So I Wrote A Book On The Scientific Method

....and if he doesn't mention prediction at all, in any context?Look, I agree that the ability to make a prediction is pretty core to the whole scientific method. But at what stage? It seems to me that it's at the testing of hypothesis stage that prediction is mostly used. 'I've observed that there are only white swans everywhere I go, I therefore predict that there are only white swans.'Simple stuff.But when you get to the theory stage the prediction issue is mostly over. What's happening there is that you have a whole pile of observations and tested hypotheses which have become established fact. Established theories - rather than hypothese - never get disproved, at best they get modified. The idea that a theory is at risk by a prediction is true, but not actually remotely likely.

That's true, but it misses the point. The theory makes the same statement as the hypothesis. So if the hypothesis is testable, so is the theory.Edited by NoNukes, : No reason given.

A real theory never has one simple, white swan hypothesis and it's never actually at risk - it's gone beyond that. We pretend that it's disprovable - we like to say tentative - but actually to become a theory - not a hypothesis - it has become the nearest thing we have to a fact that we can get because it explains a group of facts. So when we talk about theories, by-and-large the predictive element is secondary and assumed; it's its explanatory value we actually use.

In terms of actual practical usefulness quite the opposite is true.The application of any theory will involve the expectation that future events will conform to the results the theory predicts.If you build a satellite and put it into orbit (for example) you do so based on the expectation that the results predicted by the General Theory of relativity will be correct.The same is true for any other application of any other scientific theory is it not?

Yup. But it's not a prediction it's a statement of fact. If I say that the pub is open now, am I really making a prediction or am I just repeating a known fact?The prediction came long ago when we knew less and had to hypothesise, now we know it to be true and no longer have to guess.Sure, a theory is open to disproof, it has to be, but no-one really doubts that it will be.

You seem to be saying that it takes more fuel to fly from A to B than to taxi to the terminal. That's true - but it still takes fuel for both.

Passed me by I'm afraid.

Looks like this no testing idea of yours dies hard. This part, at least is true. But a theory is just the underlying hypotheses (at least the hypos and their modifications that survive testing) elevated by the scientific method to acceptance. Every test of the underlying hypotheses, is a test of the theory as well.And because all scientific theories are tentative, they are all subject to continued testing. People are still making attempts to identify gravity waves as a further test and verification of General Relativity.Edited by NoNukes, : No reason given.

I think ringo is addressing your use of the word "mostly over" when referring to the role of testing. Unless you really mean "completely over", then you've basically acknowledged the role of testing in the later stages of formulating a theory.

All theories are the best explanation of the facts we have and all are 'theoretically' falsifiable. What I'm trying to say is that in practice, anything that has been elevated to the grand state of a real theory will never be found to be totally false; they may be modified by new information and some parts may be shown to be incomplete and even wrong, but the theory will stand.We tend to use the word theory to include single ideas that are nothing more than hypothesises or are very big ideas that are waiting experimental confirmation. They are a different category of theory altogether and their fate is binary.

For example, the change from "atoms are indivisible" to "atoms aren't indivisible" required the modification of only a single word.

Doesn't that pretty much end the discussion as to whether all theories make testable predictions? I've shown where Popper acknowledges that to be the case. What else is needed?It is not a mere pedantic principle that all theories are falsifiable. It is fundamental that all scientific theories are tentative. Galilean relativity was useful for over three hundred years before anyone even suspected there was a problem with it. It is, in fact, the nature of scientific inquiry that we formulate our theories using inductive reasoning because it is not within our power to investigate everything in all realms of possibility.If it will help you to accept the central issue of tentativeness, I'd be happy to quote Popper describing exactly that issue. Or you can look on pages 3 and 4 of the book you cited and see what I would be cutting and pasting.Or you can continue on this Quixotic quest of yours for the final authority on all things scientific. I think I saw it sitting in a cabinet next to the Holy Grail and the Spear of Destiny.Edited by NoNukes, : Fix quote box

Of course all theories are thought to be tentative and of course the scientific method requires prediction and of course Popper talks of prediction. But never mind, the nuances of all this seems to escape you.

Hi Tangle,I am not completely sure what the controversy is all about, I really haven't disagreed with anything you have said or got the impression that you don't understand the scientific method. But I'll throw my two cents worth in anyway. I agree with this statement. As written, theories are pretty much descriptive. They provide a framework from which to develop hypotheses. A strong theory is basically assumed to be true and has already undergone rigorous testing - that is what has allowed it to be considered a theory.However, the thing that makes a good theory is its ability to provide a framework that makes useful predictions. "The sky is blue." may very well be a true statement and may very well be thoroughly tested, but it doesn't provide any type of framework with which to build new hypotheses and make worthwhile predictions. It would simply be a statement of fact rather than a theory.You are right, the hypothesis is where the rubber meets the road, where the work of verification and testing occurs; not the theory. Theories typically are too broad and generalized to provide a testable statement, so it is reduced to a simpler, more focused, more directed statement - the hypothesis - in the form of IF (X) THEN (Y).The problem is that hypotheses are rarely specifically spelled out and it can appear that the theory is no longer being tested and is simply assumed true. However, if you are working within a particular theoretical framework, then you ARE making predictions and testing those predictions even though they may not be blatantly obvious.An example from my own research. I am characterizing a soil borne fungus (Rhizoctonia) that causes root rot on sugar beet, dry been and corn (there are others too but those are the hosts I am focused on). As part of this characterization, I will be doing a screen for host resistance. When I write this up I will say something to the effect of "I screened each isolate for resistance on nine host varieties by rating root damage on a scale of 1 to 4." For that type of statement it may not be clear what the hypothesis is and what theory I am actually testing.However, if I wrote it like this: "IF a host variety is resistant to Rhizoctonia infection THEN it will suffer statistically significant less root damage than the control variety" and alternately "IF a host variety is not resistant THEN it will suffer an equal or greater amount of damage than the control variety." it becomes a bit clearer as to what theory I am actually testing and what the predictions are. This screen is actually working within the framework of the Germ Theory of Disease and will provide a direct test for the theory, even though it is not immediately evident to be so. In fact, part of the screen will have to be back test the infected host to confirm the organism that caused the disease was the organism being screened (Koch's Postulates). I wouldn't say that the prediction issue is "over," just serves a different role. Rather than making predictions in order to verify the hypothesis, the predictions become more applied, more useful; they allow us to work within that framework without having to justify every step. For example, I don't have to justify that it is a microorganism causing the root rot disease rather than spontaneous generation every time I do a disease screen. The Germ Theory of Disease provides the prediction I need to justify the disease screen. Agreed. Usually good, strong theories are incomplete at worst and need additional modifications to be able to better explain ALL evidence. It would be almost unimaginable to think that some discovery will come along and completely over turn any established theory. But the possibility remains that it could.HBD

I believe Tangle understands the scientific method. Nonetheless, he does not seem to think that the use of the scientific method requires that theories, like hypotheses be testable, and he does not seem to recognize statements in his references or those of others that make the conr claim. At first I thought he was just looking for references, but it turned out that just knowing that testing predictions was part of the scientific was not enough to end the discussion.At least that seemed to be his original position. Now he claims that the testability of theories is some pedantic point which can generally be ignored because once a theory is already established, then its predictions are now facts. He illustrated this position with the "theory" of the open bar.Both of those positions are wrong. But the latter position seems to me at least to be simply a matter of avoiding the fact that all theories make "predictions"; formulations which do not cannot achieve the status of accepted theory.There is no controversy. Hope this helps.Edited by NoNukes, : No reason given.Edited by NoNukes, : No reason given.Edited by NoNukes, : No reason given.