[Previous] Problem Solving While Reading | Home | [Next] David Deutsch Harassment Campaign Update for Feb 2023

Subconscious Reading; Conscious Learning; Getting Advanced Skills

Yesterday I wrote about practicing when you find any hard parts while reading. I have more to say.

First, noticing it was hard is a visible problem. What you noticed is usually under 10% of the actual problem(s). The problem is probably at least 10x larger than you initially think. So don’t ignore it. When you find visible problems you should be really glad they weren’t hidden problems, and assume they might be the visible tip of an iceberg of problems, and investigate to see if there are more hard-to-find problems near the visible problem. A visible problem is a warning something is wrong that lets you know where to investigate. That’s really useful. Sometimes things go badly wrong and you get no warning and have no idea what’s going on. Lots of people react to visible problems by trying to get rid of them, which is shooting the messenger and making any other related problems harder to find. If you have a habit of “solving” problems just enough that you no longer see the problem, then you’re hiding all the evidence of your other less visible problems and entrenching them, and you’ll have chronic problems in your life without any idea about the causes because you got rid of all the visible clues that you could.

Second, if people practiced hard reading once a day (or once per reading session) regardless of how many hard parts they ran into, they would make progress. That would be good enough in some sense even though they ignored a bunch of problems. But why would you want to do that? What is the motivation there? What part of you wants to ignore a problem, keep going, and never analyze it? What do you think you’re getting out of getting more reading done and less problem solving done?

Are you reading a book that you believe will help you with other urgent problems even if you understand it poorly? Is finishing the book faster going to be more beneficial than understanding it well due to an urgent situation? Possible but uncommon. And if you’re in that situation where you urgently need to read a book and also your reading skill is inadequate to understand the book well, you have other problems. How did you get in that situation? Why didn’t you improve at reading sooner? Or avoid taking on challenges you wouldn’t be able to do with your current skills?

Do you think your current reading, when you find stuff hard to read, is actually adequate and fine? You just think struggling while reading – enough to notice it – is part of successful reading and the solution is extra work and/or a “nobody’s perfect” attitude? Your knowledge can never be perfect so what does it matter if there were visible flaws? It could be better! You could have higher standards.

If you notice reading being hard, your subconscious doesn’t fully know how to read it. Your reading-related habits and automatizations are not good enough. There are three basic ways to deal with that:

  1. Ignore the problem.
  2. Read in a more conscious way. Try to use extra effort to succeed at reading.
  3. Improve your automatizations so your subconscious can get better at reading.

I think a ton of people believe if they can consciously read it, with a big effort, then they do know how to read it, and they have nothing more to learn. They interpret it being hard as meaning they have to try harder, not as indicating they need better skills.

What are the problems with using conscious effort to read?

First, your subconscious isn’t learning what you read well in that case. So you won’t be able to implement it in your life. People have so many problems with reading something then not using it. There are two basic ways to use something in your life:

  1. You can use it by conscious effort. You can try extra hard every time you use it.
  2. You can learn it subconsciously and then use it in a natural, intuitive, normal way. This is how we use ideas the vast majority of the time.

We don’t have the energy and conscious attention to use most of our ideas consciously. Our subconscious has 99% of our mental resources. If you try to learn something in a conscious-effort-only way, you’re unlikely to get around to ever using it, because your conscious attention is already overloaded. It’s already a bottleneck. You’re already using it around maximum capacity. Your subconscious attention is a non-bottleneck. Teaching your subconscious to do things is the only way to get more done. If you learn something so you can only do/use it by conscious effort, then you will never do/use it unless you stop doing/using some other idea. You will have to cut something out to make room for it. But if you learn something subconsciously, then you can use it without cutting anything out. Your subconscious has excess capacity.

So if reading takes conscious effort, you’ll do way less of that reading. And then every idea and skill you learn from that conscious reading will require conscious effort to use, so the reading won’t change your life much. The combination of reading not improving your life, plus taking a lot of precious conscious effort, will discourage you from reading.

It’s possible to read with conscious effort, then do separate practice activities to teach your subconscious. Even if your subconscious doesn’t learn something by reading, it can still learn it in other ways. But people usually don’t do that. And it’s better if your subconscious can learn as much as possible while you read, so less practice is needed later. That’s more efficient. It saves time and effort.

Also you can’t read in a fully conscious way. You always use your subconscious some. If your subconscious is making lots of mistakes, you’re going to make more conscious mistakes. Your conscious reading will be lower quality than when your subconscious is supporting you better. You’ll have more misunderstandings. You can try to counter that by even more conscious effort, but ultimately your conscious mind is too limited and you need to use your subconscious as an effective ally. There is an upper limit on what you can do using only your conscious mental resources plus a little of your subconscious. If your add in effective use of your subconscious, the ceiling of your capabilities rises dramatically.

Also, if you’re reading by conscious effort, you might as well use it as practice and teach your subconscious. The right way to read by conscious effort involves things like making tree diagrams. If you do that a bunch, your subconscious can learn a lot of what you’re doing so that in the future you’ll sometimes intuitively know answers before you make the diagrams.

What people do with high-effort conscious reading often involves avoiding tree diagrams, outlines, or even notes. It’s like saying “I find this math problem hard, so I’m going to try really hard … but only using mental math.” Why!? I think they often just don’t know how to explicitly and consciously break it down into parts, organize the information, process it, etc. If you can’t write down what’s going on in a clear way – if you can’t get the information out of your head onto paper or computer – then the real problem is you don’t know how to read it consciously either. If you could correctly read it in a conscious way, you could write it down. If you had a proper explicit understanding of what you read, what would stop you from putting it into words and speaking them out loud, writing them down, communicating with others, etc? It’s primarily when we’re relying on our subconscious – or just failing – that we struggle to communicate.

People don’t do tree diagrams and other higher-effort conscious analysis mostly because they don’t know how. When they try to do higher effort conscious reading, they don’t actually know what they’re doing. They just muddle through and ignore lots of problems. They weren’t just having and ignoring subconscious reading problems. They were also having and ignoring conscious reading problems. Their conscious understanding is also visibly flawed.

What should be done? You need to figure out how to get it right consciously as step one of learning a skill. Then once you’re satisfied with how you do it consciously, you practice that and form good habits/automatizations in your subconscious. This is the general, standard pattern of how learning works.

If you just keep reading a bunch while being consciously confused, you’re forming bad subconscious habits and failing to make progress. You’re missing out on the opportunity to improve your reading skills. You’re a victim of your own low standards or pessimism. If you want to be a very good, rational thinker you need to get good at reading, both consciously and subconsciously. If you don’t do that, you’ll get stuck regarding fields like critical thinking and you’ll run into chronic problems with learning, with not using and acting on what you read and “learn” (because you can’t act on what you never learned properly – or even if you managed to learn it consciously that won’t work because your conscious is already too busy – to actually do something you have to either stop doing something else or else use your more plentiful subconscious resources).

If you want to get better at reading beyond whatever habits you picked up from our culture, school, childhood, etc., you have two basic options.

Option 1: Read a huge amount and you might very gradually get better. That works for some people but not everyone. It often has diminishing returns. If you’re bad at reading and rarely read, then reading 50 novels has a decent chance to help significantly. If you’re already experienced at reading novels, then you might see little to no improvements after reading more of them. This strategy is basically hoping your subconscious will figure out how to improve if you give it lots of opportunities.

Option 2: Consciously try to improve your reading. This means explicitly figuring out how reading works, breaking it down into parts, and treating it as something that you can analyze. This is where things like outlines, grammar, sentence trees, paragraph trees, and section trees come in. Those are ways of looking at text and ideas in a more conscious, intentional, organized, explicit way.

I think people resist working on conscious reading because it’s a hassle. They read mostly in a subconscious, automatic way. Their conscious mind is actually bad at reading and unable to help much. So when they first start trying to do conscious reading, they actually get worse at reading. They have to catch their conscious reading abilities up to their subconscious reading level before they can actually take the lead with their conscious reading and then start teaching their subconscious some improvements. I suspect people don’t like getting temporarily worse at reading when trying to do it more consciously so they avoid that approach and give up fast. They don’t consciously know what the problem is but they intuitively didn’t like an approach where they’re less able to read and actually quite bad at it. Their conscious reading is a mess so they’d rather stick with their current subconscious reading automatizations – but then it’s very hard to improve much.

The only realistic way to make a lot of progress and intentionally get really good at this stuff is to figure out how to approach reading and textual analysis consciously, gain conscious competence, then gain conscious higher skill level, then teach that higher skill level to your subconscious. If you just stick with your subconscious competence, it works better in the short term but isn’t a path to making much progress. If you’re willing to face your lack of conscious reading skills and you see the value in creating those skills, then you can improve. It’s very hard to learn and improve without doing it consciously. When you originally learned to read, your conscious reading ability was at least as good as your subconscious reading ability. But then you forgot a lot of your conscious reading skill after many years of reading mostly subconsciously. You don’t remember how you thought about reading when you were first learning it and were making a big conscious effort.

You do remember some things. You could probably consciously sound out the letters in a word if you wanted to. But you don’t need to. Your reading problems are more related to reading comprehension, not about reading individual words or letters. Doing elementary school reading comprehension homework is a perfectly reasonable place to start working on your conscious reading skills again. Maybe you’d quickly move up to harder stuff and maybe not and it’s OK either way. I’ve seen adults make errors when trying to read a short story aimed at third graders and then correctly answer some questions about what happened in the story. It’s good to test yourself in some objective ways. You need an answer key or some other people who can catch errors you miss. They don’t necessarily have to be better than reading at you. If you have a group of ten people who are similarly smart and skilled to you, you can all correct each other’s work. That will work reasonably well because you have different strengths and weaknesses. You’ll make some mistakes that other people don’t, and vice versa, even though on average your skill levels are similar. There will also be some systemic mistakes everyone in your group makes, but you can improve a lot even if you don’t address that.

Doing grammar and trees is a way to try to get better at reading than most people. It’s part of the path to being an advanced reader who knows stuff that most people don’t. But a lot of people should do some more standard reading comprehension work too, which is aimed at reducing reading errors you make and getting more used to practicing reading skills, but which isn’t aimed at being an especially advanced reader. I think a lot of people don’t want to do that because of their ego, their desire to appear and/or already be clever, and their focus on advanced skills. But you’re never going to be great at advanced skills unless you go back through all your beginner and intermediate skills and fix errors. You need a much higher level of mastery than is normal at the beginner and intermediate stuff in order to be able to build advanced skills on top of them. The higher you want to build skills above a level, the lower error rate you need at that level. The bigger your aspirations for advanced stuff, the more perfect you need your foundational knowledge to be to support a bunch of advanced knowledge built on top of it.

You can think of it in terms of software functions which call other functions (sub-functions) which call other functions (sub-sub-functions). The lower level functions, like sub-sub-sub-functions, are called more times. For every high level function you call, many many lower level functions are called. So the error rate of the lower level functions needs to be very very low or else you’ll get many, many errors because they’re used so much. This is approximate in some ways but the basic concept is the more you build on something – the more you’re relying on it and repeatedly reusing it – the more error-free it needs to be. If something gets used once a month, maybe it’s OK if it screws up 1% of the time and then you have to do problem solving. If something is used 10,000 times a day, and it’s a basic thing you never want to be distracted by, then it better have a very low error rate – less than a 1 in 100,000 chance of an error is needed for it to cause a problem less than every 10 days on average.

So don’t lose self-esteem over needing to improve your basic or intermediate skills, knowledge and ideas. If you’re improving them to higher standards (lower error rates) than normal, then you aren’t just going back to school like a child due to incompetence. You’re trying to do something that most people can’t do. You’re trying to be better in a way that is relevant to gaining advanced skills that most people lack. You’re not just relearning what you should have learned in school. School teaches those ideas to kinda low standards. School teaches the ideas with error rates like 5%, and if you’re a smart person reading my stuff you’re probably already doing better than that at say a 1% error rate but now you need to revisit that stuff to get the error rate down to 0.0001% so it can support 10+ more levels of advanced knowledge above it.

For more information, see Practice and Mastery.

And I recorded a podcast: Reading, Learning and the Subconscious | Philosophy Podcast

Elliot Temple on February 14, 2023


Want to discuss this? Join my forum.

(Due to multi-year, sustained harassment from David Deutsch and his fans, commenting here requires an account. Accounts are not publicly available. Discussion info.)