This year was an interesting year. Nearing graduation, I’ve had to give reality a clear-eyed look; and being a bit of a natural idealist I was a little depressed by what I saw. I suffered a few worldview-shaking traumatic experiences. My naive principles, community building and intrinsically motivated learning, were already strained and underwent a complete failure (in their existing form). I faced a lack of confidence in my beliefs and ideas and a general gnawing anxiety. Although I really enjoyed very many new and wonderful experiences this year, the integration of these new experiences into my existing worldview crashed; my falcons wheeled in widening gyres and the center could not hold.
But, just when the careening falcons turned through every little gain back into a nauseating oscillation, and I thought this liminal state was a brutish inheritance of adulthood, I hit upon the germs of something new. Turning and pulling myself along the floor under the drifting smoke and turning again and always moving–I finally without warning tumbled down the front stairs of the burning building and drew a breath of fresh clean air.
What follows is an exposition of the fall of my worldview, and my daring escape from the wreckage.
The Spencerian Dream
Any account of why America has been special for so long has to include the widespread belief in the American Dream; that any person through hard work and a little insight can become rich and prosperous. Whether this belief was really accurate hardly mattered for its effects.
On reflection, my exploratory, extroverted, and psyched-up personality has existed symbiotically with two unspoken beliefs, which together make up the Spencerian Dream. My parents thoughtfully gave them to me, Lady Gaga style “My mama told me… /”, and they’ve kept me going when the going got tough.
The world is full of opportunity for me to do good things.
Good here is a huge category, some examples: making tea for an appreciative friend; winning a soccer game; deriving a new insight into the exponential function; enjoying the gleam of golden light on leaves in autumn; inventing a better approach to teaching linear algebra or even arithmetic so that students enjoy it; becoming quicker and stronger through spontaneous push-up sessions; making a new friendship or relationship; winning a scholarship; writing a quality essay about my new philosophy… and so on. You can start to see categories here; although forming these too rigidly is dangerous, we’ll look into some of them in a bit more detail later, like “competition” and “being present”.
Belief 2, robustness, says that in pursuing the good or awesome, you can’t go too far wrong:
I will not wreak irreperable damage or unbalanced misery by pursuing good things honestly.
A few remarks: the irreperable and unbalanced are there because damage and misery are part of life. If I cause damage or misery but make up for it with subsequent or prior improvements and delight, then I have done no worse than any other player strutting and fretting upon the stage.
This belief mostly has to do with other people, but even with myself it applies. My life is precious, full stop, and I’d struggle to proceed if I thought I could crush myself with my ambitions.
The Flicker of Darkness
In combination, these beliefs fuel exploration and boldness. The Spencerian Dream says this mode of life will pay off (Opportunity) and it will not destroy what I cannot bear to destroy (Robustness).
In the past year (and a bit before that), the Spencerian Dream faced a bipartite challenge. Opportunity fell into a vicious cycle with a strange loss of confidence in my beliefs and abilities, and Robustness came under attack as I expanded my social circle. Both challenges were ultimately unfounded, but being a person who takes challenges seriously, this took me a long time to realize.
It all started with falling confidence coming from the obsolescence of my guiding, homeschooling-derived principles. Naive community-building and Naive intrinsically motivated learning hit reality going into year 3 at the UW. In both cases, I overextended myself. I had founded so many communities, from my brunch group to the Society of Physics Students to various special-interest groups, and felt the responsibilities of all of them weighing on me. At the same time, I disconnected from the excitement of founding the communities, which soured into anxiety. I slowly and painfully shed responsibilities, facing some real and mostly imaginary rap in the process. Intrinsically motivated learning failed in a similar way–I’d joined many different fields, math, physics, computer science, entrepreneurship–each with their own ex-trinsic criteria. I couldn’t satisfy them all, and again, faced some real and mostly imaginary rap for it.
The imaginary rap was the Please Everyone delusion, which afflicts people like me and (I found in conversation) my great friend Grace Pyles, who had so much excess of both ability and social status in our childhood that we thought we could be friends with / satisfy everyone we met. Nope! Jerks and insane people exist.
The real rap culminated in a few traumatic experiences (elided for now).
So thanks be to tempestuous bureaucrats, my parents’ foresight, and the defiant desperate fortitude of Past Spencer, that I was off to the ETH in Zürich for a semester. I left my world behind, friends and communities and projects and struggles alike.
The sabbatical started out great. I rode the waves of novelty and made lots of new friends, focused on coursework, did some traveling and hiking. When the novelty wore off, though, my lack of confidence and social anxiety came through. I fought bravely, and time healed as surely as always, but the winter set in and the sun set early and everything quieted down at the house, and I struggled…
The breakthrough came through a strangely related series of readings I started while visiting Venice with friends. I had picked up a random book called The Art of Learning by Josh Waitzkin, the chess prodigy featured in *Searching for Bobby Fischer“. Waitzkin was a world-class badass who, after a very successful career in chess, missing World Youth Champion by a single match, started studying Tai Chi and within 6 years was the world champion of the martial side of that sport. I stayed up until 3am in the hostel reading it and was enthralled not only by his detailed learning methods and psychological introspection, but by the notion that such subtle introspection (one of my favorite activities) was actually so valuable to Waitzkin. The trip ended with me finishing the book on the descent into Zürich, listening to”Lose Yourself" on my headphones as Waitzkin, in his riveting account of the 2006 Tai Chi Push Hands World Championships, sat in the tension before the first match listening to the same song.
That was the beginning. A week later it was off to Thanksgiving. I read On The Road by Kerouac, Elon Musk by Vance, and crucially, the beginning of Thus Spake Zarathustra and the Stanford Philosophy Encyclopedia entry for Nietzche (hat tip Eric Erkela). And a few weeks later I went to Budapest to take the Putnam exam, and read some of Antifragile, by Taleb, on the way. Somehow, all these disparate readings came together thematically for me, inspiring a firestorm of ideas and rethinkings of old ideas. On the way back from Budapest I had a flash of insight (inspired by my brutally difficult quantum information reading) about different styles of learning which I’m still exploring. Lemme sketch a timeline of ideas:
(Introspection, Art of Learning): Introspection is valuable. It can be used, opportunistically and empirically, to guide learning on the ground.
(Metric Eternalism, AoL): Finite games (roughly, competitions) are eternalisms (in this context, it is arbitrary to believe they have value), but they are a special kind of eternalism that come with a simple measure of how well you’re doing (unlike, say, utilitarianism). Being world class at something (versus just being better than most of your friends or the people at your university) is a more restricted type of eternalism that I call a Metric Eternalism. The measure of how world class you are in some competitive discipline is not just easy to compute, it is universally recognized (by people outside the field) and locally strongly valued (by your colleagues in the field). Therefore, it is extremely motivating to pursue. I can get the highest grade on my quantum final at the UW and still feel like a n00b, but if I were (somehow) to ace the nationwide Putnam exam and become a Putnam fellow, the combination of universal recognition and local value would combine to make me feel like a total badass.
(Leroy’s Blessing, On the Road and Elon Musk): Opportunism is the antidote to optimization. Even in its most over the top form, its consequences are not as terrible as I supposed (On the Road) and its benefits are damn close to what might be called optimal (Elon’s engagement to some random actress he met after like 20 days, and his decision to start SpaceX armed with little more than a spreadsheet saying the costs of rocket building should be low). This idea has a long history for me. I used to have opportunistic ideas, then set aside a chunk of time to pursue them hard core without worrying about the balance of optimization (calling this a Leroy’s Blessing after the gaming meme Leroy Jenkins). But it needs a refinement:
(Leroy’s Blessing, Part 2): You must be able to opportunistically drop things. Otherwise, your opportunistic accretion gets you stuck in what were local optima for your past self and just suck for you now. (Example: Elon’s divorce of his first wife when it wasn’t working out). I think when I get back to the UW I’ll pick up only about 50% of the things I left behind (no SPS responsibilities in particular). Again, the consequences of this are not really so bad!
(Leroy’s Blessing, Part 3): Being present in life is a continuum-limit Leroy’s Blessing. It is ten thousand little Leroy’s Blessings, a perpetual local thirst for opportunities to exult and to learn. The Past Spencer quote: “Smile in the light of day, peer boldly into the darkness, and stay thirsty”, captures the spirit of a life lived as a Leroy’s Blessing.
(No Guilt, Nietzche): Guilt, or internalized social pressure, is a really easy emotion to detach from reality, and should be ignored waaay more often than I thought. It reinforces other people’s value systems. If you’re a contrarian or polymath trying to hang on to your own value system, you really can’t afford to be a sucker for guilt. It blocks Leroy Part 2.
(Will to Overcome Resistance (or Towards Perfection), Nietzche): What can you do if you’re not trying to please other people to avoid guilt? What’s the alternative to other people’s value systems? Just increasing your capabilities to do things in the real world. This is relatively well-defined and value-system independent. The weird thing is that Past Spencer had this right, but the idea (Towards Perfection!) got bent up by other people’s value systems (Do I have to do a perfect final project? Get a perfect grade on every single assignment and exam like I did in my first Precalc class at WCC? No, that’s being a sucker for other people’s value systems (different from leveraging my unique strengths to ace just the exam and get a 4.0). Do I have to be entirely intrinsically motivated and try to get my kicks learning group theory from Wikipedia? No, that misses the “real world” part of increasing one’s capabilities.). Well, Towards Perfection is back.
(Confidence and Competition): A strong justification for the No Guilt principle is the fact that people treat you better when you’re not actively trying to be nice to them and yielding to their social pressure, so long as you’re confident about it. This, although empirically obvious, was hard for me to accept. (Note: It doesn’t hold for people who are 100% part of not just an activity but also a social group, with totally shared groupthink expectations. I don’t have the time to belong at this level to any group. Maybe I used to with the homeschool group.)
What earns you respect instead is confidence (nearly the opposite of try-hard niceness) and winning (should be the same as try-hard collaboration but is not, especially in pickup). I had a very stark illustration of this reality in pickup basketball. When I tried to be a good team player, avoid fouling and aggression, and listen to everyone’s (mostly wrong) advice, I got more patronizing advice and got ragged on every time I made a mistake. My optimism and mental toughness was barely sufficient to stop myself from degenerating into vicious cycles of hesitation and more mistakes.
On the other hand, when I decided to focus just on the game in front of me and the satisfaction of blocking a shot or scoring a basket, ignored advice, brushed off foul calls, and took the ball to the basket aggressively, I got points, respect, high fives, and lots of people coming up after the game to double-check my name. The fact that this came as a surprise to me probably belies much of my present woes.
So, I should compete more and unabashedly enjoy winning! The pursuit of being world class per Metric Eternalism is a good narrative for this.
A final note on competition: I struggle with trash talking. When I was first testing out being more aggressive and ignoring social pressure, I tried fighting trash talking with counter trash talk. After all, I’m relatively quick verbally–I should be able to shut my opponents down, right? This didn’t work for me at all. I didn’t enjoy trashing my opponent when I beat them enough to overcome the embarrassment of trashing and losing. The solution for me was again, to focus on my game and treat trash talking as a weird sort of social expectation to be ignored and disregarded.
(The Barbell (or cuz Joseph) Strategy, Antifragile): Devote resources to saving yourself from ruin. It’s more important that you can continually take advantage of serendipity than that you take the very best advantage you can right now. A lot of people I know go so hard that they forget sleep and meals (and lose bandwidth for serendipity). I am tempted by their value systems to emulate them and feel guilt for not doing so sometimes, but this is wrong, especially given I am mostly a generalist:
(Generalism, my synthesis): There is a complicated spectrum of learning types characterized by its extremes, specialist and generalist. I realized this when I read Art of Learning, an exposition of specialism by an incredibly effective specialist, and found that although Waitzkin used introspection in the same way as I did while playing the meta-game, the way he played the game itself was totally different from my dominant approach. Waitzkin internalizes skills in order of increasing difficulty into his System 1 (fast, automatic system) leaving his System 2 (slow, deliberative, conscious reasoning system) in competition to focus only at the very topmost level and attend to a few, subtle things his System 1 can’t handle yet. His method of internalization is called “form to leave form” or “numbers to leave numbers” and involves diving deeply into technical detail with the System 2 until the lessons are internalized by the System 1. He refines skills via “making smaller circles”, or practicing little pieces of core form until they can be reapplied in any context. This style of learning stands in sharp contrast to my own, which involves building many different System 2 models. At runtime, I decide which of the models are applicable to the problem at hand, reasoning by analogy (I don’t have the mental bandwidth to exactly compare all of my models to a problem). Usually one of the models in my arsenal will seem to fit nearly exactly. Then I try to make the analogy precise by matching up the problem to the model. (I “reduce” the problem to the model). If the problem can’t be reduced completely to a known model (which is usually the case) I either break out the pieces that can’t and try to deal with them separately, or try to modify the model a bit so that the problem fits. My favorite example of this is ‘my’ (probably not original) method to show that an infinite sheet of mass exerts the same attractive force on a person no matter how high she is above the sheet (or equivalently, a sheet of mass exerts nearly the same attractive force on a person within a certain range of heights above the sheet, which is relative to the size of the sheet). The method is as follows. Fix an angle. Rotate that angle to make a circular cone, extending outwards from the person to intersect the sheet. Look at the part of the sheet that is intersected, and note that the force on the person from that part of the sheet (varying with the person’s height) doesn’t depend on the person’s height. The intersection size goes as the height squared, but the gravitational force falls off as the distance squared, so the force is constant. Taking the angle arbitrarily close to 180 degrees so that the cone covers the whole plane completes the proof (with another argument or empirical observation that the force is not infinite). Once I had this method of cones, a System 2 way of analyzing forces, I could reapply it without much trouble to the problem of showing that a person inside a spherical shell feels no force. Just show that any sufficiently small-angled cone extending in both directions from the person intersects two areas of shell whose forces cancel (these intersected areas have the same ratio as the respective distances squared), and show that arbitrarily much of the sphere’s surface can be covered with a nonoverlapping set of cones. This is a very different approach from becoming extremely facile with parametrizing the problem and constructing and solving the appropriate integrals (which is the approach nearly all of my math friends favor). Note that it makes heavy use of visualization, which is the clever use of a prebuilt System 1 module to do calculations in many different contexts (without training a specialized System 1 module for each context).
Corollary 1 comes directly from realizing that specialist learning exists and is very powerful. It’s not necessarily that specialists are cleverer or smarter than you are, they’re just better trained in their discipline. Don’t feel the need to keep bashing your ego against their problems. And Corollary 2 describes a particular way of combining the two approaches to learning.
(Generalism Corollary 1, my synthesis): If you are a generalist, don’t try to beat the specialists on their own turf. And if you can’t beat them, join them–enlist specialist friends to solve your specialist subproblems.
(Generalism Corollary 2, my synthesis): System 1 Foundations. Some models and skills are important enough (i.e. basic algebra) that their foundations should be learned with System 1. This gives you the godlike specialist facility, exactly when it is necessary. I’ve noticed an incredible System 1 facility and breadth in some of my professors and haven’t been able to explain how they do it–now, it seems obvious. I’m only exposed to the basics, and these professors have developed System 1 facility in the basics of a lot of different things. This is really helpful for accurate order-of-magnitude calculations and plausibility checks in complicated situations.
With Generalism and its corollaries, I felt that I’d reached a thrilling new philosophical high point for the first time in my present liminal passage. And I could see empirical grounding for each of the ideas I’d found. I was learning how to apply them and exulting in my renewed excitement about opportunity, zeal for competition, and pride in my natural generalism. And yet, in contrast to the products of my usual axiomatic reasoning, it was hard to see what connected the ideas. Indeed, there is no set of axioms rooting my new philosophy. Yet, the twist binding the ideas is obvious in hindsight.
The unity of the approach lies in the empirical, many-pronged generalistic rehabilitation of the Spencerian Dream.
Opportunity and Robustness, the twin banners of my cheer and bravado, had been dragged in the mud before competing eternalisms, ripped and torn by vindictive mobs, soaked and buried in the snows of hesitation and world-weary depressive thinking. It was high time they were tenderly repaired, restored, and mounted again on their flagpoles.
Introspection told me there was still Opportunity in my most natural instincts. The first Leroy’s Blessing, a Past Spencer technique, fell neatly into place, and my exile in quiet ETH had showed me the virtue of the second. The third Leroy’s Blessing brought the excitement and opportunity-seeking back into the little parts of my life. I was feeling much my old self again!
The Nietzchean doctrine of No Guilt energized me, strengthened Robustness, and made me feel secure enough in my relation to others, to make the observations of Confidence and Competition. Only then could I find Opportunity to Overcome Resistance with uncomplicated joy.
The Barbell Strategy quieted those worst parts of myself who wanted to drive me to oblivion and madness in pursuit of perfection, gave me license to build foundations, and made me feel more personally robust (I already have a decent barbell).
And Generalism was to the big picture as the third Leroy’s Blessing was to the little picture; a little culmination, a mountain sneaking upwards from the molehills. I realized the nature of my strengths as a generalist (and therefore the limitations of my weaknesses as a specialist, e.g. why I topped out in chess and can’t (really) compete with mathematicians who soak all day in abstruse details, but how I can top final exams in 2 different subjects). This gave me desperately needed hope that I would, as Past Spencer dreamed before he met reality, conquer the world, make my dent and optimize over a Metric Eternalism. I feel purposeful. I’m seeking out professors who I find particularly awesome and cold-emailing them, reading papers and books with a vengeance, working on how to organize my System 1 competencies and contemplating the shape of my mind. I’m…
emerging from the workshop, covered in sawdust and stained in sweat, carrying two bright pennants and freshly cut flagpoles. Imperfect as my own two hands, but infused with excitement and ready to meet the world. Let’s run up the Spencerian Dream!