Maybe people have asked this before - it seems a fairly basic question to ask, in a similar vein to a child's chain of "Why?" questions.
If you ask me "What is a chair?", I respond that it is an item of furniture. If you then ask "What is an item of furniture?", I reply that it is a type of object that exists in a habitable space. If you ask me "What is an object?", I reply that it is a type of "thing" that has concrete existence in the world of space and time as we know it. Ask me "What is a thing?", though, and I'm struggling!

  • Hi, welcome to Philosophy SE. Your question is a bit vague for us, more like a forum discussion topic. We take more specific, pointed and definitive questions that come up after general reading. The most general classification types are traditionally called categories, they are studied since Aristotle. There is an obvious limit to the genus-differentia type of definition you are using, base categories have to be introduced differently.
    – Conifold
    Aug 14, 2019 at 8:12
  • @Conifold Thanks for helpfully classifying my thought process! As you say the limit to this type of definition is obvious, but is there a reason for this limit? As mentioned by curiousdannii, are we left with a concept of primality beyond which it is impossible to venture further?
    – Alun Lewis
    Aug 14, 2019 at 8:51
  • The reason is that any chain humans can build must come to an end. But the "primality" only limits a particular system of concepts, we can always embed it into a larger one if the need arises. And we can describe the base concepts implicitly, in terms of their interrelations, as in axiomatic systems like Euclidean geometry or set theory; or operationally, in terms of how they are used in practice, as in sciences.
    – Conifold
    Aug 14, 2019 at 9:26

3 Answers 3


Perhaps there is. According to the research of the Natural semantic metalanguage project, the problem of circular definitions is solved through the identification of semantic primes, the basic blocks of meaning which are shared by all languages and cannot be meaningfully subdivided. Any attempt to define a prime will end up circular or more convoluted than the word itself. Part of defining non-primes is to identify what something is a KIND of, however THING is prime itself, so it's no surprise that you're struggling to abstract it.

But the NSM project is attempt to identify the primes of language, not the limits of abstraction. In addition to THING are other nouns like SOMEONE, PEOPLE, and BODY, all of which could be said to be a KIND of THING. Similarly, for the verbs, it not only has DO, and HAPPEN, the most basic actions and events, but also many specific verbs which can all be considered KINDs of DO or HAPPEN: THINK, KNOW, WANT, FEEL, SEE, HEAR, SAY. So while NSM gives us good reasons to say that there are natural limits to abstraction, it doesn't identify what all of those limits are.


This type of questioning is familiar in the Upanishads. Eventually, this will take you to Tat Tvam Asi--A Sanskrit Phrase that means "That thou art". So, the abstraction stops when one realizes Consciousness. You would find more about these types here: Mahāvākyas

  • Sounds a little bit like the Christian God's self-referential definition of Himself in the book of Exodus as "I am who I am". In that sense he cannot be classified.
    – Alun Lewis
    Aug 14, 2019 at 8:56
  • If it was about Consciousness, I mean...without any bit of ego, I believe it is the same. To classify, we need at least one more thing to arrange in classes or categories according to shared qualities or characteristics.. So there is no need of a classification. Aug 14, 2019 at 9:18
  • We have this urge to metaphorically put our concepts in boxes and label them, at least I do. Kind of difficult to do this with God. As you say, that which is unique cannot be classified. Interesting that you say that there is no need to. Maybe I shouldn't seek to classify Him but rather to know Him.
    – Alun Lewis
    Aug 14, 2019 at 9:27
  • When you say you are trying to know Him, it indirectly implies that you are excluding Consciousness or vice-versa. Aug 14, 2019 at 16:58
  • Why do you say this?
    – Alun Lewis
    Aug 15, 2019 at 6:52

We use abstractions to think about the world.

Specifically, an abstraction is meant to refer to some part of the world. There are different levels of abstraction because we can use an abstraction to refer to the part of the world already referred to by different abstractions we have already in mind. Thus, we have levels of abstractions, from the low levels to the higher levels. Abstractions on a higher level refers to a larger part of the world than each of the abstractions on the level below that they are an abstraction of.

We can analyse the world from different and a priori independent angles. However, for each particular angle, abstractions form a well-ordered structure, such that a lower-level abstraction always corresponds to only one of the higher-level abstractions at each higher level of abstraction. In other words, the part of the world referred to by the lower-level abstraction is never referred to by two different abstractions at any of the higher levels of abstraction.

Broadly, we have more general abstractions to refer to larger parts of the world.

It appears that the human brain is remarkably nimble playing that game. Each of us has our own private abstract representation, and therefore our own private collection of abstractions. We need these abstractions both for thinking about the world in the privacy of our mind, and for communicating with other human beings, which explain to a large extent the difficulty we have in understanding each other.

Shared culture, education, dogma and profession all help greatly in simplifying communication, but even when they share all of these, different people will still differ somewhat in how they abstract at least some part of the world.

This makes communication inevitably more difficult but is also essential for conceptual progress, whereby the abstract model of one individual may gradually spread to a group of people and from there possibly to the rest of society, and even to humanity as whole.

The brain is clearly very efficient in this respect. The need is to be able to define any number of well-ordered abstractions from bottom to top, starting life as a new-born with only the bottom level, our inchoate perceptions, only gradually increasing over the years the sophistication of our model of the world, through both private thinking, communication with other human beings, and also probably through the unconscious work of our brain.

The word "thing" in English has come to be used to refer to our highest-level abstraction. However, while anything all will be a thing, we only use the word "thing" to simplify communication, not to think about the world. Most of the things we can think of have no relation we could think of. Both justice and redness are things, yet we don't normally think of these two things as having any kind of relation.

The highest-level abstraction which has definite relations and to all other abstractions is reality. Reality for a new-born can only refer to the bottom level, i.e. their immediate and largely inchoate perception of the world. Life thus can be resumed as the building up of an intermediary and ever more complex structure of abstractions coming in between the bottom of our immediate perception of the world and reality as an abstraction in our mind.

From a topological point of view, then, yes, like an infinite past with a beginning, there is a limit to abstraction and no there isn't.

Reality is the highest abstraction we know of, and thus it is the upward limit to abstraction. However, it is just as true that there is no theoretical limit to the number of intermediary levels of abstractions that we can insert in between perception and reality. There is a practical limit which is of course the size of our brain and the little time we have available to think up with new abstractions.

Obviously, we all will have our own private and therefore particular abstract model of the world, and thus for each of us, at any given time, we will only have a finite number of levels of abstractions.

Another crucial limiting factor is the fact that, for fairly obvious practical reasons, we have to borrow a large part of our abstract model of the world, and this essentially through culture, education, work activity or dogma.

Most people won't waste their time building for themselves more abstraction levels. They will take for granted whatever they will be able to get from other people and be content with that. However, some people, broadly "thinkers", will apparently choose to spend a large part of the time available to them during their lives doing essentially that. For example, some philosophers, some scientists, some religious thinkers, and probably many more outside these categories.

However, broadly speaking, humanity as a whole seems to evolve--perhaps more quickly now due to the apparition of mass media, mass tourism, immigration, free trade, science, and the Internet--towards a common abstract model of the world with a particular number of abstraction levels, thus possibly giving the impression to many people that the number of levels is limited not only in practice but also in theory.

But no. You do as you please.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .