David Lymburner wrote, Your statement got me thinking about the Structural Differential (SD). In particular, I was wondering at what point in the SD, the notion of "concept formation" occurs. It seems like it should fall somewhere between the O and D, but I'm not sure. I'd be curious to hear what people know about that. Perhaps I should start a new thread.

Good idea to start a new thread.

The structural differential is an extremely abstract model. The damned thing only shows one level for all neurological processing, which we know from neurological studies has perhaps hundreds of levels. Moreover, the verbal levels are all "labeled" "labels".

In "concept formation", it depends upon whether you want to consider a "concept" as a semantic reactions prior to its formulation in words, or if you want to consider it a well-defined verbal formulation.

Both these levels operate together with the verbal level representing the output for communicating the formulation of the concept to another. I wrote A Notion of Concept a long time ago. Here are some web definitions of concept.

I am reminded of Piaget's distinction between assimilation and accommodation. "Learning takes place through adaption and Piaget believed that adaption had two aspects, assimilation and accommodation. Assimilation is the aspect that conserves the organisation of information and absorbs it into existing schema, and accommodation occurs when an object cannot be assimilated and the schemata have to be modified to include the object (Piaget, 1981)."

I would think that "concept formataion" occurs at neurological levels, but "concept formulation" occurs at verbal levels, and the latter expresses the former.

I'm inclined to agree, but none of those levels are presented in the structural differential. The structural differential only shows the event level, the object level, and verbal levels. We need a much more detailed, more up-to-date, and more comprehensive model than the structural differential. A couple of general semanticists have attempted to provide an expansion with just a couple of more levels, but, as you may have noticed, the majority of active general semanticists continue to refer back to the original. The object level encompasses all sensory and non-verbal neurological processes as one block. It's certainly not very useful. It might be a sponsored research project to try to develop various models with a few more levels.

Visualization has traditionally been thought of as ideas rather than as concepts.

In philosophy "thought" has a long tradition of being conceived of as that which I express as "interiorized imitation of language". "Thinking" was described as having a propositional structure corresponding to the language expressing it. In Frege's time such "thoughts" had "senses" that could be shared. More recntly a more formal theory called the "Language of Thought" has been published. See The Language of Thought Hypothesis, for example.

"Thought" is contrasted with "ideas", which traditionally have been conceived of as more image like. The "ideas" we have are more picture like - according to more traditional (and careful) philosophy.

I see these two notions as related in the abstraction hierarchy in terms of levels with "ideas" holding a lower level position and "thought" holding a higher or more abstract level.

I'm inclined to "think" that any concept can be readily expressed in a formulation. I would not call any image or idea from which such a verbal formulation can be abstracted a "concept".

Both the examples David provided would fall under the more traditional usage of the term 'idea'. From ideas (and notions) we abstract more coherent internalized formulations, and such a formulaton that can be clearly expressed would be entitled to be called a "concept", but not a mere image (idea), or even a connected sequence of action ideas. Concepts also have or are entitled to have a noun name (even a gerund phrase, such as "mountain climbing" or even such a particular instance as expressed by "John Doe planning to murder his wife's lover.").

Frege's examples were the Morning Star and the Evening Star. These are different concepts, but both refer to the same celestial object.

But imagining remembering seeing the first bright point of light in a bright evening sky is not the concept, although the "idea" of seeing it may likely stimulate recalling the concept - and you "think" to yourself "the evening star" or "Venus" (different concepts, same referent).

We need, in general semantics, to have a number of terms that represent brain activities but that clearly differentiat the activities in both order and level of abstraction, and we need to have clearly defined "concepts" for these terms.

Clearly 'formulating' might be the top level such term that expresses actual speaking or writing, presumably the last stage in outputing our brain responses.
Traditional "thinking" (interiorized imitation of speech with syntatical and logical structure) might be one level below formulating.
A verbal equivalent of "notion" - something less coherent than "concept", might be another, lower, level.
"Ideas" and "images" "imagining" being earlier or lower.

We can "define" the terms according to our ontology and specify clearly what kind of mental/brain activity is referred to.

Another research project?

Jabberwocky is full of such words.

Proper names of individuals also do not represent concepts (except in a few rare cases when a concept is named after somebody).

That seems like a reasonable abstraction, but I don't have a feel for what the proportion might be. I question a few (pun intended) including lots or some, speaking of which, neither is a concept.

But just to muddy the waters, is "idea" a concept? How about "notion"?

The general usage of theses is too varied and imprecise to provide the kind of precise answer we need to build a structurally adequate model.

concept idea notion

I think that, as general semanticists, it behooves us to create formulational distinctions, taking into consideration the most well thought out and articulated discussions that best serve our particular ontology. In short, we should look carefully at the time-binding associated with these terms, carefully pick the most reasoned ones, and modify them to suit our own purposes - of understanding "mental" processes in terms of levels of abstraction, and then publish a standardized technical definition for use within general semantics.

To this end I propose that we NOT consider the similarites in the use of the terms. It's far more important to clearly formulate diferences that apply across levels of abstraction.

Here's the order (without the definitions) as I see them, from high to low:
Formulation.
Thought.
Notion.
Idea.
Impression.
Percept
After-image
Image.

1. I would consider a "concept" as a particular variety or kind of "thought" - except when it is explicity formulated and written down - defined. Then we would be distinguishing the verbal level expression of it from it.

2. All of the terms above except 'formulation' would fall under the object level; 'formulation' falls under the verbal levels.

This is my point about expanding the structural differential into more detailed structures. Break the object level down into smaller levels.

3. Ontolgy would represent the things general semantics is committed to the existence of. Recent philosophy speaks of "ontological commitment", and that would apply to what we take our formulations to refer to.

There are several ontological commitments in general semantics formulations. They include the "existence" of levels of abstraction, of abstracting in nervous systems, the existence of nervous systems, sensory responses, and many more.

See also A Notion of Concept.

Incidently, the term "thought" itself has been labeled an elementalism in general semantics circles, because it also is strongly associated with "mind" also believed to be elementalistically split from "body", so we must use something like "I think-feel with my body-mind".

See Think-Feel and Know-Act.

Concepts, notions, ideas, percepts, perceptions, etc., all refer to abstractions that have no external physical visibility, so we cannot hold them up with show and tell. As such they "exist" in our virtual symbolic environment, and everyone's access to that envirenment is through their own personal unique and idiosyncratic time-binding history. The pletheora of definitions, which I've cited in previous posts, shows that there is a great deal of variation among the formulations presented for each of the terms. Because of this fluidity, we cannot have a clear picture of what each other is talking about unless we each provide our own formulational definition, and when we do, due to the variety in personal experiences, we are likely to get "arguments" as each advocates his or her own "pet view". Perhaps some will convince some others to change, as is natural with conversations and sharing. What we do NOT have as any "official" general semantics dictionary of "approved" formulational definitions for these terms. Moreover, because these are abstractions without a demonstrable physical instantiation, we cannot "get extensional" about them either. The best we can do is to research the time-binding record for various uses and definitions of the terms, and develop some criteria for which of the records should be granted more credibility, and then see if we cannot establish some preferred formulations. Should this be on the Institute's "to do" list?
"Hardening of the categories" is a problem, but so is insufficient rigidity where "structure" is needed. One is inflexibility; the other is impotence.

Perhaps the problem with "concept" might be more easily seen if you copy any particular sentence in which the word is used, substuting the word 'formulation' for 'concept' (per general semantics dictates), and then examine what the subsequent sentence seems to say.

General semantics advocates being extensional in order to reduce or eliminate communications errors. That, it seems to me, is the motivation to move from talking about "concepts" to talking about formulations. Formulations are written down and can be looked at extensionally. "Concepts", (as well as "notions", "ideas", "percepts", "perceptions", etc.) cannot be looked at extensionally. Such things are "presumed" to "exist" from a metaphical perspective, but no one can hold one up physically to show one, consequently we cannot be extensional about them in the way that we can about written words, that is to say, formulations. The difference is that "how we know", by extensional observation - is an epistemological perspective, and so is being extensional, that is, depending on empirical knowledge.

So, until we have some repeatable way to consistently measure extensional referents for "concept" (as well as "notions", "ideas", "percepts", "perceptions", etc.), they will continue to fall outside the realm that general semantics deems is the proper area for discussion.

General semantics eschews metaphysical discussion and speculation in favor of epistemological discussion and emprirical - extensional - examination.

So applying that to the original question, general semantics would say that the question is not valid for general semantics. The question must be restated without using the word "concept". That would mean going back to Piaget and looking at the activities that he described as indicating "concept formation" - that is - being extensional, and then re-asking where those activities fit into the structural differential.

I was just talking to the Devil, and I'm charged with advocating for him.

Notice that "elementalism" is "splitting apart" verbally that which empirically cannot be split.

Notice that "abstracting" is "splitting apart" and discarding, both at object levels and at verbal levels.

We cannot function as a multi-level being without abstracting - that is - splitting apart.

When we abstract we figure one split and background or discard the other.

Let's see, "You must abstract but don't elementalize."

Go figure.