For a text that self-confesses to be obscurantist, I was nodding my head in agreement an awful lot, both in terms of Scott's interpretation of what it's trying to say, but also the direct quotes. Whether that's because I've willfully misunderstood everything and am now patting myself on the back for these "insights", because I've converged on to something truth-adjacent, or because I've converged on to some Schelling point of crackpottery, is anyone's guess.
For a while now I've been drawn into thinking that humans aren't agents. And by that I don't mean the weak notion that humans don't exactly match the mathematical formalism for agents as that statement is pretty much trivially true (due to bounded computational resources imposed by laws of physics) - this is what the idea of boundedly rational agents is for. For me it's the stronger notion that claims it's almost never accurate to model human decision-making as optimizing (or satisficing, really) for their preferences. There are some trivial examples like observing that humans tend to claim to want money, and indeed they're likely to cash in winning lottery tickets or pick up hundred dollar bills from the ground (citation needed), but once you get into more complex behaviors, such as Scott's example of not asking people on a date despite yearning for a partner every single day, where exactly is the preference-satisficing behavior? One could argue that having a partner isn't people's true preference, that they actually prefer to be forever alone, or that being rejected hurts so much that even a perfectly rational agent would not risk it, but when people (I'm thinking about philosophers arguing for compatibilism in particular) talk about humans showing agenthood, they're explicitly giving some reality to people's preferences. On the other hand, to deny the reality of these preferences would seem like equating humans to rocks, that rocks have a preference to stay immobile because that's what we observe them doing, which I think is an even more radical view than my own.
My view is that the preferences people see themselves as having are real, but that accurate model of human behavior rarely invokes them and uses other concepts instead. These concepts are still endogenous (nobody's being puppeteered by an evil demon) but are often of the nature being talked about here. For example, I draw a juxtaposition from the story of the tree to ideas presented in HPMOR's chapter involving the troll attack: people often act out roles. The tree acts out the role of a mother, McGonagall a strict disciplinarian, and that way of modelling behavior might actually get you somewhere. Ditto for behaviors like virtue-signalling, or any of the other social games people unconsciously play that end up dictating their behavior.
Similarly, the idea that people want their freedoms curtailed speaks to me because I've also thought about that a lot. It seems to me that humans would indeed be the happiest in a state of "choicelessness" (I believe there's some Eastern philosophical concept for it, but if I've known one, I've since forgotten about it), always living in the moment because there's no other option. The extreme example of this would be monasticism, but this has to large extent been the general experience for most of human existence: you could in principle do something else than forage or work the fields, but then you'd starve so you don't actually have much choice now do you. Even in today's environment a lot of people seem to find the idea of not /having/ to go to work every morning abhorrent: what would they do with themselves? I've never drawn a connection between historical forms of choicelessness and modern forms like "domination by corporate HR departments", but now that the idea has been presented to me, it does seem to make sense.
(For the record, this does in no way excuse slavery: you can experience choicelessness without also being subjected to misery)
(Also for the record, I'm claiming no superiority here. In fact, I'm uncharacteristically incapable of acting on my claimed preferences)
For a text that self-confesses to be obscurantist, I was nodding my head in agreement an awful lot, both in terms of Scott's interpretation of what it's trying to say, but also the direct quotes. Whether that's because I've willfully misunderstood everything and am now patting myself on the back for these "insights", because I've converged on to something truth-adjacent, or because I've converged on to some Schelling point of crackpottery, is anyone's guess.
For a while now I've been drawn into thinking that humans aren't agents. And by that I don't mean the weak notion that humans don't exactly match the mathematical formalism for agents as that statement is pretty much trivially true (due to bounded computational resources imposed by laws of physics) - this is what the idea of boundedly rational agents is for. For me it's the stronger notion that claims it's almost never accurate to model human decision-making as optimizing (or satisficing, really) for their preferences. There are some trivial examples like observing that humans tend to claim to want money, and indeed they're likely to cash in winning lottery tickets or pick up hundred dollar bills from the ground (citation needed), but once you get into more complex behaviors, such as Scott's example of not asking people on a date despite yearning for a partner every single day, where exactly is the preference-satisficing behavior? One could argue that having a partner isn't people's true preference, that they actually prefer to be forever alone, or that being rejected hurts so much that even a perfectly rational agent would not risk it, but when people (I'm thinking about philosophers arguing for compatibilism in particular) talk about humans showing agenthood, they're explicitly giving some reality to people's preferences. On the other hand, to deny the reality of these preferences would seem like equating humans to rocks, that rocks have a preference to stay immobile because that's what we observe them doing, which I think is an even more radical view than my own.
My view is that the preferences people see themselves as having are real, but that accurate model of human behavior rarely invokes them and uses other concepts instead. These concepts are still endogenous (nobody's being puppeteered by an evil demon) but are often of the nature being talked about here. For example, I draw a juxtaposition from the story of the tree to ideas presented in HPMOR's chapter involving the troll attack: people often act out roles. The tree acts out the role of a mother, McGonagall a strict disciplinarian, and that way of modelling behavior might actually get you somewhere. Ditto for behaviors like virtue-signalling, or any of the other social games people unconsciously play that end up dictating their behavior.
Similarly, the idea that people want their freedoms curtailed speaks to me because I've also thought about that a lot. It seems to me that humans would indeed be the happiest in a state of "choicelessness" (I believe there's some Eastern philosophical concept for it, but if I've known one, I've since forgotten about it), always living in the moment because there's no other option. The extreme example of this would be monasticism, but this has to large extent been the general experience for most of human existence: you could in principle do something else than forage or work the fields, but then you'd starve so you don't actually have much choice now do you. Even in today's environment a lot of people seem to find the idea of not /having/ to go to work every morning abhorrent: what would they do with themselves? I've never drawn a connection between historical forms of choicelessness and modern forms like "domination by corporate HR departments", but now that the idea has been presented to me, it does seem to make sense.
(For the record, this does in no way excuse slavery: you can experience choicelessness without also being subjected to misery)
(Also for the record, I'm claiming no superiority here. In fact, I'm uncharacteristically incapable of acting on my claimed preferences)