Ayn Rand, and "honest" misinterpretation?

Bill: �Okay, Mr. Brendenator, I'm girded for battle. Gimme your best shot!�
Verbal fisticuffs, Bill? Not really my style. I see debate as an opportunity for a fruitful exchange of ideas, a trade for mutual benefit, not just a battle of wits. That�s why Ed and I are such good pals. Despite our differences, we share a passion for truth. I know he respects that.

��the nature of man refers to the essence of the concept of 'man.' "Objectivism holds that the essence of a concept is that fundamental characteristic(s) of its units��

And what is a unit? An object/existent viewed in a particular way. As Rand also states: �A concept is a mental integration of two or more units possessing the same distinguishing characteristic(s)��. In the case of man, the distinguishing characteristic is rationality, and this is possessed by the units, which are the individual men. In that case, it is consistent with Objectivist epistemology to claim that someone like Galt exemplifies or embodies the essential characteristic.

�Right, it's shape may be sufficient, provided that there are no art objects that look like a table but are not intended to support other smaller objects; otherwise, it may be necessary to consider its function.�

In her definition of the concept table, Rand uses the example of a pre-linguistic child forming the concept by isolating the perceived distinguishing characteristic, shape, described as �a flat, level surface and support(s)�. But in her �adult� definition of table she slips in function: ��intended to support other, smaller objects.� Why? Because without specifying the function, the definition �a flat, level surface and support(s)� could denote any number of objects, a bed, say, or a stool or a bridge. Rand needs to specify the function in order to make the definition specific enough to denote the particular objects she has in mind.

Now, I am sure that many a pre-linguistic child has observed adults eating and socialising at a table. That observation would be sufficient to form a rudimentary concept of table. Of course, that process doesn�t involve measurement omission, so it�s no good for Rand�s purposes. Even so, she finds it necessary to enlist function in order to make her definition intelligible.

�Right, because, unlike furniture, you can grasp what a table is perceptually.�

Are you contrasting 1) the concept �furniture� with the concept �table�, or 2) the referents of the concept �furniture� with the referents of the concept �table�? If (1), perception has nothing to do with it, since concepts are thought, not perceived. If (2), your above statement is false. I can quite easily compare some referents of �furniture�, eg chairs, with some referents of �table�, eg tables.

What�s the problem here? Well, certainly the concept of furniture is wider than the concept table, since it includes table. But it�s questionable that �table� is a first-level concept, that is, one where the referents can be perceived directly, and where the resulting concept does not presuppose any other concept. The reason why is simple enough: every real table is a specific object � a dining table, a coffee table etc.

In that case, if �table� is a first-level concept, the actual object that is immediately present in perception, say, a coffee table, presumably denotes a second order abstraction. But the concept table can be formed by comparing two actual coffee tables. In other words, among other things, Rand�s theory of concept formation begs the question.

An additional problem has to do with Rand�s attempts to create abstractions that correspond to percepts, as when she says: �A concept is a mental integration of two or more units�� In the case of �table�, the units are the physical existents, two or more actual tables. What is the process by which we �mentally� integrate physical objects? I can�t think how one goes about that sort of thing. It�s certainly possible to have an image in mind of a specific table, such as one�s own dining table, but not of �table� in general.

Brendan

Post 42

Friday, January 6, 2006 - 1:14pm

I wrote, �Okay, Mr. Brendenator, I'm girded for battle. Gimme your best shot!� Brendan replied,

Verbal fisticuffs, Bill? Not really my style. I see debate as an opportunity for a fruitful exchange of ideas, a trade for mutual benefit, not just a battle of wits. That�s why Ed and I are such good pals.

Well, it was Ed who dubbed you the "Brendenator," so I was just keying off of his characterization in the good natured spirit in which it was intended. I agree with you that debate is best viewed as an opportunity for a fruitful exchange of ideas, rather than an opportunity for one person to win "at the expense" of the other.

Despite our differences, we share a passion for truth. I know he respects that.

As do I. I wrote, ��the nature of man refers to the essence of the concept of 'man.' "Objectivism holds that the essence of a concept is that fundamental characteristic(s) of its units�� Brendan replied,

And what is a unit? An object/existent viewed in a particular way. As Rand also states: �A concept is a mental integration of two or more units possessing the same distinguishing characteristic(s)��. In the case of man, the distinguishing characteristic is rationality, and this is possessed by the units, which are the individual men. In that case, it is consistent with Objectivist epistemology to claim that someone like Galt exemplifies or embodies the essential characteristic.

Let's be very clear here. Galt is an example of a a man; he is not an example of rationality. It is true that he embodies rationality in the sense that he possesses it as a characteristic, but that's not the same as saying that he is an example of it. You could say that his thinking processes or the function of his mind is an example of rationality, but you cannot properly say that he is an example of it. An ice cube is an example of frozen water; it is not an example of buoyancy; its flotation is an example of buoyancy. Fido is an example of a dog; he is not an example of barking; his vocal behavior is an example of barking.

I wrote, �Right, [the table's] shape may be sufficient, provided that there are no art objects that look like a table but are not intended to support other smaller objects; otherwise, it may be necessary to consider its function.� Brendan replied,

In her definition of the concept table, Rand uses the example of a pre-linguistic child forming the concept by isolating the perceived distinguishing characteristic, shape, described as �a flat, level surface and support(s)�. But in her �adult� definition of table she slips in function: ��in tended to support other, smaller objects.� Why? Because without specifying the function, the definition �a flat, level surface and support(s)� could denote any number of objects, a bed, say, or a stool or a bridge. Rand needs to specify the function in order to make the definition specific enough to denote the particular objects she has in mind.

Now, I am sure that many a pre-linguistic child has observed adults eating and socialising at a table. That observation would be sufficient to form a rudimentary concept of table. Of course, that process doesn�t involve measurement omission, so it�s no good for Rand�s purposes.

No, I think it would involve measurement omission. Wouldn't the child have to distinguish his concept of table from (say) a chair or stool or some other piece of furniture that he is familiar with? In doing that, he would be engaged (implicitly) in what Rand calls "measurement omission." He would be observing that a table differs in shape and size from other similar objects in his environment. Brendan continued,

Even so, she finds it necessary to enlist function in order to make her definition intelligible.

She is distinguishing a child's definition from an adult's in order to illustrate how an appropriate definition depends on the breadth and context of a person's knowledge.

I wrote that "unlike furniture, you can grasp what a table is perceptually.� Brendan replied,

Are you contrasting 1) the concept �furniture� with the concept �table�, or 2) the referents of the concept �furniture� with the referents of the concept �table�? If (1), perception has nothing to do with it, since concepts are thought, not perceived. If (2), your above statement is false. I can quite easily compare some referents of �furniture�, eg chairs, with some referents of �table�, eg tables.

I meant (2), in the sense that you can understand the referents of the concept 'table'' in a way that you cannot understand the referents of the concept 'furniture.' In order to understand what the concept 'furniture' refers to you must already have formed the concepts of 'table', 'chair', 'bed', etc., since 'furniture' refers to what is common to these concepts. That is to say, the concept 'furniture' presupposes the concepts 'table', 'chair', etc.--in other words, particular kinds of furniture--whereas the concept 'table', 'chair', etc. does not presuppose the concept 'furniture'. You can understand what a table is without understanding what furniture is, but you cannot understand what furniture is without understanding what a table is. Brendan continued,

What�s the problem here? Well, certainly the concept of furniture is wider than the concept table, since it includes table. But it�s questionable that �table� is a first-level concept, that is, one where the referents can be perceived directly, and where the resulting concept does not presuppose any other concept. The reason why is simple enough: every real table is a specific object � a dining table, a coffee table etc.

In that case, if �table� is a first-level concept, the actual object that is immediately present in perception, say, a coffee table, presumably denotes a second order abstraction. But the concept table can be formed by comparing two actual coffee tables. In other words, among other things, Rand�s theory of concept formation begs the question.

No, you can't form the concept 'dining table' or 'coffee table' without first forming the concept 'table'. After you've formed the concept 'table', you can then go on to divide that classification into sub-divisions--into different kinds of tables. But you first have to form the concept 'table' before you can do that. Yes, you can form the concept 'table' by comparing two coffee tables, but at that stage of your conceptual development, you wouldn't have identified them as "coffee tables." That identification would only be possible to someone who has already formed the concept 'table' and who is then able to identify sub-classifications within that broader classification. In short, the concept 'table' is a prerequisite not only for the concept 'furniture', but also for the concept 'dining table' or 'coffee table'. Just as you cannot grasp the concept 'furniture' without first grasping the concept 'table', chair', etc., so you cannot grasp the concept 'dining table' or 'coffee table' without first grasping the concept 'table'. Brendan wrote,

An additional problem has to do with Rand�s attempts to create abstractions that correspond to percepts, as when she says: �A concept is a mental integration of two or more units�� In the case of �table�, the units are the physical existents, two or more actual tables. What is the process by which we �mentally� integrate physical objects? I can�t think how one goes about that sort of thing. It�s certainly possible to have an image in mind of a specific table, such as one�s own dining table, but not of �table� in general.

You are assuming a kind of nominalist epistemology, which Rand rejects. As she notes in her introduction to ITOE, "The 'nominalists'. . . hold that all our ideas are only images of concretes, and that abstractions are merely 'names' which we give to arbitrary groupings of concretes on the basis of vague resemblances". In the "mental integration' of physical objects that Rand is referring to, the physical objects do not coalesce into a single, unified image; that would clearly be impossible. Rather the "mental integration" she is talking about refers to a mental grouping or classification of objects based on their similarities vis-a-vis a background of difference. Consider the following, which is an example I got from Harry Binswanger in his course "Consciousness as Identification: The Nature of Cognition and Concept Formation," available from the Ayn Rand Bookstore:

A____________B_____________________________________________C

A____________B___C'

In the first row, B is closer to A than to C, so we group A with B as against C. In the second row, B is closer to C' than to A, so we group B with C' as against A. In the first row, we've "mentally integrated" A and B as against C. In the second row, we've mentally integrated B and C' as against A (even though it's the same A and B). So the mental integration Rand talks about is simply a process of grouping certain things together based on their having a closer similarity to each other than to something else.

- Bill

(Edited by William Dwyer
on 1/06, 1:56pm)

(Edited by William Dwyer
on 1/06, 1:59pm)

Post 43

Saturday, January 7, 2006 - 2:13pm

Geezus, Bill!

Holy crap, was that ever clear and to the point! That (your purposefully-productive reasoning) was ... beautiful, man! Just beautiful.

I $#!! you not,

Ed

Post 44

Sunday, January 8, 2006 - 10:21pm

Bill: �It is true that he embodies rationality in the sense that he possesses it as a characteristic��

This is the sense I was using when I said Galt is an example of the nature of man. The difference is more semantic than substantial.

�No, I think it would involve measurement omission. Wouldn't the child have to distinguish his concept of table from (say) a chair or stool or some other piece of furniture that he is familiar with? In doing that, he would be engaged (implicitly) in what Rand calls �measurement omission�. He would be observing that a table differs in shape and size from other similar objects in his environment.�

Hang on. You begin by distinguishing between the concept table, and � presumably � the concept furniture, but you end by distinguishing between the object table and other objects. Doesn�t measurement omission apply to similarities between objects, not differences between objects, nor differences between concepts? Dissimilar objects are distinguished by perceptual differences, whereas dissimilar concepts are distinguished by their definitions.

�Just as you cannot grasp the concept 'furniture' without first grasping the concept �table�, �chair�, etc., so you cannot grasp the concept �dining table� or �coffee table� without first grasping the concept �table�.�

One can also argue that �table� presupposes other concepts, such as �man-made� and �function�. But that would miss the point, which is that the object given in perception is a specific table, not the generic table of the concept. Which means that the concept table doesn�t correspond with anything outside of the mind. At best, the concept is a representation, and its relation to the objects it subsumes is indirect. Such a concept is certainly not grounded in sense data in the way that Rand wishes.

This also points up the difference between the psychological and logical understanding of concepts. In the Objectivist hierarchy of concepts, �table� subsumes �dining table� and �coffee table�, just as �furniture� subsumes �table� and �chair�. So in the logical hierarchy of concepts, �table� is the same type of abstraction as �furniture�, just at a lower level.

Much of the problem here is a matter of semantics. �Furniture� is a collective noun, whereas �table� denotes both the concept and the object, so it appears as if the latter relates directly to the object. But as above, this cannot be the case, so there�s a mismatch between the way Rand forms her concepts and the way she arranges those concepts in a hierarchy. Ordinarily, this shouldn�t be a problem, and only becomes one when combined with an understanding of concepts as based on supposed essences. This particular problem is resolved if a term such as �table� is made to represent many different characteristics rather than some essential characteristic.

�Rather the �mental integration� she is talking about refers to a mental grouping or classification of objects based on their similarities vis-a-vis a background of difference.�

According to Rand, it�s a bit more than that: �The uniting involved is not a mere sum, but an integration, i.e., a blending of the units into a single, new mental entity.� Whether the units are actual objects, sense data, abstractions, or a combination, I can�t see how this blending actually occurs.

Brendan

Post 45

Tuesday, January 10, 2006 - 2:39pm

Brendan:

Doesn�t measurement omission apply to similarities between objects, not differences between objects, nor differences between concepts?

Ed:

Measurement omission applies to similarities between objects and � operationally � to differences between concepts. It applies to the qualitative (not quantitative) uniqueness of a group of things (from all other things).

Brendan:

One can also argue that �table� presupposes other concepts, such as �man-made� and �function�.

Ed:

Not for a 3-year old.

Brendan:

� the point, which is that the object given in perception is a specific table, not the generic table of the concept. Which means that the concept table doesn�t correspond with anything outside of the mind.

Ed:

This (your conclusion) assumes that the mind is passive � and that we have to �perceive� concepts. I�ve written extensively about this thinking error (calling it: The quicksand of PDLHK). I�ve even done this directly to/at you, Brendan. How come you don�t ever �come around?� Why do you hang on to this misconception, in spite of repeated refutation? You are a creature hard to understand. If you would care to put your reasoning (above) into a syllogistic sorites (distilled to essentials), then I would be happy to show you your reasoning error(s).

Brendan:

In the Objectivist hierarchy of concepts, �table� subsumes �dining table� and �coffee table�, just as �furniture� subsumes �table� and �chair�. So in the logical hierarchy of concepts, �table� is the same type of abstraction as �furniture�, just at a lower level.

�Furniture� is a collective noun, whereas �table� denotes both the concept and the object, so it appears as if the latter relates directly to the object. But as above, this cannot be the case,

Ed:

Brendan, you�re trading on an ambiguity. Whether �table� denotes the concept, or the object, is a matter of (rational) agent intention (what we �mean� while using it). It CAN be the case that �table� relates directly to the object, when it is our intention to speak of said object (as against the generic concept: �table�). Just because a word can mean 2 things, doesn�t license you to trade on that, for a type of side-ways criticism.

Brendan:

This particular problem is resolved if a term such as �table� is made to represent many different characteristics rather than some essential characteristic.

Ed:

In Objectivism, terms (such as �table�) ARE made to represent many different characteristics, often volumes of data. You have stated, here, a non sequitur. Objectivist terms stand for volumes of data, differentiated by contextually-essential characteristics � the only way that the human mind could have differentiated the multitude of terms (thousands?, millions?) we employ.

Bill:

�Rather the �mental integration� she is talking about refers to a mental grouping or classification of objects based on their similarities vis-a-vis a background of difference.�

Brendan:

Ed:

Bill was right. A unit is a fact-based regard of 2 or more actual objects. The �blending� is the not of the Melting Pot variety, it is of the Tossed-Salad variety. Each object in the same bowl, distinctions preserved, but separated from the rest of the meal. That is how a concept can even be (by being mentally separated � distinguished � from the rest of reality). It's a mistake to view integration as a blending, where former individuals cease to exist when they coalesce into a larger whole. That's wrong.

Post 46

Wednesday, January 11, 2006 - 2:59pm

Ed Thompson already wrote a nice reply. But I'd like to respond to Brendan as well, since his comments were in response to my previous post. I wrote, �It is true that [Galt] embodies rationality in the sense that he possesses it as a characteristic�� Brendan replied,

"This is the sense I was using when I said Galt is an example of the nature of man. The difference is more semantic than substantial.
But it's not just a semantic difference. Your original point was this:
A feature of is/ought claims is that �is� statements invariably presuppose the �ought� statements they are trying to establish. Attila was a man. So was Frank O�Connor. Which one represents the nature of man? The answer to that question will determine what man ought to do, but it will also presuppose the values one is trying to establish.
As I think I've already demonstrated, a particular individual, like Atilla or Frank O�Connor does not represent or exemplify the nature of man. So I don't know why you are continuing to press the point. I would have thought the issue to have been settled by now.

I wrote, �No, I think it would involve measurement omission. Wouldn't the child have to distinguish his concept of table from (say) a chair or stool or some other piece of furniture that he is familiar with? In doing that, he would be engaged (implicitly) in what Rand calls �measurement omission�. He would be observing that a table differs in shape and size from other similar objects in his environment.� You replied

Hang on. You begin by distinguishing between the concept table, and � presumably � the concept furniture, but you end by distinguishing between the object table and other objects. Doesn�t measurement omission apply to similarities between objects, not differences between objects, nor differences between concepts?

Measurement omission applies to similarities between objects, yes, but as against a background of difference; otherwise there would be no basis for the concept. Objects are grouped together under a single concept because they are similar to each other vis-a-vis a background of difference. A concept is a mental integration of objects according to their similarities in contrast to other dissimilar objects.

Dissimilar objects are distinguished by perceptual differences, whereas dissimilar concepts are distinguished by their definitions.

Dissimilar in what respect? Everything is dissimilar from everything else in some respect, and similar to everything else in some respect. We can say, for example, that men are dissimilar to women as men, but similar to women as human beings, or that I am dissimilar to Fido as a man but similar to Fido as a mammal. Dissimilar concepts are distinguished by their definitions, but also by the (relevant) differences in their constituent units. E.g., the units subsumed by the concept �man� differ sufficiently from the units subsumed by the concept �woman� to warrant a separate classification.

I wrote, �Just as you cannot grasp the concept 'furniture' without first grasping the concept �table�, �chair�, etc., so you cannot grasp the concept �dining table� or �coffee table� without first grasping the concept �table�.� You replied,

One can also argue that �table� presupposes other concepts, such as �man-made� and �function�.
Only the adult concept of table, not the child�s.
But that would miss the point, which is that the object given in perception is a specific table, not the generic table of the concept.

There is no generic table of the concept; there is only a group of individual tables that bear a greater similarity to each other than to other pieces of furniture that are not tables. Rand rejects the idea of a Platonic archetype�of a generic instance of a particular concept.

Which means that the concept table doesn�t correspond with anything outside of the mind. At best, the concept is a representation, and its relation to the objects it subsumes is indirect. Such a concept is certainly not grounded in sense data in the way that Rand wishes.

According to Rand, concepts do not �correspond� to the units they subsume, nor is a concept a �representation� of its units or referents. A concept is simply a mental integration (a bringing together into a single group) of the units that it subsumes. A concept is �grounded in sense data� (i.e., in perceptual awareness), because ultimately all concepts must reduce to percepts. Quoting Rand, "Furniture refers to the concepts �table�, �chair� etc., �which serve as units and are treated epistemologically as if each were a single (mental) concrete�always remembering that metaphysically (i.e., in reality) each unit stands for an unlimited number of actual concretes of a certain kind.� (ITOE, p. 21)

This also points up the difference between the psychological and logical understanding of concepts. In the Objectivist hierarchy of concepts, �table� subsumes �dining table� and �coffee table�, just as �furniture� subsumes �table� and �chair�. So in the logical hierarchy of concepts, �table� is the same type of abstraction as �furniture�, just at a lower level.

You're still not recognizing the distinction that I made in my previous post. A table is not the same type of abstraction as 'furniture' in the sense that the concept �furniture� presupposes the concepts of different kinds of furniture, such as 'table' and 'chair', whereas the concept 'table' does not presuppose the concepts of different types of tables, such as 'dining table' and 'coffee table'. You can form the concept 'table' without forming the concept of different types of tables, such as 'dining table' and 'coffee table', whereas you cannot form the concept 'furniture' without forming the concepts of different types of furniture, such as 'table' and 'chair'.

Much of the problem here is a matter of semantics. �Furniture� is a collective noun, whereas �table� denotes both the concept and the object, so it appears as if the latter relates directly to the object.

It's not a matter of semantics as long as we're clear that by �furniture� we mean the concept 'furniture', and by �table,� the concept 'table' (rather than an individual table). Given that understanding, there is still the relevant difference noted above.

Ordinarily, this shouldn�t be a problem, and only becomes one when combined with an understanding of concepts as based on supposed essences. This particular problem is resolved if a term such as �table� is made to represent many different characteristics rather than some essential characteristic.

You are correct, if what you mean is that the term �table� refers to all tables, past, present and future, including all of their characteristics.

I wrote, �Rather, the �mental integration� she is talking about refers to a mental grouping or classification of objects based on their similarities vis-a-vis a background of difference.� You replied,

According to Rand, it�s a bit more than that: �The uniting involved is not a mere sum, but an integration, i.e., a blending of the units into a single, new mental entity.�

Right, the new mental entity is the concept or mental integration as symbolized by the word.�

Whether the units are actual objects, sense data, abstractions, or a combination, I can�t see how this blending actually occurs.

I believe the process is as follows: Objects are mentally integrated (regarded as the same, with their measurements unspecified) as against other objects from which they are differentiated, and this integration is capped by a word--by a visual-auditory symbol.

- Bill

Post 47

Thursday, January 12, 2006 - 12:40pm

Ed: �This (your conclusion) assumes that the mind is passive � and that we have to �perceive� concepts. I�ve written extensively about this thinking error (calling it: The quicksand of PDLHK).�

Ed, I haven�t commented on this PDHLK notion before, because, well, life is short, but I�m baffled as to how you conclude that the likes of Locke, Hume and Kant regard the mind as passive.

Certainly, Locke and Hume may have regarded perception as passive, in the sense that what they call 'ideas" are impressed on the mind. But they definitely regarded the mind as active on those ideas. Of course, you may be using the word �passive� in a special sense. That said, the mind�s activity or otherwise has little to do with my argument.

�Each object in the same bowl, distinctions preserved, but separated from the rest of the meal�It's a mistake to view integration as a blending, where former individuals cease to exist when they coalesce into a larger whole. That's wrong.�

So it looks like Rand got her own theory wrong. Or perhaps explained it badly. But that�s bye the bye. Taking your bowl-as-concept metaphor, one may ask, how do you come to be in possession of the bowl? Do you just happen to have one to hand, in which you can toss all the data?

More to the point, what does the data consist of? With �table� we�ve seen that it�s not just �shape�, but also �man-made� and �function�. And that doesn�t exhaust the list � there�s also �height�, �weight�, �colour� and so on, all of them also concepts. So it�s bowls within bowls all the way down.

Brendan

Post 48

Thursday, January 12, 2006 - 1:52pm

Brendan, you're skirting my criticisms and switching subjects (good debating techniques!).

Ed, I haven�t commented on this PDHLK notion before, because, well, life is short, but I�m baffled as to how you conclude that the likes of Locke, Hume and Kant regard the mind as passive.

Yes you have (on Independence Day, 2004) ...
http://rebirthofreason.com/Forum/GeneralForum/0170_1.shtml#30

... the mind�s activity or otherwise has little to do with my argument.

And this is the problem.

Taking your bowl-as-concept metaphor, one may ask, how do you come to be in possession of the bowl? Do you just happen to have one to hand, in which you can toss all the data?

More to the point, what does the data consist of?

Already answered -- by both Bill and I, in the last few posts. Care to take me up on that sorites request? If you'd rather that I did, then I'll create the sorites that would have had to be true & valid (ie. sound), in order for your reasoning to even get of the ground ...

Post 49

Thursday, January 12, 2006 - 5:50pm

It's too late to make any changes to my previous post. But I now realize that it is misleading to say that two objects are grouped together because they bear a greater similarity to each other than to some other object(s). A similarity between two or more objects only makes sense if there is some other object(s) from which they can be differentiated to which they bear no similarity at all. For example, suppose that there were only three objects in existence. Two of those objects could be similar to each other in a given respect as against the third object from which they are being differentiated. But in that case, it isn't that they bear a greater similarity to each other than they do to the third object, for they would then bear no similarity at all to the third object. They could bear a (lesser) similarity to the third object only in relation to a fourth object to which they would then bear no similarity, and so on. In saying that two or more objects are similar, we must always ask: similar relative to what dissimilar object(s)?

- Bill

Post 50

Thursday, January 12, 2006 - 6:48pm

Good point, Bill.

Ed

Post 51

Thursday, January 12, 2006 - 8:36pm

Ed,

In his monograph, A Theory of Abstraction, David Kelley makes this point. He quotes Rand as follows:

When a child observes that two objects (which he will later learn to designate as "tables") resemble each other, but are different from 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....

A unit is an existent regarded as a separate member of a group of two or more similar members.... Note that the concept "unit" involves an act of consciousness (a selective focus, a certain way of regarding things), but that it is not an arbitrary creation of consciousness: it is a method of identification or classification according to the attributes which a consciousness observes in reality [ITOE, p. 7]

He then comments:

Rand is not claiming, in the passaged quoted, that table A is seen as more similar to table B than the chair is; the tables are not seen as similar to the chair at all, but as different from it. Nor does she take for granted the capacity to focus selectively on the property of shape in each of the objects: the awareness of the tables' similarity in respect of shape is the result of the awareness of each table as different from the chairs in respect of shape. The similarity between A and B is brought out by their common difference from C (the chair), by the fact that they differ from C in the same way. The assumption behind this account is that the capacity to individuate and focus selectively on the respect in which A (or B) differs form C is more basic than and underlies the capacity to individuate the respect in which A and B are similar. More generally, individuating the ways in which things differ is the fundamental ability, which makes it possible to individuate the ways in which they are similar. She is also assuming a capacity to discriminate different degrees of difference (just as the comparative similarity theories assume the ability to discriminate degrees of similarity). (p. 11)

Interesting, huh!

- Bill

Post 52

Saturday, January 14, 2006 - 11:19am

Yes Bill!

My favorite part was this (requoting for emphasis) ...

The similarity between A and B is brought out by their common difference from C (the chair), by the fact that they differ from C in the same way. The assumption behind this account is that the capacity to individuate and focus selectively on the respect in which A (or B) differs form C is more basic than and underlies the capacity to individuate the respect in which A and B are similar. More generally, individuating the ways in which things differ is the fundamental ability, which makes it possible to individuate the ways in which they are similar.

We have a distinguishing capacity, we are aware of how some things are different from other things. When I independently validated Randian (ie. correct) concept-formation, I called this fact: the Veridicality of Conceptual Discernment.

Ed

Post 53

Saturday, January 14, 2006 - 11:55am

Brendan, look at this ...

The sense-perceptions which we have, are always and only of specific things outside the mind.

We don't "perceive" generality -- we conceive of it.

When we conceive of a localized (specialized!) generality, we are mentally separating things into specialized groups.

To conceive of a localized generality, to mentally regard some things as against all other things, is to form/have a concept of those things.

Our mental regard for said things, corresponds with those things -- because it is based on facts about them, as against all other things -- those things outside of the mind.

The "essentiality" of essential characteristics is not something perceived, it is arrived at via the application of logic to experience, against a background context of knowledge.

Brendan, what is your impression of this reasoning?

Ed

Post 54

Saturday, January 14, 2006 - 2:44pm

Bill: �There is no generic table of the concept; there is only a group of individual tables that bear a greater similarity to each other than to other pieces of furniture that are not tables. Rand rejects the idea of a Platonic archetype�of a generic instance of a particular concept.�

I think we�re talking past each other here. I wasn�t referring to Platonic universals, rather, that the concept is generic, the actual tables particular.

�You're still not recognizing the distinction that I made in my previous post�You can form the concept �table� without forming the concept of different types of tables�whereas you cannot form the concept �furniture� without forming the concepts of different types of furniture��

I understand the distinction you re making. �Furniture� is a wider concept than �table�, since it refers to a greater range of objects. But I don�t accept that just because some concepts can be arranged in a hierarchy, that all concepts can be arranged so, nor that they must therefore be formed according to a hierarchy.

To test this, take the essential characteristic of furniture: �movable man-made objects within a human habitation�. On the hierarchical model, one would presumably have to have formed the concept �habitation� in order to form the concept �furniture�, even though �habitation� appears to be a higher-level concept than �furniture�.

And �habitation� also subsumes �house�, �apartment�, �office block� etc. Is �house� a first-level concept? It certainly appears so, since it can be formed at the perceptual level. In that case, �house� would be at the same level as �table�. But �house� should also subsume both �table and �furniture�, even though furniture� is a higher-level concept.

This quick test shows that the hierarchical model of concepts can�t be sustained. Relationships between the various terms are complex and not easily forced into a simple ascending scale.

Brendan

Post 55

Sunday, January 15, 2006 - 3:29pm

I wrote, �You're still not recognizing the distinction that I made in my previous post�You can form the concept �table� without forming the concept of different types of tables�whereas you cannot form the concept �furniture� without forming the concepts of different types of furniture�� Brendan replied,

I understand the distinction you re making. �Furniture� is a wider concept than �table�, since it refers to a greater range of objects.

No, that's not it. A "wider" concept is not one that refers simply to a greater range of objects, but one that depends on narrower, earlier formed concepts. For instance,

The meaning of "furniture" cannot be grasped unless one has first grasped the meaning of its constituent concepts; these are its link to reality (On the lower levels of an unlimited conceptual chain, this is an illustration of the hierarchical structure of concepts.... (The chronological order in which man forms or learns these concepts is optional. A child, for instance, may first integrate the appropriate concretes into the concepts "animal," "bird," "fish," then later integrate them into a wider concept by expanding his concept of "animal." The principles involved and the ultimate choice of distinguishing characteristics will be the same, granting he reaches the same level of knowledge.)" [Rand, ITOE, p. 22]

Brendan continued,

But I don�t accept that just because some concepts can be arranged in a hierarchy, that all concepts can be arranged so, nor that they must therefore be formed according to a hierarchy.

To test this, take the essential characteristic of furniture: �movable man-made objects within a human habitation�. On the hierarchical model, one would presumably have to have formed the concept �habitation� in order to form the concept �furniture�, even though �habitation� appears to be a higher-level concept than �furniture�.

If 'habitation' were a higher order concept than 'furniture', then it would depend on the concept of 'furniture' in order to be formed. Does it? Can you form the concept of "habitation" without forming the concept of "furniture"? I think you can, for a habitation (e.g., a tent) can exist without any furnishings. Therefore, the concept "habitation" does not depend on the concept "furniture," and accordingly is not a higher order concept.

And �habitation� also subsumes �house�, �apartment�, �office block� etc.

Again, you can form the concept "habitation" without forming the concepts of particular kinds of habitations, just as you form the concept "table" without forming the concept of different kinds of tables, as I explained in my last post. So the concept of "habitation" does not "subsume" (in the sense that it depends hierarchically on) the concepts of "house," "apartment," etc.

Is �house� a first-level concept? It certainly appears so, since it can be formed at the perceptual level. In that case, �house� would be at the same level as �table�. But �house� should also subsume both �table and �furniture�, even though furniture� is a higher-level concept.

Again, the test is, can you form the concept "house" without forming the concept "furniture"? Yes, in which case, the concept "house" does not depend on the concept "furniture" and is therefore not a higher order concept.

This quick test shows that the hierarchical model of concepts can�t be sustained. Relationships between the various terms are complex and not easily forced into a simple ascending scale.

Brendan, you're not understanding what Rand means by "hierarchy" here. It's fine to criticize her, but you first need to understand what she's saying.

- Bill

Post 56

Sunday, January 15, 2006 - 3:49pm

Brendan's �movable man-made objects within a human habitation� won't work as a definition of furniture. China cabinets are furniture but not the dishes within them. Bookcases are furniture but not the books they hold. Beds are furniture but not the linens on them. There are many �movable man-made objects within a human habitation� that are not furniture.

The concept furniture refers to china cabinets, bookcases, beds, tables, and chairs (and certain other things) but it does not refer to plates or books or linens or toothbrushes. It is necessary to first know what at least two of china cabinets or bookcases or beds or tables or chairs, for example, are in order to form the concept of furniture.

That is the hierarchy.

In order to form a higher level concept you first have to be able to identify the units that it refers to and in order to identify them you have to be able to define them.

Post 57

Sunday, January 15, 2006 - 6:51pm

Rick wrote,

Brendan's �movable man-made objects within a human habitation� won't work as a definition of furniture. China cabinets are furniture but not the dishes within them. Bookcases are furniture but not the books they hold. Beds are furniture but not the linens on them. There are many �movable man-made objects within a human habitation� that are not furniture. The concept furniture refers to china cabinets, bookcases, beds, tables, and chairs (and certain other things) but it does not refer to plates or books or linens or toothbrushes.

Right. Brendan was evidently keying off of Rand's definition, but hers carried the additional qualifications: "Movable man-made objects intended to be used in a human habitation, which can support the weight of the human body or support and/or store other, smaller objects." (IOE, p. 22) Rick added,

It is necessary to first know what at least two of china cabinets or bookcases or beds or tables or chairs, for example, are in order to form the concept of furniture.

That is the hierarchy.

That's part of it, but the conceptual hierachy that Rand is referring to is the dependency of certain concepts on other, earlier formed concepts.

In order to form a higher level concept you first have to be able to identify the units that it refers to and in order to identify them you have to be able to define them.

What do you mean by "define them"? A person can identify a chair, for example, without being able to define the concept "chair." He can point one out, but he might not be able to provide a satisfactory definition, although presumably he could if he thought about it enough, since he would have an implicit understanding of the concept.

- Bill
(Edited by William Dwyer
on 1/15, 7:01pm)

Post 58

Monday, January 16, 2006 - 1:58am

Bill,

The concept furniture is a second order concept. That means that its units are other concepts rather than perceptual concretes.

The concept furniture hierarchically depends on the concepts of, eg., table, chair, bookcase, etc.. The process of concept formation is not complete without a definition. Thus the concepts of, eg., table, chair, bookcase, etc. must be defined before the concept of furniture can be defined.

Post 59

Monday, January 16, 2006 - 10:59am