Karl Popper and Reliable Science

I will admit some criticize Popper for being so dead set in favor of deductive methods and against inductive scientific methods, but his point is deductive science that tests and verifies with reliable evidence is the essential test of reliable and credible science. Falsifiability (testability) is a big and important word in the world of Popper Science. Seems Einstein agreed about experiments and evidence as controlling and dispositive.

The writings and teachings and ruminations of Popper have had a big effect on scientific endeavors and the philosophy of science and scientific epistimology.

The Popper template was a key to the Supreme Court Opinion by Blackmun in Daubert v Merrill Dow (1993) that established new and better guidance for judges on how to evaluate the admissibility of scientific testimony and evidence.

Here is a short essay on how the tautology becomes the “consensus” that shows the unreliability of the catastrophic warming/climate change/modeling crowd.

If every bit of evidence, even refutation is made into part of the proof of the thesis, it is circular tautological argumentation and not science at all.

if the theory can’t be refuted by evidence that is contrary to the premise then we are in a mythical world, aren’t we, the world of cargo cult science, where evidence is irrelevant.

About these ads

11 responses to “Karl Popper and Reliable Science

  1. As Francis Bacon put it in the early 17th century, “whatever any art fails to attain, they ever set it down upon the authority of that art itself as impossible of attainment.”

  2. Popper was full of it. His notion was that knowing the truth is impossible. He held that all that one can know is that which is false. By that self same principle, one cannot know that it is true that it was found false. The notion becomes self contradictory. That is much worse than a mere tautology, it becomes a snake eating its tail for lunch until it vanishes.

    You can’t know that it is true that a tested thing is false without relying on a principle that contradictions cannot exist. Meaning that a thing cannot both be and not be the same way at the same time. The interesting thing is that the principle of non contradiction is inductively derived. It is result of considering the totality of experience in full context. Popper relied on induction as he was attempting to invalidate it and proceeded to eat his own tail for lunch.

    • Popper was full of it? Really?

      Get control of yourself. Whatever was his weakness in rejecting inductive science, it was based on how inductive stuff is too often BS.

      Science depends on evidence that is reliable and reproducible and validated.

      What’s you version, some kind of soft and friendly agreement blanket?

      John Dale Dunn MD JD Consultant Emergency Services/Peer Review Civilian Faculty, Emergency Medicine Residency Carl R. Darnall Army Med Center Fort Hood, Texas Medical Officer, Sheriff Bobby Grubbs Brown County, Texas 325 784 6697 (h) 642 5073 (c)

      • I will ignore the insult. Science in particular depends upon a proper process of induction. Even the discovery of the scientific method was done by induction and is, in fact, a specific instance if induction in action. I offer the following as a sketch of a proper process of induction that does lead to valid and reliable knowledge.

        The fact of the matter is that all knowledge is contextual. However, for the contextual knowledge to be held to be true, no chunk of knowledge can contradict nor be contradicted by any other chunk of knowledge. This state of non contradiction must be built from the acquisition of the second chunk and be maintained by careful continued identification of non contradiction and by logical regression in the case of mistakes of identification. The process of logical regression traces the logical hierarchy from the general to the specific to validate that it is without contradiction. This is to be done for all chunks of knowledge.

        You see, induction and reduction are jointly necessary. Until their work is done, deduction has no valid part to play except by artificial restriction by definitions. For example you deduce that a particular male known as a bachelor is not married because not bachelor is defined as a male who is not married. It is believed that all you need to do is analyze the definitions.

        Yet when you are not so restricted by definition it is held that you are limited to counting specific instances. For example, all swans are white until you encounter a black swan. Then your original proposition is proven totally false. Not so. It was demonstrated to be incompletely inclusive and that whiteness was not an essential property of a swan in the same way that the colors of a black cat and a white cat are not an essential part of being a cat.

        What has happened, your context of knowledge has expanded along with the range of the necessary non-contradiction. Your identification of your concepts both expand and become more precise. White swans remain swans even in face of countless black swans.

        The error was in your original statement that ALL swans are white. You stated more than you knew, All you knew was that all the swans you have seen thus far are white. That should have been your in context non-contradictory chunk of knowledge. Induction works if and only if all propositions are restricted to what is known in context without extrapolation.

        The underlying error is known as the analytical synthetic dichotomy. Where analytical deals only with analysis of acquired definitions and is thought to be always the known to be true. synthetic deals with acquired assertions and observations and may be falsified by the next observation. There are two issues here. The definitions used by the analytical part are either arbitrary definitions not relating to reality or they are derived using a process of induction. The second issue is the insistence that your arbitrary assertion is the proper starting point of the synthetic part. It is not, if it goes beyond what is actually been shown to be non-contradictory in context.

        Ultimately all knowledge is acquired inductively and can be known to be true in context. With the addition of new knowledge, the context expands along with the range of non-contradiction. As evidence note the content, form, and function of a 20,000 BC city versus New York City. The difference is attributable to the knowledge acquired over the 20,000 years.

        Popper was full of it. QED!

  3. All I have to say is Schrodinger’s Cat.

    • Are you are so afraid of saying you don’t know something that you are willing to say we can’t know anything. All based upon a grossly misinterpreted thought experiment?

      I hold that one can know something and that the nature the existence of modern technological civilization compared to the nature of the existence of a 20,000 BC hunting and gathering tribe is proof of that fact. I am more than willing to say there are things I don’t know and perhaps even cannot know. However, that has absolutely no impact upon what I do know.

      • Westchester Bill

        You err terribly by overlooking quantitative statements and engineering. When GPS satellites had to take account of relativity, the validity of relativity was established for me.

        • Nice point. How do you adjust for relativity, if you don’t mind my ignorance?

          John Dale Dunn MD JD Consultant Emergency Services/Peer Review Civilian Faculty, Emergency Medicine Residency Carl R. Darnall Army Med Center Fort Hood, Texas Medical Officer, Sheriff Bobby Grubbs Brown County, Texas 325 784 6697 (h) 642 5073 (c)

        • Your comment is a total non-sequitur. It does not follow from anything I have written and relativity has nothing to do with the Schrodinger’s Cat in the Box Thought Experiment. The thought experiment was related to the strange interpretation of quantum mechanics in which things come into existence only when observed – with observation left largely undefined. Some magical process called “observation” caused the wave equation to collapse into a real state.

          I believe Schrodinger’s point was that the cat existed in either a live or dead state but that you could not know which state until you opened the box and looked. That you don’t know something simply means you can’t properly assert anything but that you don’t know it. The cat could be alive or it could be dead but you don’t know. That in no way means it is both alive and dead and the state is determined only when you look. It is only your knowledge of the state that is determined when you look.

          On the other hand, if only the things you know exist, then you always do know everything. Which is even more absurd than saying that the cat in the closed box is both dead and alive unless you look.

          Further, the foundation behind quantitative statements and engineering is the acquisition of a vast amount of actual knowledge that is very dependent upon the proper use of induction by someone at sometime. Without that knowledge, you can’t even start such processes except by guess and accident. Even then, you don’t have a clue what happened, why, and how. It just as well be attributed to gods or demons and was for many hundreds of thousands of years of human existence.

          • The alternative is that both theories are inherently flawed but still useful as models. The same is true for Newtonian physics and Einsteinian relativity. They are not reality but models that are useful for describing reality when kept to their proper contexts. Light is not both a particle and a wave but is sometimes best described as a particle and sometimes best described as a wave. Selecting which model best suits your purpose doesn’t change reality. If I want to show someone the relationship of all the bodies within the solar system, I’ll place the sun at the center because that’s easiest to understand. On the other hand, if I want to predict which objects will be visible from specific places on Earth, I’ll place Earth at the center. Properly constructed, both models will be equally accurate for predictions. Neither can be said to be right or wrong so long as they serve the purpose for which they were created. Quantum mechanics seems especially prone to this problem of journalists and amateurs confusing useful metaphors and models with reality.

            Why anyone would want to limit the tools available to further human understanding is beyond me. I consider myself an empiricist but that doesn’t mean I don’t believe in China because I haven’t been there. There are many things about the universe that we may never truly understand. Due to chaotic systems, poor signal to noise ratios, or even something as mundane as lack of funding for proper investigation, Inductive reasoning is often the only thing we have to make a best guess. My only objection is that qualifying terms are applied such as “this is what we think based on what we currently know.”

            Phlogiston theory was fantastic science for its day. An alchemist, against the current belief structure, managed to accurately surmise the relationship between burning, rusting, and breathing. He got the direction the particle was moving backwards, but the model still worked and allowed progress to be made for hundreds of years until measuring equipment advanced enough to realize the mass change was going the wrong direction. Inductive reasoning based on empirical observation bridged a gap in available knowledge and gave deductive reasoning a starting point from which to assist in the discovery of oxygen. Likewise, regardless of how the Great Climate Change Debate pans out, our understanding of the myriad interrelated systems that make up Earth’s climate will be improved.
            To flatly denounce inductive reasoning is like denouncing flathead screwdrivers because Phillips are better. Raising up one historical scientist, no matter how great, above others paves the way to argument from authority which is just as bad a logical fallacy than tautology.

  4. Mr. Griffith is correct: There is no “absolute value” difference between “true” and “false”. One is merely the negative of the other. And let me nip this in the bud BEFORE someone erroneously comments: NO, there are NOT “more possible false” assertions than truths. It has long been mathematically proven…but it should be obvious.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s