Broken units

[piz 0]

There are many documented cases of human abnormalities, such as three legs, two noses, two mouths, etc. But, in general, these present no more of an epistemological problem than the blind man. These are still men, but we qualify them as "a man with two noses," a man with three legs," etc. They still remain man.

Right, and I covered that in my parenthetical comment.

Now, it is metaphysically possible for a new species to develop, standing side to side with man, with enough different characteristics that there is an epistemological need to differentiate that species from "man." But then we just expand the definition of man to include a distinguishing characteristic(s) by which we further differentiate man from the new species. Where that line is drawn and the specifics by which it is accomplished is an exercise of little philosophical concern; a study of biology and anthropology provides a warehouse full of examples.

Good, and important, point.

[Betsy Speicher 0]

If we discovered that there were rational spiders on Pluto, then would you classify them as "man". The context is that we cannot travel to Pluto and will not be able to for the forseeable future. Communication with the spiders is limited to radio. Thus for a normal man, the shape of the spiders, whether their physical characteristics are like men or not, does not matter. In the context of his purpose, it makes the least of difference, whether the beings are spiders, human beings or beavers.

Would he be justified in classifying the rational spiders under the concept "man". I would say that for the sake of his cognitive efficiency, he has to classify the spiders under a different concept simply because the characteristics of the spider and man are too different.

Too different for what? The "What for?" sets the context and determines the answer to your inquiry.

Remember why we form concepts in the first place: unit economy. It allows us to deal with vast numbers of similar things and apply knowledge we already have about some entities to all similar entities. In this case, it seems the "What for?" is communication.

For the purposes of communication, whether or not someone is classified as a "man" is as irrelevant as whether he is classified as "tall" or "short" so I wouldn't even worry about that. "Conceptual being" makes more sense as a category that includes us and the spiders in this context.

[Betsy Speicher 0]

It seems that you're limiting your view of what constitutes the concept to what makes an ideal man, when it really refers to all actual men and must take into account the full range of each characteristic and not just the "ideal values."

There is no fixed category or Platonic ideal "man." There are just an enormous number of non-identical entities in the universe that we group by similarities for our epistemological convenience. How we group entities is determined by our contexts and purposes and these can differ from time to time and person to person.

[Ed from OC 0]

There are many documented cases of human abnormalities, such as three legs, two noses, two mouths, etc. But, in general, these present no more of an epistemological problem than the blind man. These are still men, but we qualify them as "a man with two noses," a man with three legs," etc. They still remain man.

Right. There's a related discussion in ITOE on the definition of "table" and whether an unusual variant still qualifies as a table. It's worth reading.

[piz 0]

There is no fixed category or Platonic ideal "man." There are just an enormous number of non-identical entities in the universe that we group by similarities for our epistemological convenience. How we group entities is determined by our contexts and purposes and these can differ from time to time and person to person.

It doesn't sound like you're disagreeing with me here, or am I missing something?

[Betsy Speicher 0]

There is no fixed category or Platonic ideal "man." There are just an enormous number of non-identical entities in the universe that we group by similarities for our epistemological convenience. How we group entities is determined by our contexts and purposes and these can differ from time to time and person to person.

It doesn't sound like you're disagreeing with me here, or am I missing something?

I'm agreeing and adding my 2 cents.

[ewv 0]

DGM: The article explicitly asserts that even the defining characteristics can be removed and a phenomenon can still be treated as an instance of the same concept.

It is not clear to me that that is what the author of the article intends. But, if so, then the assertion is most certainly wrong.

This is what he said:

One might object at this point: "...A brainless baby is not a man; a non-smelling dog is still a dog."

This objection can be refuted very simply: by pointing out that in raising the objection, one concedes that we are able to identify the brainless baby as a baby. But how? How we can identify a brainless baby as a unit of the concept 'man'? A concept whose definition is, 'rational animal'?...This objection can be refuted very simply:...

The whole article attempts to elevate an analysis of the concept 'broken' into a major epistemological issue of concept formation. The 'problem', to the extent he has identified one, that he is attempting to address has already been answered in the section on IOE on borderline cases. Whether or not the author of this article agrees with that, it was a major error to completely omit any mention of AR's discusion of that and indicates that something else is very wrong. (When he posted his thesis on another forum a few years ago I asked him why he didn't mention it and he said he had a new idea beyond what AR had written.)

[ewv 0]

A man is a man. A given man might happen not to be able to see. There's nothing about "Man" that requires that each consituent unit not be blind.

Doesn't a concept refer to all the characteristics of the units subsumed by the concept? And isn't the ability to see one of those characteristics?

The referents of the concept are the units -- all men -- not their characteristics. A thing is all of its characteristics, but the concept does not directly refer to a hodge podge of characteristics as such which happen to be possessed by one or more of the units.

It doesn't make any difference that some men can't see; that is not essential to being man even though the ability to see is true of all normal men under normal circumstances (born with normal eyes, the lights are on, etc.).

A rational faculty, on the other hand, is essential for the unit to have as a characteristic because it is the essential defining characteristic for the concept. To claim otherwise is to deny that essential means essential.

When you speak of a "brainless baby" as a human "baby" or "man" you are categorizing the creature in a particular context -- in accordance with the optional procedures for dealing with "borderline cases" -- on the basis of factors other than rationality associated with the creature and with the knowledge that it was supposed to have a brain because of them.

That special use of the term "baby" in such a borderline case does not imply a general statement that "the concept man requires that men be rational as an essential characteristic but they don't have to". When you use the term "brainless baby" you are implicitly invoking the concept "broken" in forming a compound concept (using the qualified instance as if it were concept) based on the facts on the basis of which you categorize the creature as man because of what else it is and what it was "supposed" to have because of that. It remains a borderline case treated in a special way for particular purpose with a specific context. It does not mean that Objectivist epistemology embraces "broken units" in concepts as means to circumvent the fact that essential means essential. That is one reason why the attempt to identify a formal procedure through a "syllogism" based on folding "broken units" into concepts is doomed to fail.

[DGM 0]

Sometimes it's an interesting thing to do given the time and "crow space." Sometimes it's not. Certainly always integrating everything with everything all the time would leave us with darn little time to do anything else. Fortunately, that it not necessary.

I was going to respond by pointing out that the principle came directly from Galt's speech, but when I looked it up in order to give the page number I found that I had misremembered. He actually said that no concept was valid unless integreted with the whole of one's knowledge, which is not exactly the same as no conclusion.

I'm going to withdraw what I've said for now and do some further thinking.

[Stephen Speicher 0]

He actually said that no concept was valid unless integreted with the whole of one's knowledge, which is not exactly the same as no conclusion.

But the need to integrate our conclusions is just as great, if not more so, than the need during concept-formation to integrate our concepts with all that we know. Reality is an integrated whole, and unless new knowledge is integrated with old, then we open ourselves to holding contradictions in the name of knowledge. Strictly speaking, knowledge arrived at through the continuum of evidence, as in that of which we are certain, is not really knowledge until it has been so integrated.

[tommyedison 0]

Tommy, to make sure this discussion doesn't slip into rationalism, wouldn't it be better to offer a situation taken directly from your own life?

Consider this picture. House elf. It is a house elf from a harry potter movie.

Note that this creature shares many characteristics with man - it has eyes, ears, hands, legs, a head, a torso, etc. Still on the first glance, I can tell that this creature is not a man but something different because the specific measurements of its characteristics fall out of the range of measurements for man.

Now a man without a hands (say) does not have a characteristic of man at all - he does not have hands. Now I am not denying that this existent ceases to be a man because he does not have hands.

However I do not understand why a man without hands is a qualified instance while a house elf is not a qualified instance of the concept man.

In other words, how has this epistemological issue with which you are wrestling actually emerged in your own experiences? Could you describe several instances?

If you want the actual history, I was reading Dr. Peikoff's essay on the analytic-synthetic dichotomy when I came across the sentence "Man has two eyes". That's when I thought that similarly we can say "A man has 10 fingers". Would an 11-fingered man be a unit of the concept man and if so, if so, how would that unit be a measured instance of the concept man.

In other words, can we claim that 10 fingers is a characteristic of man if some of its units DO NOT have that characteristic. Note that I am not talking of the specific measurements of the characteristics. I am talking about the characteristic itself.

[tommyedison 0]

"Too great" -- for what?

The context matters and determines what is or is not "too great."

The concept man includes the characteristic that man reasons by means of a brain.

Concepts are mental integration of two or more units possessing the same distinguishing characteristic(s), with their particular measurements omitted.

I have always taken this to mean (And please correct me if I am wrong), that units are measured instances of a concept.

If the characteristic brain is included in the concept man, then I can't understand, how an existent, who shares all its characteristics with man except that the existent does not have a brain, can be a measured instance of the concept man?

[Burgess Laughlin 0]

[...] units are measured instances of a concept.

I am unsure what you are saying. Would you give an example? Following is my attempt at an example illustrating the existence of measurable characteristics of a unit -- and their omission in the concept-formation process:

- One unit (a chair) is hard, dark colored, and five feet high.

- Another unit is soft, light colored, and three feet high.

- A third unit is medium soft, medium colored, and one foot high (for a child).

When I form the concept chair, I do so by looking for the essential defining characteristics (such as shape and function or whatever), while omitting measurements of its characteristics (color intensity, depth, height of its back, number of legs, cushiness, and so forth), both the essential ones and the inessential ones.

All the characteristics of all chairs are measurable. To form a concept, I omit the measurements and concentrate on the characteristics that both cause the entity to be what it is and distinguish it from other things of the same general kind. Once the concept is formed, as well as before concept formation, the entities subsumed have and will always have measurable characteristics.

Is that what you are saying?

[tommyedison 0]

I am unsure what you are saying. Would you give an example? Following is my attempt at an example illustrating the existence of measurable characteristics of a unit -- and their omission in the concept-formation process:

- One unit (a chair) is hard, dark colored, and five feet high.

- Another unit is soft, light colored, and three feet high.

- A third unit is medium soft, medium colored, and one foot high (for a child).

When I form the concept chair, I do so by looking for the essential defining characteristics (such as shape and function or whatever), while omitting measurements of its characteristics (color intensity, depth, height of its back, number of legs, cushiness, and so forth), both the essential ones and the inessential ones.

All the characteristics of all chairs are measurable. To form a concept, I omit the measurements and concentrate on the characteristics that both cause the entity to be what it is and distinguish it from other things of the same general kind. Once the concept is formed, as well as before concept formation, the entities subsumed have and will always have measurable characteristics.

Is that what you are saying?

I think so. To take your example, a unit of the concept chair can have any varied number of specific measurements. For example, they can have any combination of height, color, etc. Notice however that you cannot have a chair without its back. A chair without its back is not a qualified instance of chair - it becomes a stool.

The characteristic "back of the chair" is included in the concept "chair". An existent which does not have this characteristic cannot be a unit of the concept "chair".

The specific measurements such as height of the back, the color, etc. are not included in the concept "chair". An existent can have any variety of the specific measurements of these characteristics, but as long as it has all the characteristics of a chair, it remains a chair.

Mrs. Speicher - If I understand your position correctly, are you saying that provided a given existent shares the overwhelming number of characteristics (including the essential distinguishing ones) with a particular concept, but does not share all of the characteristics, then whether that existent is classified under the concept in question or not is dependent on the purpose?

(Please correct me if I am wrong)

[Burgess Laughlin 0]

The specific measurements such as height of the back, the color, etc. are not included in the concept "chair".

What do you mean by "not included in"? Does that mean the same as "not subsumed by"?

[tommyedison 0]

What do you mean by "not included in"? Does that mean the same as "not subsumed by"?

By "a characteristic is not included in the concept", I mean that the referents of the concept do not have that characteristic.

So when I say that the specific measurements of height, etc. are not included in the concept "chair" for example, I mean that the referents of the concept chair will have definite measurements of height, etc. but those measurements are not specified.

When I say that the "back of the chair" is included in the concept chair, I mean that the referents of the concept chair will have the characteristic "back of the chair".

[Betsy Speicher 0]

The concept man includes the characteristic that man reasons by means of a brain.

I have always taken this to mean (And please correct me if I am wrong), that units are measured instances of a concept.

That implies the concept comes first and that units are, or are not, within that concept.

I don't see it that way because individual concrete entities are all that exist and are the starting point of all cognition. We start with concrete things and, when we notice interesting similarities, we can organize them into one of many optional conceptual categories depending on the cognitive purpose for which we are grouping them. A unit is an individual concrete entity that is regarded as a member of a conceptual group of similar entities. That "unit perspective" is what ties a concept to reality by connecting the concept to the entities from which it was formed.

A baby's first concept of man might be "something like Mommy and Daddy and Grandma and me." The units would be Mommy, Daddy, etc. and also the little boy he just met at the park. If he finds an entity that doesn't obviously fit into his existing scheme -- like the house elf -- he may have to decide whether it serves his purpose to regard it as something like Mommy and Daddy or not. If it does, it will become a unit of his concept of "man." If it doesn't, it won't.

[Betsy Speicher 0]

Mrs. Speicher - If I understand your position correctly, are you saying that provided a given existent shares the overwhelming number of characteristics (including the essential distinguishing ones) with a particular concept, but does not share all of the characteristics, then whether that existent is classified under the concept in question or not is dependent on the purpose?

Right. It depends on the purpose of the person forming and using the concept.

WHY a particular person wants to groups certain entities together will determine how and by what characteristics he groups them into concepts. The house elf thus might be classified as a man or a creature or a servant or a nuisance etc.

[tommyedison 0]

If he finds an entity that doesn't obviously fit into his existing scheme -- like the house elf -- he may have to decide whether it serves his purpose to regard it as something like Mommy and Daddy or not. If it does, it will become a unit of his concept of "man." If it doesn't, it won't.

What do you mean by "serves his purpose"?

Could you provide a concrete situation where an existent X is regarded as the unit of a concept A (with which it shares most but not all characteristics) because it serves the purpose Y for a man?

[Burgess Laughlin 0]

What do you mean by "serves his purpose"?

Could you provide a concrete situation where an existent X is regarded as the unit of a concept A (with which it shares most but not all characteristics) because it serves the purpose Y for a man?

You have the concept animal. Do you need other concepts like dog or man? If you think you do need those concepts, then what need (purpose) are they serving? In other words, why not pet your (four-legged) animal (that barks) and look for a job working with a (two-legged) animal (that speaks and thinks)? Would the concept animal be sufficient?

So, to provide a concrete example, I consider my closest friend, Marianne, to be a man (of the female sex). I do not consider her to be only a referent of the concept animal. She is what she is, regardless of how I classify her, but classifying her as human serves my purposes very nicely, thank you.

[Betsy Speicher 0]

What do you mean by "serves his purpose"?

The purpose of consciousness, for all living things possessing that faculty, is to assist it in acting to further and maintain its life. Consciousness guides an organism toward values (like food) and away from dangers (predators, falling, etc.).

A person has many options as to how he will group and classify the entities he encounters, but ultimately his purpose is the main guiding factor. If he wants to survive on a battlefield, he will classify other persons as friend or foe. If he is looking for a wife, he may classify them as male or female, single or married, his type or not his type.

Even with regard to a generally agreed upon existing concept like "man," different purposes may result in different results. Someone wishing a high-level intellectual discussion may regard a brutish, rude, ignorant person as not human enough, while a medical doctor doing organ transplants might regard a baby born without a brain as definitely human since, for his purposes, all its other organs are the same as a normal human being's.

[PhilO 0]

So, to provide a concrete example, I consider my closest friend, Marianne, to be a man (of the female sex). I do not consider her to be only a referent of the concept animal. She is what she is, regardless of how I classify her, but classifying her as human serves my purposes very nicely, thank you.

It might be worth emphasizing the hierarchical nature of concepts, here. Identifying an entity as a concept which is a subdivision of another provides *additional* information, but one mustn't forget that that entity is fully an instance of each concept higher up the conceptual chain. That is not rationalism in any way, but in the nature of conceptual hierarchy.

So, all that we know and properly ascribe to the concept *animal* must necessarily apply to humans as well - as well as the concepts *man*, *mammal*, *life*, *entity*, etc.

[tommyedison 0]

You have the concept animal. Do you need other concepts like dog or man? If you think you do need those concepts, then what need (purpose) are they serving? In other words, why not pet your (four-legged) animal (that barks) and look for a job working with a (two-legged) animal (that speaks and thinks)? Would the concept animal be sufficient?

The difference a human being makes in one's life (whether for the good or for the bad) is extremely different from that of an animal in general. The contexts are vastly different which is why I think we need the new concept.

But once you have formed the new concept (say a table), observe that a table is not any existent which shares its overwhelming characteristics with the characterstics subsumed by the concept "table". Any unit of the concept "table" has all the characteristics subsumed by the concept "table".

[tommyedison 0]

A person has many options as to how he will group and classify the entities he encounters, but ultimately his purpose is the main guiding factor. If he wants to survive on a battlefield, he will classify other persons as friend or foe.

True enough. And friend does not mean anyone who is approxiamately a friend or someone in the middle of friend or foe. Every friend (or foe) has to share some characteristics and those characteristics must be present in all friends (or foes).

Even with regard to a generally agreed upon existing concept like "man," different purposes may result in different results. Someone wishing a high-level intellectual discussion may regard a brutish, rude, ignorant person as not human enough, while a medical doctor doing organ transplants might regard a baby born without a brain as definitely human since, for his purposes, all its other organs are the same as a normal human being's.

I don't know about the second case since it may be a borderline case, but for the first case, can you provide a few examples where a brutish person was classified as a non-man. Concepts are formed by observing similarities and differences between existents and omitting the specific measurements of certain characteristics of the concept. What measurements of what characteristics have to be omitted depends on the purpose (e.g. in a battlefield, you may omit specific measurements of gender, height, etc. while when looking for a wife/husband, you would not).

However in no case does a unit of the concept NOT share all the characteristics subsumed by the concept.

[DGM 0]

But the need to integrate our conclusions is just as great, if not more so, than the need during concept-formation to integrate our concepts with all that we know. Reality is an integrated whole, and unless new knowledge is integrated with old, then we open ourselves to holding contradictions in the name of knowledge. Strictly speaking, knowledge arrived at through the continuum of evidence, as in that of which we are certain, is not really knowledge until it has been so integrated.

I'm not disputing this. My point is that since I'm not just forgetting but actually misremembering details, it might be wise to wait until I can reread the relevent material (OPAR and ITOE) before plunging back into this. It has been years since I read some of it, and a quick refresher course may be in order.