Ayn Rand wrote a series of articles called Introduction to Objectivist Epistemology, and eventually became a book. It was only ever designed as an introduction, not even covering many aspects of epistemology. The one issue she thought was most critical was the nature of concepts. And that's what we'll discuss on this thread. I'll do my best to try to convey the essentials, but she did write a book on the topic. I'll also defer some of the sections of the book for later so we can discuss them in more detail. This book is rich in ideas, and is worth checking out for anyone really interested in epistemology.
So what is a concept? Well, we talked about induction already, and how we use it to take individual examples of something and generalize it. Concepts are the result of one kind of generalization. The goal is to take specific examples, find out that they're similar in some way, and put them together into a kind of category. When you create this abstraction, you can then use it to refer to all of the individual examples.
Let's take a simple example. Say I use the word 'book', as in I read a book the other day. Even though you haven't seen the specific item I'm talking about, you know a few things from the statement. You know it has a bunch of paper pages with writing on them. You know it conveys some information. This is information true of every book, and you've seen books before. That's why I'm able to talk about something you've never seen before, and yet you still grasp what I'm talking about. That's abstract thinking for you.
Rand gives a file-folder analogy. Imagine you're categorizing papers and putting them into a file cabinet. You would take all of the instances that fit under a specific category, and put them in that file. You have some method for deciding whether a particular item should go into that category, and so you define the category by that function. For instance, if you want to sort words by their starting letter, you would put everything starting with 'A' into the 'A' file. In the future, you could just refer to that file, since it contains every word that starts with 'A'. There are general properties that are true of everything in that file (in this case, that they start with A). But instead of going through them as a giant list, you could just say "The things in this file". Or, if you want to name the file, you could say "The things in file 'A'". And if you ever want to, you can look at each particular entry in the file.
That's the analogy, now lets look at another example. Take the word 'car'. Now imagine you're creating this mental file. You put every 'car' into it that you ever seen, ever will see, or can imagine. Now instead of talking about specific cars that you've seen, you can talk about the while category. Since all of them have similar features (4 wheels, fuel tank, steering wheel, multiple seats, windshield, etc.), you can discuss them in general. You can say that cars are expensive. You can say that cars pollute the air. And you can also convey a lot of information by saying referencing the category. By saying that you bought a car, we know all kinds of things about the object you purchased.
Remember the themes of epistemology I talked about? One was efficiency. We can't talk about every single car as if it were entirely different from every other one. By learning to generalize, we become significantly more efficient in our thinking and communication. Instead of trying to explain every detail of the car you purchased, you only have to point to it's distinguishing features. It's red, and sporty. Convertible. Bad gas mileage. Has SOLO emblazoned on the side. Whatever.
So we accomplish the efficiency need by using an abstraction. And it works great, as you should be able to see. You use them all the time. And they're not just objects, either. Here's a bunch of other concepts: jogging, fluffy, relationships, justice, introspection, quickly, democracy, important, butterflies. You get the idea. When you aren't pointing at something, you're probably using a concept.
So efficiency is good, but how about our connection to reality? That get's a little trickier. When we combine this information into a concept, we're doing it based on the qualities they have in common. What about all of the other qualities? If in our minds we think of these particulars as being exactly the same as one another, we're distorting our view of reality. We're integrating them into a single mental abstraction, but if to do that we end up losing information about the world, then the result is no longer accurate. If the price for efficiency was a disconnection from reality, it'd be a bad deal.
Fear not! We don't actually lose any of the details. Just like the file folder analogy, we're not dismissing the individual instances, we're just putting them into a category. We still have access to each individual instance. When we discuss the concept, we're discussing all of the instances, which means each and every one of them. A concept always refers to all of it's elements. There are some things they all have in common, and we can acknowledge those similarities, but we don't ignore the differences either.
I'll next talk about concept formation, which explains in more detail exactly how concepts are constructed. But as for this post, the point you should take away is that we live in a conceptual world. The bulk of our thinking is in terms of concepts. It's not surprising that an entire book is dedicated to this one aspect of epistemology.