How do our minds divide spaces and create "entities?"

Question

I've been thinking about how any object can be split into infinitely smaller pieces and how we may say that there is a particular object or entity, but it has an upper portion and lower portion. In essence, what we are doing in these cases is splitting the entity into more parts.

For example, if I were to look at a table, I could see that that table is a seperate entity than the floor or air surrounding it. I could additionally go further and say that the table has legs and a table top, thus further splitting the entity of the table into multiple parts. I could then go even further to say that there is the lower half of the table leg and the upper half of the table leg. My question though, is how it is that we make such distinctions about entities and whether there are any studies or articles that can help me better understand that idea.

I know that is has to do with comparing and contrasting qualities that an object has, but why are people prone to making certain assertions about certaint entities over others? Furthermore, wouldn't that mean that the way we categorize and group things is all relative and has no basis in true reality? Wouldn't that mean that everything is a single entity?

I may have gone a bit off topic here, but, what I'm ultimately asking is this:

1. How is it that humans group objects together and contrast them by "making new entities?"

   2. If this ability is relative, does it have a concrete basis in reality?

Answer 1

Your question is so vast that it is almost daunting. I will try to approach it from a perspective of how you go about creating a model for reality (as in science or computer simulations). That might seem reductive at first, but it might be useful.

First of all, it goes both ways: it could be that you have first have a whole (e.g.car) and as you process you realize it has parts (e.g. it has wheels, engine). Conversely, you may realize it is a part of something bigger (a company fleeet, or a batch of cars in a factory, etc.).

So how do you go about making your model? If you are following a rational process, you will try to find the "best model" which, among others:

1. Fits your needs (sufficient completeness)
2. Is the simplest you can think of (minimal number of parts => Occam's Razor)
3. Gives you a correct (accurate, sufficiently precise) calculation/prediction according to experiment.
4. Hopefully will not get you in some contradiction or dead ends (consistency with itself and the rest of your knowledge).

From the above, it is obvious that:

1. This tells you how the model should look like.
2. It doesn't tell you how to get there, except by increments (pare down and toss away elements that are not strictly necessary, add new ones that are missing, subject your model to "stress tests", by making experiments or trying to find calculation errors, inconsistencies or contradictions, etc.).
3. Most importantly, the adjudication on what your "good" model should be, is on a case by case basis, according to the function of the model you need (e.g. *what is one trying to do with the cars? Producing them, selling them, managing a company fleet, or starting a personal collection?*...).

Of course, human beings did not wait for modern science in order to develop notions about the world, a gestalt about some group of phenomena, or religious/philosophical systems. We can assume that they worked as best as they could, with intuition, observation, reasoning, imagination, piecing together elements, etc. Then a natural selection among these ideas obviously took place by social interaction, betweeen members of the same group, or different tribes or cultures. People talked or read, and thus changed their minds and adapted their ideas, and the most functional model for something tended to emerge (whatever the function of the model was).

But for an introduction on how to master the scientific process of "making new entities" (which is really modeling in order to solve a problem at hand), I would recommend the classic popular book How to Solve It, A New Aspect of Mathematical Method by mathematician George Polya (written in 1945).

Answer 2

A common abstract model for slicing sensed reality consists of:

1. A continuous stream of sensory input, including of one's own movements.

2. A set of concepts and models

3. A set of schemes, methods and principles, pertaining to how to apply the concepts and models to the sensory stream. These involve pattern matching: patterns are observed within the sensory stream, and are being translated into concepts.

A classic philosophical example can be found in Kant's system. Kant argued for a rationally apriori (and therefore necessary, universal, the same for everybody) set of twelve meta-concepts, the so called categories, that must be used for forming concepts to comprehend sensible reality:

Quantity: Unity, Plurality, Totality

Quality: Reality, Negation, Limitation

Relation: substance and accident, cause and effect, reciprocity

Modality: Possibility, Existence, Necessity

Furthermore, Kant argued that there existed rationally apriori pattern schemes, the so called schemata, to guide one in forming specific empirical concepts, i.e in applying the categories to the sensory stream.

Few philosophers today believe in rationally apriori schemes for interpreting the sensory stream. But the basic model can hold for interpretation schemes that are either biological (innate), social (learned), rational, or maybe some of each.

For a specific example of implementing a method for plucking spatial objects out of the sensory stream, see Henry Poincaré's essay "On the foundations of geometry" (1898). Poincaré believed that the models that we apply to the sensory stream are mostly innate. And he argued for a special role for groups, in the mathematical - algebraic sense, in interpreting the patterns of sensory input.