[xelxebar]

Even more concretely, the terms better and worse are only capable of encoding relative positions on a linear scale.

I find it instructive to look for those scales directly--e.g. rates of violent crime, GDP, life expectancy, etc. This also makes it more natural to recognize specific problems with such linear reductions.

[leohutson]

All those scales are valued differently by different cultures.

If I show a conservative American gun crime statistics that we agree on and we agree that are caused by American gun culture, I cannot tell them that American gun culture is inferior, because I just get back "Thats the price of freedom!"

Life expectancy, "It's not the years in your life but the life in your years!"

GDP is the most hilarious one, because it's not even per-capita. Assuming that its a high median per capita GDP: "That's materialistic! We're much happier than those greedy capitalists"

[xelxebar]

This is a good point. Any particular statistic probably doesn't capture the nuances of human values. I think statistics are quite useful though, since they make it easier to differentiate between "facts we can share" and "personal values".

In my experience, a lot of disagreement gets chocked up to opinion differences when in fact it's mainly a failure to unpack beliefs into sets of values and facts.

I like to distinguish between so called "terminal values" which are things we intrisically care about, versus "intermediate values" which are things that matter to us for their external effects. For example, I think the internet is good because of the communications it facilitates. If it did the opposite, I'd probably feel oppositely as well, so my feelings on the internet are an intermediate value. On the other hand, I probably value my mom intrinsically.

Of course, we could further unpack that. Are my feelings on "facile communication" terminal or intermediate? My mental image here is of a sort of a branching set of trees, a forest, of values where the root nodes are our terminal values and the intervening nodes are our intermediate values.

I doubt that we humans actually differ all that much on terminal values. If we take this stance axiomatically, it certainly makes a lot of sense to deconstruct intermediate values into the causal links they're composed of, i.e. facts and statistics.

[kazagistar]

I'm not sure it's so easy to differentiate, since terminal values are feelings, wishy washy and prone to contradicting each other. I would say that any terminal value can become an intermediate value when you find a contradiction between it and other terminal values. Then, you either find a way to rationalize, ignore the problem, or discover a new terminal value to make the choice and downgrade one of your terminal values to an intermediate, conditional one.

Nonetheless, it's a very useful lens. Take property: many people consider it to be terminal, but I have always seen it as intermediate, a useful tool for accomplishing good things as a society, but not some sacred absolute.

[adrianN]

There are attempts to measure "happiness". They surely have their own problems, but if we could measure that relatively objectively it seems to me that "average happiness" would be a good starting point to use to compare different cultures.

https://en.wikipedia.org/wiki/World_Happiness_Report