Mark
Practically speaking, except for a relatively small number of courses which I believe should be preserved at all costs...even if the USGA an R&A have to fund it without any say in the product, I am not terribly fussed by architectural meddling. For sure I am against most of it because I think it a waste of money when considering the target market, but its not up to me.
Theoretically speaking, before I started to make judgements based on archie intent I would want to see that written about for specific courses. An archie making a general proclamation in a mag piece is not nearly the same thing. I thought this immediately when you mentioned Colt and trees. So far as I know, tree plans were a rarity for Colt. Because he did have a specific plan on some courses doesn't mean that specificity can be applied across the board.
I don't have any answers to your questions, only my opinions...which I believe is the point people are trying to drive across on this thread. Trying to balance a desire for change (for many reasons) and an adherence to the original (or at least close to original) design tenents is a job necessarily based heavily in subjectivity. I am only trying to demonstrate that once we have some subjectivity in the picture, restoration is not the correct terminology. But, because the idea of "basing decision making in the eyes of the ODG" is easier to sell, "restoration" is the term used. I think once changes are deemed necessary (this is the step which I think often goes badly wrong) its usually a better idea to at least refer back to the original archie. That said, if a course isn't precious and if a good archie is doing the work, changes can lead to a better golf course. You roll the dice and take your chances...which is why there should be something seriously amiss before clubs start spending a ton of dough to muck about.
From my PoV, the best approach is to use old info to guide in ripping out trees, pushing out grass lines, improving drainage, enlarging greens and improving their quality. After that sort of stuff is done and settled for some time then maybe look at changes to the remaining architecture if things are not working right. More often than not no more work will be necessary.
Ciao
Ciao