Some reactions (I especially care about 2. and 2c.):
1. I think I basically share / agree with what seems to me like the “biggest” idea in the article, namely that there is progress on things like how sane and good a civilization is, and that there is an analog of a hedonic treadmill for feelings of progress.
2. I worry that the article’s emphasis on something like object-level progress (religion has less-bad beliefs than a few hundred years ago) may make the “we will always have crazy” idea sound like there is no meta-level progress, i.e., the third of the population that’s worst at epistemics is as bad at arriving at and updating towards correct beliefs as it was some hundreds of years ago, and that this will continue to be true in the future.
2a. I suspect that there is progress there too. For example, I expect that some hundred years ago, the third of the population with the worst epistemics would have been much more strident than today in shutting down people trying to speak up with calm reasonable arguments against what a powerful person says, and would have had a significantly more blatant tendency to take “but <powerful person> disagrees” to be an absolute counterargument and disqualifier for the person trying to speak up.
2b. Barring catastrophes and other huge changes, I expect the meta-level progress in epistemics to continue in the future, so I expect the hypothetical Duncan in 2200 to in fact be more able to converse with {the third of the population that is worst at epistemics} in ways that cause updates, even if the treadmill hypothesis suggests that he’ll feel no better about it.
2c. Because I think that society’s epistemics make meta progress in addition to object-level progress, I do not confidently go from “I agree that the world around me looks like this treadmill thing is true” to “I think the future will look like the treadmill thing barring catastrophe”. When progress has reached a point where something starts to come into view as a salient problem, there’s a chance that something changes about that problem that didn’t change about it before. By analogy: It seems to me like early proponents of democracy might have looked at history and thought “most countries will always be ruled by hereditary rulers”, but in fact it seems to me like the pattern of hereditary rule had a much easier time perpetuating itself before the meme that hereditary rule is bad started to get a hold in the public imagination.
3. Caveats about my agreement with the basic model of progress happening:
3a. I think historical progress has had setbacks, so you need to look at long timescales to see continuous historical progress. Furthermore, it seems to me that the timescales on which you need to look to skip over the setbacks get longer as we look backwards in time; the middle ages were pretty long compared to the Soviet Union, say. (That said, I don’t know whether the history of Ancient Greece and Ancient Rome is full of 50-year-long setbacks that are just not salient enough from the modern vantage point.)
3b. I don’t think future progress-with-temporary-setbacks is guaranteed, even conditioning on humanity not going extinct; I think technological and other progress might be able to enable a dark age that does not end (e.g. a dictatorship with sufficient surveillance to effectively root out opposition), as Bostrom suggests in his x-risk paper.
3c. I feel confused about how various forms of new media over the last 200 years have made society crazier in ways that my intuition says “are orthogonal to the point of this article”, and I also haven’t examined my confusion about it enough to confidently think that it’s orthogonal. (I am particularly thinking of newspapers with stories transmitted by telegraph; radio; TV; and social media.)
A FB friend of mine adds the following: