ewv

The Logical Leap and criticism

529 posts in this topic

I don't know if [Harriman] means that first level generalizations can be derived from perception of causal factors without first level concepts identifying them, or if he means that no concepts are required at all, at least at an early primitive level, or something else.

From reading Harriman's examples and some of his explicit statements about what is "self-evident," etc., he seems to be saying you "just see" causes and that they apply, with certainty and with no need for proof, to all the things that are similar to what you see.

At an elementary level of knowledge, once you recognize how pushing on the ball works, as you experience it and recognize the role of pushing and roundness, that is the proof, even though the child need not conceptualize it or state it formally. But that is derived and and is not "just seeing causality" as if nothing else were involved, All that would have to be carefully worked out and explained, and this still doesn't tell us what he means.

It also isn't true that the concept of "causality" -- which is not a first level concept -- is required for first level generalizations.

Yes and no. Causality is a first-level concept -- ask me to explain why -- and it is required for all generalizations on any level.

You can't have the concept causality until you recognize particular instances of it. Otherwise there are no units to integrate. Causality as identity applied to action requires the concepts identity and action and is not first level. But you could grasp causality implicitly at a preconceptual level when you have and are aware of the ingredients not yet integrated into a concept.

See my post above.

Share this post


Link to post
Share on other sites
... By applying concepts to his perceptual observations of particular causal situations, he arrives at generalizations.

And I believe this is what Harriman means by "derived" from direct experience means.

That is consistent with what I have said. Grasping directly the cause in a particular instance, derived through his experience and direct observation, and then using the generalizations inherent in the concepts of the objects to generalize his experience of the cause, is not a claim to directly perceive the causal generalization, and there is in fact substance to what he calls "derived" directly from the perceptions in the early equivalent of experiments. Thanks for supplying the extended quotes. Eventually, I will get the book.

Share this post


Link to post
Share on other sites
...

Harriman recognizes that perception of first-level causal connections is by no means all there to generalizations.

Percepts as such reveal concretes, not abstractions... By perception alone, therefore, the first-level inducer grasps causal connections only in concrete terms, as a relationship at a specific time and place between two or more particulars. ... What turns them into generalizations? What mediates the passage from a causal connection linking particulars to a universally applicable truth? ... man's conceptual faculty. The essence of concept-formation is the passage from particulars to universals.

He's stating that one can grasp causality at a pre-conceptual level? That's nonsense. At a preconceptual stage of analysis all one can hope for are reliable associations between percepts, which is how animals function.

How does grasping causation precede generalization when causation is a generalization?

By applying concepts to his perceptual observations of particular causal situations, he arrives at generalizations.

And I believe this is what Harriman means by "derived" from direct experience means.

But here we are using a separate section of text that is some seven or eight pages after the text we wished to understand the meaning of

...first-level generalization...one derived directly from perceptual observation, without the need of any antecedent generalizations.

when the text almost directly following it on p.19 (that Betsy quoted) states

Since the perceptual is the self-evident, first-level generalizations are self-evident; being the basis of inductive (and therefore deductive) knowledge, they admit and require no proof. They are available, as certainties, to anyone with the requisite simple vocabulary who takes the trouble to look at reality.

And they are available by no other means. How do you know that pushing a ball makes it roll? There is no answer, not even by Newton or Einstein, except this: look and see.

which would lead one to interpret the opposite of what you are trying to interpret.

Is this book so loosely written that one is reduced to a game or puzzle of finding text interspersed throughout later sections to validate prior meanings sections?

Share this post


Link to post
Share on other sites

Bleh, sorry. Correction:

Is this book so loosely written that one is reduced to a game or puzzle of finding text interspersed throughout later sections to validate the meanings of prior sections?

Share this post


Link to post
Share on other sites
...

Harriman recognizes that perception of first-level causal connections is by no means all there to generalizations.

Percepts as such reveal concretes, not abstractions... By perception alone, therefore, the first-level inducer grasps causal connections only in concrete terms, as a relationship at a specific time and place between two or more particulars. ... What turns them into generalizations? What mediates the passage from a causal connection linking particulars to a universally applicable truth? ... man's conceptual faculty. The essence of concept-formation is the passage from particulars to universals.

He's stating that one can grasp causality at a pre-conceptual level? That's nonsense. At a preconceptual stage of analysis all one can hope for are reliable associations between percepts, which is how animals function.

How does grasping causation precede generalization when causation is a generalization?

He grasps the causal connections in a single instance by experiencing them, not a principle of causality applied to it. He doesn't have that concept yet. Imagine yourself at an early age before you had all these concept. You don't have to know much to realize that your pushing on a ball moves it.

Share this post


Link to post
Share on other sites
...

Harriman recognizes that perception of first-level causal connections is by no means all there to generalizations.

Percepts as such reveal concretes, not abstractions... By perception alone, therefore, the first-level inducer grasps causal connections only in concrete terms, as a relationship at a specific time and place between two or more particulars. ... What turns them into generalizations? What mediates the passage from a causal connection linking particulars to a universally applicable truth? ... man's conceptual faculty. The essence of concept-formation is the passage from particulars to universals.

He's stating that one can grasp causality at a pre-conceptual level? That's nonsense. At a preconceptual stage of analysis all one can hope for are reliable associations between percepts, which is how animals function.

How does grasping causation precede generalization when causation is a generalization?

He grasps the causal connections in a single instance by experiencing them, not a principle of causality applied to it. He doesn't have that concept yet. Imagine yourself at an early age before you had all these concept. You don't have to know much to realize that your pushing on a ball moves it.

But if you're preconceptual you don't know that pushing a ball caused movement; you only identified an association between percepts.

I guess my point is that grasping causation in terms of percepts seems like a contradiction in terms.

Share this post


Link to post
Share on other sites
-------------

But if you're preconceptual you don't know that pushing a ball caused movement; you only identified an association between percepts.

I guess my point is that grasping causation in terms of percepts seems like a contradiction in terms.

Let's go back one step. When I child pushes a ball, do you have any doubt that he grasps he is pushing the ball? What does his experience consist of? Is there just an association between his experience of pushing and the pressure of the ball on his hand? What causality is there between his consciousness and his bodily action? If he does not perceive that he moved the ball by pushing it, does he perceive himself moving his hand? What does experience consist of if not the effect of reality on your consciousness. Is this not experiencing causality on a particular basis? Is it just an association that I perceive when I look at a wall and see it?

Share this post


Link to post
Share on other sites
--------

But if you're preconceptual you don't know that pushing a ball caused movement; you only identified an association between percepts.

I guess my point is that grasping causation in terms of percepts seems like a contradiction in terms.

The point that Harriman is making is that it is not grasping causation but experiencing it on a perceptual level in particular instances. When concepts are formed, the particulars become universals, and the generalization is formed.

Share this post


Link to post
Share on other sites
...

Harriman recognizes that perception of first-level causal connections is by no means all there to generalizations.

Percepts as such reveal concretes, not abstractions... By perception alone, therefore, the first-level inducer grasps causal connections only in concrete terms, as a relationship at a specific time and place between two or more particulars. ... What turns them into generalizations? What mediates the passage from a causal connection linking particulars to a universally applicable truth? ... man's conceptual faculty. The essence of concept-formation is the passage from particulars to universals.

He's stating that one can grasp causality at a pre-conceptual level? That's nonsense. At a preconceptual stage of analysis all one can hope for are reliable associations between percepts, which is how animals function.

How does grasping causation precede generalization when causation is a generalization?

He grasps the causal connections in a single instance by experiencing them, not a principle of causality applied to it. He doesn't have that concept yet. Imagine yourself at an early age before you had all these concept. You don't have to know much to realize that your pushing on a ball moves it.

But if you're preconceptual you don't know that pushing a ball caused movement; you only identified an association between percepts.

I guess my point is that grasping causation in terms of percepts seems like a contradiction in terms.

Don't push it beyond what the child himself realizes. When he grasps how he is the cause in terms of what he experiences he is not thinking in terms of your concept now of causality. He isn't thinking of it as an instance of a broader category, only of the specific connections he is grasping in his experience. He has to do that several times before he notices the similarities, let alone before forming a primitive concept of causing a ball to roll, then of other actions, and eventually a concept of causing something to happen in general, i.e., of his making something happen, and then ultimately the concept of cause by something other than himself.

Share this post


Link to post
Share on other sites
The point that Harriman is making is that it is not grasping causation but experiencing it on a perceptual level in particular instances. When concepts are formed, the particulars become universals, and the generalization is formed.

This seems to be the argument in a nutshell. But why not simply use the word 'associate' in that perceptual case. Inductions are based on associations which only become 'causal associations' in the conceptual realm, is how I think Betsy sees it. No?

Share this post


Link to post
Share on other sites
The point that Harriman is making is that it is not grasping causation but experiencing it on a perceptual level in particular instances. When concepts are formed, the particulars become universals, and the generalization is formed.

This seems to be the argument in a nutshell. But why not simply use the word 'associate' in that perceptual case. Inductions are based on associations which only become 'causal associations' in the conceptual realm, is how I think Betsy sees it. No?

It is not an undifferentiated association; he experiences his causal efficacy and realizes that he is the cause, that he made it happen with his push.

Share this post


Link to post
Share on other sites
[...] Do you disagree? If so, are you using a different definition of "directly derived" than I am? If so, what is it?

The example of the child pushing the ball and causing it to roll, seeing it roll, and supposedly, hence, experiencing a causal event is more complex than has been expressed on the thread.

The child first sees a blurry entity that may be a color, say, yellow.

The blurry entity perceived is the result of the existence of the substance of the ball.

That substance is the Material Cause (Aristotle, see defn. of MC).

Of course, the ball has properties that are integral with its subtance, e.g., the color, yellow. The child may not yet have differentiated simple colors from each other, and that is not so easy a task. The child may perceive colors, however, all that is recognized is a quality of difference, e.g., that the entity is not the same as items in the background of vision. Further exercise of the process of color differentiation requires further knowledge, and color differentiation and matching require advanced knowledge than mere perceived differences.

The child may recognize that the entity seen is a something, and that the something is locally unique.

That differentiation is a first level generalization. (See defn. of FLG.)

The child may have already experienced touch, and that things touched result in the sensation of touch.

The child may recognize that the entity touched is a something, and that the something is locally unique.

To the child the entity of the yellow ball that is no more to the child than a suspected entity.

The physical existence of the child is a second Material Cause

The physical properties of the ball and of the child are a second Efficient Cause.

The child may reach out towards the entity.

That is a major discovery being actualized: that what was experienced prior may be used to experience once again.

Already, this analysis is more complex that was previously thought, for the prior existing conditions must be taken into account.

The hypothesis by the child that an entity exists is untested.

A second order generalization has occured. (See defn. of SLG.)

The child reaches towards the yellow ball.

That the child may act to gain information is also a discovery. That's another component of the combination of Efficient Causes of the ball and the child.

That action, or the possibility to act, is probably a first order generalization.

The location of the yellow ball and the location of the child are, again, Material Causes, in that they are properties of the ball and child, or they are corollaries of same.

The possibilities of the specific action by the child, being vision, the attermpt to see, and the attempt to reach out towards the seen entity, however, are capabilities that are possible actions that may be taken towards the yellow ball by the child.

The possible properties of action (finding out what is out there via seeing, physical vision, moving towards, and touching) by the child and the properties of the possible functioning of the child are the Efficient Causes (Aristotle, see defn. of EC).

The child may see that the action of reaching and touching, taken as one process, may be directed towards the yellow entity. That an action, however exploratory at first action, may be taken towards, something that is only perceived and hypothesized as a something.

That possibility of action or tentative intention towards something possible, partially known, or implicitly hypothesized, that may have a similar result as prior vision and, also, touching, is a implicit goal.

The goal (finding out what is out there via seeing, moving towards, and touching) is a Formal Cause (Aristotle, see defn. of FC).

So far the child has not yet actually identified the entity or clearly seen the entity in terms of its properties except for the being of its substance. The existgence of the substance is unvalidated.

The child may act, and here the process of volition, or action towards a goal or value, is involved. I won't discuss that here, and more could be said on that.

The child starts and continues to act.

We may ignore the dynamic processes of the visual experience, e.g., changes of the apparent size or visual location of the yellow ball, and of the kinesthetic processes of the sense of location of the child and the parts of its body. There may be huge amounts of learning going on at that time, and much may be being written to the brain insofar as location data and actualized processess. That so far for the child seems to have worked ok.

The child acts towards the yellow ball. Just when the quality, yellow, appears as a unique visual property of the ball as differentiated from other entities may be a matter of that discovery occuring before, during, or even after, the act of reaching out towards the entity. That may be discussed separately, anthough, in other examples of the actions of children, some actions of the child may be integral with or dependent upon other actions. Those example are not here discussed.

The child completes touching the entity.

The sensations of touch occur, and the entity touched provides a tactile sensation.

Something is there that was touched, the child perceives.

That differentiation is a first level generalization. (See defn. of FLG.)

That is the Final Cause (Aristotle, see defn. of FIC).

The Final Cause is the completion of the previous causes in an integration of the causes.

What has happened is not just one thing, but several, all three being completed together simultaneously.

The child has validated the existence of the entity, and has experienced touching the substance of the entity.

The child has validated the process of acting towards something.

The child has validated that so acting, so contacting, and so experiencing, the substance of the entity, has completed a combination of causes. The child has discovered that it, me, and my actions, have resulted in the completion of a process. To the child the entity is an "it - there".

That combination or integration, is an example of a Final Cause (Aristotle, see defns. of FIC). (I have also written explanations of the FIC.)

The realization that the entity is there in substantial and sensory form is another first level generalization. (See defn. of FLG.).

The realization that the entity that is completely seen, acted towards, and touched, taken together as a single cause is a something is a discovery.

With that realization a second order generalization has occured. (See defn. of SLG.)

Based upon this evaluation, I suspect that the child may be aware, in some forms, of the existence or use of three of the Four Causes, or, more likely, in combination with one another as a Final Cause.

Based upon this evaluation, I suspect that the child may be aware in different ways of both First and Second Level Generalizations.

I suggest that the method of inquiry, at this stage, be granted an experimantal design, or a plan for further investigation.

There are the prior conditions of the child that need to be evaluated. Those, for example, may be the prior discoveries and learning by the child, the physical properties of, and the physical changes to the child.

Conditions existing at birth and at the stages of growth prior to the touching of the ball should also possibly be investigated.

All in all, I suggest that the Four Causes of Aristotle be applied to the basic philosophy of psycholgy. In that way all prior conditions may be assessed and the integrations of all relationships evaluated.

Inventor

Share this post


Link to post
Share on other sites
The point that Harriman is making is that it is not grasping causation but experiencing it on a perceptual level in particular instances. When concepts are formed, the particulars become universals, and the generalization is formed.

This seems to be the argument in a nutshell. But why not simply use the word 'associate' in that perceptual case. Inductions are based on associations which only become 'causal associations' in the conceptual realm, is how I think Betsy sees it. No?

Interesting question. What is an association? Is the awareness of 'association' perceptually derived? Association seems like a related concept to cause: one can associate things because of temporal or spatial closeness (and many other reasons) but there may not be a cause. Whereas a cause implies specific type of association, a specific sequence of interactions: the effect is always produced regardless of temporal or spatial relationships.

Harriman does not discuss this issue because I think it is not relevant to the issue of causality. Of what relevance would it be to show that something that is, in fact, a cause is an association when the association would not be the basis for a generalization or induction? According to his theory, one would not form a valid generalization based upon associations.

By the same method of grasping causality, a child would just as likely grasp an association.

Share this post


Link to post
Share on other sites
--------

The child first sees a blurry entity that may be a color, say, yellow.

----------------

We are not talking about babies who just popped out of the womb.

Share this post


Link to post
Share on other sites
The point that Harriman is making is that it is not grasping causation but experiencing it on a perceptual level in particular instances. When concepts are formed, the particulars become universals, and the generalization is formed.

This seems to be the argument in a nutshell. But why not simply use the word 'associate' in that perceptual case. Inductions are based on associations which only become 'causal associations' in the conceptual realm, is how I think Betsy sees it. No?

Interesting question. What is an association? Is the awareness of 'association' perceptually derived? Association seems like a related concept to cause: one can associate things because of temporal or spatial closeness (and many other reasons) but there may not be a cause. Whereas a cause implies specific type of association, a specific sequence of interactions: the effect is always produced regardless of temporal or spatial relationships.

Harriman does not discuss this issue because I think it is not relevant to the issue of causality. Of what relevance would it be to show that something that is, in fact, a cause is an association when the association would not be the basis for a generalization or induction? According to his theory, one would not form a valid generalization based upon associations.

By the same method of grasping causality, a child would just as likely grasp an association.

Consider one example. A mother brings a hungry child food. The child perceives the mother, the food, and his hunger. His hunger dissipates after eating. He can legitimately associate the food and the mother with his hunger dissipating. But what can he generalize later on in life? "All mothers make hunger go away?" or "All food makes hunger go away?" If Harriman is trying to demonstrate the basis for a generalization, would the former or latter example serve his purpose?

Share this post


Link to post
Share on other sites

Yes, and the reason scientists -- and non-scientists -- do induction is??

To establish general principles about the world, beyond what is known from observation of limited instances and evidence.

And the reason they do that is??

OK, I'll stop being coy. I define induction as the process of identifying causes.

Being coy serves no legitimate purpose here.

I was trying to focus entirely on Harriman's book and Harriman's arguments and leave my own ideas on induction out of this discussion, but I now realize that is impossible.

Identifying causes is essential for explanation but is only part of what is involved in the concept of induction, which is a process for establishing generalizations. That is a different concept from "identifying causes" and cannot be obliterated or ignored.

I have absolutely no intention of obliterating or ignoring generalizations. My goal is to put generalizations in their proper epistemological place. While all generalizations are causal statements, not all causal statements are generalizations. For example, "Watching Obama made me sick" is a causal statement, but not a generalization.

Identifying causes is the essential purpose of induction, not "establishing generalizations." Induction is the crucial, purposeful process of identifying what will cause us to gain, keep, or lose the values our lives depend on. It is not a rationalistic process of starting with generalizations from God knows where and then trying to justify them.

Generalizations are the conceptualization of causal relationships. A generalization integrates, into a single causal statement, the causes of similar concrete, perceived actions just as concepts integrate, into a single cognitive entity designated by a word, the similar characteristics of the concrete, perceived entities that are units of the concept.

Share this post


Link to post
Share on other sites

I define induction as the process of identifying causes.

How is that significantly different than this?

The structure of inductive reasoning, in general, on any level, is: observation; the application of one's total conceptual framework; and therefore, necessarily, a generalization.

He previously mentions

Let us start by noting that all generalizations - first-level and higher - are statements of causal connection. All assert (or imply) that an entity of a certain kind necessarily acts in a certain way under a given set of circumstances, which is the essence of the law of causality. ... there is nothing to make any generalization true, except some form of causal relationship between the two [existents].

Harriman recognizes that causes are inherent in induction, but he does not hold them as the essential, defining characteristic of induction nor as the essential purpose of induction.

Share this post


Link to post
Share on other sites
For example, "Watching Obama made me sick" is a causal statement, but not a generalization.
What a lovely example... They should really make an appearance in a grammar book; "Yes, Sally. 'Is' is the helping verb in the sentence, 'Congressman Frank is deplorable.' Good job!" :)

Share this post


Link to post
Share on other sites
Causality is a first-level concept ...

You can't have the concept causality until you recognize particular instances of it. Otherwise there are no units to integrate.

That's true. A child first recognizes particular instances of causality when he discovers that he can make his hands and feet move. He is aware, on a perceptual level, of the motion of his hand which he can both see and feel, and he can feel the effort he made to make his hand move. He experiences himself as a cause and the motions of his body as the effect. This takes him a while to learn, and begins with "hand regard" or hand-watching at 12–20 weeks leading to the ability to reach and grasp at about 5 months of age when he discovers that he can cause things outside of his own body to move. That is the perceptual basis for a child's first awareness of causality.

Causality as identity applied to action requires the concepts identity and action and is not first level. But you could grasp causality implicitly at a preconceptual level when you have and are aware of the ingredients not yet integrated into a concept.

Causality as identity applied to action requires a very sophisticated Aristotelian perspective which very few adults actually have. Instead, most people have a commonsense concept of causality based on perception and defined ostensively.

Share this post


Link to post
Share on other sites

LL (and anyone else) isn't *wrong* to observe that children have a primitive grasp of causality. But it's an error to describe that primitive grasp of causality as merely perceptual, even in a pre-conceptual infant. A pre-conceptual human infant's mental processing is still different than any other animals' (although animal consciousness itself has been underrated, a different subject.) I think an intelligent pre-conceptual infant could still make a connection between ball-rolling and its spherical nature (in deliberate conscious contrast to non-rolling cubic blocks) *without* explicitly having the concept of 'ball' or 'block'.

Until a mind actually has *some* grasp of the WHY: WHY does the ball roll when pushed, then I don't think there is any way to argue any kind of certainty at all. Certainty-by-repetition is just another way of describing induction-by-enumeration, which is a wrong method. And getting at the WHY involves just the point that Betsy is making: it's getting at the *identity* of the entity that is acting.

Conceptualization will ultimately tie to that process by permitting a generalization to be made about all particular instances of the entity - *sometimes*. But it's more rationalism to try to make a 1:1 mapping there as well. It may not be true that a cause possible to an instance of the same concept will apply to all instances. It is true that *some* chickens lay eggs. But only the female kind (in which case a subdivided concept, hen, is useful.) It is true that *some* tables can cause pain when bumped into - ones with sharp corners. But not all of them. In other words there may only be particular attributes of some particular instances which are in fact the cause of a certain effect.

Share this post


Link to post
Share on other sites
The point that Harriman is making is that it is not grasping causation but experiencing it on a perceptual level in particular instances. When concepts are formed, the particulars become universals, and the generalization is formed.

This seems to be the argument in a nutshell. But why not simply use the word 'associate' in that perceptual case. Inductions are based on associations which only become 'causal associations' in the conceptual realm, is how I think Betsy sees it. No?

Interesting question. What is an association? Is the awareness of 'association' perceptually derived? Association seems like a related concept to cause: one can associate things because of temporal or spatial closeness (and many other reasons) but there may not be a cause. Whereas a cause implies specific type of association, a specific sequence of interactions: the effect is always produced regardless of temporal or spatial relationships.

Harriman does not discuss this issue because I think it is not relevant to the issue of causality. Of what relevance would it be to show that something that is, in fact, a cause is an association when the association would not be the basis for a generalization or induction? According to his theory, one would not form a valid generalization based upon associations.

By the same method of grasping causality, a child would just as likely grasp an association.

Consider one example. A mother brings a hungry child food. The child perceives the mother, the food, and his hunger. His hunger dissipates after eating. He can legitimately associate the food and the mother with his hunger dissipating. But what can he generalize later on in life? "All mothers make hunger go away?" or "All food makes hunger go away?" If Harriman is trying to demonstrate the basis for a generalization, would the former or latter example serve his purpose?

It may take more than one instance of experience to come to a general conclusion that is correct in identifying the 'cause' in this case. One one level (associating), one could correctly say that mothers do make hunger go away. It may take an experience of being fed by a stranger that switches the association to the common denominator of food. The question for me is, is the correct association of food satiating hunger, a causal understanding? I am inclined to think that only a conceptual grasp of the association can be regarded as causal. The alternative is to have two levels of "causal". One of association, and one of understanding the nature of the entities involved.

For example:

If some primitive tribe finds that a certain herb causes them to respond a certain way, they have correctly associated that the herb causes the result. But if they think that it is a spirit living in the herb that is the cause of the result then one can see that 'cause' is need of further explanation and definition.

Share this post


Link to post
Share on other sites
--------

The child first sees a blurry entity that may be a color, say, yellow. [...]

We are not talking about babies who just popped out of the womb.

Thanks for the question. No, we are not.

I pictured a setting in which a child in a crib was crawling around the crib, and who was making his first attempts at exploring the local universe on his/her own. A world where seeing, touching, and moving are all important. To the child what may be perceived has axiomatic importance as real existents. To that child there are no epistemological existents.

Even then, I realized that there may be a lot of learning that takes place at stages earlier than solo crib time.

Piaget has something to say regarding induction, and that is that the child takes reality for granted, and the child finds that everything in his/her world exists. If something has a particular color that's the color it is, for example.

Inventor

Share this post


Link to post
Share on other sites
----------------

It may take more than one instance of experience to come to a general conclusion that is correct in identifying the 'cause' in this case. One one level (associating), one could correctly say that mothers do make hunger go away. It may take an experience of being fed by a stranger that switches the association to the common denominator of food. The question for me is, is the correct association of food satiating hunger, a causal understanding? I am inclined to think that only a conceptual grasp of the association can be regarded as causal. The alternative is to have two levels of "causal". One of association, and one of understanding the nature of the entities involved.

The question Harriman seems to be raising is, would the child be able to grasp the causal generalization 'food satiates hunger' without the experience of particular foods satiating his hunger when he is hungry, i.e., without that particular perception of the cause? I think not.

There are various levels of understanding what a cause is. It doesn't take many times for a child to observe the causal connection that eating food stops his hunger without forming a generalization. Yet even the generalization 'food satiates hunger' is not the ultimate cause: the science of digestion and chemistry is needed to explain the basic cause. Why we get hungry requires understanding also: we need energy for the body to function. After all, not all food satiates hunger: there are waste products.

For example:

If some primitive tribe finds that a certain herb causes them to respond a certain way, they have correctly associated that the herb causes the result. But if they think that it is a spirit living in the herb that is the cause of the result then one can see that 'cause' is need of further explanation and definition.

But there is no perceptual grounding for such an causal explanation, which is why the concept of association is required and why the generalization is invalid. An association is a conjunction of two events, things, actions, attributes, etc. for which no causal explanation is available.

Share this post


Link to post
Share on other sites
LL (and anyone else) isn't *wrong* to observe that children have a primitive grasp of causality. But it's an error to describe that primitive grasp of causality as merely perceptual, even in a pre-conceptual infant. A pre-conceptual human infant's mental processing is still different than any other animals' (although animal consciousness itself has been underrated, a different subject.) I think an intelligent pre-conceptual infant could still make a connection between ball-rolling and its spherical nature (in deliberate conscious contrast to non-rolling cubic blocks) *without* explicitly having the concept of 'ball' or 'block'.

Until a mind actually has *some* grasp of the WHY: WHY does the ball roll when pushed, then I don't think there is any way to argue any kind of certainty at all. Certainty-by-repetition is just another way of describing induction-by-enumeration, which is a wrong method. And getting at the WHY involves just the point that Betsy is making: it's getting at the *identity* of the entity that is acting.

Conceptualization will ultimately tie to that process by permitting a generalization to be made about all particular instances of the entity - *sometimes*. But it's more rationalism to try to make a 1:1 mapping there as well. It may not be true that a cause possible to an instance of the same concept will apply to all instances. It is true that *some* chickens lay eggs. But only the female kind (in which case a subdivided concept, hen, is useful.) It is true that *some* tables can cause pain when bumped into - ones with sharp corners. But not all of them. In other words there may only be particular attributes of some particular instances which are in fact the cause of a certain effect.

I'm not sure why you are bringing in the issue of certainty here. What do you see as the relationship between certainty and induction? Harriman does not mention the issue of certainty in his book. In OPAR, Dr. Peikoff mentions "The concept of "certainty" designates knowledge from a particular perspective: it designates some complex items of knowledge considered in contrast to the transitional evidential states that precede them." Certainty arises in the context of probable vs. possible vs. certain types of evidence. How does this relate to generalizations, at least within the context of the book? The context for generalizations assume the validity of the senses, the laws of identity and causality.

Share this post


Link to post
Share on other sites

The cause of much disagreement originates from LL's claim that a child can perceive causation, or as Harriman puts it: "The experience of rolling a ball, therefore, is the experience of causing something to happen." and "... causing, too, is an object of experience." "Thus, the experience of the quenching [thirst] is the perception of the causing." (LL, p22,23) Harriman states "The philosophic foundation of the theory presented here can be found in Chapters 1 to 5" of OPAR.

Let's see if we can find some tidbits of support in OPAR for his thesis. I offer some quotations as indications that I believe support Harriman and address several objections that have been raised in this thread. (I have not cited the specific pages for each quote. For the full context, please look it up.)

The fact of consciousness is also a fundamental starting point. ....Consciousness, to repeat, is the faculty of perceiving that which exists. ("Perceiving" is used here in its widest sense, equivalent to "being aware of.")......Entities constitute the content of the world men perceive; there is nothing else to observe. ....

.....

When a child has reached the stage of (implicitly) grasping "entity," "identity," and "action," he has the knowledge required to reach (implicitly) the law of causality. To take this step, he needs to observe an omnipresent fact: that an entity of a certain kind acts in a certain way. The child shakes his rattle and it makes a sound; he shakes his pillow and it does not. He pushes a ball and it rolls along the floor; he pushes a book and it sits there, unmoving. He lets a block out of his hands and it falls; he lets a balloon go and it rises. The child may wish the pillow to rattle, the book to roll, the block to float, but he cannot make these events occur. Things, he soon discovers, act in definite ways and only in these ways. This represents the implicit knowledge of causality; it is the child's form of grasping the relationship between the nature of an entity and its mode of action.

...........

The validity of the senses is an axiom. Like the fact of consciousness, the axiom is outside the province of proof because it is a precondition of any proof.

Proof consists in reducing an idea back to the data provided by the senses. These data themselves, the foundation of all subsequent knowledge, precede any process of inference. They are the primaries of cognition, the unchallengeable, the self-evident.

.......

Sensory experience is a form of awareness produced by physical entities (the external stimuli) acting on physical instrumentalities (the sense organs), which respond automatically, as a link in a causally determined chain. Obeying inexorable natural laws, the organs transmit a message to the nervous system and the brain.

.......

Every existent is bound by the laws of identity and causality. This applies not only to the physical world, but also to consciousness

.........

As adults, as thinkers, and even as children beyond the infant stage, what we are given when we use our senses, leaving aside all conceptual knowledge, is the awareness of entities—nothing more, but nothing less.

.......

If one seeks to prove any item of human knowledge, on any subject, he must begin with the facts of perception. These facts constitute the base of cognition. They are the self-evident and the incontestable, by reference to which we validate all later knowledge, including the knowledge that, decades earlier, when we first emerged from the womb, we experienced a brief sensation stage.

......

Up to a point, a consciousness has no option in this regard. It cannot consistently disregard relationships among its contents because consciousness by its nature involves the discovery of relationships. This is true even on the preconceptual level.

.......

A first-level concept is one formed directly from perceptual data, without the need of prior conceptualization.

...........

Knowledge, therefore, has a hierarchical structure. ... A hierarchy of knowledge means a body of concepts and conclusions ranked in order of logical dependence, one upon another, according to each item's distance from the base of the structure. The base is the perceptual data with which cognition begins.

........

Propositions too (if nonaxiomatic) must be brought back step by step to the perceptual level. They too are based on antecedent cognitions—on the chain of evidence that led to them—going back ultimately to direct observation.

.........

Proof is a form of reduction. The conclusion to be proved is a higher-level cognition, whose link to reality lies in the premises; these in turn eventually lead back to the perceptual level.

(my bold emphasis)

Share this post


Link to post
Share on other sites