JohnStOnge wrote:[quote\But you know what is falsifiable? THE FACT THAT HUMANS ARE CAUSING CLIMATE CHANGE WITH CARBON EMISSIONS.
That's not "falsifiable," YT. Surely you can see that. Even if we somehow stopped all anthropogenic carbon emissions tomorrow and monitored what subsequently happens with the climate we would have no way of confirming whether or not things would've happened in the same way or differently if we hadn't stopped carbon emissions.
All the science and understanding we have of how the climate has changed over the course of the Earth's life has been tested, as it is chemistry. We know what chemicals compose of the earth, and we know how they interact with each other based on different testing.
Certain things have been experimentally verified. Like it's been experimentally verified that carbon dioxide holds heat. But that is not the same as verifying that a complex planetary system that includes carbon dioxide in the atmosphere will warm at any particular rate because a certain amount of carbon dioxide is added to the atmosphere.
Falsifiability is the key to experimentation. experimentation is the verifying body of scientific principles, but you can't test something that isn't falsifiable.
YT, I don't know why you are so devoted to Popper's philosophical statement on science but it is a fallacy. Falsifiability is not the key to experimentation. Control is. Take the peanut butter illustration again. I mentioned this but I'll elaborate now. When the experiment to test for a cause and effect relationship between peanut butter and brain cancer at the selected level of confidence (say 95%), the correct way to state the results is, "There is not sufficient evidence, at α = 0.05, that consuming peanut butter causes brain cancer. It is NOT correct to say that the experiment showed that peanut butter does NOT cause brain cancer. The null hypothesis can never be inferred. It can only be rejected; in which case the alternative hypothesis (cause and effect) is accepted.
Ironically, the only way one could ever show that peanut butter does not cause brain cancer is if the reality is that it actually PREVENTS brain cancer. In that case, over a number of experiments, investigators would notice that there is a "significantly" LOWER rate of cancer among those who are given the peanut butter treatment. So then they might design experiments to show cause and effect in the other direction.
But if the truth is that peanut butter has no effect on brain cancer risk at all that can never be shown because over a large number of experiments the overwhelming preponderance of outcomes will be "no 'significant' difference." The hypothesis that peanut butter causes brain cancer can never be "falsified."
Now, it is possible to infer that if there IS an effect it is no larger than X. You can do that by putting confidence intervals around differences observed or by constructing a one sided confidence limit. But you can never infer that the effect = 0.
Whoever started this "falsifiability" fad as a means of trying not to get into debates about intelligent design really did some damage in terms of misinforming people. I can see that.
You are so full of shit
When we're talking about the environment, or anatomy, there's often more factors into the equation of what causes what to happen due to complex chemical structures in the economy. For instance, alcohol as a chemical will make one intoxicated, but the specific amount at which that is, is subjective due to a person's genetics, weight, experience etc. There is no set minimal amount of alcohol that everyone will drink in which they will get drunk. In other words, one beer may get person A drunk, but it may take many more for person B. This is why we look to statistical data when doing studies on drugs and other medicines. If 95% of people in the peanut butter experiment get brain cancer, we can say there is good reason to think there is a correlation. We'd observe what is different in the 5% and we could not rule out peanut butter as a cause until we take the people off of peanut butter and the rates stay similar.
Take this to global warming, we have good reason to think CO2 is the cause, thus we could test it. My point is, causation is not a simple as one thing effecting another, it's a lot more complex than that. Just testing one cause will often do nothing in terms of finding, but an accumulation of data is key.
If an assumption is falsifiable, there's no reason in pursuing it as a cause or a reason. We can only find truth through testing falsifiable claims.