Slightly different though similar YouTube version:
Effective Altruism Forum Version:
Trying is not the same as doing something. It is easy to feel like you are trying to do something, without having any chance of actually succeeding. It is important to create a clear mental model for success, test the assumptions of that model, determing the absolute single most important next action each step along the way, break down the goal into manageable steps, recognize and avoid noble obstacles, be willing to fail and iterate, and be willing to fail at less important things. Maybe we could have a social exercise to make sure we are actually working to robustly achieve our goals, rather than just “trying.”
In Eliezer Yudkowsky’s post “Trying to Try”, he reviews Yoda’s phrase “Do or do not, there is no try,” explaining that at first he found this highly dubious, sounding like what he calls “pretending to be wise”, saying something enigmatic and profound sounding which is actually superficial or meaningless.
After more deeply considering the matter, he came to the conclusion that the phrase actually is quite meaningful, as typically when people say they are going to try to do something, or in fact even when they say they are going to do something, they actually really only mean that they are going to put in the minimal amount of effort to plausibly feel like they might achieve it, or to be able to convey to others that they are working to achieve it.
He explains that it is much easier to “try to make $1 million” versus actually making $1 million, and it is much easier to “try to launch a startup” then to actually launch a startup.
I was quite floored to read this and realize that this deeply applies to myself as I want to launch a Hybrid Market which I think could completely transform the trajectory of humanity if successful, but which is insanely hard to actually succeed at, and extremely easy to “try” to succeed at.
Eliezer points out that one of the main ways we get in our own way is that rather than asking “What needs to happen for this goal to actually be achieved?” we ask “What can I do to achieve this goal?” or “How can I do my best to achieve this goal?” which are really clever ways of asking “Without making a substantial effort or disturbing my life in any meaningful way, how can I use the current spare resources at my disposal to make an attempt that plausibly moves me toward this goal?”
I found this very useful, and extrapolated that to actually achieve the goal of creating a Hybrid Market (which could be a stand-in for any ambitious goal,) I would need to create a detailed mental model of a path to create a Hybrid Market, accounting for planning fallacy and perhaps even Hofstadter’s Law (though this is of course impossible,) so that my plan is robust enough that it is not just plausible that I could achieve the goal, but eminently likely if the plan is executed properly.
This plan might include preliminary steps like using experiments to test my mental model and the underlying assumptions which need to be confirmed for the mental model to be accurate. It then requires executing on the plan at a high level over a long period of time.
Another thing I realized is that one of the biggest obstacles I will face in actually achieving my goal is that I will have to confront failure many, many times. I think this is the biggest thing which prevents me from actually achieving my goal, versus trying to achieve my goal.
When I look at my next actions list, which I often order by priority, I noticed a strong pull toward the end of the list, as I get very anxious looking at the first thing on the list which is often very difficult and presents a strong possibility of failure as well as requiring a lot of effort.
Knowing that there is a strong chance there may be little or no payoff for large effort feels very demoralizing, so it is much easier to do things that I can convince myself are urgent (even though it doesn’t really matter if they are done at all), or which plausibly could make the more important task easier by freeing up mental space or giving me some “small wins” or otherwise increasing my resources or moving me slightly toward my goal.
What I am realizing is that it takes incredible mental discipline to look at all of the things I could possibly be doing, notice all of my mental and emotional pull toward doing things that are plausibly important (also called “noble obstacles”) but that I know in my heart of hearts is not actually really that important, and have the fortitude to remember how important my most important goal is, and know that I will have to make many sacrifices of lesser important-ish goals, even just for a shot at the possibility of achieving my most important goal.
I think one thing that Eliezer misses which is not quite a counterexample but may be complementary, is that there are many incremental iterative steps to achieving the most important goal. This is not the same as doing a bunch of random small things that plausibly could move you toward the goal, but instead means finding the absolutely single most important thing you can be doing and breaking it down into small doable steps as necessary, rather than trying to achieve the whole goal at once.
I wonder how many other EAs are bogged down by “trying to try,” doing things which plausibly seem like they could be the most effective, or thinking about how to be effective, or talking and debating about how to be effective, but not actually ever being effective. I think it’s probably quite a lot of people in the movement, maybe even everyone does this sometimes. As such, I think it could be very high EV for people in the Effective Altruism movement to examine how they operate and see if they are “trying” to have the greatest possible impact, or are actually having the greatest possible impact they can have.
Perhaps this could even be done in a social event similar to “Hamming Circles,” in which EAs ask each other “What are the important problems in your field, and why aren’t you working on them?”, an event already popular among some EAs. This could basically be a follow-up question that is something like “Are you actually taking action that is exceedingly likely to solve the most important problem in your field, or are you just taking action that plausibly moves you in the right direction without too much planning and high level execution? And if so, what needs to happen for this problem to actually be solved?”
To sum up, I really liked Eliezer’s post and hope my thought processes can give some outline to how this obstacle might be overcome. Main ways of doing this are making sure you have a very clear mental model of how to actually achieve your goal, testing the underlying assumptions on which your mental models depend, continually determining the absolute single most important next action you can take and then actually taking that next action instead of noble obstacle distractions, and breaking down the absolute single most important next action into manageable bite-size actions so that this is actually feasible. Then iterating toward the goal again and again, accepting failure as part of the process. Also, accepting that you’re going to have to sacrifice and fail at lesser important goals to achieve your most important goal. Finally, I think EAs could maybe have a social exercise, an “actually succeeding” circle or something (sorry, I suck at naming things) where we support each other in seeing how we are distracted from doing the things that actually are most important for achieving our most impactful goals.
Hope this is helpful, appreciate any criticism or other ideas for how to actually succeed!