Thanks Guy, you're right, it's hard for me to prove that no trees live forever. I lean that way because of the absence of counter examples. There are plenty of indicators that death is pretty normal, even in the absence of hacks or even of certified arborist disease. Most natural forests are dynamic mosaics where life-threatening disturbances are frequent. Those disturbances set the stage for the next cohort or generation of trees. I'd be happy to be proven wrong. I take that back, I don't even need proof, just a sound argument with some evidence that immortality is reasonable to consider in a biological context. Evolution and genetic accommodation to a changing environment usually requires death of what was there before. I guess I'm not very sentimental today. I'll definitely grant you that many urban and community trees die prematurely because of human activity. But just as many wild and rural trees also die due to ecological processes like stand closure and natural thinning.
Also true that submerged or ponded wood can stay somewhat stable for a long-long time. I messed around with some dendrochronology work in Finnish Lapland, in bogs being harvested for peat fuel. The harvesters were just pitching these branches, and roots, and stems in great piles as they dug the peat. That multi-millennial pine and birch "subfossil" wood decayed pretty quickly once it was out of the muck. But sure, in anoxic environments, wood can stick around for a long time. Most of my active research in the last few years has been for dendrochronological markers of compartmentalization of fire injury. So my fire colleagues say no, not all wood rots, some of it burns. OK, true enough.
As for petrified wood, that of course is not wood. I'm looking at some beautiful petrified sections of my desk, crisp tree rings, good rays, what I interpret as compartmentalization boundaries, all fine anatomical features preserved in silicates of various composition. There is no lignin, cellulose, pectin, phenols whatever you might use to chemically identify or even suggest that something is wood.
I'm quite happy for the practicing arborist to use the CODIT model if it helps understanding. CODIT is a map at a particular scale or resolution. Compartmentalization is the actual landscape. If the arborist takes the initiative to really look closely and think about what she sees, she may find that the model walls don't really apply as well as they might. Then, knowing that CODIT is but a map to the landscape and not the landscape itself might be reassuring. It might be a cause to learn more. I don't claim to know what precisely the practicing arborist needs to know. Some want to know the minimum to get paid and I respect that. Some want to know because they are intrigued by this part of nature. I can share my perspective on the biology of the system and let the marketplace of ideas sort it out.
Central to the CODIT model are those 4 Walls. Seems like most folks just pay attention to Wall 4. But it is irritating to me that folks point to woundwood ribs or a closed wound and say "That's where the barrier zone is", and usually it is not. Wall 4 is tangential to the wound and above and below the wound, formed in the same year or the following year of injury. Wall 4 resists the outer spread of infection from reaching the vascular cambium. It is really not involved with wound closure. You can't see these walls from the outside. And on the inside, the barrier zone is the thickness of a piece of paper or construction paper. And yes, yes, this needs to be outside of the classroom, with fresh dissections, razor blades, and hand-lenses!
So that is far, far from simply a semantic argument. It is a misapplication of terms. So should that matter to the practicing arborist? That's not up to me to say. Most of the time, I'm just trying to keep the trees from being planted too deeply, the string trimmer away from thin-barked little trees, or getting wire-in-hose guy lines removed!