RoR: Forum

A famous test relevant to this topic is the Wason selection task.

Post 1

Friday, March 29, 2013 - 8:27am

34 Points and no arguments. We agee.

Merlin, thanks! I failed. I chose 8 and Red, accepting common language if-then even though I once earned an A in symbolic logic. In that context, perhaps here arguing philosophy, I adhere to the material rules, but I fell into the natural language (English language) meaning because it is, well, "natural." And I knew your intention. I looked for the falsifiability test, but did not find it in the choices offered. All that mattered was that even numbers have red backs.

I found the Wikipedia discussion informative. It is interesting that most people (more than a paltry 10%) could pick the right choices when a social rule was the subject. One of the hides nailed to my door is from Scientific American. They ran an article about people's fallacious notions about physical events, and the article included a blunder. This is not easy. We do not do well at abstract reasoning, which is why it is - and needs to be - a formal study.

(My instructor for symbolic logic was a doctoral candidate in philosophy from the University of Michigan. She was a rationalist, absolutely sure that A is A, but unwilling to bet that the sun would rise in the east tomorrow.)

Post 2

Sunday, March 31, 2013 - 6:01pm

Joe,

You did a great job at articulating the 'scientific method' (as informed by Popper's hypothetico-deductive method of falsificationism). A relevant debate, however, would be whether the scientific method is truly scientific in the first place -- whether hypothetico-deductivism is a "proper" method for science, or for acruing personal knowledge in general. Harriman, in The Logical Leap, decries the hypothetico-deductive enterprise, arguing that something more than that is needed to ground a growing knowledge base.

You outlined a few of the things that one can do when presented with a subset of numbers (4, 6, 8, 12, and 14) and given the chance to ask yes/no questions about the larger set of which it is a part. The point of that is well taken but I'd argue that that was only a few of the things that one can do -- leaving certain other, relevant things off of the table. For instance, if you presented me with those 5 numbers, I could either ask about the outcomes only (Does the superset contain the number 16? Does it contain 3? etc.) or I could take a step back and ask more basic yes/no questions about the very process of number generation for this set.

For instance, I could ask about a possible algorithm used in order to generate the numbers of the set. For example, let's say I asked these 2 questions:

1) Is an equation for generating or filtering the numbers: [all x such that x(n) = x(n-1) + 2]?
2) Is that the only equation?

Now, if I'm lucky -- or if I continue asking until I do get lucky -- then I did not arrive at a hypothesis that has the possibility of being falsified. Instead, I arrived at a full or real knowledge of the set (including numbers never seen or even imagined, let alone experimented on to determine whether they are "in" or "out"). Admittedly, it's a little harder uncovering the nature of something. It is easier to ask about an outcome or result. I'm not saying this method is easier than hypothetico-deductivism, just better. By bypassing the outcomes (e.g., 16, 3, etc.) and going straight to the source of the numbers -- the very "nature" of the set -- I "pass" the test in just 2 questions (if I'm really lucky). Also, by asking both questions, I preclude the possibility of being wrong.

This seems to me to be better than asking endless questions about endless possibilities. Using Popper's method, you never "pass" the test, because you never exhaust all that there is to be imagined.

Ed

Post 3

Sunday, April 14, 2013 - 12:26am

Thanks for the link, Merlin!

Ed, I was not suggesting that Popper's approach to science, or the hypothetico-deductive approach is correct/sufficient. I think the reason Popper's approach is so popular is because he hit upon an important truth. The inability to disconfirm a theory suggests the theory is disconnected from reality. A proper theory, properly grounded in empirical facts, and thus dependent on those facts, would be refuted if those facts changed. The theories he argued against were not dependent on any facts, and were thus impervious to all.

Falsifiability is not enough, though. It's not a real foundation of knowledge. Attempts to take one interesting insight and turn it into an entire epistemological system can't go anywhere useful.

But going to your approach to the number game, you aren't following the rules. But in this case, the rules are important because they make the game into a proper analogy. In reality, you can't go an ask nature what the equation is, or why it does the things that it does. Instead, you have to look for evidence and experiment. And that's where you have to look for disconfirming evidence. But this is the nature of induction. You don't get to start with the equations. You have to formulate them by gathering data, integrating it, speculating on causes, and testing for them (and testing against them).

Trying to ask what the equation is just attempts to avoid the messy issues of induction and turn it into deduction. You attempt to avoid Popper's constant uncertainty by choosing a method (deduction) that gives you absolute certain (or in your words "full or real knowledge). But both avoid the need for induction.

Post 4

Monday, April 15, 2013 - 4:08am

Popper offered logical negativism in response to the logical positivists, hardly something I would call insightful.

Post 5

Monday, April 15, 2013 - 10:34am

Michael Philip, you need to understand Popper's context. It is easy to pass off his work, as you did. Your quip had some merit: why did no one think of this in 400 years? Well, okay, given the existence of water-powered mechanisms such as musical instruments and door-openers in Roman Egypt of 100 BC- 100 AD, why did it take 1700 years to invent the steam engine? The only answer is: because it did. So, too, with Popper. Robert Boyle's Skeptical Chymist opened with an explanation of why the methods of experiments must be publicized, rather than being kept secret, with only the result being shown. It just took another couple of hundred years for someone else to fill in another of the missing pieces.

Moreover, Popper targeted specifically Freudian psychology, which was touted as scientific, but which cannot be falsified because its claims cannot be tested. Freud (like astrology) offered great explanations. Explanatory power is not sufficient to validate a theory.

As taught in schools today, we begin with general science courses that are not rigorous and then follow those with biology and chemistry before getting physics. That is in high school, which few enough benefit from, and hardly any choose physics as a liberal arts elective in college. But physics is the touchstone science. It is there that the process of hypothesis, deduction, and then of induction leads to testable claims.

Social science follows this model. As I have shown elsewhere, social sciences are actually more true to the scientific method than is physics as it is taught in school.

Still, as Joseph replied to Ed, the fact remains that is impossible to just "ask Nature for the equation" because the general law comes from organizing the data gathered by experiment.

This is a chicken-and-egg problem because really in science, we reason through deductively some testable hypothesis before attempting the inductive tests, but, that hypothesis can only come in the first place from the experience of the observer. You cannot have one without the other.

Post 6

Monday, April 15, 2013 - 1:43pm

I don't have much respect for Popper to begin with and I'm amazed people still flail falsification around.Popper does not correctly identify the principle that distinguishes contentless verbiage like Freudian psychology or Creationism from real scientific theories like Newtonian gravity or Evolution. His views on induction, meanings and concepts are not good. The only thing of value which Popperianism ostensibly embraces is that if you don't know the answer, you should try to find out but that's not saying much.

"If I were allowed to give some sort of modernized reconstruction, or re-interpretation of Kant ... I should say that Kant showed that the metaphysical principle of reasonableness or self-evidence does not lead unambiguously to one and only one result, to one and only one theory. Rather, it is always possible to argue, with similar apparent reasonableness, in favour of a number of different theories, and even of opposite theories."

-- Karl R. Popper, "What is Dialectic?," Mind, Vol. 49, No. 196, pp. 403-426, 1940.

When you are a rationalist who thinks you should be able to deduce rice pudding from "I think therefore I am", it is not surprising that you can deduce opposite conclusions from the content of your brain. It's also not surprising that you can confuse induction with enumeration.

Personally speaking, Mario Bunge is a far better contemporary than Popper is.
(Edited by Michael Philip on 4/15, 2:03pm)
(Edited by Michael Philip on 4/15, 3:19pm)

Post 7

Wednesday, April 17, 2013 - 11:44am

Michael, thanks for the lead. Mario Bunge is alive and well. The UT Library has 77 citations in English and Spanish. Wikipedia describes him as a physicist and philosopher.

"He was professor of theoretical physics and philosophy, 1956�1966, first at La Plata then at University of Buenos Aires. He is currently the Frothingham Professor of Logics and Metaphysics at McGill University in Montreal, where he has been since 1966.

Mario Bunge has been distinguished with sixteen honorary doctorates and four honorary professorships by universities from both the Americas and Europe."

Causality and modern science 1979
Causality; the place of the causal principle in modern science. 1959
The critical approach to science and philosophy 1964
Philosophy in crisis : the need for reconstruction 2001
Philosophy of physics. 1973
Philosophy of psychology 1987
Philosophy of science and technology 1985
Philosophy of science. Vol. 2, From explanation to justification 1998
Political philosophy : fact, fiction and vision 2009
Problems in the foundations of physics. 1971
Scientific materialism 1981
Scientific research 1967
Sense and reference 1974
Semantics 1974

Is any one of these in particular to be recommended?

Post 8

Friday, April 26, 2013 - 12:15pm

Nice job, stimulating post.

Post 9

Friday, April 26, 2013 - 12:20pm

Back when I was teaching in his region, Bunge and I had a few exchanges. I, as a naturalist, could not accept his reductionist materialism. He found, if I recall right, all value judgments empty of content.
(Edited by Machan on 4/26, 12:21pm)

(Edited by Machan on 4/26, 12:22pm)

(Edited by Machan on 4/26, 12:23pm)

Post 10

Sunday, May 12, 2013 - 12:31pm

Point taken, Joe.

However, I didn't say that my change to your analogy would transfer over to real life -- so that you could just go and ask Nature what the equation is and then Nature would politely answer you. Instead, I said that things are not that easy (i.e., that it is hard to get at the nature of something). In short, I took your analogy and added a metaphor, if you will.

You're right that induction is required (that thing to which I was referring which makes it "hard" and not "easy" to discover the nature of something), but a bit off the mark in saying I was acting so as to avoid -- even to unintentionally avoid -- the process of induction.

Ed

Post 11

Tuesday, July 16, 2013 - 9:05am