Vitalik Buterin
@vitalik.eth
I think the core of what's unintuitive about the various paradoxes of utilitarianism is the idea that utility is unbounded. This clashes with how our brains work, where there actually is a bound on how strongly we can feel (positively or negatively) about any particular situation.
33 replies
40 recasts
340 reactions
Venkatesh Rao ☀️
@vgr
This is about the shrimp and bees isn’t it? I think utilitarianism fails to model pain correctly. It needs a “nature is murder” Werner Herzog axiom. You can’t really drive down pain in complex life monotonically, only move it around. Trilemma: Pick 2 of 3 — hard-capped aggregate pain experienced by all life, sustained aggregate net utility growth, complexity of intelligence. It’s the “huntable tofu gazelles for lions” problem. Complex intelligent life capable of experiencing heights of flourishing seems to depend on an engine of causing sustained pain to simpler life to thrive. Utilitarianism has an insufficiently expressive ontology to even state this problem. Pain and flourishing need to be book-kept in non-mutually-offsettable ways.
1 reply
0 recast
4 reactions
Vitalik Buterin
@vitalik.eth
hmm I agree we require challenge for fulfillment (EY himself has written about this a lot), but I also think that's relatively a "first world problem"; there are huge amounts of suffering that don't contribute to well-being at all If we can make robot tofu gazelles optimized to make lions' lives fun then imo we should
1 reply
0 recast
0 reaction
Venkatesh Rao ☀️
@vgr
Is it first-world? I think you only land there if you start from a specific western philosophical tradition (Hegel/Nietzche/Heidegger) and assume a particular model of economic flourishing. I think it is much more basic and integral to the very fabric of nature. Ime poor and developing world environments feature similar kinds of thymos-origin suffering beyond a pretty low threshold of well-being. Buddhist accounts of suffering are not particularly linked to material condition. “Fulfillment” vs “pain relief” is a fragile and relatively shallow distinction. A bit like “need” vs”want” that bedevils UBI conversations. I don’t have good alts but I don’t trust either dichotomy to build complex policy on. Re: robot tofu gazelles, I use that example a lot because I don’t yet know what to think about it. I’m personally vegetarian going on vegan, but have low confidence in extrapolating my reasoning that far. We feed our cats all meat diets (carbs and current meat subs really mess with the digestion of obligate carnivores)
1 reply
0 recast
0 reaction
Vitalik Buterin
@vitalik.eth
Do we agree that half your kids dying before they turn age 5 is a form of suffering that does not really contribute to higher-level well-being in any interesting and not easily replaceable way? That phenomenon is pretty integral to the fabric of nature for all species, but once we just technologically wished it away, it worked out great
1 reply
0 recast
1 reaction
Venkatesh Rao ☀️
@vgr
For humans yes. For animals, depends. I’m not entirely sure brood reduction/infanticide behaviors in animals (eg geese kick weakest chick out of nest to die) can be interfered with technologically without causing more net suffering. Things like lions killing all rival cubs when they take over a pride seem genetically hard-wired and hard to mess with. But humans throwing male chicks into grinders in factory layer hen raising ops — that’s needless and horrifying.
1 reply
0 recast
0 reaction