There are different ideas on multiple trunks. Will, for example, by his own words writes: "Both are mutlistemmed and as such are scoffed at (by me). Impressive fusions of wood but not a single tree." Others non-critically look at any large trunk a a single tree whether or not they are a fused multiple trunk tree or not. I am sort of in between. I believe there is an important distinction and valid to be made between those trees that have formed a single trunk and those that have a large trunk as a result of two or more trunks fusing together. These fused trunks are, in my mind, a perfectly valid growth form that is fairly common in some species and occur commonly through natural processes. As such, they deserve to be measured and counted among the trees that represent the species, if we are to have a fair overview of the tree species population as a whole. This is the basic premise of the multitrunk classification system I proposed and have on the ENTS website. http://www.nativetreesociety.org/multi/index_multi.htm
Unlike some on the list I do not believe that multitrunk trees are simply a fusion of two or more essentially separate individuals. The idea that they can be counted as essentially individuals is not without merit. It is a useful concept that can be applied to the development of the structure of a forested setting. It can be explained largely through the basic stem exclusion process. The same competition between stems idea can even be applied to looking at competition between limbs on a single tree. Lee Frelich and I have discussed this before on the list.
I do not think these multitrunk trees should be considered the same as a single trunk tree, but I do not think they should be considered as a simple collection of individual trees that happen to be touching. They are something in between these two extremes. As you pointed out they may be sharing the same root mass. Articles like this one, mentioned when we were still on Google, talks about chemical communication and recognition between the roots of a tree with other adjacent trees. http://www.thefreelibrary.com/Root+words:+scientists+strengthen+the+case+for+subterranean+signals...-a010596772
The roots indicate that trees in a grouping may be coordinating their growth through this for of communication. They are not purely competing nor can they said to be cooperating as that implies a conscious act, but they are making accommodations for other trees in the area. This surely applies to trees which are sharing or have intermingling root systems. All out competition or to phrase it another way all out war between different trees and different trunks in a multitrunk tree might not be the most advantageous to the trees involved. Perhaps one tree would manage to overwhelm another, but at the cost of a Pyhrric victory, is that the trees best option? Evolution tends to favor things that give at least a modest advantage, so if making accommodations for other trees is better for them, that is what they do. I think this is the case for many of the species that commonly have multitrunk masses. I am not sure how to quantify this idea to see if I am correct or not, but measuring and documenting these multitrunk trees, single trunk trees, and low branching trees is surely a place to start.
PS: Scott Wade sent me this photo:
He wrote partly in jest, "Ed, evidence of multiple stemmed trees that are one tree attached!!!! I sent this to will and Dale too. I found three of these acorns this past spring. Scott"
I replied: "Scott, Wow!! You have two different trees growing from the same acorn!! Edward Frank" It is a matter of perspective.
"I love science and it pains me to think that so many are terrified of the subject or feel that choosing science means you cannot also choose compassion, or the arts, or be awe by nature. Science is not meant to cure us of mystery, but to reinvent and revigorate it." by Robert M. Sapolsky