The unreasonable effectiveness of doing, then thinking; vs thinking too much before doing
In other words, you may have looked at someone and gone:
Wow, they're really smart. Thinking really works for them.
But why does it work for them? Because first, they did. And/then they learned from it.
(Welcome back to learning about how minds work from neural networks. Today we will be covering the importance of: 'data' also known as 'a good train set'. Other methods act in an environment, but this also produces a dataset. The question of quality is resolved in the case of AlphaGo...)
“why is this rash, you’re not trying to spear-hunt a fucking bear, you're writing some code in the privacy of your own home
If you start exercising you might...become an Olympic athlete. Such a terrible fate often awaits those who head down this slippery slope. Each cries 'that probably won't happen to me'. And then it does.
Anyway, I’m not going to try to convince anyone to do then think. I can’t prove nor even properly conceptualize why you should start doing that. But I want to try and at least think about a few reasons why doing before thinking, or at least doing something that seems like doing before thinking, is so unreasonably ineffective.
Tools are created to solve a problem.
You don't become a writer by saying 'I want to become a writer'. Maybe you say 'I want to write this story'. Or maybe you don't - you just write a story.
Maybe you become a blogger by blogging.
It’s important to note that riding a bike can’t be first learned at a theoretical level.
First? Who has learned it at a theoretical level?
Anyway, we’re talking about things that are more complex than bike riding here
I'm going to guess that bike riding is complex.
mechanics
What do you mean by this?
Alas, we are usually in lack of life and death scenarios
Something is wrong with this sentence. It's still readable, but-
So it might be that doing than thinking
then
is unreasonably effective because we trade-off on wasting time and producing anxiety that might make us never take useful action, to begin with.
Yeah. A brain saying 'you're not doing anything, why think' makes sense - as does, doing things that aren't related:
Sure, it’s good to question the fundamentals of math every now and then, but maybe it’s not best to do it as part of solving a mildly annoying but certainly-doable calculation
What are the fundamentals of math?
It’s also worth thinking if you’ve gone a bit too extreme with your dieting, but doing so in the middle of a diet is bound to always result in you rationalizing “yes” to be the answer in order to splurge on sweets.
When is the diet end point?
maybe they are all bad because I didn’t spend more than 30 minutes consciously ruminating on the subject,
Earlier:
Even though we have a lot of “mental power” available at any given moment, most of what we do with it feels “unelected”, we lack understanding about why conscious thoughts flow the way they do.
Perhaps by doing then thinking you will
(not sure about the conscious bit. Also not sure why conscious thought is so important.)
Also - on tales of monkeys:
It's not obvious that circumstances have lots to do with
instead of
*Flexible enough accomdate:
Also about the cyborg story:
AGI will take over the world.
What would an elf do?
Write poetry.
Do people think that taking over the world is easy if you don't have self doubt?
It's idiosyncratic and highly variable (across people and across time and topics for a person) just how much thought is valuable compared to action (and compared to "just don't it").
I find it very useful to separate "collecting data", "planning actions", "predicting future feelings", in order to notice when I'm doing good decision-making work vs just spinning. And of course, also notice when I'm putting more work into the decision than the differential value of the result will bring.
I hesitate to give blanket advice "don't think so much", because I know a number of people who need to hear the opposite advice: "please think for a few minutes before you speak or act". I do not hesitate to give the advice "a few minutes to determine whether this is a fairly low-impact decision, or one that benefits from further thought." I'm also a big fan of time-boxing. Spend no more than 5 minutes determining how long you'll spend considering before committing.
Out of curiosity, could you give an example of such a decision where some may benefit from thinking more?
With two caveats that I left as implicit:
Just like this classic! https://slatestarcodex.com/2014/03/24/should-you-reverse-any-advice-you-hear/
I found this post interesting but I think there is something wrong with it, even though I estimate that its central point has value. My remarks focus on "this post as an advice" rather than the "phenomenon explanation" part of the post.
Perhaps the first thing I should say is that I agree some people have a tendency to think too much in certain situations. To delay the first try for too long and to waste a lot of time optimizing entire lines of though that will be revealed to be worthless after five minutes of concrete work. I am one of those people and I have spent time thinking (eh) about when and how this must be improved and corrected.
But I do not think a general cheer for the policy of "acting before thinking" is a good thing. I can cite two examples from my personal life, in the last 10 days, where a dozen hours or more were wasted because someone did not spend half an hour thinking when it was "obviously" sensible to do so. The first one involves writing code without thinking, creating bad code that had to be managed afterward (technical debt), the second involves personal relationships and is private (sorry).
I estimate that this post presents danger on the same scale as the gain it might deliver. In fact I expect this post to be a net negative in terms of direct advice. Ideally, the wisdom found in this post would have its place in a more complete framework on decision processes. As it stands, I am tempted to compare this advice to "to walk, move your right leg forward", dangerously misleading and incomplete without the complementary advice "also move your left leg".
Perhaps a better immediate advice would be.
When applicable and when your actions will not have non-trivial lasting consequences think just a little then act, make an attempt. Only after your have tried enough can you think again, if you believe it to be useful.
If we apply this advice to different timescales we get both your second and third example. We however do not get the first, insofar as there can be lasting consequences to a botched application.
I'm sure it's a common term in some corner of some subculture and it also seems pretty easy to infer what you mean, but could you explain the meaning of "left tail-risk"? Is it "events with low probability but high negative magnitude"? There are a bunch of other potential explanations, but this one seems to cover X-risk stuff well, which I assume would be the reason someone on LessWrong would use a term like you did.
A few typos.
Also: Shouldn't it be "unreasonably effective" in
a few reasons why doing before thinking, or at least doing something that seems like doing before thinking, is so unreasonably ineffective.
If there’s one lesson that I learned from almost every single smart person I know, it’s:
This is not something I’ve been able to convince anyone of, at least not with a rational argument. I might tell people things like:
Or
Or
And I can’t for the life of me convince people about doing these things because they seem “rash”... and I keep thinking (therefore I know I exist): “what the fuck are they thinking” (therefore I know they exist as a separate entity), “why is this rash, you’re not trying to spear-hunt a fucking bear, you're writing some code in the privacy of your own home, the place you masturbate in, surely the shame of failing miserably can’t be that bad”.
Anyway, I’m not going to try to convince anyone to do then think. I can’t prove nor even properly conceptualize why you should start doing that. But I want to try and at least think about a few reasons why doing before thinking, or at least doing something that seems like doing before thinking, is so unreasonably ineffective.
Something Something Subconscious Immersion
You all know the drill here.
Brilliant scientist thinks a lot about a problem, gathers all the data that doesn’t make sense and all the data that already makes sense and needs to keep making sense, struggles for a few days and comes up with nothing... then, he wakes up from a dream drooling inspiration, runs to his notebook in a trance and figures out relativity/microscopy/the periodic table/DNA structure.
If we assume that, at any given time, a large part of brain networks are inactive or below the state of conscious control, then it’s only a short inference away that the patterns of thought that’d bring them online are best triggered by actually doing the tasks and letting said networks “figure out” that they need to to help and how.
Examples of this that everyone’s noticed are learning to ride a bike, row, climb, surf, etc. Complex motor skills are learned with minimal conscious input, the conscious mind just decided to get on the bike as best it can, it receives instructions like “push with your leg in a sort of rotation motion”... and 1 hour later the network that can actually handle bike riding has learned how to do it. It’s important to note that riding a bike can’t be first learned at a theoretical level.
Rational Thought Is Best Past a Certain Data Threshold
Anyway, we’re talking about things that are more complex than bike riding here, so I get that the analogy doesn’t fit perfectly, but remember that bit about theory not being useful for beginners? I assume it’s not the same for professionals.
Professional bikers might spend hours a day discussing cycling theory, optimizing things besides bike riding like their doping regiment, training adjacent skills that they think are particularly important but hard to train “while doing the thing”. Professional sportspeople in general presumably do this quite a lot, though I expect it’s much more common at complex sports (think football, skiing) than at simple ones (sprinting, javelin throwing).
This might in part work because rational thought is good at polishing up ideas and behaviors but can’t start from a position of insufficient data, or rather, is really dumb from a position of insufficient data.
Let's think of a complex skill that people are really anxious about practicing, let’s say picking up people or interviewing with an important manager. How’s intuitively going to be better, someone that read a dozen interviewing-and-business/pickup-artistry books and watched videos and prepared bits to say... or someone that’s already done it 4 or 5 times (even if they failed)? Almost certainly the latter.
Theory and rationality seem to only help from a position where data is plentiful to think with, where there are real-world examples to think about, where you’ve already “felt” a system from the inside. Sure enough, people seem to have come up with a promising theory about mechanics much sooner than about waves, and for seemingly no fault of experimental devices, it’s just that we “do mechanics” in our day-to-day lives.
Thinking Is Meant For Dangerous Scenarios
Let’s say thinking involves several disadvantages, it wastes time, it stresses you out, it makes you less likely to do anything in the long term (because doing anything is stressful and time-consuming), and so on.
This would still make thinking advantageous if you have to go spear that bear since you want that kind of scenario to be a once-in-a-lifetime thing, so the extra stress and time is not only worthwhile for a slightly lower chance of death, it’s the appropriate signal to send: “This is really hard and dangerous, don’t do this ever again”.
Alas, we are usually in lack of life and death scenarios, and furthermore, we are trained to think too much in easy situations that should require no thought (e.g. solving math problems at school) in order to get a slight edge over someone else (e.g. during an exam).
So it might be that doing than thinking is unreasonably effective because we trade-off on wasting time and producing anxiety that might make us never take useful action, to begin with.
An important bit there is to remember that the network/self/you which thought the action is good is often not the network doing it. Maybe you did ponder the idea of getting a new job for a lot, decided on yes, and now the pesky you actually doing the job searching is thinking through every single job as though I’d be a life-and-death decision whether or not to apply. This leads me to...
We Often Think About Stuff That’s Already Decided
Sure, it’s good to question the fundamentals of math every now and then, but maybe it’s not best to do it as part of solving a mildly annoying but certainly-doable calculation. It’s also worth thinking if you’ve gone a bit too extreme with your dieting, but doing so in the middle of a diet is bound to always result in you rationalizing “yes” to be the answer in order to splurge on sweets.
Thinking often leads us to quickly rethink at the “meta” level where a decision was already made, potentially a very thoughtful and thorough decision we made a while ago.
An important note here is that this should bias us against both thinking too much while doing something that’s been decided upon and towards planning fewer things at the “meta” level, leaving the doing for some other-self.
We Can’t Control Thought Through Anything But Action
Even though we have a lot of “mental power” available at any given moment, most of what we do with it feels “unelected”, we lack understanding about why conscious thoughts flow the way they do.
But a pretty reliable way to control thought is to put yourself into a situation where thought must take a certain pattern.
If you jump through a small hole in the roof of a bear’s den, wooden spear in hand, barely not awaking the fiend from hibernation, you’ll probably have more optimal thoughts about bear-killing than if you tried to “imagine” the scene. But the trade-off is risk. Similarly, with social situations, risks of saying the wrong thing or grooming the wrong monkey could have been rather high. So we grew to love “imagining” and be afraid of action.
What we think of when not acting will be a function of what seems “most important”, thus many people end up ruminating about things such as politics, socializing and phobias. What we think of when acting will be the process itself, thus shortcutting the prioritization process.
None of this is a conclusive answer to the question, much less so a proof, but these are some of the intuitions I have about it.
As an aside, I conceptualize all of these intuitions while writing, maybe there were already formed in my subconscious, maybe they were created as a result of the imagined constraint of having to write this article, maybe they are all bad because I didn’t spend more than 30 minutes consciously ruminating on the subject, or maybe they are as good as the answers of someone that meditated upon the question his entire life... If only I could know for sure.