Properties and Characteristics

[Steve D'Ippolito]

Small digression:

Oddly enough, the kilogram is the only metric unit still defined by reference to a physical standard (the standard kilogram cylinder made of iridium and platinum, kept in Paris). The mass of that slug is *by definition* one kilogram--and it is known that the cylinder is losing mass!

Yes, Avogadro's number was originally defined on the basis of the kilogram, but there is nevertheless some talk of reversing the dependency and setting a *precise* numerical value to it, then stating that a kilogram is the mass of precisely X atoms of such and such element and isotope. (The kilogram would then no longer be tied to an actual physical object that is subject to damage, loss, or other change.) For example the mass of Carbon 12 is precisely 12 AMUs; it could be used--but they'd probably prefer an element that only has one naturally occurring isotope so that there's no need to separate isotopes to determine, in the laboratory, how much a kilogram is--it would only be necessary to isolate an absolutely pure sample of the element (which is difficult enough as it is).

This would in no way change or even contradict the fact that *conceptually* mass is prior to Avogadro's number, after all its original derivation is a historic fact. But it would make the *kilogram* dependent on it, even though historically the situation was reversed. I don't think there is a problem with reversing the concept hierarchy in the light of later knowledge, however. (It would be important for proper pedagogy though to explain this history, because there's no way to derive the new hierarchy without a lot of prior knowledge.) If that causes someone to stumble because of fear of inverting hierarchy, though, you can think of it as two different kilograms, an old kg and a new kg (legitimate because they have two different definitions), and then the concept hierarchy goes as follows: mass->old kilogram->avogadro's number->new kilogram (and clearly "atom" and "isotope" also feed into this hierarchy like this: atom->isotope->avogadro's number->new kilogram, and mass should have an arrow pointing to isotope).

OK, end digression.

[JayR]

Do you or do you not hold that characteristics refer to the dynamic and relational attributes of things? If you do, then, logically, fundamental characteristics must be the result of properties. Which means that some attributes constituting the identity of a thing aren't included in the concept about that thing.

Could you clairify this? Im still not getting it. Saying that some attributes arent included in the concept, if Im understanding, is like splitting the characteristics of a unit into groups, necessary, and contingent perhaps. Or if the fundemental characteristics are a result of the properties of an entity, entities subsumed under a single concept cant have different properties???

A concept refers to all the characteristics of its referents, known and unknown. The fact that some entities dont possess all the characteristics of other entities subsumed under a particular concept does not mean that the attributes/characteristics are not a part of the concept, the entities lacking these qualities would be broken units, borderline cases, or perhaps even qualified instances, depending on what you mean by "constituting the identity of a thing".

j..

[A is A]

Which atoms? What if different atoms have different masses? (Hint: They do.)

No kidding. It should be clear I meant atoms of a specific element. Otherwise it would make no sense.

[A is A]

The prior existence of the mass unit 'gram' and the very concept of mass is presupposed by the Avogadro constant. To use the correlation between the Avogadro number of atoms and the resulting mass in this way to deny the necessity of dealing with forces is a perfect example of a stolen concept.

But I could define any number of atoms as representing a specific mass without running into that problem. Suppose I said 100 trillion uranium 238 atoms is 1 picogram. I could use that mass as a standard to measure other masses. This definition of picogram would now be independent of force.

[JayR]

If it ever becomes possible to count atoms, all one would have to do is count an Avogadro's number of atoms and one would have that element's mass independent of gravity or forces.

The amout of atoms in an avocado is irrelevant to this discussion.

j..

[A is A]

The amout of atoms in an avocado is irrelevant to this discussion.

j..

Then don't respond to it. Apparently some do think it is relevant. And I did not bring up avocado.

Edited June 19, 2010 by A is A

[JayR]

Joke.

"...all one would have to do is count an Avocados number of atoms...."

nevermind

j..

[Vik]

Do you or do you not hold that characteristics refer to the dynamic and relational attributes of things? If you do, then, logically, fundamental characteristics must be the result of properties. Which means that some attributes constituting the identity of a thing aren't included in the concept about that thing.

Yes.

Okay.

Huh? Why? Is there something special about the fundamental characteristic that means it cannot also be the result of some other attribute?

Fundamental characteristics are supposed to explain the most others. If they are themselves explained by properties, then they explain less than the properties. Do you really think Ayn Rand meant THAT?

Measuring bond angles is measuring molecular structure, that is how you get it done. You measure the structure of a building by taking a series of single linear measurements of the exterior and interior walls, floors, ceilings windows and doors.

A structure is a configuration of constituents, not the quantitative relationships among those constituents.

Mass (rest mass) is an intrinsic property that can never be known directly but only inferred from its effects on trajectories or other signs of the force resulting from gravity. Weight and inertia are characteristics.

That's arrested knowledge. What if someone discovered that mass changed depending on certain conditions?

We used to think mass was independent of velocity. Now we have invariant mass vs relativistic mass.

[Vik]

Could you clairify this? Im still not getting it. Saying that some attributes arent included in the concept, if Im understanding, is like splitting the characteristics of a unit into groups, necessary, and contingent perhaps. Or if the fundemental characteristics are a result of the properties of an entity, entities subsumed under a single concept cant have different properties???

Recap:

There is a claim that properties are "intrinsic attributes" while characteristics are "dynamic and relational attributes"

I'm claiming that "intrinsic attribute" vs "dynamical/relational attribute" is an arbitrary distinction that leads to conclusions that are incompatible with Objectivism.

I wasn't thinking that Grames believed in the "necessary vs contingent" dichotomy.

But I do feel that "intrinsic vs dynamical/relational" sounds a lot like the distinction of "intrinsic vs dispositional". I think it's worth mentioning that the latter was exploded on pg. 282-287 of ITOE

A concept refers to all the characteristics of its referents, known and unknown. The fact that some entities dont possess all the characteristics of other entities subsumed under a particular concept does not mean that the attributes/characteristics are not a part of the concept, the entities lacking these qualities would be broken units, borderline cases, or perhaps even qualified instances, depending on what you mean by "constituting the identity of a thing".

j..

Agreed, but I don't see how it's relevant to the claim I'm objecting to or my objection to it.

[Grames]

Fundamental characteristics are supposed to explain the most others. If they are themselves explained by properties, then they explain less than the properties. Do you really think Ayn Rand meant THAT?

"Explanation" can be an ambiguous word, it either means "cause of" or "knowledge of" depending on its use. The importance of making the distinction is that we do not always know the ultimate causes.

Given a concept defined in terms of a fundamental characteristic, if we then learn that the fundamental characteristic is completely explained by an internal property then we have a basis to shift the definition to the more fundamental property. A concrete example of this would be in biology when all of the concepts of the various species were based on similarities of the bodily form until genetics, evolution, and the discovery of DNA resulted in a new understanding of what is truly fundamental in biology, resulting in new classifications of some species. The causal relation between genetics and body form has always been true, but what was considered fundamental changed when the context of scientific knowledge expanded. The scientific revolution in biology is both an example of concepts being open-ended in that new attributes were discovered for the same concepts, and an example of definitions being contextual when the new attribute was found to be fundamental. The conclusion is that a fundamental characteristic can indeed have an unknown cause ("explanation") in terms of more fundamental properties.

Sometimes even a known causal explanation can fail to justify changing the fundamental characteristic. Anything that happens in every living organism has a chemical explanation, but that level of analysis is not appropriate for understanding discrete living entities as whole organisms interacting with each other.

A structure is a configuration of constituents, not the quantitative relationships among those constituents.

The distinction you attempt here is not meaningful. A "configuration of constituents" can only exist in the form of quantitative relationships. To be is to be something particular. Existence is identity.

That's arrested knowledge. What if someone discovered that mass changed depending on certain conditions?

We used to think mass was independent of velocity. Now we have invariant mass vs relativistic mass.

This is no exception. We only know about relativistic mass because of the exact same kinds of effects on trajectories, forces, extended half-lives of decaying particles, etc. A variable mass is not an intrinsic quality of a particle but a relationship between two frames of reference.

Wikipedia quotes a physics textbook on relativistic mass:

Modern view

The invariant mass is the ratio of four-momentum to four-velocity:

p^μ = m₀v^μ

and is also the ratio of four-acceleration to four-force when the rest mass is constant. The four-dimensional form of Newton's second law is:

F^μ = m₀A^μ.

Many contemporary authors such as Taylor and Wheeler avoid using the concept of relativistic mass altogether:

"The concept of "relativistic mass" is subject to misunderstanding. That's why we don't use it. First, it applies the name mass - belonging to the magnitude of a 4-vector - to a very different concept, the time component of a 4-vector. Second, it makes increase of energy of an object with velocity or momentum appear to be connected with some change in internal structure of the object. In reality, the increase of energy with velocity originates not in the object but in the geometric properties of spacetime itself."[17]

E. F. Taylor, J. A. Wheeler (1992), Spacetime Physics, second edition, New York: W.H. Freeman and Company, ISBN 0-7167-2327-1

I'm claiming that "intrinsic attribute" vs "dynamical/relational attribute" is an arbitrary distinction that leads to conclusions that are incompatible with Objectivism.

But I do feel that "intrinsic vs dynamical/relational" sounds a lot like the distinction of "intrinsic vs dispositional". I think it's worth mentioning that the latter was exploded on pg. 282-287 of ITOE.

I agree that when I wrote about relational and intrinsic attributes that was inaccurate and misleading because those are not two kinds of attributes. The relation between an entity and its actions is not the same as that of an entity to its attributes. I made the mistake that Ayn Rand avoided by using the expression "properties and characteristics". Nevertheless, there are such things as attributes and actions and that is the distinction being made.

"Dispositional" relates to the Doctrine of internal relations (also see http://plato.stanford.edu/entries/logical-atomism/#2.1 and www.britannica.com The doctrine of internal relations and the coherence theory of truth. This is the theory that relations are attributes and it is rightly rejected.

There is evidence for one half of my interpretation in that same section:

Prof. F: But now, properties are all constitutive properties, right?

AR: Yes.

Constitutive and intrinsic both describe the same relation between entity and attribute. The actions of an entity are not constitutive, what something does is not the same as what it is. Although the two must be causally related, actions are almost always interactions between two entities and so are jointly determined by (at least) two entities that act.

[Vik]

"Explanation" can be an ambiguous word, it either means "cause of" or "knowledge of" depending on its use. The importance of making the distinction is that we do not always know the ultimate causes.

Given a concept defined in terms of a fundamental characteristic, if we then learn that the fundamental characteristic is completely explained by an internal property then we have a basis to shift the definition to the more fundamental property. A concrete example of this would be in biology when all of the concepts of the various species were based on similarities of the bodily form until genetics, evolution, and the discovery of DNA resulted in a new understanding of what is truly fundamental in biology, resulting in new classifications of some species. The causal relation between genetics and body form has always been true, but what was considered fundamental changed when the context of scientific knowledge expanded. The scientific revolution in biology is both an example of concepts being open-ended in that new attributes were discovered for the same concepts, and an example of definitions being contextual when the new attribute was found to be fundamental. The conclusion is that a fundamental characteristic can indeed have an unknown cause ("explanation") in terms of more fundamental properties.

Whether something is "fundamental" is a contextual issue, yes.

Sometimes even a known causal explanation can fail to justify changing the fundamental characteristic. Anything that happens in every living organism has a chemical explanation, but that level of analysis is not appropriate for understanding discrete living entities as whole organisms interacting with each other.

The distinction you attempt here is not meaningful. A "configuration of constituents" can only exist in the form of quantitative relationships. To be is to be something particular. Existence is identity.

I think we're talking passed each other. There is a difference between the entity itself, its components, etc., and the relationships among the attributes of those components.

>>That's arrested knowledge. What if someone discovered that mass changed depending on certain conditions?

>> We used to think mass was independent of velocity. Now we have invariant mass vs relativistic mass.

This is no exception. We only know about relativistic mass because of the exact same kinds of effects on trajectories, forces, extended half-lives of decaying particles, etc. A variable mass is not an intrinsic quality of a particle but a relationship between two frames of reference.

Wikipedia quotes a physics textbook on relativistic mass:

I agree that when I wrote about relational and intrinsic attributes that was inaccurate and misleading because those are not two kinds of attributes. The relation between an entity and its actions is not the same as that of an entity to its attributes. I made the mistake that Ayn Rand avoided by using the expression "properties and characteristics". Nevertheless, there are such things as attributes and actions and that is the distinction being made.

For Ayn Rand, shape is a characteristic (pg 12). I'm thinking that characteristics represent measurement-omissions and constitute our mental grasp of properties.

"Dispositional" relates to the Doctrine of internal relations (also see http://plato.stanford.edu/entries/logical-atomism/#2.1 and www.britannica.com The doctrine of internal relations and the coherence theory of truth. This is the theory that relations are attributes and it is rightly rejected.

There is evidence for one half of my interpretation in that same section:

Constitutive and intrinsic both describe the same relation between entity and attribute. The actions of an entity are not constitutive, what something does is not the same as what it is. Although the two must be causally related, actions are almost always interactions between two entities and so are jointly determined by (at least) two entities that act.

Would you say that causation is an interaction among entities according to their constitutive properties?

[Grames]

I think we're talking passed each other. There is a difference between the entity itself, its components, etc., and the relationships among the attributes of those components.

Perhaps. There is a thread here where the relation between a whole and its parts is discussed. There is a philosophical term for that special study, but I forget what it is now and that happens to be the thread title.

For Ayn Rand, shape is a characteristic (pg 12). I'm thinking that characteristics represent measurement-omissions and constitute our mental grasp of properties.

I found this thread on Physics Forums "property" vs "characteristic". Translating the conclusion of that thread into Objectivism, Ayn Rand may have had the distinction between similarities and differences in mind. Properties would be attributes similar to attributes of other entities, characteristics would be attributes (even the same attributes) regarded as different enough to be distinctive for identifying particular entities. This perspective would be the epistemological product of mental comparisons, rather than the more metaphysical twist I was giving it. This lines up neatly though, as properties are based on similar intrinsic attributes that lead easily to an integration omitting particulars while the characteristics refer specifically to differences which are irreducibly relational and to be retained rather than omitted. For example, shape for man in the sense of an animal or species can be regarded as a common property or as a characteristic distinguishing man relative to other kinds of animals or as individuals from each other when comparing faces or fingerprints.

Would you say that causation is an interaction among entities according to their constitutive properties?

Yes, that is good.

[Vik]

Perhaps. There is a thread here where the relation between a whole and its parts is discussed. There is a philosophical term for that special study, but I forget what it is now and that happens to be the thread title.

There is a branch of ontology called mereology.

I found this thread on Physics Forums "property" vs "characteristic". Translating the conclusion of that thread into Objectivism, Ayn Rand may have had the distinction between similarities and differences in mind. Properties would be attributes similar to attributes of other entities, characteristics would be attributes (even the same attributes) regarded as different enough to be distinctive for identifying particular entities. This perspective would be the epistemological product of mental comparisons, rather than the more metaphysical twist I was giving it. This lines up neatly though, as properties are based on similar intrinsic attributes that lead easily to an integration omitting particulars while the characteristics refer specifically to differences which are irreducibly relational and to be retained rather than omitted. For example, shape for man in the sense of an animal or species can be regarded as a common property or as a characteristic distinguishing man relative to other kinds of animals or as individuals from each other when comparing faces or fingerprints.

Existents can have the same characteristics in different measure or degree. This is key to understanding her theory of similarity. Ref page 13.

Also, abstraction from abstractions depends on omitting distinguishing characteristics from the constituent units, i.e. from the lower-level concepts. Ref page 23.

I agree that the mental isolation of shape can be either a characteristic or a property depending on what you're doing with it.

[Grames]

There is a branch of ontology called mereology.

Actually the best part of that thread I mentioned was discovering Rand had things to say that fell under mereology. They are in IOE2, pg 264 the discussion on entities and their makeup.

[Mindy]

I don't think Rand was making any distinction between properties and characteristics. It isn't usually necessary to distinguish those concepts from "feature," or "quality," etc.

That being said, I offer two examples of use that maintains the distinction.

When we say something like, "It is characteristic of him to be late to meetings," we are implying that being late is usual and typical, but not invariant. When we say something like, "Polymorphic minterals have the same molecular structure but different properties, such as specific gravity and hardness," we are referring to absolutely invariant "features" of those minerals.

Usually, though, we don't need to make such distinctions.

-- Mindy