should speakers be coupled or decoupled?

I have seen a bunch of threads that touch on this, but I wanted to ask the question directly: I don't understand the idea of putting ball bearing type decoupling devices underneath floorstanding speakers (with or without a platform). Doesn't this mean that the speaker will move around as it's woofer fires? And isn't that decidedly not good sonically?

I do understand the idea of having a very solid base for the speaker: either spikes down through carpet to the base material underneath, or a solid platform.

I'm very interested in people's thoughts on this matter.
It is not science to say that something does not exist until you have an explanation for it. Theory in physics is great but it is a broad extrapolation from reality, not reality itself. In any case there will be many effects that can affect the sound and theory by itself can not tell us which is the critical one. If we are not to depend on our hearing what are we to base our decisions on? Being convinced by a theory is no different than being convinced by some one else's glowing testimonials. Neither is definitive, science advances by discovering anomalous effects contrary to accepted theory.
I completely stumbled upon this. I my second system which is in my gym I have rubber floors over slab. When I placed my speakers on the spikes, the spikes went thru the rubber mat and touch the cement pad below. I think the 3/4 inch rubber floor really dampens the spikes and stops any subtle movements.It turns out to be a really nice base for speakers.
Stanwal's thought provoking post illustrates how observed "facts" can lead to philosophical arguments. When facts don't fit theory, or other facts, we use philosophy to reform arguments and assist in the development of new theory that better fits the facts. I put "facts" in quotes above because it is always an open question with the human variable as to whether the "facts" are, in fact, "facts" at all, other than to the person making the claim.

So, is it a "fact" that putting speakers on moveable bearings improves sound? If this fact can't be proven objectively or demonstrated to another person, then maybe it isn't a "fact" at all. Maybe it's just an opinion, and as such, only a fact to that person.

It is an observed "fact", and consistent with "theory", that the earth is round. So somebody looks out the window in Kansas and says: "My goodness, look at that, the earth is flat! That theory and those other observations about the earth being round must be incorrect. My friends here who are looking out of the window with me all agree. The earth is flat! We will have to change the theory that the earth is round to accommodate the observations that I and my friends have made about the earth actually being flat."

Anything wrong with this argument? No...Yes...???

Here's an analogy.

It is an observed "fact", and consistent with "theory", that putting speakers on spikes will improve performance. One of the theoretical bases for this is Newton's Laws of Motion. So somebody in Kansas puts their speakers on a moveable bearing and says: "My goodness, this sounds better! That theory and those other observations about fixed spikes must be incorrect. My friends here who are listening with me all agree! We will have to change the theory that the use of fixed spikes on speakers is based upon to accommodate the observations that I and my friends have made."

So, if you think moveable speaker stands improve performance, let's hear the theory upon which this is based so we can test it. If you don't have a theory, all you have is an opinion. And I prefer my opinion over yours. You see....I don't live in Kansas.
Markphd, I admire your immense patience. Of course spikes are the correct way to 'de-couple' speakers from the floor upon which they sit and I have answered many posts about placing squash balls beneath and timber platforms etc until I now just ignore the posts which continue to pop up every 2 months or so.
Your explanation of Newton's laws which explain why is again patiently put and your philosophical example of the earth being flat is a very good one.
I'm somewhat surprised at Stanwal who in the past has seemed to contribute intelligently to these discussions. Oh well, now I know. Thank you once again Markphd but, as you will quickly learn, the geese will continue to reinvent physics on every occasion in the belief that this is a 'subjective' hobby and not bound by the precepts of science.
I can claim to have some knowledge of the working of science, I did my PhD research in it's history. I reiterate that what you are describing is not science. Quoting theory will not tell you in advance how something will sound, just how it MAY sound. If our ears are not the final arbiter of sound what is? In the old days there were stories of those who never actually listened to their systems, just watched them on oscilloscopes . Maybe the stories were true. As to cones, they COUPLE components, not DECOUPLE them. The makers of the ones I use, and sell, are very specific on this point and it should be obvious with a little thought. The purpose of the cone is to transfer vibrations as rapidly as possible [ ground, as they put it] to the floor. Decoupling the components involves PREVENTING the transfer of vibrations between the component and what it is resting on. How do you think a cone does this? See Star Sound site for their discussion. Decoupling is a totally different approach; Its leading exponent is probably Barry Diament, a prominent recording engineer, he has a web site. I have debated him on this point in the past. He also can produce elegant theory as to why his approach is better. As I really believe in science I am just at the point of ordering some FIM isolators which use the principals he espouses. I intend to see for myself which sounds better. I think I am right but I would like to know for sure. That, my friends, is science, all else is hand waving.