nutritional advice was to replace butter with margarine, eliminating the saturated fat that caused high cholesterol and hence heart attacks. It turned out to be very bad advice. Saturated fats may be bad for you — the jury is still out on that, with one recent survey of the evidence concluding that they have no effect on overall mortality — but transfats are much worse. The margarine we were told to switch to was largely transfats.2
Consumption of trans unsaturated fatty acids, however, was associated with a 34% increase in all cause mortality3
If that figure is correct, the nutritional advice we were given for decades killed several million people.
Cryonic Suspension
For those of us who do not believe in religion or the immortality of the soul, the final bad outcome appears to be a certainty.
But there may be a way of, with luck, evading it, a way of playing, not safe, but safer. Arrange to be frozen immediately after you die, kept very cold thereafter.4 If you are very lucky, medical progress will eventually make it possible to revive you and cure what you died of. The odds may not be very good but they are better than with the alternative.
It looks like an unambiguous improvement in your future, at the cost of a good deal of money paid to make the arrangements. But one cannot be sure, for at least two reasons.
I do not believe in God, afterlives or an immortal soul, but I could be wrong. It is possible that after you die you go to an afterlife and, if I do not really die because I am frozen and awaiting resurrection, I don’t. It could be a very cold century.
That is the fantasy reason why cryonic suspension might be a mistake. The science-fiction reason, explored in a number of stories, depends on two facts. I do not know, if I am brought back, what sort of world, what sort of society, I will be brought back into or how the revived me will be treated. And, bad as death is, there are worse things.
Slowing Climate Change
Climate change is widely expected to have bad effects, by some to have very bad effects. Hence, it is implied, we should do what we can to reduce CO2, slow climate change. We should play safe.
There are two things wrong with that argument. One is that some things we think would slow climate change might not, might make the world worse in other ways. The second is that slowing climate change may itself have bad, even catastrophic, consequences.
An example of the first is the US biofuels program, created on the theory that converting maize to alcohol and burning the alcohol instead of gasoline would reduce CO2 output. It turned out that it wasn’t true. As even Al Gore has, to his credit, now conceded, biofuels was a mistake. At least as much CO2 is produced in the process of growing and transporting the maize and converting it to alcohol as would be produced by the gasoline it replaces.5 We still have the program because converting a third of the US maize crop, about twelve percent of world of output, pushes up the price of maize. Farmers vote.
About two hundred million people, mostly in poor countries in Africa and Asia, rely on maize as their chief food. Making the food poor people rely on more expensive has consequences — hunger and death. I do not know if any of the people responsible for initiating the biofuels program have trouble sleeping at night, but they should. The biofuels program is America’s contribution to world hunger.
Part of the mistake in the biofuels program was believing that it would reduce CO2 and so slow warming. But even if that were true it is far from obvious that the benefit of reducing future temperature by a few hundredths of a degree — there are a lot of other sources of CO2 — would be worth the cost of making food more expensive for two hundred million people.
Suppose we had some way of slowing climate change that did not have substantial bad effects; it is still not clear that doing so counts as playing safe. We are in an ice age,6 have been for more than two million years. Ice ages alternate between glacials, periods when much of the Earth is covered by ice, and interglacials, periods when the ice retreats. We are now in an interglacial.
It started, by one estimate, about twelve thousand years ago.
Over the last 450,000 years, glacials have lasted anywhere from 70,000 to 90,000 years whereas interglacials last approximately 10,000 years. (Energy Education web page)7
That raises the possibility that, absent anthropogenic warming, the current interglacial might be ending. The result, judged by past glaciation, would be half a mile of ice over the present locations of London and Chicago and sea level dropping by more than four hundred feet, leaving every port in the world high and dry.
It is very unlikely, but not impossible, that the continuation of the current interglacial depends on warming produced by CO2 emission over recent centuries. There is, however, some evidence that it is due to anthropogenic warming starting about eight thousand years ago, due not to the industrial revolution but the invention of agriculture, that if that had not happened the next glaciation would be beginning.
Bananas get sweeter as they get riper so for either a keto or low glycemic diet, freeze them before they get too ripe.
Some more recent margarines contain neither saturated fats nor transfats.
“Intake of saturated and trans unsaturated fatty acids and risk of all cause mortality, cardiovascular disease, and type 2 diabetes: systematic review and meta-analysis of observational studies,” BMJ 2015; 351 doi: https://doi.org/10.1136/bmj.h3978 (Published 12 August 2015)
The leading group doing this is the Alcor Foundation. “Frozen” is an oversimplification, they have procedures, based on technologies now used to preserve organs for transplant, that reduce, but cannot entirely eliminate, the damage done in the process of lowering your body to a temperature at which no further change will occur.
Slate on “The Ethanol Catastrophe.”
An ice age, as defined by geologists, is a time when Earth has ice on one or both poles. In popular usage, glaciations are sometimes called ice ages.
Different sources disagree on the exact definition and hence timing of interglacials; I chose that one because it makes my point. From another source whose numbers fit it less well:
“the last four interglacials lasted over ~20,000 years with the warmest portion being a relatively stable period of 10,000 to 15,000 years duration. … Work in progress on Devils Hole data for the period 60,000 to 5,000 years ago indicates that current interglacial temperature conditions may have already persisted for 17,000 years.”
My grandfather had intuited that margarine was bad, and never strayed from using butter. He always referred to margarine as “axle grease”.
Excellent article. Other possibilities? Many of the world's leading virus experts and pandemic prevention bodies decided that they should be collecting viruses from out of the way parts of the world, and studying them in laboratories. It is at least possible that this caused the worst pandemic for a century.
I don't feel I can really judge if it's correct, but I find the argument plausible that to date, AI alignment and safety work has increased AI risk, and that certain kinds of alignment/safety work would increase risk in the future.
Many of those responsible for sharing information with the Soviets that allowed them to develop the atomic bomb claimed they were motivated by concerns about the risk of only one side having nuclear weapons (though I'm not sure I believe them, given their pre-existing politics). Giving Stalin the bomb surely increased the risk of nuclear weapons being used.
The defenders of freedom granted exceptional powers to defend against a would be dictator, who then go on to use those powers to become dictators themselves, are legion.