IGS Discussion Forums: Learning GS Topics: Northrop and Concepts
Author: Ralph E. Kenyon, Jr. (diogenes) Wednesday, January 9, 2008 - 10:51 pm Link to this messageView profile or send e-mail

Northrop includes "what is directly apprehended purely inductively under concepts by intuition on page 82. Since (non-mathematical) "induction" generally means a generalization from a number of particulars, I would say that David's interpretation sounds inconsistent with Northrop.

Particular "instances" would, it seem to me, correspond to individual "object" experiences, while the "concept" would be formed by generalizing from a number of such experiences. It would be assigned to a particular term such as the name "blue" - not "bluet1", "bluet2", "bluet3", "bluet4", ... etc.

Northrop actually states that "the same" concept may represent both types. Unfortunately, that sounds to me like an identification of the result of two distinct abstraction processes. But if we distinguish them as "different", then we "connect" them by that term "epistemic correlation". It still means an individual brain, is taking two distinct abstraction pathes, one from postulates, and one from "experience" (including other non-postulate words) and finding some similarity in them. (At least in today's general semantics perspective.)

If Northrop holds a concept as inhering in an assigned "definition" of a term, then we've lost Korzybski's abstraction process as well as the three mapping laws (pre-Popperian paradigm) in Northrop's system.

Author: Ralph E. Kenyon, Jr. (diogenes) Thursday, January 10, 2008 - 07:59 am Link to this messageView profile or send e-mail




Author: Ralph E. Kenyon, Jr. (diogenes) Friday, January 11, 2008 - 09:50 pm Link to this messageView profile or send e-mail

When I was first in the Forward Engine Room after MM-A school back in my Navy days, we had a fellow who graduated with honors from MM-A school, but he could not be trusted with the machinery.

I would say his "epistemic correlation" was rather minimal.

Author: Ralph E. Kenyon, Jr. (diogenes) Sunday, January 13, 2008 - 08:27 pm Link to this messageView profile or send e-mail

David wrote, As I understand it, CBI and CBP for a particular entity will typically coexist.

I believe this phraseology exhibits identification. Concepts by intuition and concepts by postulate do not relate to "a particular entity"; they represent distinct, learned by different process, objects, not a "single entity". We "connect" the two by what Northrop calls "epistemic correlation". Northop himself "identifies" the two when he said "epistemic correlation joins a thing known in the one way to what is in some sense that same thing known in a different way". He at least qualifies it with "in some sense".

We abstract from the two distinct learnings a commonality.

As I have often noted in the past, an observer may choose his or her level of abstraction sufficiently high as to emphasise the similarity at the expense of the difference or his or her level of abstraction sufficiently lows as to emphasize the differences at the expense of the simililarities. To call two objects "one" or to say that two approachs lead to "one" takes us beyond "epistemic correlation" to the realm of identification.

We must try to remember that that which we learn by intuition and that which we learn by postulates instantiate differently in our brains, and that we can correlate them. We can even abstract them both to a single brain response, but that requires a high enough level of abstraction that the source differences are excluded; they merge into "it" which is neither a concept by intuition nor a concept by postulation, but one undifferentiated by source. A "pure" concept? :-) Or do we simply "identify" them? Can we even actually do such a thing? Perhaps what we call a "concept by intuition" has some sort of supremacy in our ablity to hold objects. Perhaps whichever we learned first ends up being stronger.

I'm not confortable with the notion that a concept by intuition and a concept by postulation could represent "a particular entity". At least not if these indicate responses in brains.

Author: Ralph E. Kenyon, Jr. (diogenes) Monday, January 14, 2008 - 12:06 pm Link to this messageView profile or send e-mail

Throughout most of this discussion "Concepts" are referred to ambigously as either "formulations" or as "semantic reactions" in brains.

General semanticists frequently claim "meaning" is only in people. Of what use is a dicussion of "concepts" by intuition as a "formulation" or a "term" if meaning in indeed in people? Clearly Northrop's process of acquisition does not allow for indidivual meaning if a "concept" is represented by a term. Otherwise, the "meaning" of such a term varies from individual to individual. But that's not what the concept by intuition of "concept" seems to mean. Is a "concept" supposed to be relatively invariant from individual to individual?

I do not think we can define "concept" as a semantic reaction, because these are idiosyncratic to each individual. If we do not, then we are left with formulations - dictionary definitions - or formal technical definitions. But these seems to go against Northrop's description of the acquisition process for concepts by intuition.

All these questions... I'm inclined to follow Kendig's prohibition.

Author: Ralph E. Kenyon, Jr. (diogenes) Monday, January 14, 2008 - 01:38 pm Link to this messageView profile or send e-mail

I'll make it simpler.

"Concepts" as the union of "concepts by intuition" and "concepts by postulation".

Spoken of as a "Term" or a "response in a brain".

Author: Ralph E. Kenyon, Jr. (diogenes) Monday, January 14, 2008 - 08:15 pm Link to this messageView profile or send e-mail

David wrote "Basically, you would have no concept with which to associate the sound of the word."
David, this sounds like you are using the word 'concept' as indicating a brain response that the listener associates with the sound of a word. Is this how you are using the term 'concept'?

Author: Ralph E. Kenyon, Jr. (diogenes) Tuesday, January 15, 2008 - 12:11 am Link to this messageView profile or send e-mail

David, that is what I said earlier, that you and others have been talking about "concepts" as essentially being "semantic reactions" while also talking about Northrop's assertion that they are 'terms' with assigned "meanings".

Each of these "assumptions" has logical consequences.

A "logical consequence" of "defining", characterizing, explaining, etc., a "concept" as a semantic reactions, a nervous system plus brain response, a response in brains, etc., take the "assigned meaning" of any such concept out of the possibility of public inspection and verification, because every brain responds differently, and we have no direct way to compare one brain's response to another brain's response.

This is precisely why "The Institute" under Kendig eschewed the use of the term 'concept'; written words can be publically inspected, so we have one level at which agreement is capable of being measured. If Korzybski was in agreement with Kendig, and I suspect he was, then Korzybski's agreement with Northop may have simply been the acceptance of Northrop's "definition", characterization, etc, of both of his concepts as "terms". That would agree with the Institute's preference for formulation.

Clearly Northop's concept by postulation as a formal definition of a term, directly or as the logical consequence of postulates, squarely agrees with the Institute's preference for 'formulation', which I noted earlier.

The unfortunate situation, however, involves the impossibility of having such a terminological definition for "concept by intuition", as this depends on the individual forming an appropriate semantic reaction induced, generalized, abstracted, etc., from a limited set of examples.

If you have not been trained to use the color word 'aquamarine' and you are presented with this color, you may ambigously choose 'blue' or 'green' as the appropriate word, and even train others to under-discriminate in lerning color names.

Consequently, even as simple a concept by intuition as a primary color name depends on an individual being trained by others to utter the same word when presented with a paradigm case example. We cannot measure directly any person's semantic reaction, and we cannot compare them directly. We can only compare utterances after abstraction. (And then we are actually comparing our own private individual response to the words heard after we have abstracted them into our own private semantic reactions.)

Because of "time-binding", consisting of training each person to utter "the same words" in the presence of stimuli deemed "the same" by the trainer, we have some commonality of usage among subsets of the population. We therefore get "dictionary" (intensional) definitions for individual words, each of which stands for a "concept by intuition" that is itself instantiated differently in each individual.

It is the fact of relative agreement among a significant portion of the population that provides "relative invariance" from individual to individual for any particular "concept" - with the proviso that concepts vary significantly in their cohesiveness and variability. You might classify these by the shape of the normal distribution. A very small standard deviation would be apparent for a "well defined" or "well understood" "concept" (for example primary color names), whereas a very large standard deviation would be apparent for a "poorly defined" or "poorly understood" or even ambiguous "concept" (for example, "mind", "idea", "thought", etc.).

In "concepts by postulation" there can be no variation between (correct) definitions or usages except in so far as alternate definitions are proven quivalent as theorems.

This critera cannot hold for concepts by intuition, as they are generalized by each individual through exposure to examples which ultimately go back to direct experiences of the simple naming of categories. You may think I may deviate slightly from Northrop when I include secondary verbal (intensional) "definitions" for more abstract notions, but Northrop includes these possibilities in his chapters V and VI.

Korzybski, Kending, the institute, asks us to extensionalize by using the term 'formulation' and to speak of specific formulations. These present public time-binding records that each of use can examine independently.

Formulations express "concepts" in a many-to-many relation where proponents and opponents argue with respect to which (formulation) is best for which (concept) precisely because each person forms his or her own semantic reaction (which we may call his or her instantiation of each "concept") by his or her own intuition based on his or her own experience with more or less training in public usage.

This many-to-many relationship exists between the "concepts" and the private semantic reactions. If we do not talk about "concepts", but talk about formulations instead, that which is discussed becomes one-to many, significantly reducing the difficulty.

One formulation goes to many semantic reactions.

But a concept (by intuition) is not represented by a single formulation; it is expressend by multiple formulations, some concepts more cohesive, some less.

So we have multiple formulations each representing a normal distribution clustering about "a concept" where each formulation is one to many (semantic reactions).

There may be a conventional "best" formulation to express some concepts, but not for all "concepts by intuition", and even so, that is only the modal formulation most agree represents their understanding of the "concept" expressed by the set of formulations.

Formulations relate to semantic reactions on a one formulation to many semantic reactions basis (as many as individuals at a minimum).

A "concept", however, having multiple formulations to express it, has the relation one concept to many formulations, and each formullation to many semantic reactions, and this conposite relations ultimately has a many formulations to many semantic reactions structure.

The only way around this is to say that a concept is a term - a formulation - then it become external public property but that only works for concepts by postulation. It does not work for "concepts by intuition".

And this brings me back to my classification scheme earlier suggested.

"Formulation by postulation" and "semantic reaction by abstraction". Unambiguously defined terms and ambiguously "defined" or "described" terms. Terms which are wholly and completely specified by formulations and terms which are inherently uncertain in varying degrees because they depend upon abstraction, (non-mathematical) induction, intuition for one person to "guess" at the "intension" of the other person.

We do pretty, considering, for "concrete" classification terms, but the more abstract the term, the more variation and the more uncertainty creeps in, and wholly unobserbables such as "though", "idea", "mind", "consciousness", end up used extremely variably.

Where does Korzybski agree with Northrop? In extensionalizing to formulations ("terms"). What is lost by Northrop? Individualization of semantic reactions in "concepts by intuition"

Even though it sound like Northrop may be talking about individual brain responses when he is speaking about concepts by intuition, he is not, because he is assuming that such a concept is essentially the same for all abstractors. This is a consequence of treating it as a term rather than as a brain response. It's a very simplistic structure that is essentially consistent with Frege's "sense of a term" in which the sense of "the morning star" is not the same as the sense of "the evening star", even though both refer to the same celestial body. It is presumed that these "senses" are shared commonalities that each of us understands in the same way. This does not allow for individual semantic reactions or brain responses. Northorop may have been moving slightly in that direction when defined "concept by intuition", but he failed to realize these are unique to each brain; he treated them as one thing each of us is capable of learning, and this is doubly corroborated when he defines it as a 'term' with an assigned meaning.

To get "commonality" of meaning we have to look at the whole structure of communications between individuals, and measure their pointing and naming acts, find statisticals clusterings, and then postulate "hidden" entities corresponding to these clusterings - entities which can be apprehended by multiple individuals in more or less the same way - as measured by ussage in communication. The collection of such theoretical abstractions is what I have called our semantic environment. We connect to it by time-binding (communicating), and we learn to utter what our parents uttered when we would like someone to pass the potatoes, sometimes altering the process as time goes on.

Our semantic environment presents a structure enforced by everyone we speak to. I'm influencing that environment in a misiscule way by writing my interpretation here now, to the extent that my words have any influence whatsoever on your thinking and subsequent use of terms. The vast majority of my usage is relatively unambiguous use of relatively unambigous terms with a long history of cohesive usage, and every such usage reinforces your usage of the same terms. Common words. These form the bedrock on which we stand when we dispute about high level abstractions. Are 'the', 'an', 'and', 'but', 'so', 'in', 'large', 'many', and many others terms naming primarily 'concepts by intuition'?

We stand on frozen words to argue about the melted ones. But try to pick up and examine any particular word, including the frozen ones, and it melts in our hands. Many sublime away to vapor. Those "defined" by postulates take a little longer, because we cannot pick them up without getting a handful of others, and the others begin to melt first.

Author: Ralph E. Kenyon, Jr. (diogenes) Tuesday, January 15, 2008 - 10:04 am Link to this messageView profile or send e-mail

David, my long experience with general semantics, the philosophy of science, and philosophy in general disagrees with all your disagreements.

I will entertain any exact quotes of Northrop together with a precise (source and page) citation and your explanation of why you think that particular quote supports your claim. But simple denials and your assertions amount to little more than straw-men and red herrings.

I can see that you are not thirsty, and I see no reason to continue.