Over the last few centuries, as mankind has gained interest in exploring his own history through archaeology, we have destroyed an incalculable quantity of data about that history, solely for the purpose of being able to manipulate these artifacts for analysis, on the basis that it is necessary for us to do so in order to learn anything.
While that assumption has been justified in times past, it no longer is and that has been clear for some decades now. While it is true that, for the most part, we still need physical access to historical artifacts in order to get information out of them, it’s clear that sensor technologies will eventually advance to a state where fully non-destructive analysis will become available over the next few decades. There is no excuse for continuing our current digs and physical manipulation of artifacts, knowing what we know, except impatience.
It may well be that we have to give up all or most archaeology for a generation or two, but that’s as it should be. If we have reason to believe that a particular location contains, for example, samples of a plague that we need to protect ourselves against, or we know that a location will be destroyed by construction/a river/etc. then it is fair to dig and extract a bare minimum of what needs to be gotten to on a fast schedule. But outside of that, history is a curiosity, not a need. It can wait 50 years while we advance scanners that can tell us every molecule and its high-def, 3D coordinate. Trying to go in and get stuff, without that technology, knowing that technology will come, is just willfully destroying history. And while it may be that history is just a curiosity, impatience is still not a viable excuse for destroying most of it.
We should have laws that ban destructive archaeology, except in case of emergency (as noted), and we should have test structures and materials constructed and layered under different materials, to act as a testing ground for sensor technology. Test archaeology should be the limit of most of our work for the next few decades.