Concept-formation

Metaphysics and Epistemology
1

I’m getting hung up on the description of concept-formation in OPAR (the first part of Chapter 3 on unit-perspective.) Are differentiation, abstraction, and isolation the same thing basically? It seems so to me, and why I’m confused is as follows:

Particulars are considered as members of a class (units) based on observed similarities. This means you have to observe some particulars in reality first. Then you have to observe some similarities between some particulars, which involves isolating the characteristics within each unit, which involves the mental act of abstraction. However, the only way to isolate is by differentiating because if everything were the same, there would be nothing to abstract and no way to form any concepts.

So this means that in order to perceive some particulars as similar, you have to have something to compare them against, something different. Inherent in perceiving similarities is perceiving differences between the similar things and everything else. Also similarity itself is just “less different.” So it seems all three of these terms (differentiation, abstraction, isolation) are the same thing.

Second problem, is with what differentiation, abstraction, and isolation pertain to. On what am I performing the mental act? The characteristics of the units, or the group itself?

OPAR and ITOE both describe differentiation as mentally isolating a group (on the basis of similarities) from other particulars. But in order to regard the group as a group you have to regard the members of the group as units meaning you have to isolate the characteristics of the group, which has to come first before unit-perspective is achieved. But Rand and Peikoff seems to be describing comparing the similar particulars together already, focusing in their differences to other particulars, not on their similarities. “All concepts are formed by first differentiating two or more existents from other existents.” But I have to observe their similarities first. But as above, similarities are just things that are “less different relative to something else that is more different,” so which one does “differentiation” refer to? Observing the similarities of the units (i.e. isolating/abstracting [is isolation and abstraction not the same thing?!!] the similar characteristics of the particulars which allows me to regard them as members of a group in the first place), or observing the group’s differences from all other particulars? Or do both take place at the same time?

2

Particulars are considered as members of a class (units) based on observed similarities. This means you have to observe some particulars in reality first.

The particulars are entities. The entities are differentiated from the background and each other. You don’t have the unit perspective yet upon observing some entities.

Then you have to observe some similarities between some particulars, which involves isolating the characteristics within each unit, which involves the mental act of abstraction.

Similarities at the first level are simply perceived, and the selective attention to the entities and characteristics that are similar is not abstraction.

After the differences and similarities are perceived then the isolation/abstraction and unification/integration can take place.

However, the only way to isolate is by differentiating because if everything were the same, there would be nothing to abstract and no way to form any concepts.

So this means that in order to perceive some particulars as similar, you have to have something to compare them against, something different. Inherent in perceiving similarities is perceiving differences between the similar things and everything else.

To perceive similarity requires two entities against a background. The entities must differ from the background to be discriminated. The entities must be the same or nearly the same in some characteristic which is the basis of the similarity. As discrete entities they will have different locations.

Also similarity itself is just “less different.”

Now you confuse yourself. Similarity refers to the experience of sameness or near sameness. Similarity is detected neurologically on the basis of the same sensory neurons firing when presented with the same stimuli, or the same shape or sound pattern being detected. Similar colors cause the same response from the rods and cones of the retina. Similar tones excite the same small group of nearby hair cells in the inner ear.

Similarity in more complex patterns such the shape of a raspberry bush or a song melody are also detected by the activation of the same small set of neurons corresponding to that pattern.

So it seems all three of these terms (differentiation, abstraction, isolation) are the same thing.

They are not the same. Rand describes the isolation of certain characteristics in concept formation as a process of abstraction. For first level concepts differences and differentiation is a perceptual given and is distinct from isolation/abstraction.

1 Like
3

So this means that in order to perceive some particulars as similar, you have to have something to compare them against, something different. Inherent in perceiving similarities is perceiving differences between the similar things and everything else.

I don’t think this is necessarily true if you are the “first person” grasping to form a new abstraction. When I introspect about instances in my own thinking, I find that the first step is often classifying an existent as “sort of” being in some existing category. Imagine a doctor who has been taught about (say) 5 patterns of heart-beats. He knows that heart-beats don’t fall exactly into those patterns, but the classification is nevertheless useful. Sometimes he sees a concrete pattern that is hard to classify: it is closest to Pattern B, and he might classify it as such, but he recognizes that it is one of the less prototypical ones. There are often outliers like this in all sorts of categories: e.g. the Fed pumped tons of money into the economy and broad prices are still not budging. Assuming such a difference is noted, the thinker may still classify the instance as being part of the broader class, while noting that certain special rules and considerations apply to some concretes.

Imagine the doctor sees about 5% of outliers in all the five categories of beats. After a while, he might start to notice that certain outliers actually have some commonality. (Or, the economist goes back to history and discovers a few more instance where the massive creation of money did not create upward price movements the way one would think.) This might still not lead to a brand new category, but it might.

4

So does the act of abstraction refer to focusing on the distinguishing characteristics of the units, or in mentally isolating the units from all other entities and considering them as a class? It seems to describe isolating and abstracting as focusing in on and considering separately particular characteristics of entities, then it describes isolating that group of like entities from other un-like entities. Are these two separate acts of abstraction or one in the same? Or is the second one integration? And what role does differentiation play in abstracting then?

5

Reference to an example might clarify some ideas.

There was a time when I didn’t know what graphite or ink was.

I knew that if I gripped a pencil and pressed the black end down on paper it would leave a gray/black mark that I could erase. So the similarity between the two pencils was perceptual and distinguished them from other things with entirely different uses such as paper, desk, and so on.

Later I noticed some differences between pencils. Some pencils were shorter than others, some pencils were hexagonal while others were round, some marks were darker than others. The pencils I had observed possessed the same characteristics (length, color, shape, etc.) but in different degrees. I didn’t have a word for length, but I was preparing my mind for abstracting “length”. Likewise for “color”, “shape”, etc.

Then I learned about pens. I knew that pens left (relatively) permanent marks. So I differentiated pens from pencils on the basis of the permanence of the mark. The permanence of the mark was a common denominator. It served as a sort of “axis” along which I could distinguish pens and pencils from each other. At the same time I distinguished the marking substance of the pen from the marking substance of the pencil. I learned of two types of materials: graphite vs ink. I learned that regardless of whether the particular ink involved was blue or black or red, ink was more permanent than graphite. I abstracted “graphite” and “ink” as such by mentally separating (isolating) certain attributes from all things possessing those attributes.

So first I grasped a similarity among pencils. That similarity enabled me to mentally isolate pencils from the situations in which they appeared. Retention of the similarity helped me differentiate pens and pencils from each other.

Then I abstracted “writing”, “graphite”, “ink”.

Similarity is a relationship between two or more existents possessing the same characteristics but in different degree.

Differentiation is the mental process of distinguishing two or more existents from all other existents.

Abstraction is the mental separation of a certain aspect of reality from the existents it is an aspect of.

6

So does the act of abstraction refer to focusing on the distinguishing characteristics of the units, or in mentally isolating the units from all other entities and considering them as a class? It seems to describe isolating and abstracting as focusing in on and considering separately particular characteristics of entities, then it describes isolating that group of like entities from other un-like entities. Are these two separate acts of abstraction or one in the same? Or is the second one integration? And what role does differentiation play in abstracting then?

To abstract or to perform abstraction is defined as a verb meaning to take away or to remove. The isolation Rand referred to was mentally removing the particulars under consideration from their background context by selective attention to the trait that differentiates them from that background. It is also an act of abstraction to perform measurement omission where it is the details of particulars themselves that are taken away or removed so they can be integrated together. There are two acts of abstraction which are described as differentiation and integration.

7

So if I can understand it now as this: abstraction is the mental act or process involved in differentiation and integration. Differentiation is the same thing as isolation, meaning to examine the attributes that differ from one thing to another, so as to distinguish them apart. Integration is mentally associating things together based on their observed similarities.

So basically differentiation is looking at the differences, and integration is looking at the similarities?

Is that right, and if so, do they go in that order? It seems to me I might look at the similarities first. Or I get hung up on considering both looking at similarities and differences as “isolation” and then mentally associating the things based on the isolated similarities and differences as a separate act of integration.

Sometimes Rand and Branden speak of “isolation and integration” instead of “differentiation and integration” so this confuses me. For example, Branden encapsulates the concept-formation process as basically: observing the similarities and differences among things, abstracting the similarities as distinguishing characteristics, isolating similar things from all others, and integrating them into one group. Where is the “differentiation” in that?

In the first chapter of ITOE Rand encapsulates it basically like this: when a child observes that two objects (tables) resemble each other but are different to four other objects (chairs), his mind is focusing on a particular attribute of the objects (their shape), then isolating them according to their differences, and integrating them as units into separate groups according to their similarities. Wait, why didn’t he look at the similarities first?

So does looking at similarities play any part in “differentiation”? Earlier Rand describes the third stage of the implicit concept “existent” consisting of grasping relationships among entities by grasping the similarities and differences of their identities.

Peikoff describes the same thing as after having grasped the identities of particular entities, you go on to grasp relationships among them by grasping the similarities and differences of the identities of the entities. Wait so you look at the similarities and differences together? Then he goes on to say there are two main processes involved in this: differentiation and integration. But differentiation is just grasping the differences? So when does grasping similarities come in? Is that integration? But don’t I have to observe the similarities first? Do you see how I’m confused now, how I came to think of observing similarities as inextricably involving differentiation? (E.g. So to form the concept blue, we don’t just need blue objects. We need other colored objects as well. We have to differentiate the blue ones from something else. By seeing that the blue ones are not very different with respect to each other as they are with respect to other colors like red, white, orange, green, and purple, we are able to place them in their own category. So the ability to integrate referents based on their similarity is actually part of the differentiation process. Meaning all these words are running together in my mind: abstracting = isolating = differentiating = looking at similarities = looking at differences = differentiating = integrating = looking at similarities = isolating attributes = abstracting = looking at differences = differentiating = integrating. You see why I am stuck?)

I get that point that concepts are formed by observing the similarities and differences of things and then classifying them based upon those observed relationships. That seemed simple enough. But I’m having trouble telling what is what. Which part is the “differentiation” and which part is the “integration” and do both parts involve abstraction (apparently they both do from Grames’ last post.) And where did “isolation” come from? That is involved in the differentiation process? But don’t you have to isolate something to consider it an integrated class as well? So in that case is “isolation” not just the same thing as “abstraction” i.e. to selectively focus and consider something as separate?

8

Similarity refers to the experience of sameness or near sameness. Similarity is detected neurologically on the basis of the same sensory neurons firing when presented with the same stimuli, or the same shape or sound pattern being detected. Similar colors cause the same response from the rods and cones of the retina. Similar tones excite the same small group of nearby hair cells in the inner ear.

Similarity in more complex patterns such the shape of a raspberry bush or a song melody are also detected by the activation of the same small set of neurons corresponding to that pattern.

Would these be considered physiological processes, or strictly neurological?

In regard to perception then, difference and similarity is “given” or automatic.

9

In the first chapter of ITOE Rand encapsulates it basically like this: when a child observes that two objects (tables) resemble each other but are different to four other objects (chairs), his mind is focusing on a particular attribute of the objects (their shape), then isolating them according to their differences, and integrating them as units into separate groups according to their similarities. Wait, why didn’t he look at the similarities first?

It would not be possible to look at similarities first. Similarities of what? You need to to isolate two or more discrete entities to reach that point. By isolate I don’t mean any fancy terminology, I mean simply the ability to consider one entity at a time as opposed to many at once. To take an extreme example, imagine I had been placed in a completely white room, floor, ceiling and walls all looking the same. There is effectively no way to pick out wall from ceiling or any kind of funny splotches on the wall. Everything is similar as can be: all white. You can’t isolate any discrete entity because perceptually, everything is identical. Consequentially, there is no way to pick out more than one entity because there *is* no difference to observe. To be able to isolate that wall as a discrete entity, you need to differentiate something perceived from EVERYTHING else.

If somehow a portion of the wall lit up, a green square section, I would then have some way to differentiate my percepts; there is now a ground (all the white areas) and foreground (the green square sections). More importantly, once I pick out that there is a perceptual difference (differentiation), I can then pick out discrete entities (isolation). I do wonder actually if concept formation can be done with only two entities. I need two entities that are different, I know that much, but to see similarities in SOMETHING, I need a third entity that is similar to one of those two different entities. In any case, I’ll say there are three entities, all the white areas, and two green squares. I notice that these two squares are similar as percepts, so I mentally combine them into a unit. This part is integration.

I would say abstraction is just the wider category of all these steps. Each of these steps are a type of abstraction.

10

It would not be possible to look at similarities first.

While this sounds right to me as relates to very fundamental things – e.g. childhood conceptualization – I think similarities can be the trigger for some later abstractions. One might spot similarities among a sub-set of instances within a larger set that one previously thought of as being uniform. This could then be the trigger for further analysis.

11

So I’m putting more on it than needs be? Differentiation is just isolating a particular entity from its background (including other entities) and integration is just grouping similar entities together based on their similarities. Observing the similarities is just a part of the integration process then, not a separate mental act?

12

So if I can understand it now as this: abstraction is the mental act or process involved in differentiation and integration. Differentiation is the same thing as isolation, meaning to examine the attributes that differ from one thing to another, so as to distinguish them apart. Integration is mentally associating things together based on their observed similarities.

So basically differentiation is looking at the differences, and integration is looking at the similarities?

Is that right, and if so, do they go in that order?

Yes that is right and they go in that order. Given a uniform field of blue it is possible to discriminate a single entity that is not blue based on that difference. Without that difference it would not be possible to be aware of the not-blue thing at all.

Sometimes Rand and Branden speak of “isolation and integration” instead of “differentiation and integration” so this confuses me. For example, Branden encapsulates the concept-formation process as basically: observing the similarities and differences among things, abstracting the similarities as distinguishing characteristics, isolating similar things from all others, and integrating them into one group. Where is the “differentiation” in that?

The psychologist Branden skipped over differentiation, which makes the account defective or at best a summary.

So does looking at similarities play any part in “differentiation”?

See conceptual common denominator

. . . Do you see how I’m confused now, how I came to think of observing similarities as inextricably involving differentiation?

They are both necessary and inextricable, but differentiation comes first.

(E.g. So to form the concept blue, we don’t just need blue objects. We need other colored objects as well. We have to differentiate the blue ones from something else. By seeing that the blue ones are not very different with respect to each other as they are with respect to other colors like red, white, orange, green, and purple, we are able to place them in their own category. So the ability to integrate referents based on their similarity is actually part of the differentiation process.

Actually it comes strictly after the differentiation process, so is distinguishable from it.

Meaning all these words are running together in my mind: abstracting = isolating = differentiating = looking at similarities = looking at differences = differentiating = integrating = looking at similarities = isolating attributes = abstracting = looking at differences = differentiating = integrating. You see why I am stuck?)

You forgot ‘separation’ (see IOE2 bottom of page 143, the conversation about the 710-15 tires).

Which part is the “differentiation” and which part is the “integration” and do both parts involve abstraction (apparently they both do from Grames’ last post.)

Differentiation is the first part, integration is the second part. Slapping a word on it would be the third part.

Abstraction is the genus of which differentiation and integration are species. Abstraction is not a separate act from differentiation and integration, it is those acts.

And where did “isolation” come from? That is involved in the differentiation process?

Yes.

But don’t you have to isolate something to consider it an integrated class as well?

Yes, which is why differentiation has already occurred if you are given a class to consider. Differentiation logically and chronologically precedes integration.

So in that case is “isolation” not just the same thing as “abstraction” i.e. to selectively focus and consider something as separate?

Pretty much. Isolation particularly refers to the initial differentiation which makes it possible to consider things in a new light which may never have been noticed before. I would discourage using the word “isolation” to describe the selective focus on the similarity/essence because the concept refers to whole entities not just specific attributes (the essence) in isolation.

13

So I’m putting more on it than needs be? Differentiation is just isolating a particular entity from its background (including other entities) and integration is just grouping similar entities together based on their similarities. Observing the similarities is just a part of the integration process then, not a separate mental act?

Differentiation means distinguishing units from other things/aspects. Those units aren’t necessarily entities. When you form wider concepts, you treat previously formed concepts as units.

Integration means doing something with units so the result is a new mental thing to be used as a single unit of thought. Similarities are certainly involved when you integrate concepts into wider ones.

But similarity is also necessary for making finer differentiations. When you subdivide a concept into subcategories, the distinguishing characteristic of the original concept is taken as a common denominator. The act of subdivision simply involves specifying measurements or adding on a category of measurements. “Fountain-pen” specifies how the ink is transferred to the tip. But “pen” leaves THAT aspect of the writing implement unspecified.

14

So differentiation is not simply isolating entities, as in “this” entity versus “that” entity. It’s isolating the members of the group (the soon-to-be units) according to their distinguishing characteristic(s) from all else in existence? What are the “distinguishing characteristics” then? Aren’t they the similarities (and the CCD)? Where do the differences play in this then?

Branden: We start by observing the similarities and differences among things, abstracting the similarities as distinguishing characteristics, isolating similar things from all others… .

Peikoff: We begin the formation of a concept by isolating a group of concretes. We do this on the basis of observed similarities that distinguish these concretes from the rest of our perceptual field.

Peikoff: The first step of the method is the mental isolation of a group of similars.

Gotthelf: In the formation of a concept, as we have seen, the units are first isolated from certain other objects by focusing on a distinguishing characteristic or characteristics.

Rand: All concepts are formed by first differentiating two or more existents from other existents.

Rand: In the process of forming [the first concepts] a child’s mind has to focus on a distinguishing characteristic [is this not a similarity??] in order to isolate one group of entities from all others.

WTF

I’m still having trouble seeing how noticing the differences come first, and what exactly differentiation is, and what I am differentiating from what.

15

WTF

I’m still having trouble seeing how noticing the differences come first, and what exactly differentiation is, and what I am differentiating from what.

If one is beyond the most basic childhood level, one doesn’t have to notice differences first. One notices similarities and/or differences. Imagine you see five dots, and they’re not all identical. You might see the similarity between two of them, isolate that particular attribute and differentiate them from the other three dots based on that attribute. The various quotes you provided are all compatible. For instance, these could be the way one thinks about each (of course without using many of the words that are being used to describe the introspection)

Branden: We start by observing the similarities and differences among things, abstracting the similarities as distinguishing characteristics, isolating similar things from all others… .

They’re all round, they’re all the same size, but they aren’t the same in this other way – color [abstract that attribute out] – that makes dot-A similar to dot-D.

Peikoff: We begin the formation of a concept by isolating a group of concretes. We do this on the basis of observed similarities that distinguish these concretes from the rest of our perceptual field.

I’ll group dot-A and dot-D. They both have this one similarity, color.

Peikoff: The first step of the method is the mental isolation of a group of similars.

I’ll group Dot A and dot-D because they have this particular similarity.

Gotthelf: In the formation of a concept, as we have seen, the units are first isolated from certain other objects by focusing on a distinguishing characteristic or characteristics.

Dot-A and Dot-D are similar to each other and different from the other dots, in this one way – color.

Rand: All concepts are formed by first differentiating two or more existents from other existents.

Dot-A and Dot-D are different from the others,let’s give this type of attribute a name: color.

Rand: In the process of forming [the first concepts] a child’s mind has to focus on a distinguishing characteristic [is this not a similarity??] in order to isolate one group of entities from all others.

What’s this thing that (focusing on one attribute here) that makes these two seem to belong together in the same group (this attribute distinguishing them from the other three)

All those can describe the exact same concrete instance of a person looking at five dots and isolating two by color. They don’t necessarily describe 5 different ways in which a person might approach the issue. They describe the same exact mental process.

16

So let me try to start back at the beginning and see if I can untangle myself. So differentiation is just the simple act of distinguishing entities from their background and from each other. Just looking out at the world and seeing separate things, rather than an undifferentiated blur or nothingness. Is that right?

Okay so then how does that connect to isolation? Now that I can see distinct entities, it then allows me to focus on the entities and compare their attributes. I notice some are different from others, and some are similar to others. The differences are what makes it possible for me to see similarities. All this is under the category of “differentiation”? When does “isolation” come into the picture?

Is that when I selectively focus on and consider separately a group of similar entities?

If so, this seems like a distinct third step in between differentiation and integration, and so I’m still confused.

17

If so, this seems like a distinct third step in between differentiation and integration, and so I’m still confused.

Imagine you observe a single sui generis object clashing with its background. It is singular, so you don’t need to form a concept. You might name it or label it anyway, in order to refer to it in conceptual thought but there is no similarity present. You have isolated it when you gave attention to it. That is the case for lumping isolation with differentiation: it occurs even in the absence of integration. In making it a distinct step of its own that would imply you could differentiate something without paying attention to it. Maybe there is something to theories of sub-conscious and subliminal awareness (I think they have merit) but they are not theories of concept formation.

18

So what exactly does isolation refer to? Just looking at an entity, any entity, and considering it by itself (distinguished from its background), or does it only happen when you look at all the units and consider them as similars, or both?

19

Would these be considered physiological processes, or strictly neurological?

In regard to perception then, difference and similarity is “given” or automatic.

Neurological processes are a type of physiological process, so I will take your question to pose neurological and non-neurological physiological processes as the alternatives. The evidence is for both, referencing the way light obeys the principles of optics in traversing the physiology of the eye before it gets to the optic nerve and the way sound is detected by the cochlea of the ear via sorting sound waves by their wavelength.

That difference and similarity is perceived is the position of Rand in ITOE, Peikoff in various lectures, and Kelley in his book on perception. (Consensus!)

20

So what exactly does isolation refer to? Just looking at an entity, any entity, and considering it by itself (distinguished from its background), or does it only happen when you look at all the units and consider them as similars, or both?

It refers to the separation of the object of attention from the background, regardless of number. “Isolation” is just another way to differentiate “differentiation” from “integration”, which are both methods of abstraction.