There’s almost nothing I love more than seeing a new scientific tool solve an old problem of life. Curing diseases, solutions to obesity, artificial wombs, brain uploading, cryonics, the journey from scarcity to abundance to immortality. I am infinitely interested in applying empirical rational tools to nearly every domain of life.
There seems to be an important distinction to be made about general life optimization, which I describe as hacks, heuristics, and frameworks. A hack is a solution to one problem; a heuristic is, in some ways, a reusable hack; and a framework is something that tells you which problems are worth solving and why.
Why should we try to optimize sleep? Because better sleep leads to better subjective experience and better performance. And in general, better performance is desirable, and better subjective experience is good, or something like that. Why should we date? Maybe because social relationships almost always make people happier, maybe also because dating is conducive to marriage, and marriage is a stable institution suitable for raising children. Why should we raise children? Maybe because demographic collapse is a real problem, maybe there are selfish reasons to have more kids, and maybe you don’t want any of these things but it is good to have options, and ways to go about it when you do.
Hacks don’t replace frameworks. They smuggle in implicit frameworks while pretending to be framework-free. These “super basic notions that we have all internalized as obviously good” are residues of intellectual traditions. “Health is good” is based on the modern affirmation of ordinary biological life; stoics were indifferent to bodily states and medieval Christians valued mortification of the flesh. “Time is a finite resource and saving it is good” draws on a Protestant-capitalist conception of time as something to be spent wisely, which would have been incomprehensible to most pre-modern cultures with cyclical time. “ Individual agency over your life is possible and desirable” relates to Enlightenment autonomy and Romantic authenticity. Charles Taylor calls these notions constitutive goods, moral sources that power our evaluations that we’ve so thoroughly internalized them that we take them as neutral facts about the world.
It is important to note that the genealogical observation (these values have specific historical origins) and the metaethical claim (some moral truths are objectively correct) are not necessarily in conflict — but the relationship between them is complicated. The tension isn’t between genealogy and objective morality. It’s between the first-order moral claims (health is good, suffering is bad — these might well be objective) and the ordering of those claims relative to each other, the absence of any shared, explicit ordering — even if the individual values being ordered might be correct.
Since they can no longer combine in the service of a supreme value, [the value-systems] claim equality one with the other: like strangers they exist side by side, an economic value-system of 'good business' next to an aesthetic one of l'art pour l'art, a military code of values side by side with a technical or an athletic, each autonomous, each 'in and for itself,' each 'unfettered' in its autonomy, each resolved to push home with radical thoroughness the final conclusions of its logic and to break its own record.
— Hermann Brooch, The Sleepwalkers
The modern secular person has implicit frameworks (eg. human rights, individual dignity, the value of flourishing), plus an enormous and growing stack of hacks built on top of it, but has lost explicit frameworks that would organize these implicit goods into a hierarchy.
If all genuine goods are harmonious, you don’t need a framework — you just need more hacks. Optimize health and productivity and relationships and meaning and agency, and they’ll all reinforce each other. The hack ecosystem implicitly assumes this: every “10 things to improve your life” list treats its items as additive, or as inspirational ideas you can freely adopt or ignore. There’s a related notion that hacks are just random things that might or might not work for you, and you should simply try them and keep what sticks. But if goodness does compete — if real goods genuinely conflict, and pursuing one requires sacrificing another — then hacks are structurally incapable of helping you at the decisions that matter most. And it is obviously true that you can neither implement all 50 suggestions to your life, or pick all options between every decision. If we put such effort into solving object-level problems — optimizing sleep, nootropics, workflows — it seems odd to make value-level decisions arbitrarily.
This is probably also why the EA framework resonates with people even when they might disagree with its conclusions. It at least attempts an ordering. Neglected causes matter more than popular ones. Expected value calculations can guide your choices when goods conflict. Et cetera. You can argue with any of those claims, but at least there are claims to argue with.
Ulrich took it as a matter of course that a man who has intellect has all kinds of intellect, so that intellect is more original than qualities. He himself was a man of many contradictions and supposed that all the qualities that have ever manifested themselves in human beings lie close together in every man’s mind, if he has a mind at all. This may not be quite right, but what we know about the origin of good and evil suggests that while everyone has a mind of a certain size, he can still probably wear a great variety of clothing in that size, if fate so determines.
And so Ulrich felt that what he had just thought was not entirely without significance. For if, in the course of time, commonplace and impersonal ideas are automatically reinforced while unusual ideas fade away, so that almost everyone, with a mechanical certainty, is bound to become increasingly mediocre, this explains why, despite the thousandfold possibilities available to everyone, the average human being is in fact average. And it also explains why even among those privileged persons who make a place for themselves and achieve recognition there will be found a certain mixture of about 51 percent depth and 49 percent shallowness, which is the most successful of all. Ulrich had perceived this for a long time as so intricately senseless and unbearably sad that he would have gladly gone on thinking about it.
— Robert Musil, The Man Without Qualities
Part of the enthusiasm about agency emerges from a concern that people don’t really know what their future looks like, and they desire to control or lay claim to it in a way they hope agency will provide. In other words, the agency trend might itself be a symptom of framework loss. When we run out of answers to “what should I do with my life?”, the next best thing is to get really good at doing things in general.
Man ... is driven out into the horror of the infinite, and, no matter how he shudders at the prospect, no matter how romantically and sentimentally he may yearn to return to the fold of faith, he is helplessly caught in the mechanism of the autonomous value-systems, and can do nothing but submit himself to the particular value that has become his profession, he can do nothing but become a function of that value — a specialist, eaten up by the radical logic of the value into whose jaws he has fallen.
— Hermann Brooch, The Sleepwalkers
I’m not sure what can serve as a genuine framework. This seems like one of the most difficult problems imaginable. But I am confident that there exist a failure more of letting hacks or heuristics harden into pseudo-frameworks. We take something that is actually an empirical claim bundled with a value judgment and compress it into a theoretical principle that feels like it can organize everything. “Passions are malleable” “you can just do things” or “government power tends to expand, so be cautious” can itself become a mind virus if you start treating it as a universal principle rather than a contextual heuristic. When a contextual truth gets promoted to a universal ordering principle and becomes load-bearing for your entire worldview, challenging it feels like pulling a thread that might unravel everything.
It is probably worse to filled the framework-shaped hole with something that doesn’t actually fit, but that sits comfortably enough to make the searching feel unnecessary; than admitting to having no framework at all.
(crosspost from my substack)
There’s almost nothing I love more than seeing a new scientific tool solve an old problem of life. Curing diseases, solutions to obesity, artificial wombs, brain uploading, cryonics, the journey from scarcity to abundance to immortality. I am infinitely interested in applying empirical rational tools to nearly every domain of life.
How to sleep better. How to have better skin. How to parent. How to prevent colds. How to pick college courses. How to be good at sex. How to find a partner. How to use data to decide who to marry. How to think. How to make your room as bright as the outdoors. How to read self-help. How to optimize. How not to optimize — and what to optimize for. How to solve problems in general.
This sometimes accumulates to a genre of writing I love just as much:
There has never been a better time to have a problem, and if you think there is a problem you cannot solve, maybe you’re not actually trying.
There seems to be an important distinction to be made about general life optimization, which I describe as hacks, heuristics, and frameworks. A hack is a solution to one problem; a heuristic is, in some ways, a reusable hack; and a framework is something that tells you which problems are worth solving and why.
Why should we try to optimize sleep? Because better sleep leads to better subjective experience and better performance. And in general, better performance is desirable, and better subjective experience is good, or something like that. Why should we date? Maybe because social relationships almost always make people happier, maybe also because dating is conducive to marriage, and marriage is a stable institution suitable for raising children. Why should we raise children? Maybe because demographic collapse is a real problem, maybe there are selfish reasons to have more kids, and maybe you don’t want any of these things but it is good to have options, and ways to go about it when you do.
Hacks don’t replace frameworks. They smuggle in implicit frameworks while pretending to be framework-free. These “super basic notions that we have all internalized as obviously good” are residues of intellectual traditions. “Health is good” is based on the modern affirmation of ordinary biological life; stoics were indifferent to bodily states and medieval Christians valued mortification of the flesh. “Time is a finite resource and saving it is good” draws on a Protestant-capitalist conception of time as something to be spent wisely, which would have been incomprehensible to most pre-modern cultures with cyclical time. “ Individual agency over your life is possible and desirable” relates to Enlightenment autonomy and Romantic authenticity. Charles Taylor calls these notions constitutive goods, moral sources that power our evaluations that we’ve so thoroughly internalized them that we take them as neutral facts about the world.
It is important to note that the genealogical observation (these values have specific historical origins) and the metaethical claim (some moral truths are objectively correct) are not necessarily in conflict — but the relationship between them is complicated. The tension isn’t between genealogy and objective morality. It’s between the first-order moral claims (health is good, suffering is bad — these might well be objective) and the ordering of those claims relative to each other, the absence of any shared, explicit ordering — even if the individual values being ordered might be correct.
— Hermann Brooch, The Sleepwalkers
The modern secular person has implicit frameworks (eg. human rights, individual dignity, the value of flourishing), plus an enormous and growing stack of hacks built on top of it, but has lost explicit frameworks that would organize these implicit goods into a hierarchy.
If all genuine goods are harmonious, you don’t need a framework — you just need more hacks. Optimize health and productivity and relationships and meaning and agency, and they’ll all reinforce each other. The hack ecosystem implicitly assumes this: every “10 things to improve your life” list treats its items as additive, or as inspirational ideas you can freely adopt or ignore. There’s a related notion that hacks are just random things that might or might not work for you, and you should simply try them and keep what sticks. But if goodness does compete — if real goods genuinely conflict, and pursuing one requires sacrificing another — then hacks are structurally incapable of helping you at the decisions that matter most. And it is obviously true that you can neither implement all 50 suggestions to your life, or pick all options between every decision. If we put such effort into solving object-level problems — optimizing sleep, nootropics, workflows — it seems odd to make value-level decisions arbitrarily.
This is probably also why the EA framework resonates with people even when they might disagree with its conclusions. It at least attempts an ordering. Neglected causes matter more than popular ones. Expected value calculations can guide your choices when goods conflict. Et cetera. You can argue with any of those claims, but at least there are claims to argue with.
— Robert Musil, The Man Without Qualities
Part of the enthusiasm about agency emerges from a concern that people don’t really know what their future looks like, and they desire to control or lay claim to it in a way they hope agency will provide. In other words, the agency trend might itself be a symptom of framework loss. When we run out of answers to “what should I do with my life?”, the next best thing is to get really good at doing things in general.
— Hermann Brooch, The Sleepwalkers
I’m not sure what can serve as a genuine framework. This seems like one of the most difficult problems imaginable. But I am confident that there exist a failure more of letting hacks or heuristics harden into pseudo-frameworks. We take something that is actually an empirical claim bundled with a value judgment and compress it into a theoretical principle that feels like it can organize everything. “Passions are malleable” “you can just do things” or “government power tends to expand, so be cautious” can itself become a mind virus if you start treating it as a universal principle rather than a contextual heuristic. When a contextual truth gets promoted to a universal ordering principle and becomes load-bearing for your entire worldview, challenging it feels like pulling a thread that might unravel everything.
It is probably worse to filled the framework-shaped hole with something that doesn’t actually fit, but that sits comfortably enough to make the searching feel unnecessary; than admitting to having no framework at all.