Pondering Hierarchical Knowledge
by Joseph Rowlands

In many ways, a theory that knowledge is hierarchical is obviously true. But what exactly does it mean? There are multiple ways in which we could talk about knowledge being hierarchical, and they don't really fit together.

One of the easy examples of hierarchy exists in how concepts are formed and related to one another. This theory says that our concepts ultimately are based on abstraction from perception. We see a bunch of ducks and we formulate a concept called ducks. We see a bunch of hamburgers and we formulate the concept hamburger. These first-level concepts are ones that are abstracted from our perceptions. These are the kinds of concepts that are often easier to point at then to try to describe.

Once first-level concepts are formed, we can then abstract based on similarities between the first-level concepts. So we might have a concept of green, blue, red, and yellow, and then we abstract further and formulate the concept color. And we can look at concepts like color, length, brittleness, and weight, and then formulate the concept quality. These abstractions of abstractions build an ever deeper hierarchy as our concepts become more abstract.

Abstracting further is not the only way to create concepts, though. We can divide the concepts as well. We might formulate the concept of 'group' to describe a gathering of individuals. We might then subdivide this concept into different kinds of groups, such as teams or bands or nations. We might also subdivide based on function, and formulate the concept of government. Or we can subdivide further by structure, and come up with democracy, republic, monarchy or the like.

Once you start subdividing concepts, it is possible to abstract again with a different focus of abstraction. For instance, you might abstract from specific kinds of food to the more general concept 'food', and then decide to subdivide based on the location of origin for each kind of food, like 'Italian' or 'Chinese'. We could then generalize again based on geographic region or the contents of the food.

This already complicates the hierarchical theory. It would be simple if abstractions always became more generalized as they moved away from perception. But since we are able to subdivide, you get very complicated concepts that aren't as abstract as others. The concept of 'art' or 'justice' is more complicated than the concept of 'living things'. By more complicated, I mean that it takes more steps to get to that concept.

And this is just the structure of concepts. We can turn to knowledge in general.

Imagine being in an argument about a political program. You start off arguing details of the program, but you quickly realize that your argument is more fundamental than that. You realize that they are using a different criteria for judging the political program. So you start having a more basic conversation about the standard they are using, but quickly find out that they seem to disagree one cognitive power of emotions. So you have to go to an even more basic point about reason versus emotions. But from there you start having a disagreement that stems form a different view of the validity of perception.

Maybe you don't go quite this far, but the example highlights how knowledge is hierarchical in one sense. Certain ideas count on far more assumptions than others. If you are arguing about politics, you are each bringing a vast number of more basic assumptions to the table, and you might find many of these are in disagreement. You can either end the discussion, start shouting at one another, or start moving towards ideas that depend on less.

There's an idea of hierarchy here in that more advanced topics are logically dependent on earlier, more fundamental topics. And it is recognized that you can move to one of these more fundamental topics and not have to debate the derivative topics. Once you start talking about sense perception, it doesn't matter that you disagree about immigration policy. The ideas of sense perception are not derived from immigration policy, and so the topic can be ignored. The few the assumptions, the more that can be ignored.

The interesting thing about this hierarchy is that the most basic assumptions, like that everything obeys the Law of Identity, are actually very abstract in terms of perception. If knowledge is rooted in perception, then it seems that there are two kinds of hierarchies here. One suggests that the most wide-reaching abstractions, like the primacy of consciousness, the Law of Identity, Existence Exists, and so forth, are the basic building blocks that everything else is connected to. If you argue long enough, you will eventually reach these most fundamental ideas.

And yet, these ideas are not close to the perceptual level at all. If we are talking about perceptual level ideas, we might point at the fact that objects tossed in the air fall to the ground. That is very simple, pretty concrete, and easily connected to perception. And yet, if someone doubted its validity, it would lead to discussions of perception, objectivity, identity, causality, etc.

From the standpoint of perception, the Law of Identity is a vastly abstracted concept. It may not have the conceptual complication of something like 'art' or 'justice', but it integrates all available perceptual data. So if we imagine a tree-like structure of knowledge, we could see all of our perceptual data combining to form these fundamental concepts.

The root of the concepts is the perceptual data, and these concepts are the conclusions based on this huge number of inputs. But when we talked about arguments, these concepts were the starting points, with a huge number of additional concepts dependent on them. The tree is reversed. One has these fundamental concepts as the source or root of other concepts, and the other sees them as the product of other concepts. The hierarchies seem to be reversed. Both of these appear correct, which makes the topic more complicated.

In some ways, this isn't surprising. We recognize that even very basic thinking done by children rely on certain assumptions, like the Law of Identity. Children may not think about it, almost certainly haven't heard of it, might not be able to understand it if you presented it to them, but at the same time implicitly rely on it. When they look for identity, and treat identity as a constant in time, and otherwise seek knowledge, they are acting on the assumption that knowledge is possible. They depend on the validity of the senses, their own free will, the ability to conceptualize information, the world around us is real, etc.

It's not until you start arguing with people, or otherwise need to identify these assumptions and justify them, that you start filling in the details and forming or learning these concepts. And it would be challenging without a vast amount of other information. But while the justifications come late, the assumptions are used early. This is why arguments can devolve into questions of metaphysics or epistemology. The assumptions really are fundamental. But the explicit concepts aren't formulated until much later. So they can appear as both a foundation of a hierarchy, or as the tip of a hierarchy.

This would seem to explain why they reside in very different locations depending on the kind of hierarchy you're looking at. If you're talking about the foundations of knowledge, these fundamental concepts are the implied justification. If you're talking about what it takes to explicitly formulate them, it takes a vast amount of information to get there.

So in one sense hierarchical knowledge is taken as a way of describing how arguments work. When disagreements are found, you move towards less specific details, which means towards greater generalizations. Eventually you get to something like A is A, and this is treated as if it is the foundation. But in the other sense, the hierarchy clearly begins with sense perception, which means more concrete, and works its way towards more abstract.

Discuss this Article (0 messages)