Hacker News Reader

> If our ancestors would have made the rational assessment that there is unlikely to be a predator hiding behind the bush, that would have worked only as long as it worked, until one day they got eaten.

That wouldn't have been a rational assessment, because it wouldn't have been an accurate assessment of the risks of being wrong, and the behavior required to avoid them.

If there's only a 1% chance that a predator is behind a bush, and that predator might eat you, it's absolutely rational to act as though there is a predator. You'll be seeing lots of bushes in your life, and you can't escape from those 1% chances for long.

The same thinking is why it would have been rational to try and avoid global warming 30 years ago. Even if the science was not settled, in the worst-case scenario, you'd have "wasted" a bunch of money making green energy production. In the best-case scenario, you saved the planet.

It's not actually rational, let alone long-term optimal, to act as though there is a predator behind every bush given a 1% (in reality it's probably a couple of orders of magnitude less likely, but we'll ignore that). If you need water and head for the local watering hole, avoiding bushes will most likely result in you not getting water since bushes tend to grow where there is water. I may be 1% likely to get eaten by something hiding behind the bush but I am 100% likely to die if I don't drink water.

Avoidance of all possible risk is a recipe for paralysis. Part of being rational is evaluation of risks vs rewards as well as recognizing the dangers of unintended consequences and the fact that nearly all meaningful decisions are made with incomplete information and time limits.

Somehow you took their response to mean "the rational thing to do is avoid all bushes, forever, no matter the consequences".

The OP merely stated you should adjust your behavior to the 1% chance. That would include weighing it against the risk of dying from dehydration, in your example.

Humans operate by doing, then rationalizing, and much of the attempts at rational thought here demonstrate how easy it is to fool ourselves into thinking we are being rational, when really we are acting on feelings and delusions and then constructing what feels like a rational argument that we originally had - but falls apart upon analysis.

In the past, it is a rational concern to be worried about being jumped by a predator from behind a bush, and if you don’t know if or if not there is a predator, it is perfectly rational to be worried about such a concern!

Same with diseases and causes when you don’t know what is causing them, etc.

It’s a tendency to dismiss older concerns from a time when there was a severe lack of information as irrational, where when you know your limits and see the results, there is no other rational way to behave except to be concerned or avoid those things. While also not rational to believe clearly contradictory religious dogma that covers the topic, it is rational to follow or support it when it has clear alignment with visibly effective methods encoded in it for avoiding disease and other problems.

> In the past, it is a rational concern to be worried about being jumped by a predator from behind a bush, and if you don’t know if or if not there is a predator, it is perfectly rational to be worried about such a concern!

I think we agree, but I also think you are using "rational" here in the colloquial sense to mean the "smartest" thing to do.

The article, and my comment in response, uses the traditional definition of "rational" as something derived from logic, and not from impulse or instinct.

The two definitions are not the same (not that one is better than the other, they just mean different things).

Nope, explicitly using logic. We didn’t invent thinking about things in the last hundred years after all.

If you don’t know what is behind x thing, and every y times someone walks by a thing like x thing they get jumped by a leopard, then only walk by x thing when the risk is worth it. Which it rarely is.

If you’re referring to formal logic, then sure - but almost no one in that thread seems to be using that definition either. Formal logic is incredibly expensive (mentally), and only a few percent of folks even now can afford to use it with any regularity.

This is also captured in the “midwit phenomenon”, where people who are just smart enough to start applying “rationality” make worse decisions than stupid people. This is because stupid people are operating off of hard-earned adaptations (encoded as traditions, folk wisdom, etc.). Midwits are smart enough to realize that the putative justifications for these adaptations are wrong, and therefore they toss out the adaptations. People who think about it even harder realize that these adaptations were mostly there for good reasons, and getting rid of them isn’t a good idea even if the relevant just-so stories explaining them don’t hold up to “rational” scrutiny.

Midwits (which we all are to one degree or another) can be mostly fixed by applying Chesterton's Fence principle though. We just need a knock or two in both directions to better estimate a relative weight of that rule as a heuristic.

Just because there's a 99% chance the bush has no predators behind it does not make it rational to assume there are no predators.

In Bayesian decision theory, you'd choose the action (walk directly by the bush; walk by the bush but have your guard up; steer clear of the bush) that minimizes your loss function (e.g. probability of dying or probability of your blood line dying out). You'd end up picking a path that balances the risk of being eaten by a lion with the cost of having to walk further (and thus having less time and energy to gather food; or tripping and cutting yourself and dying of infection; or whatever).

I like this, but I also like the even simpler explanation. Our bodies/minds require energy and "skipping" to conclusions (irrationally) expends less energy.