Faith writes:
The idea was that a theory about what happened in the past like the Geo timetable and ToE are not testable,...
All theories analyze the past and make predictions about the future. In the case of automobile accidents, the laws of physics are used to analyze the past, for instance, measuring skid marks and figuring out how fast the car was going based on its weight and the coefficient of friction of its tires with the ground. In the case of satellites, the laws of physics are used to predict the future, for instance to determine the necessary velocity vector to place a satellite in a stable orbit.
But notice that the laws of physics (laws are just theories, by the way) are not about the past or the future. They're generalized principles that can be applied to analyze the past or predict the future. Evolution is the same as physics in this regard in being generalized principles. Simply stated, evolution is descent with modifification and natural selection. These principles in and of themselves say nothing about the past or the future, yet they enable us to analyze the past, just like the laws of physics, and they enable us to make predictions about the future, just like the laws of physics.
Please let me know if my posts are too long or too difficult to understand or if something else is getting in the way, because I've explained this before in earlier posts, though not in the exact same words, and the point doesn't seem to be getting across.
Once comment about the past. Everything we're aware of takes place in the past. Nothing ever happens in the present. By the time we're aware of it, the event is in the past. If you look at something about a foot away, that light takes a nanosecond to reach your eye. The chemical reactions in your retina, the signals along the optic nerve to the brain, the interpretation by the brain itself, take far longer, probably in the area of an additional 100 million nanoseconds (.1 seconds). By the time you're aware of an event, it is already in the past. This may seem like a niggling point to you, but it highlights a key issue: if being too far in the past is an impediment to applying the scientific process, how far in the past is too far?
It's a rhetorical question you're not expected to answer. Its only purpose is to make clear there can be no line of demarcation between events young enough to analyze and events too old to analyze. Evidence, no matter how old, is still evidence, and evidence can be interpreted and analyzed. The more sparse the evidence, the more damaged or decayed the evidence, the more complex the analysis must be in order to tease out the implications. You can never conclude that something was too far in the past to study. That can never be a valid reason for not studying something. The only valid reason for inability to study something is the absence of evidence. The age of the evidence will only affect its quality and usefulness. Mere age by itself can can never preclude the possibility of study.
... that certain observations that support those ideas may be testable but in fact you can't test the theories themselves and can never verify {edit: meant "falsify"} them.
But of course we can test them. Let me provide an example, one you've already seen before. Evolution predicts that the order of the fossils we find in the geologic layers should be increasingly different from modern forms with increasing depth. We can test this prediction by examining fossils from different geological layers. When we do this we find that, just as evolution predicts, the fossils are increasingly different from modern forms with increasing depth. Thus, evolution has made a successful prediction.
This same test of the prediction could have falsified the theory of evolution, which is why the theory of evolution is considered falsifiable. Had we found that the geological layers did not contain the fossil progression predicted by the theory of evolution, this would be a falsification. Failing such a fundamental test would call the theory of evolution into serious question.
You list evolution along with "Relativity, gravity, wave/particle duality of light, quantum theory, radioactive decay, Maxwell's laws, cosmological expansion," but what I would point out about these things is that they are not about historical events but about ongoing physical facts that are exactly what CAN be tested.
Theories are general principles, not lists of historical events. Like all the scientific fields I listed, evolution consists of general principles, not a list of what has occurred throughout evolutionary history. The laws of physics can tell us when there were lunar eclipses in the first century AD, but the laws of physics are things like F=ma and F=Gm
1m
2/r
2, not lists of times of astronomical events. In the same way, evolution is the principles of descent with modification and natural selection, not a list of which dinosaurs lived in which eras. The theory of evolution has played a key role in helping us figure out the evolutionary past, and that's why its a valuable interpretive framework (which, by the way, is what a theory is).
My earlier example of the automobile accident with the skid marks is a good example. Let us say that the automobile accident happened in Pompei, and just by sheer bad luck Mount Vesuvius decided to pick just that moment to erupt and bury the entire accident scene in miles of lava and volcanic ash. Thousands of years from now archeologists unearth Pompei (again) and uncover the accident scene with the skid marks. Just for the heck of it, the grad students on the archeological team decide to measure the skid marks to figure out how fast the automobile was going at the time of the accident, and discovering that he was over the limit (preserved on a sign nearby) by 20 km/hour, they issue the offenders skeleton (also preserved) a belated speeding ticket.
The point of this story is to show how this auto accident that happened thousands of years in the past for these archeologists is still open to analysis because the relevant evidence has been preserved. Your division of the past into two eras, one era that's recent enough to analyze, and another era that's too long ago to analyze, does not correspond to the real world. The key factor governing whether some past event is open to analysis is how much evidence has been preserved, not how long ago it happened.
Here's another example illustrating that your approach of using age to determine whether something is open to analysis or not is flawed, but this time in the other direction. In particle physics, things happen fast. Real fast. Huge devices accelerate particles up near the speed of light and then have them collide with one another to produce a spray of subatomic particles. Often these particles have extremely short lifetimes. In some cases if you wait even a nanosecond to measure the produced particles they'll already be gone. This is yet another counterexample to your view of things, in which a nanosecond ago is the virtual present and therefore open to study. But in this case a nanosecond ago is too far in the past. Your criteria of recent things being open to study again fails, this time in the opposite direction.
It may not be possible to arrive at proof in many senses with those theories too,...
You've again used the word proof, and I'm still not certain which definition of the word you intend. Do you mean "verified beyond doubt", or do you mean "supported by persuasive evidence"?
... but they are in principle replicable as their content is ongoing, always available for observation.
Historical events on the other hand are by definition past and cannot be replicated.
You're correct, they can't be replicated, but I think everyone is puzzled why you think replication is necessary. An automobile accident that happened today can be analyzed by archeologists thousands of years in the future to figure out the speed of the automobile without replicating the accident. The operative factor is the availability of evidence, not the practicality of reenacting the entire event, and not how long ago in the past it happened.
The whole approach to proving past events...
There's that word "prove" again. Could you at some point clarify how you're thinking of the word? If by "prove" you mean "verify beyond doubt", then we agree with you, past events can't be proved. Of course, all events happen in the past, so that would mean no events can be proved. It's far more practical, and it is in fact the practice within science when speaking casually, to define the word "proof" to mean "persuasively supported by evidence."
--Percy
This message has been edited by Percy, 04-11-2005 01:42 PM
™ Version 4.2