Some people trying to learn philosophy start with big concepts like: precision, clarity, high content, social roles, individualized, chosen roles, pre-made knowledge, static memes, traditional knowledge. this is a mistake

properly, these concepts are integrations of many ideas. an integration means combining several things together. people should first know lots of details, concretes, etc. then get big concepts like these second (via integration).

thinking should go more like this (which, btw, is part of Objectivism):

you learn from little things. you learn 25 little things, and you notice that 5 are related. you integrate those together into a concept. then you learn some more little things. then you notice that several are related and you integrate them together into a single concept.

you repeat this lots of times. instead of dealing with every little thing individually, you create some bigger concepts that deal with multiple little things.

then the same process repeats with those concepts. when you have integrated little things together 25 times, so you have a bunch of concepts, you then have a bunch of stuff to keep track of. and you may figure out some of those concepts are related. so you integrate together several concepts into bigger concepts.

to a first approximation, thinking proceeds in a hierarchy like this. you start with little things at level 1. you combine some to make bigger stuff – level 2. you combine some level 2 stuff to get level 3 stuff. you combine some level 3 stuff to get level 4 stuff.

this lets you build up to bigger ideas, rather than starting with them.

combining ideas has a purpose. it's too hard to keep track of everything individually. combining things makes them more manageable. integration of stuff into broader concepts makes one's knowledge more usable and useful.

you can't think about too many separate things at once. your knowledge needs to be organized. combining things lets you hold more in your head, actively, because each individual thing has more knowledge.

there is a trap many learners fall into. once people have big concepts, e.g. level 20 concepts, they use them. and the learners see these great concepts and go "cool! i want that!". then learners sometimes start using the words from the level 20 concepts. but they don't have all the underlying knowledge. it's just words – vague abstractions, minus most of the content.

knowledge doesn't have to always be created in this organized way. people who try to skip ahead and understand a level 20 concept someone else used don't know nothing. it's not a complete zero. they can pick up some sense of what it is, how it relates to reality, how it's used in discussion, how one uses it in life, etc...

but these people who skip ahead are usually missing most of it and end up using it poorly. this is a similar issue to cargo culting.

it's important for people to build up their own knowledge. you have to create your own understanding. you have to go through your own learning and integrating process. other people cannot think for you.

one of the advantages of learning through this integration process is you can unwrap concepts. you can go back down the hierarchy and see where an idea came from, and where each underlying idea came from, and so on. normally you deal with the integrations, but you can also go examine the lower level stuff when there's a problem or other reason to. (another reason would be to explain it to someone else.)

this has a roughly pyramid structure. having the whole pyramid lets you think and discuss at any level. it lets you take new information at any level and integrate it into your thinking. and it lets you figure out if a new fact contradicts you or not, because you actually have all the underlying ideas that your high level worldview is integrated from. if you only have the top layer, you don't really understand it.

knowledge isn't strictly like a pyramid though. that's a pretty loose approximation. like Karl Popper discussed, it can be more like a spiderweb. that means no up or down directions, just lots of connections between different ideas in any direction.

leaving the metaphor behind: a level 4 concept doesn't have to be integrated only from level 3 concepts. you could integrate together some level 2 concepts, some level 3 concepts, and a level 6 concept.

and ideas don't have to be labelled with what level they are. the levels are just approximations. they are just there to help you keep track of things, not as an important claim. what matters is which ideas were integrated from which. keep track of that and you're ok. pyramid levels are a visual aid to understand the concept. but when you do it, it won't all divide up neatly into levels. it gets a bit messy. which is no big deal.

don't be rigid by trying to organize all your ideas into levels. and don't try to only ever think from smaller ideas to bigger ones. it's also ok to sometimes use a big high level idea to learn about a lower level. actually that's pretty common once you know a lot.

it's rare at first to go from big ideas to little ones because you don't start out with high level ideas when you're born. and when you look around, you see things like a table or iPhone, not a big concept like justice or virtue. but once you have big ideas, then you can use them all over the place. e.g. you can combine some low level experimental evidence with a high level idea about proper methods of science to reach the conclusion that the evidence is no good and should be ignored.

all this, btw, mirrors how programming works. programmers write code. when they notice several parts of their code do roughly the same thing, they refactor it into one bigger concept – a e.g. function, method, class or library. those can themselves get refactored into bigger things if several are similar. dealing with everything individually is too hard, so programs build up layer on layer on layer, with each layer helping make things more manageable and reducing the raw amount of stuff to deal with.

programmers also notice several lines of code work well together and think of them as a conceptual unit. if the code does these 5 steps, then the result is to save the document. so they make a save document function and put those steps there. or sometimes programmers just put related code adjacent with some whitespace or comments before and after.

it makes sense that programming mirrors epistemology since the main task of a programmer is to organize knowledge. (and btw i'm not just talking in theory about how programming should work. this is what people actually do. many programmers who don't read philosophy would do a process like this and be able to roughly explain what i just said.)

people have different world views, different pyramids, different ideas differently organized. so to communicate you can't just talk about the top of your pyramid. it's the lower parts of the pyramids where people have more in common and understand each other better. at the top, even if they use lots of the same words, their thinking is usually rather different because it's integrated in a different way to get to those big concepts.

a way to help communicate better is with examples. this helps because the examples bring up some of the lower level ideas, rather than just talking about the highest level stuff (higher level means integrations of integrations of integrations of integrations). when an example is itself a high level concept, then it doesn't work well because it doesn't mean the same thing to different people. examples should be kept simple so people have a better shot to take away the same meaning from them.

(Did you like this? Click for a list of all the essays.)

By Elliot Temple, Dec 2015 | Join the discussion group, receive the newsletter, or send comments to

Acknowledgments: some ideas presented are modified from, or inspired by, ideas from Karl Popper, Ayn Rand, David Deutsch and William Godwin. Thank you!