Talk:Logical fallacy

Manufacturing facts from a theory

"...is the source of many controversies in modern astronomy, some of them strikingly bitter"

Umm...really? Such as?

In any case, this is pretty weak. The entry's written broadly enough to cover basically all scientific inference. Strictly speaking invalid from a deductive point of view, but still. Tsumetai 10:04, 6 March 2007 (EST)

You asked, "Such as?" And I'll give you some examples. Start with the Oort cloud, the so-called "cloud of comets" where comets reside until something knocks them into the near solar system. Why do we hear of the Oort cloud?

1. Comets lose their brilliance and become invisible to the naked eye after 100 near-sun passes.
2. A large number of comets are "short period" comets that are still brilliant. (Example: Comet Halley, which makes a pass ever 76 years, and heralded the birth and death of Samuel L. Clemens)
3. Therefore, the solar system can be no older than 7600 years.
4. But this is "impossible" since radiometric dating "says" that the earth itself is 4.6 billion years old.
5. Therefore, comets reside in a "cloud of comets" in the outer reaches of the solar system, and do not actually start making long, thin orbits of the sun until some passing asteroid perturbs their orbits and knocks them "down."

The Oort cloud is, therefore, a fact manufactured to support the theory of an age for the solar system measurable in billions of years.

Back to Sir Isaac Newton: He made a definite prediction about the strength of gravity and the orbits of the planets. But one planet did not cooperate: Mercury. When it reaches perihelion, it deviates significantly from a path that a strict Newtonian calculation would predict.

1. Newton's Law of Gravitation predicts that Mercury should pass in a certain way.
2. Mercury deviates from that path.
3. Therefore the solar system contains another planet (named "Vulcan") inside of Mercury.

Centuries later, Albert Einstein issued a correction that explained Mercury's orbit without resort to any "new innermost planet."

The inferred planet Vulcan was, therefore, a "fact" manufactured to support a theory that was clearly inadequate. Astronomers insisted that this planet must exist, even though no man had ever seen it, until Einstein finally solved the riddle.

That's not such a bitter example. The bitter examples today involve anomalies in all the planets--things that no theory of "spontaneous generation" of planets can possibly explain. So in each and every case, astronomers manufacture an historical "fact"--that a large asteroid--not in evidence today--collided with the planet and brought about the anomaly.

This sort of thing stands in violation of Occam's razor, which states that one must not multiply unsupported inferences without sufficient reason.--TerryH 12:55, 6 March 2007 (EST)

Twoproblems

Newton's laws are not an example of an overgeneralization, but an example of an inferrence that didn't generalize as much and so shouldn't be in that set. The argument from authority section also misses that experts are only relevant if they have access to more resources of some form not accesible to the people in question (although strictly speaking, from a formal logical perspective, such an argument is never valid- but we seem to be not distinguishing between formal and informal logic here). JoshuaZ 10:36, 6 March 2007 (EST)