David Hume's argument contra validity of induction

[Michael Stuart Kelly]

If you weren't claiming that Popper would classify such a statement as "Some swans are white" as being an induction, I'm at a loss to figure out what you were claiming.

Ellen,

Well, that's a start in the right direction, at least. It's always a good idea to torch the strawman and say you just don't know. It gets rid of cognitive clutter and makes for a better exchange of ideas.

(1) Do you or do you not classify "Some swans are white" as an induction?

Yes.

(2) Do you or do you not believe that Popper would have classified "Some swans are white" as an induction?

Not from what I read and remember. I just looked it up and became so sure it hurts:

Induction simply does not exist, and the opposite view is a straightforward mistake.

It doesn't get any clearer than that. If that isn't an induction disparager, I don't know what is.

There's a third question and qualification that makes all this even clearer, but I'll let you mull that over for a while.

Michael

[Brant Gaede]

You can't use induction "for timeless conclusions."

Sorry, Brant. "Timeless conclusions" is just what "induction," traditionally understood, is about: How do you go from X observations that Y to the generalization that ALL observations of Y will be the same? (How do you go from a set of particulars to a universal?)

Ellen

Well, you don't, do you? You merely come up with a falsifiable theory. An inference. A conclusion. A "universal."

--Brant

[Ellen Stuttle]

You can't use induction "for timeless conclusions."

Sorry, Brant. "Timeless conclusions" is just what "induction," traditionally understood, is about: How do you go from X observations that Y to the generalization that ALL observations of Y will be the same? (How do you go from a set of particulars to a universal?)

Ellen

Well, you don't, do you? You merely come up with a falsifiable theory. An inference. A conclusion. A "universal."

--Brant

Right, Brant. No, you don't. That's just what Popper argued, contra the expectation that a solution would be found according to which you could go from a set of particular observations to a guaranteed conclusion about instances NOT observed. That's why Popper said that the traditional "problem of induction" can't be solved. You are accepting Popper's argument, while showing no awareness of where "the problem" came from to begin with.

I'll add, yet again, the O'ist idea of "contextual certainty" agrees with Popper that there's no way you can go to an incontestably true statement about instances not yet observed from instances thus far observed. On this point, O'ism and Popper agree. But O'ists are wont to call Popper Mr. Bad for saying just what they say, in different terminology.

Ellen

___

[Ellen Stuttle]

If you weren't claiming that Popper would classify such a statement as "Some swans are white" as being an induction, I'm at a loss to figure out what you were claiming.

Ellen,

Well, that's a start in the right direction, at least. It's always a good idea to torch the strawman and say you just don't know. It gets rid of cognitive clutter and makes for a better exchange of ideas.

(1) Do you or do you not classify "Some swans are white" as an induction?

Yes.

(2) Do you or do you not believe that Popper would have classified "Some swans are white" as an induction?

Not from what I read and remember. I just looked it up and became so sure it hurts:

Induction simply does not exist, and the opposite view is a straightforward mistake.

It doesn't get any clearer than that. If that isn't an induction disparager, I don't know what is.

There's a third question and qualification that makes all this even clearer, but I'll let you mull that over for a while.

Michael

Michael, sometimes I despair of ever communicating with you on philosophic issues. Are you completely missing that Popper is defining "induction" differently than you are? And that this point is what I'm trying -- apparently with no success -- to get through to you? He wasn't disparaging what you call "induction," since he didn't mean by "induction" what you mean by "induction." He would have had no objections I'm aware of to the statement that "Some swans are white," a statement well attested by observation. But such a statement isn't what he meant by "induction."

By "induction" he meant an argument, proceeding from premises to a conclusion. E.g., proceeding from the premise that "some swans are white" to the conclusion that "all swans are white," or, more cautiously, "most swans are white." He didn't mean, as you do, an identification of a set of observations. (A "set" there might mean only one example. One trustworthily attested observation of one white swan would make the statment "some swans are white" true. The problem Popper was talking about was how to get beyond that to an incontestable statement about a wider set of observations which haven't been made.)

See my post above to Brant.

Ellen

___

Edited September 2, 2008 by Ellen Stuttle

[merjet]

By "induction" he [Popper] meant an argument, proceeding from premises to a conclusion. E.g., proceeding from the premise that "some swans are white" to the conclusion that "all swans are white," or, more cautiously, "most swans are white."

Yup. Popper was quite capable of playing the word games that Mr. Barnes delights in accusing Rand of. Rand idiosyncratically defined "selfishness" and "altruism" and ruled anybody else's conception of those terms "out of court." Popper idiosyncratically defined "induction" and "valid" and ruled anybody else's conception of the terms "out of court."

Edited September 2, 2008 by Merlin Jetton

[tjohnson]

Yep, no one's perfect, eh?

[Michael Stuart Kelly]

Michael, sometimes I despair of ever communicating with you on philosophic issues. Are you completely missing that Popper is defining "induction" differently than you are? And that this point is what I'm trying -- apparently with no success -- to get through to you? He wasn't disparaging what you call "induction," since he didn't mean by "induction" what you mean by "induction." He would have had no objections I'm aware of to the statement that "Some swans are white," a statement well attested by observation. But such a statement isn't what he meant by "induction."

By "induction" he meant an argument, proceeding from premises to a conclusion. E.g., proceeding from the premise that "some swans are white" to the conclusion that "all swans are white," or, more cautiously, "most swans are white." He didn't mean, as you do, an identification of a set of observations.

Ellen,

Incredible! That has been my point all along.

You have constantly claimed that the Objectivist definition is not valid so it practically doesn't exist. I have claimed it is and that dictionaries exist so two meanings of the same word can exist, that many people use that definition, etc., etc., etc. Would you like me to cite posts? I have only said this several times in the last 2 days alone.

I have also stated that in conceptual terms, Popper and Rand were on the same wavelength on several things, but they used different jargon. I have only said this several times also ever since I read "Two kinds of Definitions" and reported on it. (You vehemently disagreed with me at the time. Now you are agreeing.) Want more posts?

Talking with you about this has been difficult because you want to restrict the word induction to mean the propositional version only and bash everything else as ignorance, especially "Rand's errors." I don't mind bashing them when they are actual errors, but not when the issue is semantics and she was consistent with her meanings.

What I have been trying to say all along is (pointing) this is Rand's meaning here and that is Popper's meaning there (or whoever), and both use the word induction. The differences are these and the similarities are those.

Maybe we can now lay that particular issue at rest and simply be clear on which meaning we are using at a specific moment, rather than bash one or the other for ignorance.

About the traditional proposition per se, it rests on the premise that full time (past, present and future) is not a component of the existence of a thing, that a thing does not exist in full time, it only exists in partial time, thus one part of time can be excluded from the reasoning process. This is what allows for the idea that just because the sun was observed coming up a gazillion times in the past, this says nothing about if it will come up tomorrow.

If you exclude time, that's easy. But not if you believe that the rotation of planets exists within the time-space continuum. Then it is contradicting the nature of planetary rotation to exclude full time from reasoning. Ironically, to do that you have to make a statement of fact about the future, i.e., that since you cannot know if the sun will come up tomorrow, the fact exists that it might not come up tomorrow, in other words, that the future will present unknowable information. (I wonder how that fact can be known if observation of the past is ruled out as a standard and something needs to be falsified instead.) If you believe that the rotation of planets exists only in the past and the present, then this argument makes some kind of sense. But that rests on an unfounded proposition itself, that time exists as two different continuums.

The problem of induction boils down to pretending that the present and the past exist as one thing, and the future exists disconnected from them. (The buzzword that signals that this is the problem is "predictability" when metaphysics and not just science is being discussed.)

When we observe something, we observe its existence in time, which means, past, present and future. It is true that we cannot see into the future, so we cannot see what its future will be (as if we were observing the present) because we also exist in the same time-space continuum, but we can see what its future was from examining past instances and following the timeline. This is how causality is identified.

The main difference I see between Rand and Hume (on which Popper's theory rests) as regards their agreement (that the future contains information presently unknown) is one of focus. The Humean version uses the premise that the future is disconnected from the past and present, and Rand's premise is that we are not omniscient, so we cannot observe all that exists. This is what makes one side claim that we cannot know with certainty if the sun will rise tomorrow and the other call this nonsense, even though it admits the possibility of something presently unknown interfering with it.

Another way of saying this is that one side claims that the future contains unknowable information, that we can only presume information about the future if we cannot falsify a proposition at the present, and the other side claims that the future contains both known information (based on causality) and unknown information (based on man not being omnipotent).

Michael

[BaalChatzaf]

You have constantly claimed that the Objectivist definition is not valid so it practically doesn't exist. I have claimed it is and that dictionaries exist so two meanings of the same word can exist, that many people use that definition, etc., etc., etc. Would you like me to cite posts? I have only said this several times in the last 2 days alone.

The Objectivist claim that induction is a valid method of getting true general statements from a finite set of particular statements is flat out false.

Ba'al Chatzaf

[Michael Stuart Kelly]

The Objectivist claim that induction is a valid method of getting true general statements from a finite set of particular statements is flat out false.

Bob,

The Objectivist claim that induction is a valid method of getting true general statements from a finite set of particular statements is flat out true.

Michael

[BaalChatzaf]

The Objectivist claim that induction is a valid method of getting true general statements from a finite set of particular statements is flat out false.

Bob,

The Objectivist claim that induction is a valid method of getting true general statements from a finite set of particular statements is flat out true.

Michael

Counter example: Europeans having seen only white swans concluded inductively that all swans are white. Then the got to Australia and they found black swans there. So the general statement all swans are white is false. It is an example of a false generality inferred from several true statements. This counterexample is sufficient to show that induction is NOT generally an valid form of inference.

Inductive reasoning gets general statements (universally quantified statements) from a finite set of particular (non-quantified) statements. Sometimes these generalities are true and sometimes they are false. The existence of a false inductive generalization is sufficient to prove that induction is not a valid mode of inference.

Logic 101 at work.

Q.E.D.

Ba'al Chatzaf

[Michael Stuart Kelly]

Bob,

We are losing the idea in ham-handed repetition.

You are another who does not believe in dictionaries.

Michael

[Brant Gaede]

The Objectivist claim that induction is a valid method of getting true general statements from a finite set of particular statements is flat out false.

Bob,

The Objectivist claim that induction is a valid method of getting true general statements from a finite set of particular statements is flat out true.

Michael

I really don't see how reasoning can be befurcated into inductive-deductive. The human mind takes the ball of evidence and tosses it back and forth between the two to come up with "truth." Some truths are more tentative than other truths. Some truths are so obviously true we don't even think of them in any other way, but the more abstract the more wobbly, generally speaking. The truth in a proposition is confirmed by things that literally work until something doesn't work comes along, then the proposition has to be discarded or recast. That's not the only way a proposition gets thrown out, of course. If there is a "problem" with inductive reasoning then there is a problem with scientific methodology. The "problem" is a philosophical problem that philosophers toss back and forth amongst themselves. It can't be solved because a solution doesn't and can't exist. The reason bifurcation is analogous to the trifurcation of "time" into past, present and future. The first two of the three don't exist. The future is all that exists of those three. Time is only the measurement of motion. The "present" is the (lived in) future. The future of future is no more real than the present or past. The past was only once the future. The present can't even claim that; it never was. However, by examing what we know of the past we can make predictions about what will be which will thus inform what we do now.

--Brant

Edited September 2, 2008 by Brant Gaede

[BaalChatzaf]

Bob,

We are losing the idea in ham-handed repetition.

You are another who does not believe in dictionaries.

Michael

Induction is a form of thinking or inference that goes from a set of true particular statements to a universally quantified statement (which may be true or false). Since the truth of the generalization is NOT guaranteed,. the process of producing the generalization is not valid.

See

http://en.wikipedia.org/wiki/Inductive_reasoning

and

http://plato.stanford.edu/entries/logic-inductive/

to get a start on the subject.

Ba'al Chatzaf

[Michael Stuart Kelly]

Bob,

Thank you. I already have a start on the subject.

In this discussion, there are 2 different definitions being used for induction.

Why is that so hard to understand? It has only been mentioned a gazillion times.

You can say, according to definition one, yada yada yada. Or according to definition two, yada yada yada.

But not total blank-out as if two definitions were not being discussed. It's like there is this colossal nothing where an idea should be along with constant repetition of the obvious.

Michael

[Paul Mawdsley]

Is it possible that there are two forms of induction that correspond with two modes of thinking about reality: scientific thinking and metaphysical thinking? One operates in bottom-up or particular to whole thinking. The other operates in top-down or whole to particular thinking. One is Popper's and one is Rand's. Is it possible that this is causing much confusion here? I believe Rodney was making a similar point toward the end of the Popper thread previously, suggesting the one Rand applied had more the holism of an art and produces contextual certainty (truth).

Paul

[Michael Stuart Kelly]

Paul,

That is precisely it.

It's gratifying to see someone understand this.

Michael

[BaalChatzaf]

Is it possible that there are two forms of induction that correspond with two modes of thinking about reality: scientific thinking and metaphysical thinking? One operates in bottom-up or particular to whole thinking. The other operates in top-down or whole to particular thinking. One is Popper's and one is Rand's. Is it possible that this is causing much confusion here? I believe Rodney was making a similar point toward the end of the Popper thread previously, suggesting the one Rand applied had more the holism of an art and produces contextual certainty (truth).

Paul

Contextual certainty and $5.89 will get you a can of cashew nuts at the CVS Pharmacy.

What goodies has contextual certainty delivered unto us? What inventions? What items of wealth? What assets? In short what are some of the RESULTS of contextual certainty, whatever that is. X is certainly true except when it isn't?

Ba'al Chatzaf

[Dragonfly]

The point is that Randians are always busy bashing Popper, but they can hardly blame him for using another definition (that was also used by Hume) than the Randian definition. If Objectivists claim to have solved the problem of induction, they should use the definition that was used in stating the problem, i.e. Hume's and Popper's definition, not their own, new definition, because then they haven't solved anything, they've merely sidestepped the problem.

[BaalChatzaf]

The point is that Randians are always busy bashing Popper, but they can hardly blame him for using another definition (that was also used by Hume) than the Randian definition. If Objectivists claim to have solved the problem of induction, they should use the definition that was used in stating the problem, i.e. Hume's and Popper's definition, not their own, new definition, because then they haven't solved anything, they've merely sidestepped the problem.

The Re-definers think they have defined the problem away. That have not. Redefinition is evasion.

Ba'al Chatzaf

[tjohnson]

If this were mathematics it would be so much simpler.

Def: A swan is a stately heavy-bodied aquatic bird with very long neck white plumage as adult

Theorem: All swans are white

Let X be an adult swan, then it is white by definition.

[Paul Mawdsley]

The different perspectives are not contradictory. They are paradoxical. Think wave particle duality in physics. Wave is what you get when you measure properties of the whole system. Particles are what you get when you measure properties of the parts. Both are real. Both are....errr, valid. It should not be either, or. It should be a matter of stepping back, accepting both views are real, accepting both views can be part of one person, and finding synthesis. Synthesis doesn't come from choosing sides. Synthesis comes in the form of understanding the reciprocity of part-whole relations and in understanding how the mind can process such relations.

Paul

[Michael Stuart Kelly]

Paul,

Brant's not off, either. All the reasoning methods fit together. One doesn't exist without the other.

Michael

[BaalChatzaf]

If this were mathematics it would be so much simpler.

Def: A swan is a stately heavy-bodied aquatic bird with very long neck white plumage as adult

Theorem: All swans are white

Let X be an adult swan, then it is white by definition.

Swans are defined by their shape and by their habits, not by their color. Color is an accidental property of a swan. To use Objective-Speak color is not part of the conceptual common denominator of the set of swans. This was true before the first non-white swan was found in Australia.

Ba'al Chatzaf

[Paul Mawdsley]

Brant's not off, either. All the reasoning methods fit together. One doesn't exist without the other.

Michael,

We are getting closer to what I mean by orientations of consciousness and psychological/philosophical lenses. We can be more than any one orientation or lens. The ego does not have to identify with one and disown another. It needs to find a more inclusive context--i.e.: synthesis.

I need to think more about Brant's post. It seems I have as much difficulty understanding him as he has understanding me. It's just a matter of finding the right lens.

Paul

[Michael Stuart Kelly]

Paul,

Once again, I have to mention Ken Wilber. A synthesis like what you are saying is very much the theme of An Integral Theory of Consciousness, especially the parts dealing with his quadrant.

People may take issue with some of the ways he applies this quadrant, but it is extremely useful is many areas. I normally do not like comparisons of this kind, but I think it puts the DIM Hypothesis in its hip pocket and spits out change.

Michael