I'm quite a bit more skeptical toward arboriculture than you are. Forests are going to be very unwieldy systems to work with, and I don't see arboreal-sapient society emerging as a simple extension of currently existing primate lifestyles.
To expand the forest, just keep killing or driving off the herbivores that destroy immature growing seedlings. If you can make even a 5% difference, then within a few centuries, North America is a forest from one end to the other.
Driving off herbivores isn't going to turn the Great Plains into a forest: you have to increase the availability of water. In ecological parlance, forests are late-succession ecosystems: they always outcompete and eventually replace grasslands when water is in high enough supply to support trees. This means that any location that can support a forest is probably already going to be a forest before our sapients get there. In order to get forests to grow elsewhere, you have to change a region from one that can't support forests, to one that can. Basically, you have to increase the availability of water.
So, irrigation is the key to expansion, which means the arboreal sapients will be restricted geographically to the forests they originally lived in until they either develop enough construction technology to make canals or aqueducts, or learn how to adapt to tree-less landscapes. So, irrigation on one hand, or artificial tree-like structures (bamboo jungle-gyms) on the other.
DValdron said:
Ah, but that's the key to intelligence. Not every tree is productive. Not every productive tree is productive at a given time - most aren't. Intelligence is required to know which tree to seek out to harvest and when.
I'm a bit dubious on this part. Primate intelligence already seems sufficient to handle the spatial and temporal complexity of fruit harvests, so I don't see this aspect of primate ecology generating much drive to increase their intelligence. I think we'd need to look elsewhere for the "sapient spark": carnivory/dietary diversification, sexual selection, survival in the face of ecological disaster, etc. Maybe all of the above.
An arboreal civilization would not want to get a monoculture at all. That would mean that all your foodstuffs would come online in a short period, say one month out of twelve, and you would be scrambling to try and preserve the vast majority of it, or starving over the eleven months.
I could see an arboreal civilization optimizing through local monocultures, say exclusively one kind of tree for a particular soil or moisture condition. But overall, I'd think the strategic thinking or the trial and error solution would be a package of tree species.
Honestly, I don't think tree fruits are going to be the key here. Trees are high-risk, high-reward ventures: they produce a lot of fruit, but they also take a lot of time, effort and foresight to manage properly, and the lack of resilience means losses will be absolutely devastating. This means tree-based societies are going to be highly failure-prone. Don't get me wrong: I have no doubt that "Paleolithic" arboreals will drastically alter the composition of forests by their actions and preferences (there is
precedent for things like this, even among unintelligent species), but I don't see it as a realistic means to a real breakthrough in social development.
Vines (e.g., cucurbits, legumes, grapes, ivies, and morning glories) and epiphytes (orchids and bromeliads) are just as frequently utilized by primates as trees are, and, because they grow faster and often have longer production windows than trees, they would be easier to figure out, experiment with and manipulate, with lower risks involved. These characteristics make their cultivation and domestication much more likely than arboriculture to emerge and develop early. So, vine and epiphyte cultivation is the more likely impetus toward real agriculture.
So, the arboreal sapients can move from patch to patch, promoting vine and epiphyte growth in their "orchards" to diversify each patch and increase the productivity window. Increasing the productivity window would allow more sedentarism, to tend and manage the fruit trees and improve cultivation techniques. So, learning vine and epiphyte cultivation first would not only be more plausible, but also improve arboriculture in the end.