[Note: backported from LessWrong]

This is the third post in a series discussing my recent bout of productivity. Within, I discuss two techniques I use to avoid akrasia and one technique I use to be especially productive.

Deregulating Distraction

I like to pretend that I have higher-than-normal willpower, because my ability to Get Things Done seems to be somewhat above average. In fact, this is not the case. I'm not good at fighting akrasia. I merely have a knack for avoiding it.

When I was young, my parents were very good at convincing me to manage my money. They gave me an allowance, perhaps a dollar a week. When we would go to the store, I'd get excited about some trite toy and ask my parents whether I could buy it.

Their answers were similar. My mother would crouch down, put a hand on my shoulder, and say "Of course you can. But before you do, think carefully about how much you will enjoy this after you've bought it, and what other things you would be able to buy if instead you saved up."

My father was a bit more direct. He'd just shrug and say "It's your money", with the barest hint of derision.

I rarely spent my allowance.

I now use a similar technique when dealing with distractions.

(It's worth noting that it's always been very easy to put me into far mode, perhaps in part because I decided at a very young age that I wasn't going to die.)

As Kaj Sotala and a few others noted, assigning guilt to non-productive tasks is not especially healthy. Nor is it, in my experience, sustainable. In a few different cases, I experienced scenarios where I wanted to do something but couldn't will myself to do it. I suffered ego depletion and hit a vicious cycle of unproductivity and depression. I never fell completely into the self-hate death spiral, but I flirted around at the edges. It became clear that I needed a new strategy.

To break the cycle, I decided to stop fighting myself.

The world is full of distractions, and I have plenty of vices. I am just as susceptible as anyone to binging on TV shows or video games or book series. Instead of trying (and often failing) to stop myself from indulging, I decided to allow myself to indulge whenever I really wanted to.

"It's your time", I told myself.

This changed the game entirely. I no longer willed myself to avoid temptation: I weighed temptations alongside my other options, took their pros and cons into account, and made an informed decision. Did I need to distract myself? Sometimes, the answer was yes.

Knowing that I could no longer trust myself to bail me out if I got addicted to new media, I took special care in removing as many distractions as I could from my environment. Because I'd resolved not to spend willpower to cancel addictions, I became much more cautious at the point of entry. These days, I ignore recommendations about new TV shows and books, preferring not even to learn the premises, thus dodging the temptation entirely.

By allowing distractions a place in my mental calculus I allowed myself to choose between them with more care: I am able to watch movies instead of TV shows, to read standalone books instead of entire series.

I know full well that my resolution against spending willpower against myself means that once I get addicted to something, it has to run its full course before I can be productive again. This is a nuclear option: because I know that I won't stop, I am very leery of lengthy media. I avoid open-ended addictions (ongoing online games, chemical addictions, etc.) like the plague.

I refer to this strategy as "playing chicken against myself": because I know that I'll let long addictions run their course, I seldom have to.

From another perspective, you could say that I deregulated a black market on distractions: By lifting the mental ban on entertainment, I was able to price it accurately and weigh the tradeoffs. If there is a new book I want to read, the answer is not an outright and unenforcible "No". Rather, it's "can we afford to be underproductive for the next few days?". And when the answer is negative, it's significantly easier for me to postpone gratification than to resist the temptation entirely. The end result is that I have much more control over when I indulge in escapism.

Finally, I've found that this feels a lot better than feeling guilty about being unproductive. It's a healthier state of mind, and it's led to a general increase in happiness.

Moving Towards the Goal

My teachers used to tell my parents that I have two modes of operation: I either put in the minimum possible effort or I blow expectations completely out of the water. They claimed I have no middle ground.

This isn't quite accurate. The truth is, I always put in the minimum effort. Anything else would be wasted motion. The discrepancy they observed was not due to some whim of passion, it was an artifact of how our incentives were misaligned.

In school I was incentivized to ace classes with minimal work. I was very good at obeying the letter of the law while blatantly flouting the spirit, and I had a knack for knowing exactly how far I could push my luck. My teachers had… polarized opinions of me, to say the least. I was an arrogant kid.

Yet when my schoolwork happened to align with some personal goal — mastering a new technique, figuring out new secrets of the universe — then I was relentless, shattering expectations with apparent ease. A number of my teachers took it upon themselves to press upon me just how much I could do if I actually applied myself. I didn't bother correcting them. If they weren't going to invent a grade higher than 'A', why should I waste my efforts in the classroom? I had better things to do.

Like I said, I was an arrogant kid.

This experience in school had two important repercussions. First, it taught me to seek out the gap between the intended rules and the actual rules. I developed a knack for it, and this has served me well in many walks of life. Noticing the space between what you meant and what you said is a fundamental skill for programmers. Math is a tool designed to narrow such gaps. Logical incompleteness theorems are statements about the gap between what logic can say and what mathematicians want to say.

Secondly, and more relevant to this post, school helped me make explicit the virtue of putting in the minimum possible effort. Authority figures parroted the value of hard work, but that's only half the story. You should always be putting forth the least amount of effort that it takes to achieve your goals. That's not to say that you should never do hard work: in many situations, the easiest way to achieve your goals is to do things right the first time. I'm not condoning shoddy work, either: if quality is part of your goal then you'd best do things correctly. If you're trying to signal competence, then by all means, put in extra effort. But you should never expend extra effort just for effort's sake.

This leads us to my second trick for avoiding akrasia: I am not Trying Really Hard. People who are Trying Really Hard give themselves rewards for progress or punishments for failure. They incentivize the behavior that they want to have. They keep on deciding to continue doing what they're doing, and they engage in valiant battle against akrasia. I don't do any of that.

Instead, I simply Move Towards the Goal.

I don't will myself to study. It is not a chore, it is not something I force myself to do. That's not to say I enjoy studying, per se: it's hard work, and the reward structure is pathetic compared to programming. If I had to force or convince myself to study lots of math continously, I don't think I'd get very far.

That's not how I operate. I don't Try Really Hard. I simply Move Towards the Goal.

This is where the previous post ties in. I've mostly eliminated the guilt I feel while unproductive, but I've maintained two very important things from that era of my life:

1. In my head, long-term satisfaction is linked to productivity.
2. I have maintained habitual productivity for years.

Between these two points, I know that once I've settled on a goal, I'm going to more towards it.

This is, internally, an immutable fact, made so both by habit and by crude Pavlovian training. None of this is explicit, mind you, it's just the nature of goals. I can change the goal and I can drop the goal, but I can't hold the goal and not pursue it.

I never decided to study really hard. You can "decide" not to watch the next episode of that TV show only to sternly berate yourself three episodes later. My decision to study hard was made on a lower level, it's been internalized. Acting on goals is the thing that System 1 does regardless of what System 2 "decides".

System 2 controls things by picking the goals. It was a long and arduous process to internalize my most recent set of goals, the ones that have driven me to study hard and become a research associate and so on. It took a few months and a bit of mindhacking, and that's a story for another day. But once the goal was chosen, marching towards it was out of my hands.

System 2 isn't in control of whether I move towards the goal. Instead, it spends its time doing something it's very good at: finding the most efficient path. Minimizing effort.

I don't actively force myself to study hard. Rather, the structure of the environment is such that the shortest path to the goal requires hard studying. I merely follow that path.

Moving Towards the Goal might look a lot like Trying Really Hard from the outside. Superficially, the two are similar. On the inside, though, they feel very different. I've Tried Really Hard before, and I'm not good at it. It requires exertion of willpower and results in depletion of ego.

When I'm Moving Towards the Goal, I don't worry about whether things will be done. I've outsourced that concern to habit. Instead, mental effort is spent looking for the shortest path, the easiest route. Difficult paths do not require additional willpower, because the internal narrative is not one of expending effort. If anything, a difficult path is worth extra points, because it means I'm pursuing admirable goals. Internally, I'm not Struggling Against Akrasia. I'm Finding an Efficient Route.

Don't get me wrong, studying math at high speed for five months was hard. However, I have built myself a headspace where hardness is not an obstacle to overcome but a feature of the terrain. I am going to march on regardless. System 2 doesn't have to spend effort convincing System 1 to move forward, because System 1 is going to move forward come hell or high water. Thus, System 2 spends its time making sure that the march is as easy as possible.

This leaves me free to try new techniques to achieve my goals more effectively, and that leads us to our final trick for the day.

Level Hopping

I started doing NaNoWriMo in 2011, and I noticed something interesting: a vast majority of winners barely made it to 50,000 words. The goal of NaNoWriMo is to write 50k words in a month, so I wasn't particularly surprised. However, from my interactions with others I found that a vast majority of these winners felt like they were pushing themselves to the limit, even though many of them were probably psychologically anchored below their actual limits. After all, in my experience, the hardest part of NaNoWriMo is writing every day: the most difficult part of being productive is switching contexts, once you get rolling it's not difficult to keep rolling.

It seemed clear that if the goal had been 60k, many of the same people would have eeked out a victory with similar margins and the same narrative of butting against their limits. The natural conclusion was that I can't trust myself to feel out my own limits.

This is when I decided to start hopping to higher levels of productivity. These days, I occasionally throw wrenches into my study plans when I think I'm growing complacent.

"Those set theory and category theory books were easy", I'll say, "Let's try skipping introductory logic and going straight to model theory".

Or, "All this studying is great, but I bet I could keep it up and also do a NaNoWriMo for 75k words".

Often, this fails spectacularly. Sometimes, I am at or near my limits, and skipping an intro logic textbook to dive straight into Model Theory is a really bad idea. Other times, I find out that I actually was just hovering around an anchor point, seduced by a narrative of linear improvement.

This is not an original idea, by any means. In fact, there's a relevant Bruce Lee quote:

There are no limits. There are plateaus, but you must not stay there, you must go beyond them. If it kills you, it kills you. A man must constantly exceed his level.

- Bruce Lee

My point, more broadly, is that this is the type of thing that occupies my mental narrative. I'm not wondering whether I will be able to convince myself to study each day. Instead, I'm gauging whether I'm reading the most effective material. I'm noticing that it won't be enough for me to just learn the material, I also have to signal that I've learned the material (and that I should start doing book reviews). I'm monitoring to see when I've grown complacent and looking for ways to keep me on my toes. This is process is doubly useful: It helps me sidestep akrasia and it also helps me become more effective.

These are my three Light Side tools:

1. I've constructed an environment in which productivity is habitual. In the absence of distractions, I trust myself to get things done.
2. I've lifted my mental ban on distractions, and trust myself to use them wisely.
3. My mental narrative is one of expending minimal effort, not one of trying to succeed: instead of worrying about whether I can continue, I worry about how to perform better.

Most of these tricks are likely familiar: I do not claim originality; this is merely an account of the methods that I use, the things that work for me. Consider this to be evidence that these techniques work for people who share my personality (which I've tried to illustrate along the way).

You now have a broad sketch of how I maintain productivity, but it may seem somewhat unstable, difficult to maintain indefinitely. The next post will detail my Dark Side tactics: tricks I use to remain unrelenting and sustain my vigorous pace, but which may make rationalists uncomfortable.

After that, I'll tell the story of a kid who decided he would save the world for reasons completely unrelated to existential risk, and how he came to align himself with MIRI's mission. This will help you understand the source of my passion, and will conclude the series.