|« Why Wittgenstein is Wrong about Language and Meaning||The Nature of Concepts »|
The Conceptual Atoms
The Conceptual Atoms
This post continues the series on language and meaning. The first post asked the question of how language has meaning and how we can come to know what words and sentences mean. In the several posts since, I have laid out the evidence that shows that the meaning of language ultimately comes from the ideas that one has in mind when they speak. This meaning is not objective because it is not possible to read other people's minds, at least not at that level of detail. This meaning is instead intersubjective, which essentially means that we have a certain level of mutual understanding of things that we can only know firsthand from personal experience. Last week explained a little about the nature of pure concepts and their relation to language. For today's post, I will take a shot at providing a rough approximation of the conceptual atoms from which all complex concepts, and thus any idea that we can have in mind and/or talk about, must ultimately be derived.
One of the most important factors that allow any degree of mutual understanding of concepts to be possible is the similarity of human minds. People who can form mutual understanding through communication must have an innate understanding of certain conceptual atoms that all complex concepts are constructed from. We can conclude from communication amongst each other that there are certain atoms that are virtually the same for anyone who is capable of conveying meaning to another through speech.
The main influences for this theory are Immanuel Kant's transcendental idealism, Bertrand Russell's logical atomism, and Jerry Fodor's language of thought. To briefly define transcendental idealism, Kant argued that there is certain synthetic a priori knowledge that is innate to each mind (Kant, Critique of Pure Reason, 3.B.(2).b). According to Kant, the four “forms of judgment” are quantity, quality, relation and modality, each of which are subdivided into three “categories of understanding” and a further three “pure categories”, for a total of twelve of each. Kant argued that any token experience that one may have is determined by the effect that external sense data has on these innate mental categories. Side note: there are other aspects of transcendental idealism that I don't agree with, but I won't get into that right now.
Kant's work was groundbreaking at the time, but not all of the “categories of understanding” or “pure categories” that he lists seem atomic. Notably, the categories of understanding within the form quality are affirmative, negative, and infinite, while the pure categories are reality, negation, and limitation. These six categories do very little to categorize all of the qualitatively different experiences one can have. The quality of an experience must be far more complicated than these six categories. Despite its limitations, Kant's basic framework is relevant for this new theory because any idea one can have in mind and try to convey to others is ultimately constructed from their experiences, which can be categorized.
The theory of logical atomism was briefly in a prior post. Logical atomism is relevant to this new theory because it makes sense that if there are conceptual atoms that meaning ultimately is constructed from then it should be possible to come up with symbols that are approximations of these. The meaning behind the symbols that are a used within predicate logic, and perhaps other forms of logic, are probably close representations of at least some of the conceptual atoms that are innate to the mind.
The theory I am proposing is based on Fodor's language of thought theory. Fodor pointed out that the conceptual atoms, whatever they are, must have some sort of symbolic representation within the mind when they are in the form of active thoughts so that the mind can carry out computations. Fodor argued that mental computations can only be possible if the mind creates thoughts by bringing together token instances of these symbols into a regular syntax.
The following is a working list of the atomic categories. It is difficult to pinpoint each conceptual atom because the ideas that we have are so complex that breaking them down to their atoms requires intensive phenomenological research, but this is what has been established thus far:
- Indivisible substance – This is probably the most important building block of thought. Whenever we conceptualize something that is supposed to be indivisible, this is the atom that is being tokenized within the mind. For example, if one has a conception of an electron being indivisible but having certain laws that govern its behavior, then this idea is constructed from a token of this atom along with others, such as cause, time, and space.
- Conjunction – Tokens of this operate on two or more ideas (which are themselves comprised of tokens of conceptual atoms). The English language word that most often invokes this atom is “and”. When someone thinks something like “I want apples and oranges” there is a token of this atom in their mind, along with the ideas that it operates on.
- Disjunction – Similar to conjunction except that the English language word that most often invokes this atom is “or”.
- Negation – Tokens of this operate on a single idea. Tokens of this cannot exist independently. It is possible to think about the negation of any thought that is conceivable, but it is not possible to just have a thought about nothing.
- Existence – We can conceive of imaginary things and we also can know the difference between fantasy and reality. The difference between a thought about fantasy and one that is about reality (at least as far as one understands it) is that the latter has a token of existence.
- Space – Many of the ideas we have involve something being extended in space, especially collections of indivisible substances.
- Cause – Thoughts about one state of affairs necessarily leading to another make use of tokens of this atom.
- Time – Thoughts about one state of affairs being chronologically before or after another make use of tokens of this atom.
- Various types of sensual and emotional data – This is not one atom, but many. There must be one for each type of sense experience and each type of emotional experience. Memories about how something felt or any sensual experiences make use of instances of this kind of token. To list a few examples: the sound of a trumpet, the softness of a pillow, the feeling of pain in the leg. Any token thought or memory of anything like this is comprised of tokens of atoms that are about the specific types of basic experiences.
- Another thought – It is possible to think about another thought. Any reader that is considering the atoms of thought has tokens of this atom in their mind. For example, if one is considering the properties of the statement “apples and oranges”, then this will result in a token of this atom that refers to a thought about apples and oranges. This is also possible in a higher degree, which would then be a thought about a thought about a thought. Readers who understood sentences in this paragraph likely had such a chain of thought references in their mind. This is how the analysis of language and thought is possible. No thought, however, can refer to itself. It has to refer to another thought. If it seems like any thought might be self-referential, what is actually going on is that multiple token thoughts are similar and they each refer to the same thought. It is impossible for a thought to be self-referential.
This list is only a rough approximation and there are probably many more conceptual atoms than these. I suppose it is possible that one or more of the items in this list could turn out to not be atomic but instead to be comprised of even simpler concepts. The most important point to take away from this is that all our ideas and everything we think about and talk about must be comprised of extremely simple concepts. Another important point is that the most effective way of analyzing what these conceptual atoms are is through phenomenology, which I have used in this series of posts to get to this point and I will continue to use for the forthcoming posts as I work towards the conclusion of this series on language and meaning.