False Dichotomies, Package Deals, and Karl Popper

[Harrison Danneskjold]

I think it all boils down to the fact that Popper's epistemology consists of "guess and check".  Within that context it's very good and very effective at the checking (although checking your guesses requires positive induction, itself) but the entire realm of "guess" is vacant.

 

The human brain is not a randomizer.  We learn from experience and form hypotheses about the future; none of it is arbitrary [even what Peikoff considers arbitrary is actually specific and intentional- just not in accordance with reality].

For instance, Pascal's Wager is not an example of arbitrary guessing.  It has nothing to do with the facts of reality. . . But it does follow a specifically logical pattern.

 

Popper's ideas could be integrated with Rand's, but only if one takes his "checking" methodology and allows Rand to fill in what is absolutely not a "guess".

[tadmjones]

Is there a negative induction?

[Plasmatic]

Harrison said:
 

Popper's ideas could be integrated with Rand's, but only if one takes his "checking" methodology and allows Rand to fill in what is absolutely not a "guess".

No it couldn't and doesn't need to. Given the Oist position on active mindedness nothing like what you call "checking" is missing.
 
edit: removed last comment for errors

Edited July 8, 2013 by Plasmatic

[curi]

DonAthos,

 

When you describe my guess as being "good," I take that as implying that some guesses are better than others... And on what basis are we able to make this distinction?

 

I meant a good guess like worth trying, a good guess to make and consider. I didn’t mean good like true or partly true.

 

It’s also good in the sense of being plausible to me. Maybe you have the answer. I don’t see any reason it’s false in the first minute of thought, which is a good start.

 

Also, the important thing is that this sense of “good” has no authority. It is a casual, loose usage, not an important epistemological pronouncement. It has no solid meaning or impact, it was just meant to communicate. It has no particular consequences or implications.

 

When it comes down to it, being a “good guess” doesn’t matter. What really matters – what is decisive – is whether we have any criticisms or not. (In my view.)

 

"good" (as opposed to a guess that it was Miss Scarlet, on the HMS Bounty, with a bronzed pineapple)

 

But I have criticisms of those guesses. Why guess it was Scarlet when you found mustard? Why guess it was a location other than where the body was found? These issues might be answerable. But the simple version where you just suspect Scarlet without giving any answer to these issues is wrong, criticized, refuted.

 

as evidence (i.e. that which provides a "positive" basis for some particular "guess" as to what has happened)

 

Evidence is the observation data we have to work with. The fingerprints are evidence. But the way to use evidence is to look at what it contradicts. Evidence can be combined with some ideas to form a criticism and rule something out. Like, “It wouldn’t be Scarlet because the fingerprints have a special type of mustard that isn’t sold in stores, and she had no access to.” That combines the evidence (mustard found at scene) with some ideas to criticize and rule out a possibility.

 

(Note: Criticisms are open to counter-criticisms. None may succeed but the attempt is always allowed. So you might point out that actually a mustard depot was broken into recently, so maybe Scarlet both broke in there and did the murder, and was trying to frame Mustard. Then she becomes a suspect again. But then you catch the guy who broke in and find all the stolen mustard and none is missing, and thanks to this new evidence she’s ruled out again.)

 

To take a simpler example, if you see 3 apples on a table, that rules out the table being empty. The evidence can be used (via some thinking and ideas) to rule something out. Not only that, it rules out all numbers of apples besides 3 being on the table. So you can conclude the one remaining possibility: there are 3 apples on the table.

 

If it didn’t rule everything out -- 7 apples wasn’t ruled out for some reason -- then you absolutely better not conclude it’s 3 apples, no matter how much evidence and “support” you have. If all your evidence and support doesn’t rule out the 7 apples possibility, then what good is it? (Answer: well it’s good for ruling out 4 apples, 5 apples, etc, everything but 3 and 7. But it provides no legitimate support/authority/status/etc for 3 over 7, since it leaves them both as open possibilities.)

 

If even one other thing isn’t ruled out, that’s a really big deal. Why isn’t it ruled out? Why is option A so great when B isn’t ruled out? etc

 

it seems to me that I am operating more on the basis of drawing conclusions from the mustard fingerprints (in that they are presumed to point to the murderer) rather than contemplating Mrs. Peacock-as-murderer and finding some criticism of that theory.

 

Because you know the finger prints basically rule out everyone else. No one else has mustard prints.

 

How else would you even know what was “supported”? What conclusion would you draw, if your evidence didn’t rule anything out? Or if it only ruled out everything but ten things, then what? I think how much is ruled out is really the key factor. Each case is dramatically different by how much the evidence rules out.

 

Whether I'm able to explain to you "what constitutes how much of a basis for what, for all cases," to your satisfaction, or etc., does that make it improper for me to draw conclusions in the sorts of scenarios we're discussing, according to a positive approach?

 

This is a very common view. Induction and support are an unsolved problem, but one day we will solve them. That is, indeed, Rand’s view.

 

I don’t expect it. For one thing, your examples where we do stuff (like walk) without a full explicit understanding can be explained without induction or support. We could have a partial critical understanding, largely unconscious, and walk. I don’t think the walking example really helps one side over the other.

 

Another thing is the arguments against induction and support are not like “there’s a few gaps to work out”. They are more along the lines of “here are 5 reasons it’s impossible and misconceived root and branch, which no one has any answer to”.

 

Maybe we should go into that?

 

It remains a bit of an open question for me, whether Popperian epistemology seeks to describe what we already do, or what we ought to do

 

Both. Popper says no one has ever done induction. None, ever; that is a myth. Because it’s impossible. Some people thought they did it, but they were mistaken and didn’t understand what they were actually doing. Understanding the right approach could help people do it better.

 

People decide between which approach to try to do. Trying to do the wrong thing that doesn’t work can lead to a lot of wasted effort and mistakes.

I don't yet see any way around drawing a connection between mustard fingerprints and Col. Mustard -- which I would describe as being "positive"

 

What is the connection, exactly?

 

I see that they finger prints are compatible with Mustard being guilty and are (via a few arguments) incompatible with others. What kind of connection is there other than the compatibility with one option and incompatible with other options? (As always, this is fallible contextual knowledge, open to revision and criticism when it’s discovered that Scarlet has a jar of mustard in her purse or whatever.)

 

I'm trying to approach this literally, and imagine the actual scenario playing out. My daughter, when she is at the proper age, will sit down to a right triangle with smaller sides of 3 and 4, and she'll be asked to find hypotenuse x. Now, on the one hand, she could "guess" a number and then see whether it "contradicts her math knowledge" (though this would seem to me to possibly beg the question of how that "math knowledge" is acquired in the first place, if not in some positive manner)... maybe her first guess is 5. Or maybe it isn't -- maybe it is 6 or 7 or 8 or 5.1 or 5.2 or 5.3, and she rules them out, one by one, until (hopefully) she guesses 5.

 

Criticism does not mean one by one. Criticisms often rule out categories or sets of things. We can rule out the set of everything except 5 with some mathematical arguments.

 

And guessing doesn’t mean guessing at random. Do the math and work out 5 (you are allowed to get your guesses/ideas in any manner whatsoever, no problem using the Pythagorean theorem). Now 5 is one of the suggested possibilities (call it a “guess” or “idea” or whatever else you like). And if you try to say anything but 5 that’s easy to criticize. But 5 hasn’t got anything known to be wrong with it.

 

It’s like the apples on a table example earlier. If 7 wasn’t ruled out, you’d be in trouble saying it was 5. No matter how much positive basis 5 supposedly had, it wouldn’t matter at all if 7 wasn’t ruled out. If 7 isn’t ruled out, you don’t know the answer, you have a contradiction, you better look over things again. But if 7 is ruled out, then 5 is wonderful, you’re golden, no problem. For the status of 5, everything depends on whether or not 7 (or any other number) is ruled out or not. Either 5 is the one idea we have that isn’t ruled out, or it isn’t.

 

That's right -- there is a particular context to my claim that I'm typing on my keyboard, and my claim is (only) certain within that context. If tomorrow I found that I had... I don't know... lost my mind utterly a few weeks back, I might have to revisit this, insofar as I were able. But until I have good reason to entertain such a notion, I mostly likely won't. (Which might be a third sticking point? For how could I have "good reason" for anything?) It suffices for me to understand that knowledge and certainty are contextual.

 

I agree. No sticking point. Revisit it when you think it’s worth revisiting and have no criticism of doing so.

 

Again, I fear extending myself too much when I don't believe I quite grasp all of the matters at play, but on "imagination" and "creativity," I don't see these as being bad things, or incompatible with my views on epistemology, or how I approach the actual matters of my life.

I don’t think imagination and creativity have any incompatible with you or Objectivism. I bring them up sometimes because they are important to my approach.

 

at some point, I don't believe that "guessing" continues to be an appropriate description of how I come to a certain conclusion. Given a right triangle, with sides (in order of length) of 3, 4, and x, I am not "guessing" that x = 5.

 

The reason I prefer the word “guess” (and also “idea”) is because they are words with no status or authority.

 

I don’t believe, at any point, does one’s guesses gain any authority from their methods. The results must always be evaluated by a critical consideration of the content of the guess/idea/whatever, and nothing else. The source is not relevant to our critical evaluation and bestows zero special privileges.

 

I experience these two processes differently, and that seems to be an important distinction to maintain in our concepts and language. One is "guessing," the other is not.

 

The difference is that in one case you had knowledge and in the other you didn’t. (More precisely, in the second case where you guess what number I’m thinking of, you have knowledge that the right answer is a number, and some things like that, but that leaves open a very big set of possibilities that aren’t ruled out and you do not have knowledge of which of those it is.)

 

Suppose we are discussing an idea that "the planet Neptune exists." When you say that an idea "must be judged on its content and not its source," what does that mean here? If I were to see Neptune and lay claim to its existence on that basis, am I in error?

 

So you judge by whether you can find anything wrong with the idea that Neptune exists. (Including if there is a contradictory rival idea, such as “Neptune does not exist”, it would have to rule that out. If it doesn’t, that’s something wrong.) You do not judge by whether you first came up with the idea in a dream. That has no bearing on whether it’s true. It’s not a criticism of it; it doesn’t rule it out.

 

I may not be interested in considering what you dream. You may not be either. You need not consider everything you could consider. There’s a million things in life and we have limited time and attention and focus, so we have to be selective. We have to try to understand what is problematic and focus attention there.

 

But if you dream an idea, and then you consider it, and you don’t see anything wrong with it, then who cares that it came from a dream? It survived 5 minutes of you trying to criticize it. So did some other idea you got in another way. They have the same status now: they are ideas which you don’t see anything wrong with. That’s it, nothing else matters. (Ideas you haven’t yet critically considered at all, I have no interest in, again regardless of the source. If I wanted ideas like that, I could create plenty of my own.)

 

That I cannot criticize it? But there are a million possible planets, all fictional, for which I would have equal "criticism" as the proposed Neptune.

 

If a claim is arbitrary, say “it is arbitrary” and that is a criticism of it. Yes you have equal criticism of all of them, but you do have a criticism and have just given it.

 

But what "content" can a planet (or ultimately, anything) have apart from my experience of it?

 

For example, the currently prevailing idea of Neptune implies that if telescopes built to certain specifications are pointed at certain places at certain times, they will detect certain wavelengths of light that are rather different from what they detect when pointed in most directions at most times. In other words, they will detect something different than empty space with the occasional distant star. The telescopes could easily be computer controlled and programmed to turn on a light if Neptune is there. We could then look at the light and watch it turn on. That light turning on is part of the “content” of the Neptune idea – if it didn’t turn on we would have a criticism of Neptune. This is something other than your experience of seeing Neptune. (But, as always when dealing with reality, some kind of perception has to be involved somewhere.)

[Eiuol]

"But the simple version where you just suspect Scarlet without giving any answer to these issues is wrong, criticized, refuted."

 

The point Don seems to be making is that there is no good reason to suspect Miss Scarlet committed the murder, which you agree with. "There is evidence Mustard committed the murder because his fingerprints were at the scene of the crime." and "Mustard's fingerprints on the scene of the crime rules out Miss Scarlet" means exactly the same thing, though. Neither is a claim to what actually happened. In both cases, Mustard is presented as the person who makes other scenarios less likely  Evidence doesn't mean proof, it really just provides a reason to determine likelihood. I can concoct wild scenarios, but if you want to investigate, your best bet is to see what Mustard knows. This is how Bayesian inference works roughly where you use likelihood to make some reasonable hypotheses based upon likelihood, often taking into account evidence, but not able to generate "whodunit" on its own. I doubt you want a math lesson right now about Bayesian inference, but the terminology you're using seems to be the only issue, not the content of what you said.

[curi]

I know Bayesian math. Bayesian epistemology is false and refuted by Popper though. I'm certainly not agreeing with it. Objectivism also disagrees with Bayesian epistemology so I'm not sure why you bring it up.

[Eiuol]

I said Bayesian inference, I didn't say the use of Bayesian epistemology as a way to, on its own, determine what is true. It is useful insofar as you can generate some kind of hypothesis or starting point, but that's it. If even your claim is that Bayesian inference is totally invalid in all ways, I think you are contradicting yourself, since you are implicitly suggesting it is valid by the very means you rule out Miss Scarlet - that is, there is greater reason to investigate Mustard over Scarlet.

Edited July 8, 2013 by Eiuol

[curi]

"This is how Bayesian inference works roughly where you use likelihood to make some reasonable hypotheses based upon likelihood" -- this is epistemology, not math. the math formulas do not generate hypotheses, they only adjust the probabilities of hypotheses you already have given a bunch of other information (like prior probabilities, and more) that you already have. (and the math adjustments only apply to stuff where probability literally applies. extending it where it metaphorically applies isn't math)

[Eiuol]

Actually, it's used to determine how strong to "believe" something, it isn't necessarily probability. It really just assigns some kind of weight as a matter of likelihood to determine at least what is most sensible to look into first. That basis is nothing but the status of your knowledge about the world without abandoning multiple conjectures, hypotheses, or whatever else. You did the same thing with Miss Scarlet, except not nearly as explicitly. Indeed this isn't math, but that is a tool of thinking. I'm not saying you are literally doing Bayesian inference, only that the principle is the same.

[Harrison Danneskjold]

Is there a negative induction?

 I have never before seen Bigfoot; I know of nobody who has and I am not aware of any real evidence for Bigfoot.

Therefore Bigfoot probably does not exist.

The same method behind any other generalization, but to negate instead of affirm.  Or if there's something wrong with that then I used a redundant word and must come to terms with that, in time.

 

No it couldn't and doesn't need to. Given the Oist position on active mindedness nothing like what you call "checking" is missing.

 

 I never said that Oist epistemology needs Popper's addition; it really doesn't.  But it could be integrated with some modifications.

 

There's nothing wrong with standard Oist epistemology, exactly as Rand left it, except for the fact that it's incomplete.  It could use some fleshing out.

It has no contradictions.

Edited July 9, 2013 by Harrison Danneskjold

[Harrison Danneskjold]

 Curi: the only functional difference I can see between "guess and check" and simple inference is that the latter would allow us to simply leap to the obvious conclusion; the former requires that you guess, I criticize and then you counter-criticize at minimum before accepting anything as true.

If you'd prefer to call it something other than induction then fine; call it whatever you like.  But both approaches would reach the same results, right?

The only difference, then, is that one method is simpler, faster and more elegant while the other provides a built-in correction mechanism.  Am I correct?

 

If so then [Plasmatic] do we really want to sit here and declare that nothing foreign to Objectivism may taint the wisdom of Rand, or do we want to identify our common ground, identify our respective disagreements- and objectively find the best solution possible?

 

It's okay to analyze Rand critically; even posthumously her logic speaks for itself.  Furthermore, let's stop and consider how much of a difference is actually involved, here.  Popper's system of "observe, guess, criticize" differs mainly on the issue of induction. . . Which he codenamed "guess" just to be cheeky.

So how wrong is he really saying that Rand is?

Edited July 9, 2013 by Harrison Danneskjold

[Plasmatic]

Harrison said:

"If so then [Plasmatic] do we really want to sit here and declare that nothing foreign to Objectivism may taint the wisdom of Rand, or do we want to identify our common ground, identify our respective disagreements- and objectively find the best solution possible?

It's okay to analyze Rand critically; even posthumously her logic speaks for itself. Furthermore, let's stop and consider how much of a difference is actually involved, here. Popper's system of "observe, guess, criticize" differs mainly on the issue of induction. . . Which he codenamed "guess" just to be cheeky."

Nothing Ive said can be construed as me claiming that one should accept Oism uncritically and I dont know what "taint"ing "the wisdom of Rand" even would mean. I, unlike most in this discussion, know what Popper actually has said and he got almost nothing correct. He rejected every fundamental tenet that ANY valid philosphy MUST rest on.

By the way he didnt say "observe, guess and criticise", he said guess arbitrarily, then make observations/test and never stop making them because youll never be able to know truth in the sense that non skeptics mean it.

Edited July 9, 2013 by Plasmatic

[Plasmatic]

Id like anyone who thinks any of either Popper or Curi's ideas have compatibility with Oism to state what they are as succinctly as possible.

[Mikee]

I think nicholas Dykes illustrated some similarities at the end of his critique of Popper

[Plasmatic]

Yes he did but only after noting the facts that made those similarities ironic.

Edit: thought Id add it:

"This paper will therefore conclude by conjecturing that when Popper said, "in science there is no 'knowledge'... in the sense which implies finality" [OSE2 12] what he may have been after was Rand's insight that concepts are open-ended.50

For if Rand had been confronted with Einstein's rewrite of Newton; or a black swan where there had only been white ones; or the discovery of a new kind of water; she would not have said, as Popper did, that our previous knowledge had been "overthrown" or had "come to grief" or that "the belief in scientific certainty... is just wishful thinking" [OSE2 374]. Rather, she would have said simply that our knowledge had been expanded.

The description of concepts as 'open-ended' does appear to be the Philosopher's Stone which Popper sought but never found. He correctly saw that there is a problem with most people's idea of certainty, yet never quite fought his way through to an acceptable solution.

But be that as it may. Whatever one may think of Popper, or of Rand, the open-endedness of concepts certainly seems to be a more fruitful, less fraught, and more commonsensical qualification of certainty that "We never know what we are talking about." "

Edited July 9, 2013 by Plasmatic

[DonAthos]

When you describe my guess as being "good," I take that as implying that some guesses are better than others... And on what basis are we able to make this distinction?

 

I meant a good guess like worth trying, a good guess to make and consider. I didn’t mean good like true or partly true.

 

Hmmm, okay. But I'll admit that's not how I took it, and I don't believe I would tell someone that they have made a "good guess" if I only meant that it was good of them to make a guess. If I asked someone to guess at the speed of light, and they came back with 3mph, I would not say to them, "good guess!" (Apart from being ironic, perhaps, or condescending.)

 

It’s also good in the sense of being plausible to me.

 

Yes. This is closer to the sense in which I took your comment.

 

Maybe you have the answer. I don’t see any reason it’s false in the first minute of thought, which is a good start.

 

Also, the important thing is that this sense of “good” has no authority. It is a casual, loose usage, not an important epistemological pronouncement. It has no solid meaning or impact, it was just meant to communicate. It has no particular consequences or implications.

 

And here, I believe that we disagree. While I agree that there is some informality in something like "good guess," and agree that it is therefore "casual, loose," I think that there is yet important epistemological meaning (which provides the content of that which you "just meant to communicate") and has particular implications, especially related to our conversation.

I don't wish to make too much of this small use of language, but I also must observe that I think it speaks to the point I'm attempting to make, regarding how "evidence" does seem to provide support of a given hypothesis. I think relating fingerprints to a particular murder suspect makes the hypothesis that this murder suspect is guilty a better hypothesis than others -- a "good guess". And I think this is fairly described as being positive support.

 

When it comes down to it, being a “good guess” doesn’t matter. What really matters – what is decisive – is whether we have any criticisms or not. (In my view.)

 

I understand. I don't want to overreach and say that I don't think "criticism" matters here -- I'm sure that it does. But the identification of a "good guess," I believe, matters as well.

 

"good" (as opposed to a guess that it was Miss Scarlet, on the HMS Bounty, with a bronzed pineapple)

 

But I have criticisms of those guesses. Why guess it was Scarlet when you found mustard? Why guess it was a location other than where the body was found? These issues might be answerable. But the simple version where you just suspect Scarlet without giving any answer to these issues is wrong, criticized, refuted.

 

But hang on... If I were to try to describe the fingerprints as pointing to Col. Mustard (i.e. say that they are "evidence" of/for something), don't you think it likely that you would ask me something akin to "Why guess it was Mustard when you found mustard?", in order to demonstrate the unsuitability of using evidence as support... or something like that, at any rate? Because it seems to me like you're implying here that Mustard is a good guess because I found evidence consonant with that hypothesis -- because I can answer the question "Why guess Mustard?" in a way I cannot answer "Why guess Scarlet?" -- which is... precisely the point I've been making!

 

as evidence (i.e. that which provides a "positive" basis for some particular "guess" as to what has happened)

 

Evidence is the observation data we have to work with. The fingerprints are evidence. But the way to use evidence is to look at what it contradicts.

 

I do not agree that this is how we actually operate epistemologically, or use fingerprints in point of fact. I believe that detectives use fingerprints to find a "match." If a lawyer were presenting fingerprints (as "evidence") to a jury, I believe he would phrase it as indicating the presence of a particular suspect, not disqualifying the presence of others. I think his phrasing would not be accidental, but would speak to how we come to ideas and assess hypotheticals.

 

Evidence can be combined with some ideas to form a criticism and rule something out. Like, “It wouldn’t be Scarlet because the fingerprints have a special type of mustard that isn’t sold in stores, and she had no access to.” That combines the evidence (mustard found at scene) with some ideas to criticize and rule out a possibility.

 

I agree that the process you describe here has an important function in critical thinking and assessment. I just think that I disagree that such a process is the only one we employ; that we use evidence in this strictly negative fashion, as claimed.

 

To take a simpler example, if you see 3 apples on a table, that rules out the table being empty. The evidence can be used (via some thinking and ideas) to rule something out. Not only that, it rules out all numbers of apples besides 3 being on the table. So you can conclude the one remaining possibility: there are 3 apples on the table.

 

I cannot agree that this is how people work or ought to work. If I see three apples on a table, my process (insofar as I am able to describe it) is emphatically not to consider every possibility (in the world...? five plums? eleven and a half pears? twenty rodents?) and rule them out. It is the direct assessment that the three apples I see constitutes ("positive") evidence of three apples.

 

If it didn’t rule everything out -- 7 apples wasn’t ruled out for some reason -- then you absolutely better not conclude it’s 3 apples, no matter how much evidence and “support” you have. If all your evidence and support doesn’t rule out the 7 apples possibility, then what good is it? (Answer: well it’s good for ruling out 4 apples, 5 apples, etc, everything but 3 and 7. But it provides no legitimate support/authority/status/etc for 3 over 7, since it leaves them both as open possibilities.)

 

If even one other thing isn’t ruled out, that’s a really big deal. Why isn’t it ruled out? Why is option A so great when B isn’t ruled out? etc

 

This "ruling out" needs to be delved into more, perhaps. I don't think it's my primary aim, in any event, to "rule things out" when I see three apples on a table and describe what I've seen as "three apples on a table." Am I simultaneously swearing that there are not more apples on the table that I haven't yet seen (perhaps they are really small, and situated behind the three)? I don't believe that I am, though I may be surprised to find them. Could they be plastic apples, such that when I try to take a bite based on the conclusions I've drawn from my initial observation, I am bound to be disappointed? Yes, perhaps. But the character of the evidence, initially, I believe is positive: I am led to construct a "hypothesis" or an "idea" or a "guess" or however you'd like to phrase it, on the basis of the evidence that I have found. Not because I am systematically eliminating other possibilities.

 

it seems to me that I am operating more on the basis of drawing conclusions from the mustard fingerprints (in that they are presumed to point to the murderer) rather than contemplating Mrs. Peacock-as-murderer and finding some criticism of that theory.

 

Because you know the finger prints basically rule out everyone else. No one else has mustard prints.

 

How else would you even know what was “supported”? What conclusion would you draw, if your evidence didn’t rule anything out? Or if it only ruled out everything but ten things, then what? I think how much is ruled out is really the key factor. Each case is dramatically different by how much the evidence rules out.

 

I suspect that there are situations and scenarios where coming to a given answer entails ruling other things out: where it is as clear as, since I have concluded for X, not-X must be wrong, or vice-versa. But in those cases, I don't believe that the preferred operation, then, is necessarily to find every not-X false. I believe that there may yet be some method for finding X.

I think that to say that the fingerprints "basically rule out everyone else" is to concede this point fundamentally, even if you would rather describe it in a way that speaks to Popperian epistemology. I think it is a recognition that the fingerprints "solves for X" and simultaneously rules everything else out (as opposed to the process of comparing the fingerprints against every other "possibility"/hypothetical and somehow finding them wanting, until we are left with Col. Mustard).

 

Whether I'm able to explain to you "what constitutes how much of a basis for what, for all cases," to your satisfaction, or etc., does that make it improper for me to draw conclusions in the sorts of scenarios we're discussing, according to a positive approach?

 

This is a very common view. Induction and support are an unsolved problem, but one day we will solve them. That is, indeed, Rand’s view.

 

Well... I don't know that I was necessarily making claims about induction or support being unsolved, or asserting that they will be solved in the future. I was asking whether it is improper for me to use a process when I'm unable to explain/describe that process in full? I likened it to an implicit use of bio-mechanical processes, and indeed, mathematical computation, that a child employs without a whit of understanding of those processes, those computations.

 

I don’t expect it. For one thing, your examples where we do stuff (like walk) without a full explicit understanding can be explained without induction or support.

 

I wasn't arguing that walking relies on induction or support (in the sense we're discussing; walking does take a lot of support in another sense! ), and thus constitutes proof of induction or support. I was using it as an analogy to describe how a man may do something without conscious knowledge of all of the particulars of the means by which he does it.

 

We could have a partial critical understanding, largely unconscious, and walk. I don’t think the walking example really helps one side over the other.

 

Except that isn't it possible that we could have a "partial critical understanding" of how support functions, even if some portion of it remains "unconscious"? That we understand that Col. Mustard's fingerprints indicate Col. Mustard's presence, even if/when we are not able to describe that process in full?

 

Another thing is the arguments against induction and support are not like “there’s a few gaps to work out”. They are more along the lines of “here are 5 reasons it’s impossible and misconceived root and branch, which no one has any answer to”.

 

Maybe we should go into that?

 

Yes, absolutely!

 

It remains a bit of an open question for me, whether Popperian epistemology seeks to describe what we already do, or what we ought to do

 

Both. Popper says no one has ever done induction. None, ever; that is a myth. Because it’s impossible. Some people thought they did it, but they were mistaken and didn’t understand what they were actually doing. Understanding the right approach could help people do it better.

 

We haven't really gotten into induction yet, and I'd rather try to keep this from bloating further at the moment. But soon, perhaps; as soon as your next reply, if you'd like.

 

I'm trying to approach this literally, and imagine the actual scenario playing out. My daughter, when she is at the proper age, will sit down to a right triangle with smaller sides of 3 and 4, and she'll be asked to find hypotenuse x. Now, on the one hand, she could "guess" a number and then see whether it "contradicts her math knowledge" (though this would seem to me to possibly beg the question of how that "math knowledge" is acquired in the first place, if not in some positive manner)... maybe her first guess is 5. Or maybe it isn't -- maybe it is 6 or 7 or 8 or 5.1 or 5.2 or 5.3, and she rules them out, one by one, until (hopefully) she guesses 5.

 

Criticism does not mean one by one. Criticisms often rule out categories or sets of things. We can rule out the set of everything except 5 with some mathematical arguments.

 

And guessing doesn’t mean guessing at random. Do the math and work out 5 (you are allowed to get your guesses/ideas in any manner whatsoever, no problem using the Pythagorean theorem). Now 5 is one of the suggested possibilities (call it a “guess” or “idea” or whatever else you like). And if you try to say anything but 5 that’s easy to criticize. But 5 hasn’t got anything known to be wrong with it.

 

But it seems to me that saying that "all numbers apart from 5 have something wrong with it" in the given example is a roundabout way of admitting to the positive correctness of both 5 and the Pythagorean theorem generally. And my understanding of that, and other like theorems, is that they are themselves constructed on "support." The method of acknowledging that there is preexisting "knowledge" against which new claims may be compared (i.e. "criticized"), it seems to me, ignores the (positive) process by which that original knowledge was accrued.

Out of curiosity, does Popperian epistemology allow for the original construction of a geometrical proof, or the Pythagorean theorem, or similar? Because those seem to be arguments in themselves, and to rely upon evidence (as being "for" a thing), and to be "positive" in character in that they demonstrate how to come to one answer, rather than disqualifying/"criticizing" others.

 

It’s like the apples on a table example earlier. If 7 wasn’t ruled out, you’d be in trouble saying it was 5. No matter how much positive basis 5 supposedly had, it wouldn’t matter at all if 7 wasn’t ruled out. If 7 isn’t ruled out, you don’t know the answer, you have a contradiction, you better look over things again. But if 7 is ruled out, then 5 is wonderful, you’re golden, no problem. For the status of 5, everything depends on whether or not 7 (or any other number) is ruled out or not. Either 5 is the one idea we have that isn’t ruled out, or it isn’t.

 

Because 7 is ruled out (logically) at the same time 5 is reached does not mean that 5 is ruled out via ruling out 7. Indeed, I don't think we arrive at 5 by ruling out every other possibility; I think we rule out every other possibility by arriving at 5.

 

I may not be interested in considering what you dream. You may not be either. You need not consider everything you could consider. There’s a million things in life and we have limited time and attention and focus, so we have to be selective. We have to try to understand what is problematic and focus attention there.

 

Yes, agreed. And the means by which we are "selective" in this manner, I believe -- why we consider some things and not others, out of the entire range of "everything you could consider" -- are according to the positive character of evidence. Finding Col. Mustard's fingerprints at the scene of the crime leads us down a very particular path, and lends support to certain "ideas/guesses/whatevers" that others do not have, which is why we describe the former as being "good guesses" and the latter as being "arbitrary," and not worth our limited time and attention and focus.

As to Neptune, I'm afraid I must leave off for now to attend to the matters of the day. I thank you greatly for the time and courtesy you've extended in the discussion thus far, and I hope that you've been enjoying it as I have.

[Mikee]

 "if you see 3 apples on a table, that rules out the table being empty. The evidence can be used (via some thinking and ideas) to rule something out. Not only that, it rules out all numbers of apples besides 3 being on the table. So you can conclude the one remaining possibility: there are 3 apples on the table." Curi

 

 

recognizing nothing else is there does not require a systematic ruling out of the particular existents that are absent. Broad abstractions such as "all" and "anything" cover all that is substantive, omitting every particular and specific measure, and requiring only that the referents possess existence. From this we can grasp the opposite concepts that are "nothing", "none" and "non-existence" without the itemization process proposed.

Edited July 9, 2013 by Mikee

[Plasmatic]

Correct, the negative approach would undo all concepts because we we could never rely on the generality obtained via measurement omission. It follows from this approach that many who actually attempt it end up either neutralizing words with devices like scare quotes or attempt definition by negatives, which defeats the purpose of definition and measurement omission.

Edited July 9, 2013 by Plasmatic

[Harrison Danneskjold]

"Not only that, it rules out all numbers of apples besides 3 being on the table."  -curi

The process of elimination applied to numbers?  When have all the alternatives been eliminated?

[Harrison Danneskjold]

Id like anyone who thinks any of either Popper or Curi's ideas have compatibility with Oism to state what they are as succinctly as possible.

 

 I think that "observe, guess, criticize" is compatible with Objectivism (and that's what I've been referring to) so long as you put "induct" in place of "guess".  But considering that induction is necessary for any sort of thought at all, I figured it properly belongs there anyway.

And I still think that "observe, induct, criticize" is entirely compatible with Objectivism. . . Because that wouldn't really change anything, anyway.

 

However, in retrospect, I have been making a lot of broad generalizations about Popper based on minimal information (curi's comments and an hour or two of research) which led me to form untrue inferences about your comments, as well.

Several of my assertions were completely baseless and I'm sorry for that.

 

But so "observe/induct/criticize" should be compatible with Oism but I'm not sure how much of that is actually Popperian.

Edited July 9, 2013 by Harrison Danneskjold

[curi]

DonAthos,

 

If I asked someone to guess at the speed of light, and they came back with 3mph, I would not say to them, "good guess!"

 

Well that would be a bad guess. It’s totally different.  You have immediate criticism of 3mph speed of light. I didn’t have immediate criticism of mustard being the culprit.

 

I don't wish to make too much of this small use of language, but I also must observe that I think it speaks to the point I'm attempting to make, regarding how "evidence" does seem to provide support of a given hypothesis.

 

Which hypothesis does which evidence support? Does it equally support all hypotheses which it doesn’t contradict? (If so, that’s actually binary.) Or something else?

 

I think relating fingerprints to a particular murder suspect makes the hypothesis that this murder suspect is guilty a better hypothesis than others -- a "good guess". And I think this is fairly described as being positive support.

 

I understand what you’re saying, but that is one of the main epistemological mistakes Popper refuted and corrected.

 

There is no relationship between the evidence and the idea you claim it positively supports other than logical compatibility. It has exactly the same relationship with all other ideas that it doesn’t contradict.

 

You disagree. You think there is a relationship. OK, what is it?

 

Because it seems to me like you're implying here that Mustard is a good guess because I found evidence consonant with that hypothesis

 

No that is not what I mean. It’s a good guess to start with because 1) no immediate criticisms  2) no immediate non-criticized rival ideas.  if you quickly get to a situation where you only have one candidate idea, that’s a good start.

 

I do not agree that this is how we actually operate epistemologically, or use fingerprints in point of fact. I believe that detectives use fingerprints to find a "match." If a lawyer were presenting fingerprints (as "evidence") to a jury, I believe he would phrase it as indicating the presence of a particular suspect, not disqualifying the presence of others.

 

You’re treating evidence a bit strangely (from my point of view). The only good explanation of unique finger prints being there is that the guy with those fingers was there. That’s an idea I don’t have a criticism of, and which has no uncriticized rivals. So, in this way, it does indicate the particular suspect was there, because him being the prevailing idea that explains the evidence and isn’t contradicted.

 

(I’m trying to play along with the example btw. IRL i don’t think fingerprint analysis is as reliable as is commonly believed.)

 

I agree that the process you describe here has an important function in critical thinking and assessment. I just think that I disagree that such a process is the only one we employ; that we use evidence in this strictly negative fashion, as claimed.

 

I think it has to be the only one because it’s the only known one that actually could ever work at all. All the others are refuted. (For purposes of our discussion, this is pending some questions I raised above. But I thought giving my answer on this topic -- the conclusion if you agree with lots of my stuff -- would help clarify.)

 

cannot agree that this is how people work or ought to work. If I see three apples on a table, my process (insofar as I am able to describe it) is emphatically not to consider every possibility (in the world...? five plums? eleven and a half pears? twenty rodents?) and rule them out. It is the direct assessment that the three apples I see constitutes ("positive") evidence of three apples.

 

I don’t think you’re adequately taking into account how much people automate their thinking and do it lightning fast (like Rand explains). i’m largely talking about how people think unconsciously, the underlying way people can figure anything out. people’s accompanying conscious thinking is often largely irrelevant and silly, or highly incomplete (relying on the underlying unconscious analysis), or many many other possibilities. what you think you’re doing is not my focus.

 

also your statement of the critical method is not accurate. you don’t have to consider every possibility. you may consider whatever possibilities you want. (what if you choose by whim? arbitrarily? what’s to save reason? answer: criticism of your method of choosing possibilities to consider.) you CANNOT consider every possibility. if you are not interested in a possibility, ok don’t consider it. shrug. in some specific case i might think that was a mistake, but in many many cases it’s fine.

 

i think you’re somewhat wrong about what is intuitive or common sense too. the reason i think you don’t care to consider the other possibilities is you know you could rule them out if you considered them. you’re able to immediately estimate what the result of considering them would be. (you lighting fast realize the whole category is refuted. fallibly as always, but that’s fine).

 

if you thought that if you took the time to go through all the numbers and fruits you’d be able to rule out most but not all, you’d be in trouble! that’s no good. if there’s even one other possibility you don’t know how to rule out – have no criticism of – then you must not ignore it and it’s irrational to claim the evidence supports the one you chose to consider over the one you can’t rule out but arbitrarily ignore.

 

if there is something you couldn’t rule out but you never thought if it, had no inkling of it, fine, no problem, you knowledge is limited. as long as you made a reasonable effort to think, appropriate to the situation, then your ignorance is forgivable. however if you had any idea that there was any possibility you didn’t know how to rule out, then you’d better stop and consider it, not ignore it.

 

whether you could rule everything out is the crucial factor. i think this is totally intuitive. if you thought there were 3 apples, and could rule out all amounts of apples except 3 or 8, then i think you immediately know that concluding “there are 3 apples” is a big error. that 8 – that one alternative you don’t know how to rule out – deserves attention. in the real scenario you do know how to rule out 8 too so it’s ok and that is why it’s ok.

 

This "ruling out" needs to be delved into more, perhaps. I don't think it's my primary aim, in any event, to "rule things out" when I see three apples on a table and describe what I've seen as "three apples on a table."

 

But isn’t it? If you had no idea how to rule out there being 8 apples not 3, wouldn’t that be a huge problem? no matter how many positive supporting reasons you could give for claiming it’s 3, if none of your arguments rule out 8 then it’d be dumb to conclude it’s 3, wouldn’t it?

 

(my takeaway: positive arguments don’t matter, they are useless (as in this example where they make no difference), and to the extent anyone actually uses positive supporting arguments they are thinking irrationally)

 

(but often when people think they are using positive supporting arguments, or claim to, they aren’t actually.)

 

But the character of the evidence, initially, I believe is positive: I am led

 

Both Popper and Objectivism strongly disagree that you are led. Evidence does not and cannot lead you. You must lead yourself, you must have an active mind, etc

 

(There’s also the issue of: where does any given evidence lead one? Why there and not somewhere else? basically any piece of evidence contradicts some possibilities and doesn’t lead there, and does not contradict some other possibilities and equally well leads to all of them, so focussing on one in particular is arbitrary. also you have to figure this out, it’s not really leading.)

 

I was asking whether it is improper for me to use a process when I'm unable to explain/describe that process in full?

 

That is not improper. However, some processes are improper (like induction and support). And it’s harder to explain that when you don’t understand what you’re even doing. The vagueness helps partially immunize your position from criticism. But also, in any case, the more you don’t fully understand what you’re doing, the more you should consider that it might not actually have anything to do with induction or positive support!

 

Yes, absolutely! [go into refutations of induction and support]

 

OK. Can you tell me a little about your familiarity with the subject? FYI many books by inductivists concede this. it’s common knowledge. e.g. i read some bayesian stuff not too long ago and they were happy to concede it and were able to correctly state some of the unsolved problems with induction and support.

 

in the fabric of reality, david deutsch argues that actually most inductivists today are characterized by thinking that induction not working is a big problem. not by claiming to be able to actually rationally defend induction.

 

so far what i’ve run into with objectivists is they don’t seem to be familiar with any of this (they only read objectivist material maybe), and they make claims that i think are ridiculous like that peikoff solved the problem of induction in OPAR. (the word “induction” is in OPAR a total of 8 times btw... and on a quick skim through them i don’t see any explanation of the problem of induction, let alone a solution)

 

anyway, can you indicate a bit about where you’re coming from, where you stand? also are you willing to read things or do you just want to discuss?

 

also above i started raising some of the problems.

 

But it seems to me that saying that "all numbers apart from 5 have something wrong with it" in the given example is a roundabout way of admitting to the positive correctness of both 5 and the Pythagorean theorem generally.

 

i don’t think it’s roundabout. i think it’s necessary. if there wasn’t something wrong with 6, and you concluded 5, you’d be in big trouble!

Out of curiosity, does Popperian epistemology allow for the original construction of a geometrical proof, or the Pythagorean theorem, or similar? Because those seem to be arguments in themselves, and to rely upon evidence (as being "for" a thing), and to be "positive" in character in that they demonstrate how to come to one answer, rather than disqualifying/"criticizing" others.

 

We don’t have a problem with the Pythagorean theorem or any geometry “proof” or anything like that. (I don’t like the word “proof” and in the fabric of reality, david deutsch, who is a popperian, has a chapter explaining that math “proofs” are fallible. many people seem to think they are infallible.)

 

but anyway the ideas themselves are fine. lots of mathematical methods are great. use them. and then make criticisms like “all other numbers are incompatible with this method. so unless you can point out something with this method, there’s only one viable non-refuted conclusion”. (i’m being particularly explicit to be clear. so sure it sounds a little weird. it doesn’t necessarily matter if you use other terminology, even positive terminology, what really matters is the actual thought processes and how they work.)