Comment It should be MODELS of nature. (Score 1) 729
There seems to be a lot of confusion about what is really meant by "laws" of nature; that is apparent in the article and some of the comments here.
Try substituting the word "model" for "law" in the texts. There is nothing absolute about the models as they are described; they are just the best models we have been able to describe so far. There may be better models waiting to be discovered.
Consider Newton's model of gravity. Since it's a model, how would we go about deciding if the model is wrong? Well, you might propose that the model does not hold for, say, bananas. We can then go off and conduct an experiment where we drop bananas in vacuum chambers and measure if the time to impact the ground is the one predicted by the model. If it's not, the model isn't correct.
As we know, it turns out that Newton's model doesn't give the full picture - once we get into extremely dense objects (black holes) or objects moving at high speed (i.e. near the speed of light) - so better models are introduced.
The basis of science, then, is to describe models that fit with the phenomena we can observe. The models need to be testable (I can construct an experiment to test my model), reproducible (you can do it as well), and falsifiable (we can conceive of an experiment that would prove the model wrong).
The last point is crucial. Otherwise, we could end up with a model of the universe like: "Everything happens because the plant om my desk decided it should be that way". There is no way of disproving this model, but it is useless because we cannot predict anything from it; any outcomes are equally likely based on this model.
Try substituting the word "model" for "law" in the texts. There is nothing absolute about the models as they are described; they are just the best models we have been able to describe so far. There may be better models waiting to be discovered.
Consider Newton's model of gravity. Since it's a model, how would we go about deciding if the model is wrong? Well, you might propose that the model does not hold for, say, bananas. We can then go off and conduct an experiment where we drop bananas in vacuum chambers and measure if the time to impact the ground is the one predicted by the model. If it's not, the model isn't correct.
As we know, it turns out that Newton's model doesn't give the full picture - once we get into extremely dense objects (black holes) or objects moving at high speed (i.e. near the speed of light) - so better models are introduced.
The basis of science, then, is to describe models that fit with the phenomena we can observe. The models need to be testable (I can construct an experiment to test my model), reproducible (you can do it as well), and falsifiable (we can conceive of an experiment that would prove the model wrong).
The last point is crucial. Otherwise, we could end up with a model of the universe like: "Everything happens because the plant om my desk decided it should be that way". There is no way of disproving this model, but it is useless because we cannot predict anything from it; any outcomes are equally likely based on this model.
