Chris Bolt on Hume and Induction
Chris Bolt writes:
One would presumably think that if Hume is outdated and his problem easily dealt with it would not require so much writing to provide an answer to my questions.
But I’m hoping to provide some pointers here for Bolt, though I’m supposing much of it will go over his head given his unfamiliarity with Objectivism.
Recall from what Dawson has written that I am still awaiting a response from him. What he has provided thus far does not suffice. His comments indicate that he has more to say in order to try and answer my questions.
Unfortunately, the lack of a conceptual understanding of induction does not stop presuppositionalists from promoting Hume’s problem of induction as a topic of debate with non-Christians. They apparently think not only that Hume’s conception of the problem needs to be addressed on Hume’s terms (which assumes that Hume’s analysis of induction is faultless), but also that belief in the Christian god somehow overcomes the problem. The operative implication of the inductive version of TAG (the “transcendental argument for the existence of ‘God’”) goes essentially as follows: if non-Christians can produce no satisfactory answer to Hume (again, taking the validity of Hume’s analysis of induction for granted), then Christianity is vindicated. Why? Because only a supernatural being such as the Christian god can guarantee the uniformity of nature.
As Brian Knapp writes:
In the nature of the case, the answer to the question of why it is reasonable to assume nature is and will continue to be uniform must originate from outside nature itself; that is, outside of man and his experiences. Any answer that originates from within nature will always ultimately be justified through the use of induction, as for any solution to apply to the unexperienced realm requires applying a conclusion drawn from experience to that which has not yet been experienced. (“Induction and the Unbeliever,” The Portable Presuppositionalist, p. 131)
My view is not only that thinkers need to make a more critical examination of Hume’s conception of induction, but also that we should recognize the inductive implications which concept-formation provides even before our very first inductive inference. I don't know why this should be so controversial, unless someone is trying to hide something.
This approach seems completely foreign to the presuppositionalist mindset, as if it had never been considered. Apologist James Anderson, for instance, has written an essay titled Secular Responses to the Problem of Induction. In this essay, Anderson reviews several of the more popular attempts to address the problem of induction, including those endorsed by Frederick Will, Max Black, Karl Popper, Hans Reichenbach, et al. Curiously, Anderson does not consider Kelley’s response to Hume. Indeed, I don’t think it would serve Anderson’s ends very well if he had. What is common to all the attempted answers to the problem of induction which Anderson does survey, is the fact that none of them points out that Hume’s analysis of induction is faulty. Additionally, none of those attempted solutions addresses induction’s nature as a conceptual process. When I pointed out to Anderson that
I'm always surprised, when reading a paper that attempts to deal with induction, that there is no discussion of concepts, the nature of their forming, or their relationship to inductive generalization, as if these issues did not matter
Well, it's not immediately obvious to me how the nature of concept formation bears either on the description of the problem of induction or on the development of cogent solutions.
it is evident that there presently exists no satisfactory solution to the problem of induction from a secular perspective. (Secular Responses to the Problem of Induction)
In short, the solution to the problem of induction involves two fundamental tasks: first, correcting the errors in Hume’s understanding of induction (Hume did not understand induction as a conceptual process; neither do presuppositionalists); and second, recognizing how Rand’s theory of concept-formation provides the working model of inductive generalization (as well as answers many of the misconceptions which attend the conventional understanding of the problem of induction, such as when Hume’s analysis of inductive reasoning is taken for granted).
But in my opinion, Bolt is not anywhere near ready for any of this, as his own worldview has burdened his mind with much unnecessary baggage. For instance, he seems not to have understood one of the points which Justin Hall proffered in his own 29 Sept. comment:
A things identity in totality really does not change, for that very identity includes all the ways in which it can change, and if it changes in a way not included in our identity of it, well we expand and improve of conceptualization of its identity.
But when it comes to a proper understanding of induction, where would Bolt go to find this? To the bible?
Please, this is a serious matter.
It would seem that a presuppositionalist – i.e., someone who feigns to be concerned about “presuppositions” – would have more appreciation for the foundations of inductive reasoning than Bolt displays. That is why I asked him to make it clear whether or not he disputes the truth of the Objectivist axioms.
Specifically, this is what I asked Bolt in my 17 Sept. comment:
In the meantime, I wanted to ask you if you dispute the truth of the Objectivist axioms. They are the following:
1. The axiom of existence: This is the axiom which states “existence exists.” It is the explicit recognition that something exists, that there is a reality.
2. The axiom of consciousness: This is the axiom which states “consciousness is conscious of something.” It is the recognition that, to be conscious of the fact that things exist (the axiom of existence), one must be conscious.
3. The axiom of identity: This is the axiom which states “to exist is to be something” (as opposed to “nothing”). This is the recognition that a thing which exists is itself, that to exist is to have a nature, an identity, that A = A.
4. The primacy of existence: This is the recognition that “existence exists independent of consciousness,” i.e., that the nature of an entity is what it is independent of the activity of consciousness.
If you dispute the truth of any of these axioms, it is important for your understanding that you make your contentions known before going any further. If your own understanding of the Objectivist position is not important to you, then I would ask that you make this clear.
As for the so called “Objectivist axioms”; it is necessary for Dawson to show in a much more specific manner how they are at all relevant to the discussion. So far he has been unable to do so. I find the axioms to be incoherent. It may be that I just do not know enough about them. In either case I rightfully have difficulty accepting them.
The reason why I asked Bolt whether or not he disputes the truth of the Objectivist axioms, was not specifically to draw out their implications for inductive reasoning per se, but to make it clear where he stands. If he denies the truth of the Objectivist axioms outright, then I want to know this before wasting any more time trying to educate him on the topic of induction. As for their relevance to induction, this should not be difficult to see. Induction is a mental process about objects of one’s awareness. As such, induction presupposes the truth of the axioms; it presupposes the fact that there is a reality (the axiom of existence), that the one performing inductive inferences is in fact aware of objects (the axiom of consciousness), and that the objects of one’s awareness have a specific nature (the axiom of identity). Induction also presupposes a relationship between consciousness and its objects, which is identified by the primacy of existence. If Bolt thinks that these axioms are not true, and/or fails to recognize their fundamental importance to inductive reasoning, then I would wager that he is in sore need of substantial remedial tutoring before he would be in any position to understand, let alone appreciate, the Objectivist analysis of induction and its answer to Hume. Indeed, that Bolt needs all this spelled out to him explicitly, only proves my suspicion that he is simply not ready for a crash course on the Objectivist analysis of induction, that in fact he should start with the basics, beginning with a primer in the axioms.
In the same breath, Bolt acknowledges that he may simply not understand the axioms and their relation to induction sufficiently. If that’s the case, then indeed he requires much schooling on the matter, and I do not know why this is my responsibility, especially if he insists on being unteachable on the matter. For all I know, it may be the case that he does not even recognize the fact that he assumes the truth of the Objectivist axioms every time he thinks, speaks and acts. Helping him understand this would be first-order business, long before we ever get to the conceptual mechanics of induction.
For example if “existence exists” is “something exists; there is a reality” then I do not understand why the tenet would be expressed in such vague language.
The language here is not vague. The axiom ‘existence exists’ identifies a fundamental truth using a single concept. If the axiom used more than one concept, we would be left asking: which is more fundamental? Rand avoided this by stating her irreducible primary as a single-concept axiom. In Objectivism, the concept ‘existence’ is a collective noun denoting everything that exists, which has existed, and which will exist.
It is wholly important that we not miss out on the purpose which the Objectivist axioms fulfill. They explicitly identify a relationship which is fundamental to all knowledge, as Porter explains:
Axiomatic concepts [‘existence’, ‘consciousness’, ‘identity’] are metaphysical concepts, identifying the fundamental distinction and relation between consciousness and existence, between the knower and the known, between epistemology and ontology. (Ayn Rand’s Theory of Knowledge, p. 223)
It hasn’t been explicitly articulated, so philosophers feel no discomfort in straddling it. (Porter, Ayn Rand’s Theory of Knowledge, p. 199)
I think the primacy of existence is the most important issue in philosophy. I think it’s the real axiom of Objectivism. (Ibid., p. 198)
In mulling over the meaning of the axiom of existence, Bolt himself acknowledges its truth:
Do I believe that something exists? Yes
Then he hastens to state:
God exists, for example.
Do I believe that there is a reality? I suppose that would depend upon how one defines “reality”.
Bolt goes on:
If “existence exists” is the same thing as “something exists” then “existence” must be “something”, but what is it and how is it known? Such vague language being utilized in the expression of an axiom makes me wary and raises suspicion that much more may be smuggled in somewhere down the line.
Hopefully my points above will put Bolt’s fears to rest.
But what about the other axioms? Bolt only kicks around on the axiom of existence. He does not indicate whether or not he disputes the truth of the axioms of consciousness, identity and the primacy of existence. Instead, he wants to discuss higher-level issues pertaining to induction. But if Bolt disputes the truth of the Objectivist axioms, it’s unlikely that he’ll understand (let alone accept) anything I have to say on induction, since the Objectivist theory of induction which I hold presupposes the truth of the Objectivist axioms. As Brian Knapp puts it:
The issue at hand is truly presuppositional in nature. (“Induction and the Unbeliever,” The Portable Presuppositionalist, p. 121)
In my 29 Sept. comment responding to Justin Hall, I wrote:
This 'makes sense' given their acceptance of the Humean conception of causation. I have discussed the problems with this conception of causation here.
Have you actually read Hume? He offers something quite like what you are presenting here. I am afraid that the solution is not as easy as saying that causation in Objectivism is drastically different so as to avoid Hume’s concerns. :) I have already addressed this attempt at a way out in my questions and plan to write more on it.
All events seem entirely loose and separate. One event follows another; but we never can observe any tie between them. They seem conjoined, but never connected. (An Enquiry Concerning Human Understanding)
If the mind of God does not sovereignly determine the relationship of every event to every other event according to His wise plan, then the way things are in the world and what happens there are random and indeterminate. In that case, there is no intelligible basis for holding that any experience is like any other experience, there is nothing objectively common to the two of them, and there is no causal connection between any two events – and thus they are meaningless and undescribable. (P. 110n.64)
The Objectivist conception of causation is radically different from the Humean view in that it (Objectivism) views causation as a relationship between an entity and its own actions rather than merely a relationship between “events” (however the term may be defined). The Objectivist view of causation is essentially the application of the law of identity to action (since actions exist, they have identity), and constitutes the recognition that an entity’s actions depend on its nature (hence it is a necessary relationship). This is axiomatic. It is also significant to a proper understanding of induction. And no, I never suggested that this is all there is to it, as Bolt seems to think. Objectivism does not say: “our conception of causation is different from Hume’s, and that alone solves the problem of induction.” The proper conception of causality is indeed very important to the matter, but it is not the only factor. There is also the theory of concepts. As Rand pointed out:
The process of observing the facts of reality and of integrating them into concepts is, in essence, a process of induction. (Introduction to Objectivist Epistemology, p. 28)
This is why, more and more, I have come to view concept-formation as providing the working model for inductive reasoning.
While presuppositionalists try to solve the problem of induction by (a) accepting Hume’s analysis of induction and (b) pointing to an imaginary creator which somehow guarantees the uniformity of nature (which is stipulated in that imagination), Objectivism takes a radically different approach, including (a) correcting Hume’s faulty analysis of induction and (b) understanding the conceptual nature of inductive reasoning explicitly.
Hume made numerous mistakes in his epistemology. But nowhere do I see Van Til or Bahnsen pointing this out when they deploy the inductive version of TAG. On the contrary, they seem to be counting on the potential that both apologists who wield TAG and those whom such apologists seek to engage, are simply unaware of Hume’s mistakes. For instance, in his essay “Induction and the Unbeliever” (The Portable Presuppositionalist, pp. 118-142), presuppositionalist Brian Knapp makes no effort to point out Hume’s mistakes. Indeed, doing so would be counter-productive to the intended outcome of TAG, which is essentially to elicit the response: “Duh, I donno! Must be God did it!”
Typically they believe that in order to use knowledge of the present to inform our projections of the future, we have to prove that nature is uniform. But this ignores several key facts, such as: (i) proof presupposes the uniformity of nature, and (ii) the uniformity of nature is essentially the consistent application of the axioms
Ignores? I do not think it ignores these things at all.
If the challenge is that one prove that nature is uniform (as my comment noted), then the challenge does in fact ignore the fact that proof presupposes the uniformity of nature. The uniformity of nature is not established by proof. To require a proof is to commit the fallacy of the stolen concept. But in his essay, Knapp doesn’t seem to recognize this either.
Bolt goes on to ask:
What difference does it make that the uniformity of nature must be presupposed?
The difference which this makes is the difference between knowledge and fallacy. Axiomatic truths are not truths which must be established by means of proof. It is not the case that the uniformity must merely be presupposed – certainly not for its own sake. It is a precondition of meaning. But since this is ultimately axiomatic, that is not a problem for those who are willing to govern their cognition according to the constraints delineated by the axioms.
The skeptic is essentially asking, “Why presuppose it?”
The very question “Why presuppose it?” itself presupposes it. The question would not be meaningful without it. That’s all we need to point out to the skeptic. If he doesn’t like it, he can pound sand. His likes and dislikes do not alter reality. Nor do they constitute a lien on man’s cognition.
It has not been shown how “consistent application of the axioms” solves the problem either, regardless of how many times Dawson repeats the “Objectivist axioms” as though they are philosophically insightful.
Nowhere did I say that the consistent application of the axioms “solves the problem [of induction],” but rather that ”the uniformity of nature is essentially the consistent application of the axioms”. Take a look at the axiom of identity. It is the recognition that to exist is to be something, to have a nature. If something exists, it is itself. As Rand rightly put it, “Existence is Identity” (Atlas Shrugged). How one could deny this truth and yet affirm the uniformity of nature is beyond me. One would need (very) good reason not to integrate new units into his knowledge according to this recognition once it’s been made explicit. The skeptic is cognitively impotent at this point, since any attempt he makes to validate the move to jettison this recognition will itself assume the truth of this recognition. So he can only commit the fallacy of the stolen concept in asking us to pursue with him his skeptical course.
For skeptics, ‘the future’ is merely a stand-in for ‘the unknowable’…But for rational individuals, the concepts ‘past’, ‘present’ and ‘future’ are merely temporal designations.
The term future is not synonymous with the term unknowable
and there is no need for a skeptic to assume that it is.
Of course “past”, “present”, and “future” are temporal designations. So what? You have not provided anything that would lead one to believe that “preconditions” must therefore be “affirmed” at these different “times”.
Another, very crucial point about temporal designations (as which Bolt agrees the concepts ‘past’, ‘present’ and ‘future’ qualify), a point which most treatments of induction tend to miss, is the fact that in forming concepts of entities (and keep in mind that entities are what act), time is an omitted measurement. Moreover, so is location. This is significant.
In his essay “Induction and the Unbeliever” (in The Portable Presuppositionalist), Brian Knapp suggests that
an appeal to past experience in drawing conclusions about the future is the very definition of inductive reasoning (p. 126)
Although induction is primarily thought of in the relation of past events to future events, it is also relevant to the way in which a given event will occur in a different location. ...induction has both spatial as well as temporal applicability. (p. 122n.5)
While Knapp is correct to say that we make inferences about future happenings as well as about happenings in different locations, what he fails to grasp explicitly is the fact that time and place are omitted measurements. This has profound implications for induction. In fact, this is what makes such inferences possible in the first place. Specifically, the fact that a concept integrating like entities into a single unit omits (i.e., de-specifies) temporal and spatial measurements, is what makes them available to our inferences about an entity’s actions regardless of time and place. But the knowledge necessary for such inferences, i.e., the information which we need to inform such estimations, is already available to us in our formation of concepts of entities. Gotthelf summarizes as follows:
The integration distinctive of concept-formation begins with multiple perceptual grasps of a small number of individuals (for example, a child’s noticing of some tables similar to each other and different from some nearby chairs), and moves to an open-end grasp of all relevantly similar individuals, past, present and future (for example, a grasp of all tables, past, present, and future). (Ayn Rand on Concepts)
So not only do temporal designations themselves presuppose certain fundamental preconditions in order for them to have any meaning (let the skeptic affirm their meaning while denying their genetic roots), the fact that time is an omitted measurement in the formation of concepts of entities eliminates the skeptical hurdles which the Humean conception of induction arbitrarily imposes on human cognition (let the skeptic affirm that the formation of concepts does not omit measurements). The skeptic’s angle simply implodes on itself. Meanwhile, the concept ‘future’, properly understood, simply does not have the adverse significance for inductive reasoning that the skeptic assumes it has.
“The future,” then, refers to a continuation of the reality which exists from the present.
No Dawson. How do you know that reality will continue from the present?
Bolt then asks:
How do you know that it will be the same? Are you saying that reality never changes, that specifics of reality never change, or what? Be careful lest you head down the same road as Justin! :)
[The concept ‘future’] does not, therefore, refer to some alien universe whose physics constitute a reversal of those which apply in the reality which exists.
Presuppositionalists point to Hume as if his conclusion regarding induction were sound. But they never show that it is sound. They simply assume that it is, and with this assumption they endorse all of Hume’s relevant mistakes.
You can hand waive all day but it will not make the arguments and questions go away. [sic]
At any rate, observing that an argument is faulty, is all one needs to do to “make it go away.” If an argument is unsound, why should anyone need to pay it any mind? Unfortunately, Bolt has not shown that Hume’s argument is error-free. He can wave his hand all day, but that will not make the errors which an argument commits go away.
by Dawson Bethrick